CURRENT MAILING ADDRESS
- 格式:pdf
- 大小:112.70 KB
- 文档页数:18
Teaching Digital Design to Computing Science Students in a Single Academic TermJos´e Nelson Amaral,Paul Berube,Paras MehtaDepartment of Computing ScienceUniversity of Alberta,Edmonton,Canadaamaral,berube,paras@cs.ualberta.caMailing Address:Jos´e Nelson AmaralDepartment of Computing ScienceUniversity of AlbertaEdmonton,AlbertaCanada,T6G2E8Phone:(780)492-5411Fax:(780)492-1071AbstractHow should digital design be taught to Computing Science students in a single one-semester course?This paper advocates the use of state of the art design tools and pro-grammable devices and presents a series of laboratory exercises to help students learn digitallogic.Each exercise introduces new concepts and produces the complete design of a stand-alone apparatus that is fun and interesting to use.These exercises lead to the most challengingcapstone designs for a single semester course of which the authors are aware.Fast progress ismade possible by providing students with pre-designed input/output modules.Student feed-back demonstrates that the students approve this methodology.An extensive set of slides,support teaching material,and lab exercises are freely available for downloading.Keywords:Digital logic design,digital systems,teaching laboratory,Field-Programmable Gate Arrays(FPGA).1IntroductionThe teaching of digital design in Electrical and Computer Engineering(ECE)curricula is well established.However,the teaching of digital design to Computing Science(CS)students has not been discussed at length in the engineering education literature.Additional constraints ina CS program include(1)reduced number of hours dedicated to hardware-related subjects;and(2)incoming students’lack of background on switching theory,analog circuits,electronics,and limited exposure to the concepts of concurrency,feedback,and timing.Even when incoming CS students lack what would be considered essential prerequisites in an ECE program,a well planned,one-semester course on digital design can produce adequate state-of-the-art training on digital design,expose students to key digital design principles,and allow the students to design and implement non-trivial apparatuses that work.This paper presents a methodology for the teaching of digital design in a single semester course in a Computing Science program.Section2examines alternative approaches to teaching logic design.Lecture material and teaching methodology are discussed in Section3.Then Sections4,5,and6present the laboratory environment and exercises,and Section7presents results from the student evaluation of the class.2Related ApproachesThis section reviews experiences and reflections about the transition from plug-boards to pro-grammable logic.The digital logic education literature provides extensive arguments against starting design labswith complex devices such as FPGAs.Kleinfelder et al.,Areibi,and Nickels advocate that when transitioning from low density logic devices mounted in plug-boards to programmable logic, digital design classes should retain some component of non-programmable small or medium-scale integrated circuits[1,2,3].Newman et al.chose to make the transition from plug-boards to programmable logic devices(PLDs),which are simpler than FPGAs[4].Nixon argues that the complexity of FPGAs are inappropriate for afirst course on logic design[5].Most of these observations stem from experiences within an ECE program where more time(in comparison with a CS program)is dedicated to digital design.In order to reach the state of the art in one term,the authors forgo the“touch and feel”experience awarded by lower scale integration.With careful planning and support material,an aggressive schedule of increasingly complex designs can be successfully implemented in one semester.Although the often extolled debugging of wire connections is absent from this lab environment,the use of an external keyboard,audio-set,Light Emitting Diode(LED)displays and pushbuttons gives the students an appropriate level of interaction with the devices that they design.Once FPGAs are selected to be used from the start of the course,a set of design tools must be chosen.Calazans and Moraes suggest that the availability of design tools and FPGAs allows the combined teaching of computer architecture and digital design early in a CS program(third term), and they recommend a combination of industry-grade and educational tools[6].Although some authors of educational tools are their strongest advocates(such as Rodr´ıguez-Pardo et al.[7]), others argue for industry-grade tools[8].In a curriculum with a single digital design course,the use of industry-grade tools is recommended because they afford the students the best training fora career that might include logic design.Instructors interested in switching to FPGA-based digital design laboratories will alsofind relevant description of experiences in[9]-[10].The course plan described here is distinct from these experiences because it focus on transitioning from very basic design skills to challenging capstone designs in one term.This goal is accomplished through the use of supporting modules for I/O operations.3Lecturing MethodologyIn-class time is effectively used by relying on an extensive set of slides for the lectures[11]. These slides are carefully designed to allow in-class interaction with the students.1Animation techniques are often used to allow in-class quizzes.Students are asked to put printout of slides aside and to take out paper and pencil to solve these quizzes.Logic gate construction follows Yale Patt’s abstraction of light switch to represent a transistor[12].The initial chapters of the adopted textbook,by John F.Wakerly,discuss number systems and electronic technology[13].This material is made into reading assignments with scheduled10-minute in-class quizzes.Homework assignments are regularly assigned,but in order to rationalize Teaching Assistants’(TAs)time the solutions are not collected for grading.Instead,short,scheduled,in-class quizzes test the same material at the homework due-date.The use of FPGAs requires the inclusion of VHSIC Hardware Description Language(VHDL)2 learning in the one-term course.The lab exercises were designed to minimize the in-class time(a)XSA-50+XStend,in custom enclosure(b)XSA-50+XStendFigure1:Development boards as used in the laboratory,from XESS Corp dedicated to VHDL.The students are told on thefirst day of class that they are expected to learn VHDL on their own.3Once this level of expectation is set,only two50-minute lectures are required to discuss the most important principles of VHDL modeling in class.The careful use of technology and old fashioned common-sense allows the development of a challenging but exciting one-term digital design course for a CS curriculum.Students now look forward to,and praise(Section7),this third year optional class in the program.4Laboratory EnvironmentThe exercises where used in a laboratory with20workstations,each workstation with a XSA-50FPGA board and an XStend daughter-board.4Desk fastened custom plastic enclosures min-imize wear and tear,accidental damage,or unsupervised removal and allows24-hour operation for the lab.Figure1(a)shows the enclosed boards connected to a parallel port of a host computer, to a power supply,to a PS2keyboard,and to a headphone set.Figure1(b)is a closer view of the XStend and the XSA-50outside the plexiglass encasing.4.1Hardware and Software EnvironmentsThe XStend boards have an8-segment bar-graph LED used to display binary values,a7-segment digital LED used as an alphanumerical display,and a stereo codec used for audio output. Inputs are accepted through the buttons on the XStend board,through the parallel port data lines, and from a keyboard attached to the PS2port.Students are provided with a module to interface with the keyboard.Once learned,this interface is re-used in several exercises.The XSA’s8MB Dynamic Random Access Memory(DRAM)and a128KB Flash memories are used to create designs that function without constant input from the user.The Xilinx Integrated Software Environment(ISE)5.2i is used for design entry.ISE presents a hierarchical view of the design process which clearly illustrates the dependencies and sequenc-ing of implementation tasks.VHDL Simili from Sonata Electronic Design Automation(EDA)is used for testbench-based simulation.VHDL Simili is more intuitive and user friendly for be-ginner students than the industry-strong ModelSim from Mentor Graphics.Testbench writing is challenging to the novice designer,thus students are provided rudimentary testbenches for all but the capstone labs.4.1.1TestbenchesWhen using a testbench the student can detect errors and identify signals that are assigned incorrectly.Testbenching also allows for easier grading of lab exercises.Teaching assistants (TAs)are provided with testbenches that are not published to the students,and hence can do a thorough,fast and fair comparison of designs.A simplified testbench driver is supplied to allow students to perform initial testing,and to provide the basis for a more complete testbench. Testbench inputs for self-checking testbenches are supplied as a textfile,and the corresponding correct outputs are provided in anotherfile.If simulated outputs do not match supplied outputs, an error message is generated.4.1.2DemonstrationSuccessfully simulated designs may not necessarily work in hardware since timing issues and hazards may not be simulated completely.Also,complex behaviors,such as the PS2protocol, are not simulated;instead,the PS2output is simulated.Hence,if the keyboard interface is not correctly connected to the keyboard pins,this error would not be simulated.Conversely,unsuc-cessful simulations do not imply total failure.To allow partial marks for partial functionality, demonstration is the only way to assess the extent of the student effort.5Term Lab ExercisesThis section provides only brief descriptions of thefirstfive plete descriptions and VHDL modules are found in the course web-page.Initial lab exercises familiarize the students with the integrated development environment and with the design of combinatorial circuits. Simple Alarm System.Design the logic for the control circuit of an alarm from a natural lan-guage specification.The students are required to build both a schematic diagram and a VHDL design so that they can contrast the advantages of both input formats.Parity Checker.Design a circuit to read a hexadecimal digit from a keyboard and calculate its parity.The parity is displayed as a numerical digit on a7-segment LED,while the input is represented in binary on a bar-graph LED.This lab introduces the board’s input/output features,structural VHDL,and the VHDL Simili simulator.Treehouse Encryption.Implement a bit-scrambling encryption algorithm.A standard PS2 keyboard is used as plete modules in VHDL and partially completed VHDL code are provided for the students to examine.Portions of the design are provided as spec-ifications that require the students to build a hierarchical design using submodules. Scrolling Message Display.Produce a scrolling message in the two neighboring7-segment LEDs of the XStend boards.The message to be displayed is stored in an Off-Chip Memory in ASCII format.The end of the message is detected by a special ASCII code.When the end of the message is reached,the message has to be scrolled in reverse order until the begin-ning of the message is reached,whereupon it will scroll forward again.A PS2keyboardFigure2:Block diagram for the Scrolling Message Display system.is used to control the speed of the scrolling.A block diagram for this lab is shown in Fig.2.Besides the VHDL code for the input and output interfaces,the students are also givena skeleton for the design of the Address-LED Multiplexer and for the Controlpletion of the code in these skeletons will produce the basic functionality of the simple scrolling device.However,to obtain the continuing forward-reverse scrolling function,the Control Unit must be re-designed.With a functional forward scrolling system,the use of a hierarchical design for the new Control Unit should be natural.Also,the control of the scrolling speed through the keyboard interface requires that the students implement a variable counter in the Control Unit to determine how often a new character has to be read from the Off-Chip Memory.This exercise introduces counters, multiplexing,and component reuse,and practices hierarchical design.Multi-Mode Calculator.Design afinite state machine that implements a two-digit hexadecimal calculator that handles input expressions in prefix,infix,or postfix form.The PS2interface is used both to select the operation mode,and to input the operations to be performed bythe calculator.The7-segment LEDs display the result of the calculations and the current operation mode.Numerical inputs are two digits.Addition,subtraction,and multiplication operations are supported in all modes.This calculator can be implemented as a single largefinite state machine(FSM)in which numerical inputs,arithmetic operators,and commands to change mode of operation are the inputs.However,a better strategy is for the student to break this FSM into smaller ones.For this lab no block diagrams are provided,and the students are free to build their design however they desire.6Capstone DesignsThefinal lab exercise is the capstone for the class.A collection of different exercises can be rotated as the capstone design in order to keep the course interesting and introduce some variation in the undergraduate program.Currently there are two capstone designs:a music recorder and an interactive game.6.1Music RecorderThe Music Recorder is the design of a musical keyboard played through the audio codec whose output can be recorded to and played back from the RAM.The design must also be able to transpose the music up or down by as much as one octave.The amount of transposition is set from the keyboard.Students are given only a natural(i.e.,English)language specification of the design.Students are provided with the keyboard interface and a series of VHDL modules that produce an audio interface with the codec that translates Musical Instrument Digital Interface(MIDI) note signals into sounds.Students are responsible for the design of the Control Modules. As a bonus,the students may expand their design to display the transposition of notes on the LEDs.This requirement demands that shared lines between the LEDs and the RAM memory be multiplexed.Important features of this lab include:(a)tones must be held while a keyboard key is de-pressed and stopped when the key is released(therefore,both key presses and releases must be monitored);(b)RAM must be accessed for both reading and writing;meaning control signals must be properly generated;and(c)control inputs can arrive from the keyboard asynchronously.A successful design depends on the proper interaction of several fairly complex state machines.6.2Type,Type,Revolution!Type,Type,Revolution!is based on a popular arcade game of similar name,“Dance,Dance, Revolution!”This assignment involves designing an interactive game using FPGAs.Students are given only high-level design specifications for this game.The game itself is fairly simple:the system reads an instruction from memory,selected from a set of directional keys(up,down,left,right),and displays the instruction on two7-segment LEDs on the XStend board.The user must then press the key corresponding to the displayed instruction before the next instruction is displayed.Three mistakes or missed keys are allowed before the game ends.At the end of the game the number of sequences completed is displayed.When the sequence ends,the user is credited one additional allowable mistake(to a maximum total ofthree),and the sequence repeats.Students are also encouraged to develop additional functionality for bonus marks.Suggestions include increasing the game speed after each completed sequence and making the length of time for each key variable.7Student FeedbackThis section presents results from a standard student evaluation of courses and a summary of a qualitative evaluation of the labs through written comments in the lab reports.Year#of Students Registered#of Responses5The size of the cohort for some questions and years was not published by the university,but it should be similarYear StronglyDisagree Neutral AgreeStronglyPercentile Disagree AgreeIn-class time was used effectively200011210650%of8510200101320750%2002010191150%2003001102775%Overall,the quality of the course content was excellent20001137850%of8523200100719550%200211615850%2003004151975%Table2:Student’s reaction to various statements evaluating the coursethe answers are compared with all sessions evaluated in that term at the University of Alberta. The disruption caused by the adoption of VHDL and FPGAs caused a significant reduction in the number of students who consider the lab environment appropriate in2001and2002.However,on the third edition of the new course(2003),the student responses were more positive than before these changes.As glitches were eliminated from the lab experiments,the software environment and lecture material were improved.The instructor acquired experience teaching the class,thus the evaluation of the course improved significantly.The most significant changes are in the number of studentswho consider the workload appropriate and the course content excellent.The workload of this course is higher than in similar courses in other universities,and it is also higher than similar courses in the same term at the University of Alberta.However,over the three-year period in which the changes were implemented,the expectation of the students changed,and the additional effort required for the course is now considered“normal”by the students.A clear evidence of the students’approval of the changes and of their benefit from the course is that this elective class and the enrollment has not dropped in spite of the additional effort that the class requires.Review of Student CommentsStudents are requested to include in the report submitted with each lab a brief evaluation of the lab exercise.In this evaluation they discuss the lab specification and indicate any problems that they had with the lab presentation.During the annual updating of the labs,these comments are reviewed to make corrections for the next year.Here is a summary of the comments about the capstone labs(Music Recorder in2001;Type,Type,Revolution!in2002and2003): In2001,students were given less guidance in early labs;thus they were exposed to the process of conceptualizing a design from a natural language specification earlier.They found Lab6challenging but not overburdening.In2001,students frequently wrote that they were excited to have designed and implemented an apparatus“that does something”(play music).In2002,block diagrams and conceptual designs to the earlier labs were added in an attempt to help out the students.This addition proved to be a wrong decision because the studentsfelt that there was a major gap between the earlier labs and Lab7.Students felt that timing issues were not discussed in class to the extent needed in the lab.This class is offered to CS students,and they have a hard time working with difficult to reproduce behavior during testing.The problem was corrected in2003by including one extra lecture on timing and by warning students at the start of the term that they could encounter non-deterministic behavior in the lab.In order to address these problems less design information(no block diagrams)was provided in earlier labs in2003,and the discussion of designflow and timing issues in the classroom was expanded.Also,test benches were introduced to allow more comprehensive testing during simulation,and the twofirst lab exercises were redesigned to emphasize VHDL-based,as opposed to schematic,design input.Although sometimes students spend frustrating hours in the lab because of the challenging nature of the exercises,student comments are very positive and indicate an enjoyment of the exercises.Students were included in the process of improving the labs,and they often offered constructive feedback.While reading the comments one cannot fail to feel the sense of accom-plishment and triumph over a challenge that many students expressed.8Final RemarksThis paper addresses the problem of introducing Computing Science students to state-of-the-art digital design.This task often falls to ECE faculty or to CS faculty with ECE background. The experience described here demonstrates that with careful planning and well developed lecturematerial and lab exercises,CS students can be offered an engaging and rewarding digital design experience.The material presented in this paper is publicly available in the course webpage in the hope that it will be of benefit to colleagues elsewhere.AcknowledgementsStudents were supported by Summer Fellowships from the Natural Sciences and Engineering Research Council(NSERC)of Canada.Xilinx Co.donated the boards and software for the lab.References[1]S.Areibi,“Afirst course in digital design using VHDL and programmable logic,”in31stASEE/IEEE Frontiers in Education Conference,vol.31.ASEE/IEEE,October2001,pp.19–23.[2]M.W.Kleinfelder,M.D.Gray,and L.-C.G.Dudevoir,“A hierarchical approach to dig-ital design using computer-aided design and hardware description languages,”in29th ASEE/IEEE Frontiers in Education Conference.ASEE/IEEE,November1999,pp.18–22.[3]K.Nickels,“Pros and cons of replacing discrete logic with programmable logic in intro-ductory digital logic courses,”in Proceedings–2000ASEE Annual Conference.ASEE, 2000.[4]K.E.Newman,J.O.Hamblen,and T.S.Hall,“An introductory digital design course using alow-cost autonomous robot,”IEEE Transactions on Education,vol.45,no.3,pp.289–296, August2002.[5]M.S.Nixon,“On a programmable approach to introducing digital design,”IEEE Transac-tions on Education,vol.40,no.3,pp.195–206,August1997.[6]N.L.V.Calazans and F.G.Moraes,“Integrating the teaching of computer organization andarchitecture with digital hardware design early in undergraduate courses,”IEEE Transac-tions on Education,vol.44,no.2,pp.109–119,May2001.[7]L.Rodr´ıguez-Pardo,M.Moure,M.Vald´e s,and E.Mandado,“VISCP:A virtual instrumen-tation and CAD tool for electronic engineering learning,”in1998Frontiers in Education Conference.ASEE/IEEE,1998.[8]G.Puvvada and M.A.Breuer,“Teaching computer hardware design using commercial CADtools,”IEEE Transactions on Education,vol.36,no.1,pp.158–162,February1993. [9]E.I.Boemo,“Computer-based tools for electrical engineering education:Some informalnotes,”in Proceedings of Computer Aided Engineering Conference1999,1999,pp.7–13.[10]A.Leva,“A hands-on experimental laboratory for undergraduate courses in automatic con-trol,”IEEE Transactions on Education,vol.46,no.2,pp.263–272,May2003.[11]J.N.Amaral,“Cmput329webpage,”http://www.cs.ualberta.ca/amaral/courses/329,2003.[12]Y.N.Patt and S.J.Patel,Introduction to Computing Systems:from bits&gates to C&beyond.McGrawHill Press,2001.[13]J.F.Wakerly,Digital Design Principles&Practices.Prentice Hall,2002. BiographiesJos´e Nelson Amaral received the Ph.D.in Electrical and Computer Engineering from the Uni-versity of Texas at Austin,in1994,the M.E.from the Instituto Tecnol´o gico de Aeron´a utica, S˜a o Jos´e dos Campos,SP,Brazil,in1989and the B.E.from the Pontif´icia Universidade Cat´o lica do Rio Grande do Sul(PUCRS),RS,Brazil,in1987.He is a Professor in the Department of Computing Science at the University of Alberta,Canada.He was a post-doctoral researcher at the University of Delaware from1998to2000,and was Associate Professor of Electrical Engineering at PUCRS before that.His previous research includes theory and applications of Artificial Neural Networks,Combinatorial Optimization Prob-lems,Parallel Architectures for Symbolic Processing,and Multi-Threaded Architectures and Programming Models.His current research interests include Compiler Design and Op-timization,Cache-Conscious Algorithms,Applications of Programmable Logic,and High-Performance Computer Systems.Dr.Amaral is a Senior Member of the IEEE,an associate editor for the IEEE Transactions on Computers,and served in the organization of many international conferences.Paul Berube received a B.Sc.in Computing Science from the University of Alberta in2003.He was awarded a competitive Natural Sciences and Engineering Research Council(NSERC)of Canada graduate fellowship in2003and is currently pursuing a M.Sc.in Computing Science at the University of Alberta.His research interests include Compiler Design and Optimization,Computer Architecture,and Programmable Logic Devices.Paras Mehta received the B.Sc.in Computing Science from the University of Alberta in2004.He received a Undergraduate Student Research Awards(USRA)from NSERC in2003and 2004.Mr.Mehta is a recipient of the Terence Holowach Memorial Prize and of the Amdahl Academic Achievement Scholarship in Computing Science.His currently research interest is on parallel programming systems.。
英文信件模板:取消通知模板1Date:Name of Authorized Person/Cancellation DepartmentName of Company/Service ProviderMailing Address/PO Box no.City, State, Zip CodeRe: Cancellation of Membership/Subscription No: #_(specify number)_Dear Sir/Madam,I am sending this written notice to request the cancellation of my Subscription/membership with effective date _(mm/dd/yy)_. I would appreciate if you send me a written confirmation of the same within 30 days. I am enclosing a bank draft worth US$ _(specify amount)_ towards full and final payment of my outstanding membership/subscription fees. Please ensure that no further payments are applied to my account henceforth.Thank you for your prompt attention to this matter.Sincerely,-s/dSender's NameSender's Mailing addressSender's Contact number模板2Date:Name of Authorized Person/Cancellation DepartmentName of Company/Service ProviderMailing Address/PO Box no.City, State, Zip CodeRe: Membership Number: #_(specify number)_Dear Sir/Madam,This letter serves to request an immediate cancellation of my membership at your vacation club/gym/golf club, etc. All obligations have been fulfilled by myself with regard to this membership. I request you to waive off any cancellation charges applicable, considering my long association with your company. Please confirm the receipt of this request by return mail at your earliest convenience.Sincerely,-s/dSender's NameSender's Mailing addressSender's Contact number。
程序员面试的问题1.Are you the principal applicant?2.What is your marital status? When did you get married?3.Do you remember the date of your marriage?4.What is your current mailing address?5.Do you have any children?6.Do you have a passport?7.When did you graduate from the university?8.What is the major of your university degree?9.What are the major courses of your university major?10.When did you start working?11.What is your current occupation?12.When did you become an assistant engineer?13.When did you become an engineer? 14.What do you do at your current position?15.When did you change your job to the currentposition?16.Do you have an immigration agent?17.Which city in Canada do you intent to move in 18.How much money do you earn each month? 19.How much money do you have to bring over to Canada?20.Why did you choose Toronto as your immigration destination?21.How many workers in your work unit?22.How many employees are there in your group?23.How many engineers and clerks are there in your group?24.Could you explain your first job on a daily basis?25.What are the names of projects that you had finished in the past few years?26.Could you explain your academic research at the institute?27. Why did you want to immigrant to Canada instead of other countries, such as USA, or Australia?28.How are you going to find a job in Canada?29.IF you could not find a job in the first years, what are you going to do ?30.Have you done anything for your preparation of immigration to Canada?31.Do you know anything about the Canadian labor market?32.Do you have any friend in Canada?33.How much money do you have that is transferable to Canada?34.Do you know anything about Toronto?35.Please explain in details of how you do research on N.C system?36.What kind of computer language do you know?37.What kind of automatic control system do you research?38.What is your wife’s occupation? 39Do you have any relatives in Canada?39.What is the name of your friend in Canada (if you have)?40.Are you going to study in Canada?41.Could you write down what are you going to do tomorrow in English?42.What’s a computer?43.Can you explain the concept of a programming language to me?44.What’s a software package?45.Can you give me a definition of a network?46.Can you explain the concept of client/server architeture?47.Please illustrate the formal steps of developing a user needed software system.48.What’s the Internet?49.What’s C programming language?50.What’s objet-oriented method?51.What can you do with a computer?52.How much programming language do you know?53.What’s an operating system?54.Can you give me the definition of Windows?55.Can you give me a definition of DOS? 57.How much operating system do you know?56.What’s a protocol?57.Can you explain TCP/IP?58.Can you explain FTP?59.What’s an email?60.What’s word processing?61.Would you please to write a simple program in c/pascal/basic?62.What’s a database?63.What’s a process? 66.What’s a thread?64.What’s the main differences between C and BASIC?65.What’s a UNIX system?66.How many years have you used computer? What have you done?67.What’s a user interface? Can you give me an example?68.What’s a web site?69.What’s a homepage? 73.What’s www?69.How to write a homepage?70.What’s HTTP?71.What’s a database system?72.What’s java language?[程序员的问题]相关文章:。
英文地址书写格式报名参加各类考试,都有要填写:Mailing Address 1:Mailing Address 2:Mailing Address 3:都怎么填呀?英文地址从小到大写与中文的向反英文地址翻译参考:常见地址参考:***室/房Room ******单元Unit ******村*** Vallage***号楼/栋*** Building***号No.******公司*** Com.或*** Crop或***LTD.CO***号宿舍*** Dormitory***厂*** Factory***楼/层***/F***酒楼/酒店*** Hotel***住宅区/小区*** Residential Quater***县*** County甲/乙/丙/丁A/B/C/D***镇*** Town***巷/弄Lane ******市*** City***路*** Road***省*** Prov.***花园*** Garden***院*** Yard***街*** Street***大学*** College***信箱Mailbox*** ***区*** District**表示序数词,比如1st、2nd、3rd、4th……如果不会,就用No.***代替,或者直接填数字!另外有一些***里之类难翻译的东西,就直接写拼音*** Li。
而***东(南、西、北)路,直接用拼音也行,写*** East(South、West、North)Road也行。
还有,如果地方不够可以将7栋3012室写成:7-3012。
201室Room 20112号No.122单元Unit 2长安街Chang An street3号楼Building No.3长安公司Chang An Company南京路Nanjing road宝山区BaoShan District赵家酒店ZhaoJia hotel钱家花园Qianjia garden孙家县Sunjia county李家镇Lijia town广州市Guangzhou city广东省Guangdong province中国China地址翻译实例:宝山区南京路12号3号楼201室room 201,building No.3,No.12,nan jing road,BaoShan District如果地方不够可以将3号楼201室写成:3-201宝山区示范新村37号403室Room 403,No.37,SiFan Residential Quarter,BaoShan District中华人民共和国民政部政策研究中心北京河沿大街147号No.147# HeiYian Street the policy center of civil administration department the People’Republi c of China虹口区西康南路125弄34号201室Room 201,No.34,Lane 125,XiKang Road(South),HongKou District北京市崇文区天坛南里西区20楼3单元101Room 3-101 building No.20,TianTan-NanXiLi Residential ChongWen District BeiJing City江苏省扬州市宝应县泰山东村102栋204室Room 204 building No.102, East TaiShan Residential BaoYin County JiangSu Province473004河南省南阳市中州路42号王坤Wang KunRoom 42,Zhongzhou Road,Nanyang City,Henan Prov.China 473004中国四川省江油市川西北矿区采气一队1 Team CaiQi ChuanXiBei Mining Area JiangYou City SiChuan Province China中国河北省邢台市群众艺术馆The Masses Art Centre XinTai City HeBei Prov.China江苏省吴江市平望镇联北村七组7 Group LiBei Village PingWang Town WeJiang City JiangSu Province434000湖北省荆州市红苑大酒店王坤Wang KunHongyuan Hotel,Jingzhou city,Hubei Prov. China 434000473000河南南阳市八一路272号特钢公司王坤Wang KunSpecial Steel Corp,No.272,Bayi Road,Nanyang City,Henan Prov. China 473000528400广东中山市东区亨达花园7栋702 王坤Wang KunRoom 702, 7th Building,Hengda Garden, East District,Zhongshan, China 528400361012福建省厦门市莲花五村龙昌里34号601室王坤Wang KunRoom 601, No.34 Long Chang Li,Xiamen, Fujian, China 361012361004厦门公交总公司承诺办王坤Mr. Wang KunCheng Nuo Ban, Gong Jiao Zong Gong Si Xiamen, Fujian, China 361004266042山东省青岛市开平路53号国棉四厂二宿舍1号楼2单元204户甲王坤Mr. Zhou WangcaiNO. 204, A, Building NO. 1,The 2nd Dormitory of the NO. 4 State-owned Textile Factory, 53 Kaiping Road, Qingdao,Shandong, China 266042--------------------------------------------------------------------------------英文地址的写法[英文地址]中国MPA在线的英文地址:Rome1232,Youth building,No.34 West Beisanhuan Road, Beijing,China一、寄达城市名的批译:我国城市有用英文等书写的,也有用汉语拼音书写的。
英文求职信地址格式[Your Name][Your Address][City, State, Zip Code][Email Address][Phone Number][Date][Hiring Manager’s Name][Company Name][Company Address][City, State, Zip Code]Dear [Hiring Manager’s Name],I am writing to express my interest in the [position title] position at [company name], as advertised on [where you found the job listing]. With [number] years of experience in [relevant field or industry], I believe that I have the skills and qualifications necessary to excel in this role.I graduated with a [degree] from [University Name], where I honed my [relevant skills such as communication, problem-solving, teamwork, etc.]. During my time at [University Name], I also completed internships at [Company Name], where I gained valuable hands-on experience in [specific tasks or projects relevant to the position].In my current role at [Current Company Name], I havesuccessfully [mention specific accomplishments or projects that demonstrate your skills and qualifications]. I have a proven track record of [mention any relevant achievements or awards]. I am confident that my experience and abilities make me a strong candidate for the [position title] position at [company name].I am particularly drawn to [company name] because of [specific reasons why you are interested in the company, such as its reputation, values, products, or services]. I am impressed by [specific examples of the company’s work or accomplishments]. I am excited about the prospect of contributing to such a dynamic and innovative team.I am passionate about [specific interests or skills that are relevant to the position], and I am eager to bring my expertise to [company name]. I am confident that my background and experience make me a strong fit for the [position title] role.Thank you for considering my application. I look forward to the opportunity to discuss how my skills and experiences align with the needs of [company name]. Please find my resume attached for your review. I am available for an interview at your earliest convenience.Thank you for your time and consideration.Sincerely,[Your Name]。
通讯录用英语怎么说通讯录是一般在日常生活中用笔记录,然而现在很多都被手机电脑等取代了。
你知道通讯录用英语怎么说吗?下面随店铺一起来学习一下关于通讯录的英语知识吧。
通讯录的英语说法address listaddress book通讯录相关英语表达秘密通讯录 Secret Contacts校通讯录 School address book共用通讯录 global address list通讯录的英语例句1. That list of addresses is outdated, many have changed.那个通讯录已经没用了, 许多地址已经改了.2. He walked by my table and, casually as you please, picked up my address book.他走到我的桌旁,非常随意地拿起了我的通讯录。
3. I forgot the address so I must look in my address book.我忘了地址,得查通讯录.4. Remove my name from your mailing list.把我的名字从你的通讯录上划掉.5. Communicator is unable to import this file into the address book.通信器无法把此文件导入到通讯录中.6. Our " MOYU address book " has registered over 100 members.五、本帮的《魔域通讯录》,现已有一百多名成员已登记.7. All Address Book database workarounds are implemented within this component.通讯录的所有数据库的替代方法是实施此元件.8. Imports an address book from a tab - delimited or comma - separated ( CSV ) text file.从以TAB键分隔或逗号分隔 ( csv ) 的文本文件中导入通讯录.9. We can remove his name from the mailing list.我们可以从通讯录上将他的名字除掉.10. The man who stole my wallet took my address book as well.偷走我钱包的那个人把我的通讯录也拿走了.11. Then click OK to complete the group and close the address book.然后按ok完成群组并且关闭通讯录.12. Imports mail and address books from Outlook '97 or Outlook '98.从outlook'97或outlook'98导入邮件和通讯录.13. Error importing address book % S, all addresses may not have been imported.导入通讯录%S出错, 所有地址不能导入.14. I have a phone directory but I'm afraid it's out - of - date.我是有个通讯录,但恐怕年代久远没什么用了.15. ACCESS database using ODBC to connect the address book program.使用ODBC连接ACCESS数据库的通讯录程序.通讯录英文阅读:iPhone被盗小偷寄回手写版通讯录A best man from central China whose iPhone was stolen after a stag night was “stunned” when the thief posted him a handwritten list of more than 1,000 contacts stored on the device.在中国中部的一个城市,一个小伙子在参加单身派对时iPhone被偷了,但使他“震惊”的是,小偷居然把他电话里存的1000多个联系人的号码全部手抄了一遍,然后给他寄了回去。
address的用法和例句Address,英语单词,主要用作名词、动词。
作名词时意为“地址,住址;网址;电子邮箱地址;演讲,演说”。
作动词时意为“演说,演讲;处理,设法解决;在(信封、包裹等)上写姓名和地址,致函”等。
单词释义英[əˈdres] 美[əˈdres]n. 地址,住址;网址;电子邮箱地址;演讲,演说;(电子表格中某个单元格的)位置;<古>(对他人的)谈吐;<古>谦恭(或脉脉含情)的方式;<旧>机敏,迅速;位置存储编码v. 演说,演讲;处理,设法解决;在(信封、包裹等)上写姓名和地址,致函;<正式> 向……讲话;称呼;<正式>(向某人)提出(陈述、投诉等);(高尔夫)就位击(球)短语搭配email address 电子邮箱信箱ip address 网络地址,IP地址address book 通讯簿;住址名册home address 家庭住址;标识地址;内部地址mailing address 邮寄地址network address 网络地址address space 地址空间address as 称呼e-mail address 电子邮件地址public address 公用地址;扩音装置address system 地址系统;称谓系统inaugural address 就职演说address translation 地址转换;位址翻译public address system 扩声系统;有线广播系统;公用地址系统correspondence address 通讯地址;通信地址specific address 专用地址full address 全地址physical address 物理地址;实际地址permanent address 永久地址;永久住址;原籍mail address 邮件地址;信件地址双语例句1、They had no permanent address. 他们没有固定的地址。
Application for T emporary Resident Visa Made Outside ofCanada (IMM 5257)Who must fill out this application form? This form must be completed by each person applying for a visa to visit Canada.Completing the form You must answer all the questions on this application form unless indicated otherwise.Download and fill out the application form on a computer.You also have the option of saving your form and completing it later. Note: Completing the form electronically is easier and reduces the risk of errors that can slow down the application process.In order to help you fill out the application form, read and follow the steps below.Family Information (IMM 5645)Who needs to complete this application form? This form must be completed by each person, 18 years of age or older, applying for a Temporary Resident Visa from outside Canada.Use of a Representative (IMM 5476) Who may usethis form?Complete this form only if you are appointing a representative.Who is a representa-tive? A representative is someone who has your permission to conduct business on your behalf with Citizenship and Immigration Canada.When you appoint a representative:∙you authorize CIC to share information from your case file to this person,∙your application will not be given special attention nor can you expect faster processing or a favourable outcome,∙they are authorized to represent you only on matters related to the application you submit with this form, and∙you can appoint only onerepresentative for each application you submit.Note: You are not obliged to hire a representative. CIC treats everyone equally, whether they use the service of a representative or not.You must notify us if your representative‟s contact informationchanges or if you cancel the appointment of a representative.Types of representa- tives There are two types of representatives.Unpaid representatives can be:∙friends and family members who do not charge a fee for their advice and services,∙organizations that do not charge a fee for providing immigration advice or assistance (such as a non-governmental or religiousorganization),∙consultants, lawyers and Quebec notaries who do not, and will not, charge a fee to represent you.Paid representatives:If you want us to conduct business with a representative who is, or will be charging a fee to represent you, they must be authorized. Authorized representatives are:∙immigration consultants who are members in good standing of the Canadian Society of Immigration Consultants (CSIC), ∙lawyers who are members in good standing of a Canadian provincial or territorial law society and students-at-law under theirsupervision,∙notaries who are members in good standing of the Chambre des notaires du Québec and students-at-law under their supervision.If you appoint a paid representative who is not a member of one of these designated bodies, your application will be returned. For more information on using a representative, visit our website at:http://www.cic.gc.ca/english/information/representative/index.aspRelease of information to other individuals To authorize CIC to release information from your case file to someone other than a representative, you will need to complete the form Authority to Release Personal Information to a Designated Individual (IMM 5475) which is available at Canadian embassies, high commissions and consulates, or on our website at:www.cic.gc.ca/english/information/applications/release-info.asp.The person you designate will only be able to obtain information on your case file, such as the status of your application and not act on your behalf with CIC.。
全国英语等级考试(公共英语PETS)全国英语等级考试第二级PETS-2备考过程中,听力是不可忽视的一部分,今天开始每天进行一点一滴的练习与积累,相信假以时日一定会提高自己的英语听力能力,并顺利通过全国英语等级考试第二级PETS-2!该练习来自于全国英语等级考试第二级PETS-2的标准教程中,其语言难度以及词汇语法考查知识点在一定程度上都切合全国英语等级考试第二级PETS-2的大纲要求,因此,大家在此基础上不断练习积累即可。
====================================================================You are going tohear a telephone conversation between an American college advisor and a Chinesestudent. The advisor is looking at a form and filling it in as he speaks to thestudent.你将要听到的对话发生在美国一所大学的顾问和一位中国学生之间。
顾问一边与学生谈话,一边在看并填着表格。
在听的过程中请注意以下4个问题:Questions:1.For what does LiXiaoqiang come to meet the advisor?李晓强为什么来见这个顾问?2.Where does Li now live?李现在住在哪儿?3.Who advised Li to attendthe college?谁建议李来这所大学?4.What is your impression of the student’slistening ability?你对这位学生的听力的印象是?听力原文及译文:Advisor: What’s your last name? 顾问:你姓什么?Li Xiaoqiang: Ah ... you mean family name? 李晓强:你指的是家姓吗?Advisor: Yes. 顾问:是的。
PREQUALIFICATION FORMFOR SUPPLIERS / CONTRACTORS供应商调查表A)PARTICULARS OF THE FIRM 企业概况: of the Firm公司全称:2.Registered Address注册地址:3.Mailing Address邮寄地址(If different from 2)4.Telephone No.电话:5.Fax No.传真::6.Web-site Address公司网址:7.E-mail Address 电子邮件:8.Type of Company公司性质Public/Private/ Partnership/ Sole Proprietor/Joint Venture/Others (Please specify)国营/私营/股份制/独资/合资/其他(请列出)9.Country of Incorporation国家:10.Registration No.营业执照号::11.Date of Registration/Incorporation注册日期NB : To attach a copy of the latest Business Instant Information from Registry ofCompanies请附上最近的公司登记情况表12.Registration with Professional, Trade or Quality Control Organisation, board and/orAssociation etc.在有关专业团体,贸易机构和质量控制机构的登记证书:(Please attach a copy of the certificate of registration请附上营业执照附本)13.Other Affiliations, Subsidiaries or Joint Venture (with indication of % owned)其他从属关系或合资关系(请用%指出)B)PARTICULARS OF HEAD QUARTERS (IF ANY) 总公司概况(如有): of the Firm公司全称:2.Registered Address注册地址3.Telephone No.电话:4.Fax No.传真::5.Web-site Address公司网址:6.Type of Company经济性质7.Country of Incorporation国家:C)FINANCIAL CAPACITY 财务概况: of Bank and Address银行全称和地址:2.Credits Facilities / Overdrafts信用设备/透支款项3.Financial Standing of the Last 3 Financial Years:过去三年里财务状况Year 20__ Year 20__ Year 20__ Authorized Capital 注册资金Paid up Capital 已缴资金Net Worth 净值Turnover 交易额Net Profit / <Loss> after Tax 除税后净利/损Current Assets 流动资金Current Liabilities 流动债务NB : To attach complete annual reports with corresponding Statements of Profit and Loss Account certified by accountants and a certified copy of the audited Balance Sheet for the last 3 years.请会计员附上过去3年完整的相应年度亏赢报告和审计对比报告。
address用法Address是一个英语单词,它的意思是地址。
在日常生活中,我们经常需要用到address这个单词来描述一个人或者一个地方的位置。
在英语中,address作为名词和动词都有着广泛的应用。
首先,address作为名词时,它可以表示一个人或者机构的地址。
例如,“What is your address?”(你的地址是什么?)“My address is 123 Main Street.”(我的地址是123 Main Street.)除此之外,address还可以表示信函、邮件等物品的收件人地址。
例如,“Please write the recipient's address on the envelope.”(请在信封上写上收件人的地址。
)其次,address作为动词时,它可以表示解决问题、处理事情、发表讲话等意思。
例如,“The company needs to address the issue of employee satisfaction.”(公司需要解决员工满意度问题。
)“The president w ill address the nation tonight.”(总统今晚将发表全国讲话。
)除了以上常见用法外,address还有一些特殊用法:1. Email Address:电子邮件地址2. IP Address:IP 地址3. Mailing Address:邮寄地址4. Home Address:家庭住址5. Business Address:商业地址6. Shipping Address:送货地址7. Billing Address:账单寄送地址8. Postal Address:邮政地址9. Web Address/URL: 网址总之,address在英语中是一个非常重要的单词,它包含了很多不同的用法。
熟练掌握这些用法,可以帮助我们更好地理解和运用英语。
修改港澳通行证邮寄地址流程1.申请人可以将所需材料寄往指定地址。
Applicants can send the required documents to the designated address.2.寄出前请确认所有资料齐全。
Please make sure all documents are complete before sending.3.填好申请表格并随同其他要求文件一同寄去。
Fill out the application form and send it along with the required documents.4.邮寄地址将在办理申请时提供。
The mailing address will be provided when applying.5.检查邮寄地址是否正确无误。
Check if the mailing address is correct.6.在信封上清楚标注收件人姓名及地址。
Clearly label the recipient's name and address on the envelope.7.使用安全的包装材料以防文件损坏。
Use secure packaging materials to prevent damage to the documents.8.快递公司可提供跟踪服务以确保文件安全送达。
The courier company can provide tracking services to ensure the safe delivery of the documents.9.将所有文件一并寄出以避免遗漏。
Send all documents together to avoid any omissions.10.在邮寄前复印所有文件以备份。
Make copies of all documents before mailing for backup.11.邮寄费用将自行承担。
Fuzzy logic based improvements in efficiency optimizationof induction motor drivesJuan M. Moreno-Eguilaz✝, M. Cipolla-Ficarra✝, P.J. da Costa Branco✝✝, Juan Peracaula✝✝ Power Electronics Division Department of Electronic Engineering Polytechnic University of CataloniaDEE-ETSEIB-UPCBarcelona, SPAIN✝✝ Departamento de EngenhariaElectrotécnica e de Computadores Secçao de M. Eléctricas e E. de Potencia Instituto Superior Técnico, (IST)Lisbon, PORTUGALMailing address:Juan M. Moreno EguilazUniversidad Politécnica de CatalunyaDEE-ETSEIB-UPCDiagonal 647 (ETSEIB), Planta 9E-08028 Barcelona, SPAIN.Tel.: +34-3-4016649Fax: +34-3-4017785E-MAIL : eguilaz@eel.upc.esAbstract:Induction motor are, without any doubt, the most used in industry. Motor drive efficiency optimization is important for two reasons: economic saving and reduction of environmental pollution.In this paper, advantages of using fuzzy logic in steady-state efficiency optimization for induction motor drives are described. Experimental results of a fuzzy logic based optimum flux search controller are presented..For transient states, a new original idea is introduced: a fuzzy logic based controller, actuating as a supervisor, is proposed to work with reduced flux levels during transients to optimize efficiency also in dynamic mode. Two different rule tables are designed, for torque transitions and for reference speed changes. With this controller, efficiency can be improved in transients, and also search controller convergence speed is increased. Experimental results with a 1.5 kW induction motor drive demonstrate the validity of the proposed methods. Keywords: Induction Motor, Efficiency Optimization, Fuzzy Logic based Supervisor.1.- INTRODUCTIONIt is estimated that more than 50% of the world electric energy generated is consumed by electric machines [1]. Improving efficiency in electric drives is important, mainly, for two reasons: economic saving and reduction of environmental pollution [2], [3].Induction motors have a high efficiency at rated speed and torque [4]. However, at light loads, the iron losses increase dramatically, reducing considerably the efficiency [5].The main induction motor losses are usually split into 5 components: stator copper losses, rotor copper losses, iron losses, mechanical and stray losses. To improve the motor efficiency, the flux must be reduced, obtaining a balance between copper and iron losses [6].Basically, there exist two different approaches to improve the efficiency [7]:A. Loss Model based approachIf a motor loss model is available [7], [8], [9], the loss minimization optimum flux is computed analytically. The main advantage is the simplicity of this method, not requiring extra hardware. However, it is mandatory an accurate knowledge of motor parameters, which change considerably with temperature, saturation, skin effect, etc.B. Power Measure based approachAn optimum flux search algorithm is used, and the drive power consumption is measured [2], [7], [10], [11], [12]. Obviously, this approach does not require the knowledge of motor parameters. However, it is only efficient at steady-state condition and with high powers, the method has convergence problems.Although there exist many search algorithms [13], their main limitations are speed of convergence and undesirable oscillations around the minimum loss point.In this work, advantages of using a fuzzy logic based search controller to optimize efficiency at steady state are described [2]. For transients, a fuzzy logic based controller, actuating as a supervisor, is proposed to optimize efficiency with reduced flux levels in dynamic mode. Experimental results with an indirect field oriented control are obtained using a standard 1.5 kW induction motor at our laboratory.2.- PHILOSOPHY OF SEARCH CONTROLLERSAt steady state, the input power of an induction motor drive is a flux dependent convex function. The philosophy of search controllers is to decrease the motor flux as a function of the power consumption until the minimum loss point is found (point A in Fig. 1), obtaining a balance between copper and iron losses.Figure 1 shows the experimental results obtained using this approach with an indirect field oriented control for a 1.5 kW induction motor drive. At steady state, (speed and torque constant), the flux current i sd* is decreased gradually, while the torque current is increased, until the optimum flux is found.Number of Search StepsC o n s u m e d P o w e r , (W )0102030405060708090100024681012140.511.522.533.544.55i s d *, i s q *, (A )isd*isq*Consumed Actual Powerminimum loss pointAFlux current Torque currentFigure 1. Philosophy of search controllers Experimental results at 25% rate speed with no load.3.- EXPERIMENTAL EQUIPMENTThe experimental equipment (Fig. 2) used in the implementation consists of a power converter and a 1.5 kW induction motor. The rectifier is based on power diodes and the three-phase inverter is a MOSFET current-controlled voltage converter. The induction machine is coupled to a DC generator with an electronic load,working as brake.The power measurement is done at the drive input, sampling ac voltage and current with a transformer and a Hall-effect current sensor, respectively. A software routine computes the consumed power with a digital average.An 80486 microprocessor, running at 66 MHz, is the control computer.Fig. 2. Indirect vector-controlled induction motor drive with a search controller and a fuzzy logic basedsupervisor to optimize efficiency at steady state and dynamic mode.4.-FUZZY LOGIC APPLIED TO SEARCH CONTROLLERSThere exist many search algorithms to optimize efficiency for induction motor drives using the power measure based approach [13]. Their main limitations are speed of convergence, perturbations in the electromagnetic torque and oscillations around the optimum flux.Fuzzy logic can help avoid these problems, improving the performance of the search controller. This idea of using fuzzy logic for search controllers was recently reported in [2].The FL controller antecedents are the power change ()∆P n and the last flux current change ()∆i sd n *−1and theconsequent is the new flux change ()∆i sdn*.(see [2] for more details).Table 1: Fuzzy rules to find the optimum flux (output ()∆i sdn*).()∆∆P in sd n \*−1N P PB PM NM PB: Positive BigPM PS NS PM: Positive Medium PS PS NS PS: Positive Small ZE ZE ZE ZE: ZeroNS NS PS NB: Negative BigNM NM PM NM: Negative Medium NBNB PBNS: Negative SmallWith this mentioned controller, some improvements may be obtained (see Figs. 3.a, 3.b). First, the speed of convergence increases due to the adaptive size of the flux steps imposed to the machine. Moreover, when we are approaching to the optimum flux, the step size is reduced dramatically, avoiding unnecessary movements around the minimum loss point.Finally, the fuzzy logic based search is infinitum, i.e., the algorithm is always working, avoiding deviations of the optimum flux due to motor parameter changes by temperature effect, and so on.It can be demonstrated the fuzzy logic based search method consists of a gradient based search [13] with adaptive flux steps and infinitum search with non-null flux step around the minimum loss point.The only disadvantage of this approach is that the search controller requires adjustable gains (speed and torque dependent), which depend on the motor power.time, (s)R o t o r S p e e d , (r p m )050100150200250300350400024681012141612345678i s q *, (A )Speedisq*Fig. 3.a. Rotor speed and reference torque current i sq *evolution(Experimental results using the FL search controller).time, (s)I n p u t P o w e r , (W )050100150200250024681012141612345678910i s d *, (A )Consumed Powerisd*Fig. 3.b. Consumed input power and reference flux current i sd*(Experimental results using the FL search controller).5.-FUZZY LOGIC APPLIED TO EFFICIENCY OPTIMIZATION IN TRANSIENTSSo far, we have been only considering optimization at steady-state condition. The above fuzzy logic based search controller is only effective with constant speed and torque. When a torque perturbation or command speed change is produced, the search process stops. All the works in recent literature [2], [10], [11], [12] propose to establish the rated flux during a dynamic transition. This permits to have the maximum torque capability in order to get the optimum transient response.However, the transient torque and speed response may not be so important for some practical applications (it is the case of an elevator or a crane). For example, if the process has many small load torque perturbations. In these cases, efficiency optimization can take priority. In this paper, we propose to work with reduced flux levels during dynamic transitions, optimizing power in the dynamic mode and accelerating the search process at steady state. For this, a fuzzy logic controller, actuating as a supervisor, is designed. During transients, the controller increases the flux, depending on the speed error and its derivative. If the transient is very large or the speed error is very large, the flux is established at rated level, in order to let the drive track the reference command.The proposed controller is very simple, containing two tables of heuristic rules, distinguishing a torque transition and a reference speed change. The controller only provides positive flux increments, actuating as a supervisor, since it supervises the right command tracking, incrementing the flux if necessary.The antecedents are the speed error eω and its derivative eω and the consequent is the flux current increment * (Fig. 4). Table 2 and 3 shows the rules for a positive torque transition and for a positive reference speed ∆isdchange, respectively.Fig. 4. Diagram of the proposed fuzzy logic supervisor to increment the flux current.Table 2. Rules for a positive torque transition.\ ZE PS PM PBe eωωZE ZE ZE PS PMPS PS PS PM PBPM PM PM PS PBPB PB PB PM PBTable 3. Rules for a positive reference speed change.\ NB NM NS ZEe eωωZE ZE ZE ZE ZEPS ZE ZE ZE PBPM PS PS PS PBPB PS PM PM PBThe product-sum was used as inference method and the center of gravity was implemented as defuzzification method to get a fast algorithm in real time.Figures 5, 6, 7, 8 show the advantages of using the fuzzy logic supervisor during a torque transition. As can be seen, efficiency during the transient is improved considerably, although the transient speed response is not so good and fast (Fig. 6).time, (s)R o t o r S p e e d , (r p m )250260270280290300310320024681012141618Torque perturbationFig. 5. Speed evolution with a torque perturbationusing rated flux in the transition.time, (s)R o t o r S p e e d , (r p m )250260270280290300310320024681012141618Torque perturbationFig. 6. Speed evolution with a torque perturbationusing reduced flux in the transition.time, (s)I n p u t P o w e r , (W )020*********12014016024681012141618Transition with rated fluxTransition with reduced fluxTorque perturbationFig. 7. Comparison of consumed powerwith rated and reduced flux during the torque transition.time, (s)i s d *, (A )0.511.522.533.5024681012141618Transition with rated fluxTransition with reduced fluxTorque perturbationrated isd*Fig. 8. Reference flux currents for the torque transition (Figs. 5, 6, 7).During the transient due to a change in speed command (Figs. 9, 10), energy can also be optimized (Figs. 11, 12)by using the fuzzy logic supervisor, demonstrating the validity of the proposed method for transients.time, (s)R o t o r S p e e d , (r p m )250300350400450500246810121416Change in speed commandFig. 9. Speed evolution with transition at rated flux.time, (s)R o t o r S p e e d , (r p m )250300350400450500246810121416Change in speed commandFig. 10. Speed evolution with transition at reducedflux.time, (s)I n p u t P o w e r , (W )0204060801001201401600246810121416Speed TransitionTransition with rated flux Transition with reduced fluxFig. 11. Comparison of consumed power for the speedtransition with rated and reduced flux.time, (s)i s d *, (A )00.511.522.533.50246810121416Transition with rated flux Transition with reduced fluxSpeed Transitionrated isd*Fig. 12. Reference flux current i sd *for the speed transition (Figs. 9, 10, 11).6.-CONCLUSIONSIn this work, advantages of using fuzzy logic in efficiency optimization for induction motor drives have been described.A fuzzy logic based search controller improves the speed of convergence and reduces oscillations around the optimum flux using a simple rule table. However, the controller requires adjustable speed and torque gains, which depend on the motor power.A new fuzzy logic based controller, actuating as a supervisor has been proposed to optimize efficiency during transients. Two different rule tables have been designed for torque transitions and for reference speed changes.Experimental results of the proposed controllers with a 1.5 kW induction motor drive demonstrate the validity of the described methods.Present research is focused on how to design an autonomous efficiency controller, using the energy consumed in a mechanical cycle to adjust the controller gains, in order to get the optimum response from the point of view of efficiency. And important open question is how to quantify the benefits of the proposed solution for all operation points of the machine. And obviously, the compromise between efficiency optimization and a good dynamic response always depends on the kind of application.REFERENCES[1]W. Leonhard, “Controlled AC drives, a successful transfer from ideas to industrial practice”, CETTI95, Brazil, September 11-12, pp. 1-12, 1995.[2]G.C.D. Sousa, B.K. Bose, J.G. Cleland, “Fuzzy logic based on-line efficiency optimization control ofan indirect vector-controlled induction motor drive”, IEEE Transactions on Industrial Electronics, vol. 42, no. 2, pp. 192-198, April 1995.[3]H.R. Andersen, C.B. Rasmussen, E. Ritchie, J.K. Pedersen, “Efficiency comparison of electricalmotors for energy optimized variable speed low power and low cost household equipment”, 6th European Conference on Power Electronics and Applications, EPE’95, vol. 3, pp. 3381-3386, Seville, Spain, 1995.[4]J.M. Bodson, “Digital control improves variable speed drives”, EPE Journal, vol. 2, no. 4, pp. 243-248, December 1992.[5] A. Kusko, D. Galler, “Control means for minimization of losses in ac and dc motor drives”, IEEETransactions on Industry Applications, vol. 19, no. 4, pp. 561-570, July/August 1983.[6]P. Thogersen, M. Tonnes, U. Jaeger, S.E. Nielsen, “New high performance vector controlled ac drivewith automatic energy optimizer”, 6th European Conference on Power Electronics and Applications, EPE’95, vol. 3, pp. 3381-3386, Seville, Spain, 1995.[7]J. Abrahamsen, J.K. Pedersen, F. Blaabjerg, “State-of-the-art of optimal efficiency control of low costinduction motor drives”, PEMC’96, vol. 2, pp. 163-170, Budapest, Hungry, 2-4 de September 1996.[8]G.O.Garcia, J.C. Mendes Luís, R.M. Stephan, E.H. Watanabe, “An efficient controller for anadjustable speed induction motor drive”, IEEE Transactions on Industrial Electronics, vol. 41, no. 5, pp. 533-539, October 1994.[9]I. Kioskeridis, N. Margaris, “Loss minimization in induction motor adjustable-speed drives”, IEEETransactions on Industrial Electronics, vol. 43, no. 1, pp. 226-231, February 1996.[10]J.C. Moreira, T.A. Lipo, V. Blasko, “Simple efficiency maximizer for an adjustable frequencyinduction motor drive”, IEEE Transactions on Industry Applications, vol. 27, no. 5, pp. 940-945, Sep/October 1991.[11]I. Kioskeridis, N. Margaris, “Loss minimization in scalar-controlled induction motor drives withsearch controllers”, IEEE Transactions on Power Electronics, vol. 11, no. 2, pp. 213-220, March 1996.[12]G. Kim, I. Ha, M. Ko, “Control of induction motors for both high dynamic performance and highpower efficiency”, IEEE Transactions on Industrial Electronics, vol. 39, no. 4, pp. 323-333, August 1992.[13]J.M. Moreno-Eguilaz, J. Peracaula, “Induction motor drives energy optimization in steady andtransient states: a new approach”, sent to 7th European Conference on Power Electronics and Applications, EPE’97, Trondheim, Norway, 8-10 de September 1996.[14]M. Cipolla, J. M. Moreno-Eguilaz, J Peracaula “Fuzzy control of an induction motor withcompensation of system dead-time”, IEEE Power Electronics Specialist Conference,PESC'96, Baveno, Italy, 24-27 June 1996.。
email address是什么意思含义我们给客户发送邮件,首先要有的就是email address。
下面是店铺给大家整理的email address是什么意思,供大家参阅!email address是什么意思电子邮件地址email address的英语例句1. You can contact me at my email address , at 87645218.您可以发email给我,也可以拔找电话87645218与我联系.2. Thee - mail address and email address ( Reenter to confirm ) do not match.邮件地址和认证邮件地址不一致.3. Please send us your inquiry to the email address below, thanks.请用以下的电子邮件地址和我们联络, 谢谢.4. Group manager of account and email address list files.集团总经理的帐户和电子邮件地址名单的文件.5. You have to enter something in the email address field.您必须在电子邮件地址框中输入内容.6. This Email Address will be used to activate your Account.此邮件地址将用来启动你的帐户.7. The Email Address and Inbound Directory must point to the same place.电子邮件地址和入站目录必须指向相同地点.8. Enter your email address and click the Submit button.输入您的电子邮件地址,并点击提交键.9. Zuckerberg's Facebook friends access to his email address and his cell - phone number.如果在Facebook上成为了扎克伯格的朋友,就可以看到他的邮箱地址和电话号码.10. Your email address. Please provide a valid email address.您的电子邮件信箱,请提供一个正确的信箱,以方便我们与您连络.11. Please enter all fields and a valid email address. Thank you.请填入所有的空格并使用一个正常操作的邮件地址. 多谢.12. Miyoko typed in her telephone number, address and email address.美代子填入了自己的电话号码、地址和电子邮件地址.13. Please enter a valid email address. Email confirmation does not match.请输入一个有效的邮件地址. 邮件确认不匹配.14. I would appreciated if he can give me his email address.希望他可以给我一个他的电子邮箱!15. Include your full name, mailing address, phone number and email address.包括你的全名、通信地址、电话号码和电子邮件地址.电子邮件地址email address的双语例句1. First you need to get an e-mail address.首先你需要申请一个电子邮件地址。
Thank you for contacting us.如果有人写信来询问公司的服务,就可以使用这句句子开头。
向他们对公司的兴趣表示感谢。
Thank you for your prompt reply.当一个客户或是同事很快就回复了你的邮件,一定记得要感谢他们。
如果回复并不及时,只要将“prompt”除去即可,你还可以说,“Thank you for getting back to me.”Thank you for providing the requested information.如果你询问某人一些信息,他们花了点时间才发送给你,那就用这句句子表示你仍然对他们的付出表示感激。
Thank you for all your assistance.如果有人给了你特别的帮助,那一定要感谢他们!如果你想对他们表示特别的感激,就用这个句子,“I truly appreciate … your help in resolving the problem.”Thank you raising your concerns.就算某个客户或是经理写邮件给你对你的工作提出了一定的质疑,你还是要感谢他们。
这样你能表现出你对他们的认真态度表示尊重及感激。
同时,你也可以使用,“Thank you for your feedback.”在邮件的结尾在邮件开头表示感谢一般是表示对对方过去付出的感谢,而在邮件结尾处表示感谢是对将来的帮助表示感谢。
事先表示感谢,能让对方在行动时更主动更乐意。
Thank you for your kind cooperation.如果你需要读者帮助你做某事,那就先得表示感谢。
Thank you for your attention to this matter.与以上的类似,本句包含了你对对方将来可能的帮助表示感谢。
Thank you for your understanding.如果你写到任何会对读者产生负面影响的内容那就使用这句句子吧。
邮寄格式英语作文Mailing FormatProper mailing format is an essential aspect of effective communication and professional correspondence. Whether you are sending a personal letter, a business document, or an important package, following the correct mailing format can ensure that your item reaches its intended recipient promptly and efficiently. In this essay, we will explore the key elements of mailing format and discuss the importance of adhering to established standards.The first and most crucial element of mailing format is the address. A properly formatted address should include the recipient's full name, the street address or post office box number, the city, state or province, and the zip or postal code. It is essential to ensure that the address is accurate and complete to avoid any delays or misdirection in the delivery process. Additionally, the return address, which includes the sender's name and address, should be clearly displayed, typically in the top left corner of the envelope or package.Another important aspect of mailing format is the placement and formatting of the address information. The recipient's address should be centered on the envelope or package, with each line of the address separated by a single line space. The return address should be positioned in the top left corner, with a single line space between each line of the address. It is crucial to maintain consistent spacing and alignment to ensure that the mailing information is easy to read and process by postal services.The placement of postage is also a critical component of mailing format. Postage should be affixed to the top right corner of the envelope or package, ensuring that it is clearly visible and accessible for postal workers to process. The type and amount of postage required will depend on the size, weight, and destination of the item being mailed. It is essential to research and comply with the current postal regulations to avoid any delays or additional fees.In addition to the basic address and postage requirements, there are other elements of mailing format that may be necessary depending on the type of item being sent. For example, if you are mailing a package, you may need to include a customs declaration form or other shipping documentation. If you are sending a registered or certified letter, you may need to include additional forms or labels to indicate the special handling required.The importance of adhering to proper mailing format cannot be overstated. Proper formatting not only ensures the timely and accurate delivery of your item but also demonstrates professionalism and attention to detail. In a business context, following mailing format standards can help to establish a positive image and build trust with clients, customers, and business partners.Moreover, proper mailing format can also help to prevent common issues that can arise during the delivery process. For instance, if the address is not clearly legible or the postage is not properly affixed, the item may be delayed or returned to the sender, leading to frustration and potential financial consequences.In conclusion, mailing format is a crucial aspect of effective communication and professional correspondence. By understanding and adhering to the established standards for address formatting, postage placement, and additional documentation requirements, you can ensure that your items are delivered promptly and efficiently, while also demonstrating your attention to detail and commitmentto professionalism. Whether you are sending a personal letter or a business document, mastering the art of mailing format can be a valuable asset in your communication toolkit.。
COERCED MARKOV MODELS FOR CROSS-LINGUAL LEXICAL-TAG RELATIONS Pascale FungComputer Science DepartmentColumbia Universitypascale@CURRENT MAILING ADDRESS:1.INTRODUCTIONThe analysis,transfer,and synthesis paradigm for machine translation is readily amenable to statistical methods(Brown et al.1993).Since the transfer stage is designed to exploit mapping knowledge about different linguistic relationships between the source and target languages,statis-tical information can be incorporated into this stage.Typical types of mapping relations include sentence to sentence,word to word,or part-of-speech(POS)tags to tags.Statistical algorithms generally model the word to word lexical relations between a pair of sentences in the target and source languages with probabilities(Brown et al.1993;Dagan et al.1993;Dagan&Church1994; Fung&McKeown1994;Wu&Xia1994;Fung1994).These probabilities help in the transfer stage to constrain or prune the search for an optimal sequence of translated words.Linguistic information such as part-of-speech has also found to be useful for constraining this search.(Chang&Chen1994; Papageorgiou et al.1994).In this paper we investigate an underutilized source of constraints,namely,the mapping be-tween words in the source language and parts-of-speech in the target language.Such information would also constrain search in the translation model.We believe the mapping relations can be automatically learned from bilingual corpora.However,to our knowledge no such attempt has been made,perhaps to the modeling difficulties in the problem.We introduce a Coerced Markov Model(CMM)representation that accommodates mapping relations between source-words and target-tags in a statistical framework.Although there has been work on mapping between source language tags and target language tags,(Chang&Chen1994;Papageorgiou et al.1994),this mapping might not be meaningful or sufficiently helpful for translation.In the common scenario,texts of both languages are tagged by their respective POS taggers.A tag to tag mapping between the two languages is obtained from the tagged text.However,most part-of-speech classes are determined by human according to the linguistic knowledge in that particular language.It is not evident that there should be1a direct correspondence between POS classes in two different languages,especially in language pairs which do not share any common root such as English and Chinese.The relationship we derive from English and Chinese part-of-speech mapping is therefore not necessarily a good constraint for translation search.On the other hand,source language words are capable of giving much more discriminative information about target tags than source tags are.Moreover,a reliable tagger for source languages such as Chinese may not be available in thefirst place.We propose to capture the correlation between source words and target tags with the Coerced Markov Model.As we discuss below, CMMs are a particular case of discrete,first-order,hidden Markov models such that the state sequence is determined by coercion from some second state sequence from outside the model.One application of the CMM is that it can predict the English tag sequence corresponding to a given Chinese sentence.This tag sequence can be used as a constraint to the pruned search of the transfer model for the production of an English lexical sequence.Since a transfer model produces an English translation sentence by choosing the individual English words corresponding to the individual words in the Chinese sentence,it can produce a number of translation hypotheses.An alternative application of CMM is to provide a measure of the goodness of the hypotheses.In the following sections,wefirst define the CMM formalism,and then describe its training and testing stages.2.COERCED MARKOV MODELSMarkov chains are widely used for characterizing parametric random processes.The basic theory of Hidden Markov Models(HMM)was proposed by Baum&Petrie(1966);Baum&Egon (1967)as early as the1960s.It was later adapted by Baker(1975);Jelinek et al.(1975)for processing speech signals.The fundamental assumption of using a Markov model for a linguistic2mapping(in our case between words in one language and tags in the other language)is that the mapping is a stochastic process and its parameters are estimable.A Markov chain describes the changes of states of a system.For example,at time,the system is in state,it changes to state at time,then there is a state transition from to with certain probability.First order Markov chain assumes the probabilistic dependency of a state is only on its preceding state.i.e.At a given state,there is an output from the state.This output can be continuous,such as the spectral signal of speech in a speech recognition system,or discrete,such as the weather condition of a meteorology system.If we regard the mapping between Chinese word sequences and the tag sequence of its corresponding English translation as a stochastic process,the Coerced Markov Model for the process is discrete.A Markov model is hidden if its states are not deterministically observable.Given an observation sequence,the underlying states are non-deterministic.Hidden Markov Model(HMM)are typically used in speech processing where the underlying states of a model do not correspond to something explicit such as a phoneme or a word.CMM states are also non-deterministic and therefore hidden because the same output sequence can be generated from different state sequences given a particular model.For our application in Chinese-English translation,the CMM is coercing English tags into Chinese language modeling.In other words,the CMM says that English tags cannot just follow the rules in English language models,they also have to consider the fact that they are now“partners”of Chinese words which also have their own rules.CMM is modeling the“adaptation”of English tags to Chinese word orders.This is a step beyond monolingual language modeling such as word3N-gram or class N-gram computation.CMM’s purpose and strength is to model cross-lingual class1.N-grams.An example of a four-state CMM is illustrated in FigureFormally,we define the following elements:1.the state variable:Hidden Chinese states,with coerced English tag class values2.the observable symbol:Chinese wordsOur definition of and is to optimize the both the modeling and the discriminative power of the CMM.If we had chosen individual lexical items to be the states,there would be a lot of cases where word never follows word given any in the dictionary.There would be many restrictions on which state interconnects with which and makes our model neitherflexible nor powerful.Instead,it is logical to use POS classes as the states of a CMM because these classes have some physical(or rather,linguistic)significance.In addition,since presumably any POS class can follow any other POS class given a large enough corpus,there can be interconnection between any two given states.This means that the CMM is an ergodic model.This makes CMM potentially4moreflexible and powerful.It follows that,the observable symbols should be the lexical items in the other language.Once we have defined the nature of and,we have to choose which language and should come from.We choose to be the English POS classes,because English POS taggers are readily available and there has been enough agreement in thefield as to which once the basic “good”English POS classes are.On the other hand,due to the short history of Chinese natural language processing,most Chinese taggers are still under research,and there is still a lack of a general paradigm for POS classes determination in Chinese.It follows that is the set of Chinese words in the dictionary.Referring again to Figure1,each state in the CMM corresponds to an English POS class.For our experiments,we use Brill(1993)’s tagger of English tag classes.Given any two states,there is a weighted transition going in either direction from one to the other.Each state can also transit into itself.The output from a state is an array of Chinese words with different weights.The next three sections of this paper discuss methods and experiments for three fundamental problems of CMMs:1.Estimation:Given a CMM(i.e.,its topology),estimate its parameters so as to best describean observed training sequence.2.Path recovery:Given a CMM,its parameters,and a test observation sequence,determine theoptimal hidden state sequence.Can be used to suggest constraints on translation hypotheses.3.Scoring:Given a CMM and its parameters,determine the probabilistic score of a sequenceof states.Can be used to score translation hypotheses.It may be helpful,in order to understand these three problems,to note a certain parallel between them and the three fundamental problems of HMM(Rabiner&Juang1993),although the cross-lingual coercion leads to substantial differences.We will see that problem(1)is the parameter5estimation process for a CMM,and that problems(2)and(3)can be used for two different translation applications that each yield an experimental evaluation.3.ESTIMATIONIn this section we describe how we estimate1.the transition probabilities2.the output probabilitiesgiven a word-aligned parallel corpus.Remember that the objectives of training the CMM are,first, to best model the stochastic process of Chinese word sequences co-occuring with their English tag counterparts,and second,to supply the most useful constraints possible to help prune the search process in a statistical transfer model.Transition probabilities We have defined the state and output symbols of CMM,now we need to train the parameters of CMM.In this section,we describe how to compute the transition probabilities where and are any two states in CMM.To use an example,the Chinese sentencehas the English alignmentThese arrangements enhance our ability<to>maintain monetary stability.with their POS tags as shown in Table1.The tag sequence$contains as null tags since there is no English word alignment to the Chinese word at that position.According to this Chinese sentence and its aligned English words,there is a transition from initial state to,to,to,….Here,the6English tag sequence is coerced into modeling the Chinese word sequence.If our training data had this single sentence only,then we would get a total of13transitions and each transition probability would be=1/13.The null tag state comes from the particular phenomenon in Chinese/English translations where many Chinese words are not aligned to any English words due to a relatively large linguistic difference between the two languages.We believe these null alignments give highly unreliable information.In our experiments we penalize the transitions into and out of the null state by assigning a very low probability to them.Thefinal transition probabilities are converted into the logarithmic form for computational purpose.In general,since the probabilities are less than one,the logs become negative numbers,therefore we take the negation of the logarithm probabilities for computation.So the formula for transition probabilities is:number of transitions from i to jTable1:Training data formatAlignment position English POS1DT2NNS4VB5PRP$8VB9JJ10NN6NN16.Using a similar negative logarithmic form withflooring as transition probabilities,the output probabilities we get are:number of Chinese word k observed when in state j4.English word alignment to the individual Chinese words were found by using a Estimation-Maximization model(Wu&Xia1994)5.Filtering of the training corpus was done by applying criteria described in(Wu1995)We obtained a total of1885Chinese sentences with aligned English words and English POS tags as our training corpus.An example of the training corpus format is shown in thefirst four columns of Table1.Using this training data,we trained CMM as follows:pute initial probabilities:pute transition probabilities:there are1969null transitions probabilities out of atotal of11236transitions.pute output probabilities:4.OPTIMAL PATH RECOVERYWe use two different evaluation methods corresponding to the solutions of problem(2)and problem(3)in CMM design.Evaluation one was to produce a English tag sequence from a Chinese sentence.We use a Chinese sentence from the corpus which was not included in the training set as the test sample.The Chinese and its corresponding English aligned words and their tags are shown in Table2.We use the Chinese part of sentence in Table2as input to this test,and compare the output to the English tag sequence.Tofind the solution for predicting the best state sequence,i.e. English tag sequence from the observation sequence,i.e.the Chinese sentence of length C,we use a Viterbi algorithm(Viterbi1967;Forney1973)as the9following,considering that transition probabilities and output probabilities are in the negative logarithmic form:Initializationwherewhere probability of being the initial statewhereRecursionwhereTerminationViterbi scorestate sequencePath reconstruction10Table2:Test sentenceEnglish word Transitions </s>We PRP2provide VB22additional JJ5CD090CD011<>14homes NNS18attention NN17places NNS23The state sequence obtained is compared to the tag sequence in the corpus as follows:Viterbi tag sequence PRP MD IN NN JJ<>CD CD CD CD NNS NN:CC<>:<>. Corpus tag sequence PRP MD VB JJ JJ<>CD CD CD CD<>NN NNS CC NN NNS NNS. Mismatchings*******We can see that our tag sequence output corresponds mostly to the original one.All the mismatchings are due to either the Chinese word not being found in the dictionary or there being no English word alignment for a Chinese word.This illustrates the fact that CMM can generate English tags from Chinese words when Chinese word was correctly segmented and found in the dictionary.However,when we actually apply CMM to constrain a translation model,we can easily deal with these two cases by applying a null CMM constraint default.i.e.:111if not found in dictionary||no English word alignment2=translation model probability;3else4=+translation model probability;5.SCORING TRANSLATION HYPOTHESESAnother way of using CMM for translation is in the solution to problem(3):given an English state sequence,we try to score it by CMM.Statistical machine translation models can generate a number of translation hypotheses sentences according to the EM-based transfer model.This is analogous to the N-best algorithm used for speech recognition and is found to be more optimal in choosing the best candidate sentence().This sentence would be the best translation in our case.Given a hypothesis English sentence with length,we obtain a tag sequence by the following way:Scorewhere the Chinese word aligned to the English wordand length of the English sentenceFor our test,we manually generated a list of13-best translation hypotheses according to the Chinese words in the sentence500012Since the Chinese character sequence can be segmented in different ways into word sequences,the total number of Chinese words in a sentence can be different.For each Chinese sentence with a particular length,we manually generate an alignment English word to the individual Chinese words. Some Chinese words can be aligned to multiple English words leading to multiple hypotheses. Each of these hypothetical sentence is tagged by Brill’s tagger.We score the tag sequence of each hypothesis by summing the logarithmic transition probabilities from one tag to the following one,normalized by the length of the sentence.The English hypotheses,their tag sequences sorted by CMM scores are shown in Table3.The lowest score indicates the best translation.The best candidate was chosen to be We will provide the age and additional5000home and care home places.which is indeed the reference translation for the sentence in the original corpus.Note that CMM scoring cannot choose between two sequences which differ only in their lexical items but not tag sequences.For example,sequence(9)and(10)differ only by theirfinal word places versus seats,these two words are both tagged as,therefore the scores for(9)and(10) are the same.However,this lexical choice is obviously a problem of English language modeling, and we can hope that the synthesis part of the statistical translation model will make an intelligent decision between the two.6.DISCUSSIONA problem of CMM which might deserve more research is how to better model the null states. Since there are many null alignments of Chinese words to English,one can try to come up with a more powerful model by looking at the classes of Chinese words which typically have null alignments or other patterns for these alignments.We used a single English POS class to represent a state in the CMM,it would be worth experimenting with a more complex state such as POS bigrams.POS bigrams are a feature of monolingual language modeling and their inclusion can possibly render CMM more powerful.13Table3:13-best translation hypotheses and their CMM scores11.364408We will provide the aged an additional5000home and attention home places.PRP MD VB DT JJ DT JJ CD NN CC NN NN NNS.11.932087We will provide old people in addition5000old people home and attention home places.PRP MD VB JJ NNS IN NN CD JJ NNS NN CC NN NN NNS.11.982643We will for the old people increase5000old people homes and attention attention homes places.PRP MD IN DT JJ NNS NN CD JJ NNS NNS CC NN NN NNS NNS.12.153794We will for the aged an additional5000home and attention attending home places.PRP MD IN DT JJ DT JJ CD NN CC NN VBG NN NNS.12.219560We will for the aged add5000home and attention home places.PRP MD IN DT JJ VB CD NN CC NN NN NNS.12.342510We will provide the aged additional5000home and attention home places.PRP MD VB DT JJ JJ CD NN CC NN NN NNS.12.766230We will for the aged increase5000aged people home and caring and attention home places.PRP MD IN DT JJ NN CD VBN NNS NN CC NN CC NN NN NNS.12.827581We will provide the aged an additional5000aged home and attention home places.PRP MD VB DT JJ DT JJ CD VBN NN CC NN NN NNS.12.928034We will provide the aged increasing5000old people home and attention attention home places.PRP MD VB DT JJ NN CD JJ NNS NN CC NN NN NN NNS.12.928034We will provide the aged increasing5000old people home and attention attention home seats.PRP MD VB DT JJ NN CD JJ NNS NN CC NN NN NN NNS.13.120893We will provide the aged an additional5000the aged home and attention home places.PRP MD VB DT JJ DT JJ CD DT JJ NN CC NN NN NNS.13.371675We will provide the aged adding5000aged home and attention home place.PRP MD VB DT JJ NN CD VBN NN CC NN NN NN.13.402670We will for the aged addition5000home and caring attending old people home places.PRP MD IN DT JJ NN CD NN CC VBG VBG JJ NNS NN NNS.We have used a predefined English POS classes for training our CMM,it might be worthwhile to investigate how different POS class definitions can affect CMM.Finally,we would like to point out that there is another similarity between CMM and HMM which is that CMM can also be regarded as a generator of observations:given,,,,and a sequence of English tags as input,CMM can generate a observation sequence of Chinese words. This would seem to be an application in English to Chinese translation.147.CONCLUSIONWe have seen that the Coerced Markov Model is effective in modeling the relationship between lexical sequence of a sentence in one language and part-of-speech sequence in its translated version. The model coerces the English tag sequence into modeling Chinese word sequence structure,and can be seen as a form of cross-lingual language modeling.We have formally specified the CMM states,transitions,and output symbols.A method was given for estimating its parameters from a word-aligned training corpus,corresponding to the solution to thefirst fundamental problem of CMMs.We have shown two applications to improving the statistical transfer model,corresponding to the solutions of fundamental problems(2)and(3)of CMMs:first,we showed that CMM can predict a English tag sequence given a Chinese sentence, providing tag constraints to the search of best English lexical sequence as translation;second,we showed that CMM scoring of a N-best list of translation hypotheses can help to select the best one.ReferencesB AKER,J.K.1975.The Dragon system–An overview.IEEE Trans.Acoustics,Speech,SignalProc.,23(1):24–29.B AUM,L.E.&J.A.E GON.1967.An inequality with applications to statistical estimation forprobabilistic functions of a Markov process and to a model for ecology.Bull.Amer.Meteorol.Soc.,73:360–363.B AUM,L.E.&T.P ETRIE.1966.Statistical inference for probabilistic functions offinite stateMarkov chains.Ann.Math.Stat.,37:1554–1563.B RILL,E RIC,1993.A corpus-based approach to language learning.University of Pennsylvaniadissertation.B ROWN,P.F.,S.A.D ELLA P IETRA,V.J.D ELLA P IETRA,&R.L.M ERCER.1993.The mathematicsof machine translation:Parameter putational Linguistics,19(2):263–311.15C HANG,J YUN-S HENG&H UEY-C HYUN C ing partically aligned parallel text andpart-of-speech information in word alignment.In Proceedings of the First Conference of the Association for Machine Translation in the Americas,16–23,Columbia,Maryland.D AGAN,I DO&K ENNETH W.C HURCH.1994.Termight:Identifying and translating technicalterminology.In Proceedings of the4th Conference on Applied Natural Language Processing, 34–40,Stuttgart,Germany.D AGAN,I DO,K ENNETH W.C HURCH,&W ILLIAM A.G ALE.1993.Robust bilingual word alignmentfor machine aided translation.In Proceedings of the Workshop on V ery Large Corpora: Academic and Industrial Perspectives,1–8,Columbus,Ohio.F ORNEY,G.D.1973.The Viterbi algorithm.In Proceedings of the IEEE,volume61,268–278.F UNG,P ASCALE,1994.A pattern matching method forfinding noun and proper noun translationsfrom noisy parallel corpora.In submission.F UNG,P ASCALE&K ATHLEEN M C K EOWN.1994.Aligning noisy parallel corpora across languagegroups:Word pair feature matching by dynamic time warping.In Proceedings of the First Conference of the Association for Machine Translation in the Americas,81–88,Columbia, Maryland.F UNG,P ASCALE&D EKAI W U.1994.Statistical augmentation of a Chinese machine-readabledictionary.In Proceedings of the2nd Annual Workshop on V ery Large Corpora,Kyoto,Japan. J ELINEK,F.,L.R.B AHL,&R.L.M ERCER.1975.Design of a linguistic statistical decoder for the recognition of continuous speech.IEEE rmation Theory,21:250–256.P APAGEORGIOU,H.,L.C RANIAS,&S.P IPERIDIS.1994.Automatic alignmnet in parallel corpora.In Proceedings of the32nd Annual Conference of the Association for Computational Linguistics Student Session,331–333,Las Cruces,New Mexico.16R ABINER,L AWRENCE&B ING-H WANG J UANG.1993.Fundamentals of speech recognition.Signal Processing Series.Englewood Cliffs,NJ:Prentice-Hall.V ITERBI,A.J.1967.Error bounds for convolutional codes and an asymptotically optimal decoding algorithm.IEEE Transaction on Information Theory,13:260–269.W U,D EKAI.1994.Aligning a parallel English-Chinese corpus statistically with lexical criteria.In Proceedings of the32nd Annual Conference of the Association for Computational Linguistics, 80–87,Las Cruces,New Mexico.W U,D EKAI,1995.Grammarless extraction of phrasal translation examples from parallel texts.In submission.W U,D EKAI&P ASCALE F UNG.1994.Improving Chinese tokenization with linguisticfilters on statistical lexical acquisition.In Proceedings of the4th Conference on Applied Natural Language Processing,180–181,Stuttgart,Germany.W U,D EKAI&X UANYIN X IA.1994.Learning an English-Chinese lexicon from a parallel corpus.In Proceedings of the First Conference of the Association for Machine Translation in the Americas,206–213,Columbia,Maryland.17。