下载文档原格式
Autodesk®Moldflow® Insight Plastics made perfect.1Autodesk® Moldflow® Insight software, part ofthe Autodesk® solution for Digital Prototyping,provides injection molding simulation tools foruse on digital prototypes. Providing in-depthvalidation and optimization of plastic parts andassociated injection molds, Autodesk MoldflowInsight software helps study the injection moldingprocesses in use today. Used by some of thetop manufacturers in the automotive, consumerelectronics, medical, and packaging industries,Autodesk Moldflow Insight software helps toreduce the need for costly mold rework andphysical prototypes, minimize delays associatedwith removing molds from production, and getinnovative products to market faster.Autodesk Moldflow Insight Product LineAutodesk is dedicated to providing a wide rangeof injection molding simulation tools to help CAEanalysts, designers, engineers, mold makers, andmolding professionals create accurate digitalprototypes and bring better products to marketat less cost.Validation and Optimization of Plastic PartsWith the use of plastic parts on the rise in almost every industry, and the pressure to reduce costs and cut time to market, the need for simulation tools that provide deep insight into the plastic injection molding process has neverbeen greater.12Hot Runner SystemsModel hot runner system components and set up sequential valve gates to help eliminate weld lines and control the packing phase.Plastic Flow SimulationSimulate the flow of melted plastic to help optimize part and mold designs, reduce potential part defects, and improve the molding process.Part DefectsDetermine potential part defects, such as weld lines, air traps, and sink marks, and then rework designs to help avoid these problems.Thermoplastic FillingSimulate the filling phase of the thermoplasticinjection molding process to help predict the flow of melted plastic and fill mold cavities uniformly; avoid short shots; and eliminate, minimize, or reposition weld lines and air traps.Thermoplastic PackingOptimize packing profiles and visualize magnitude and distribution of volumetric shrinkage to help minimize part warpage and reduce defects, such as sink marks.Validate and optimize plastic parts, injection molds, and theinjection molding process.Feed System SimulationModel and optimize hot and cold runner systems and gating configurations. Improve part surfaces, minimize part warpage, and reduce cycle times.Gate LocationIdentify up to 10 gate locations simultaneously. Minimize injection pressure and exclude specific areas when determining gate location.Runner Design WizardCreate feed systems based on inputs for layout, size, and type of components, such as sprue, runners, and gates.Balancing RunnersBalance runner systems of single-cavity, multicavity, and family mold layouts so parts fill simultaneously,reducing stress levels and volume of material.Mold Cooling SimulationImprove cooling system efficiency, minimize part warpage, achieve smooth surfaces, and reduce cycle times.Cooling Component ModelingAnalyze the mold’s cooling system efficiency. Model cooling circuits, baffles, bubblers, and mold inserts and bases.Cooling System AnalysisOptimize mold and cooling circuit designs to help achieve uniform part cooling, minimize cycle times, reduce part warpage, and decrease manufacturing costs.WarpagePredict warpage resulting from process-inducedstresses. Identify where warpage might occurand optimize part and mold design, materialchoice, and processing parameters to help controlpart deformation.Core Shift ControlMinimize the movement of mold cores by determin-ing ideal processing conditions for injectionpressure, packing profile, and gate locations.Fiber OrientationControl fiber orientation within plastics to helpreduce part shrinkage and warpage across themolded part.CAE Data ExchangeValidate and optimize plastic part designs usingtools to exchange data with structural simulationsoftware. CAE data exchange is available withAutodesk® Algor® Simulation, ANSYS®, andAbaqus® structural simulation software to accountfor the effects of processing on the performanceof fiber-filled, injection-molded plastic parts whensubjected to service loading.Rapid Heat Cycle MoldingSet up variable mold surface temperature profilesto both maintain warmer temperatures duringfilling to achieve smooth surfaces, and also reducetemperatures in the packing and cooling phases tohelp freeze parts and decrease cycle times.Shrinkage & Warpage SimulationEvaluate part and mold designs to help controlshrinkage and warpage.ShrinkageMeet part tolerances by predicting part shrinkagebased on processing parameters and grade-specificmaterial data.34Thermoset Flow SimulationSimulate thermoset injection molding, RIM/SRIM, resin transfer molding, and rubber compound injection molding.Reactive Injection MoldingPredict how molds will fill with or without fiber-reinforced pre-forms. Help avoid short shots due to pre-gelation of resin, and identify air traps and problematic weld lines. Balance runner systems, select molding machine size, and evaluate thermoset materials.Microchip EncapsulationSimulate encapsulation of semiconductor chips with reactive resins and the interconnectivity of electrical chips. Predict bonding wire deformation within the cavity and shifting of the lead frame due to pressure imbalances.Underfill EncapsulationSimulate flip-chip encapsulation to predictmaterial flow in the cavity between the chip andthe substrate.Leading-Edge Simulation ToolsUse leading-edge simulation tools to solve design challenges.Insert OvermoldingRun an insert overmolding simulation to helpdetermine the impact of mold inserts on melt flow, cooling rate, and part warpage.Two-Shot Sequential OvermoldingSimulate the two-shot sequential overmolding process: one part is filled; the tool opens and indexes to a new position; and a second part is molded over the first.BirefringencePredict optical performance of an injection-molded part by evaluating refractive index changes that result from process-induced stresses. Evaluate multiple materials, processing conditions, and gate and runner designs to help controlbirefringence in the part.Specialized Molding ProcessesSimulate a wide range of plastic molding processes and state-of-the-art process applications.Gas-Assisted Injection MoldingDetermine where to position polymer and gas entrances, how much plastic to inject prior to gas injection, and how to optimize size and placement of gas channels.Co-Injection MoldingVisualize the advancement of skin and core materials in the cavity and view the dynamic relationship between the two materials as filling progresses. Optimize material combinations while maximizing the product's cost-performance ratio.Injection-Compression MoldingSimulate simultaneous or sequential polymer injection and mold compression. Evaluate material candidates, part and mold design,and processing conditions.5CAD Interoperability and MeshingUse tools for native CAD model translation and optimization. Get geometry support for thin-walled parts and thick and solid applications. Select mesh type based on desiredsimulation accuracy and solution time.CAD Solid ModelsImport and mesh solid geometry from Parasolid ®-based CAD systems, Autodesk ® Inventor ® software, CATIA ® V5, Pro/ENGINEER ®, and SolidWorks ®, as well as IGES and STEP universal files.Error Checking and RepairScan imported geometry and automatically fix defects that can occur when translating the model from CAD software.Centerline Import/ExportImport and export feed system and coolingchannel centerlines from and to CAD software to help decrease modeling time and avoid runner and cooling channel modeling errors.Autodesk ® Moldflow ® CAD DoctorCheck, correct, heal, and simplify solid models imported from 3D CAD systems to prepare for simulation.3D SimulationsPerform 3D simulations on complex geometry using a solid, tetrahedral, finite element mesh technique. Ideal for electrical connectors, thick structural components, and geometries with thickness variations.Dual Domain TechnologySimulate solid models of thin-walled parts using Dual Domain™ technology. Work directly from 3D solid CAD models, leading to easier analysis of design iterations.Midplane MeshesGenerate 2D planar surface meshes with assignedthicknesses for thin-walled parts.Results Interpretation & PresentationUse a wide range of tools for model visualization, results evaluation, and presentation.Automatic Reporting ToolsUse the Report Generation Wizard to create web-based reports. Prepare and share simulation results more quickly and easily with customers, vendors, and team members.Microsoft® OfficeExport results and images for use in Microsoft®Word reports and PowerPoint® presentations. Autodesk® Moldflow® Communicator Collaborate with manufacturing personnel, procurement engineers, suppliers, and external customers using Autodesk® Moldflow®Communicator software. The Autodesk Moldflow Communicator results viewer enables you to export results from Autodesk Moldflow software so stakeholders can more easily visualize, quantify, and compare simulation results.Material DataImprove simulation accuracy with precisematerial data.Materials DatabaseUse the built-in materials database of grade-specific information on more than 8,000 plasticmaterials characterized for use in plastic injectionmolding simulation.Autodesk® Moldflow® Plastics LabsGet state-of-the-art plastic material testingservices, expert data-fitting services, and extensivematerial databases.Productivity ToolsUse extensive help to boost productivity.HelpGet help on a results plot, including informationon what to look for and how to correct typicalproblems. Learn more about solver theory,interpreting simulation results, and designingbetter plastic parts and injection molds.Automation and CustomizationAutomate common tasks and customize AutodeskMoldflow software for your organization.API ToolsApplication programming interface (API) toolsexpand the functionality of Autodesk Moldflowsoftware by enabling you to automate commontasks, customize the user interface, work with third-party applications, and help implement corporatestandards and best practices.WorkspacesCustomize the user interface and applicationfeatures for your team. Set up profiles to guide newusers through the simulation process and identifycommon problems. Define other profiles to giveadditional functionality and flexibility to moreexperienced users. Results Evaluation and Productivity ToolsVisualize and evaluate simulation results, and use the automatic reporting tools to share the results with stakeholders. Take advantage of features such as a materials database and customizable workspaces to further boostproductivity.6。
操稳特性快速评估及其在飞机设计中的应⽤南京航空航天⼤学硕⼠学位论⽂操稳特性快速评估及其在飞机设计中的应⽤姓名：张帅申请学位级别：硕⼠专业：飞⾏器设计指导教师：余雄庆20081201南京航空航天⼤学硕⼠学位论⽂摘要飞机总体设计阶段需要对飞机设计⽅案的操稳特性作出快速评估。
在采⽤主动控制技术进⾏飞机总体设计时，还需要考虑飞⾏控制系统的作⽤，对包括飞控系统在内的全机操稳特性进⾏快速评估。
针对以上需求，本⽂主要完成了以下研究⼯作：1）对应⽤飞机操稳分析程序（DATCOM）计算⽓动系数和导数的⽅法进⾏了研究，提出了⼀些实⽤的输⼊⽂件建模⽅法，为DATCOM程序开发了输⼊与输出接⼝程序，提⾼了DATCOM 程序的使⽤效率，⽅便了程序与其它分析系统的集成。
2）应⽤Simulink对⽓动数据进⾏插值处理，并建⽴了⾮线性模型；应⽤MATLAB的功能函数实现了⾃动提取线性模型以及根据线性模型分析飞机本体的操稳特性。
3）利⽤ Simulink 的建模及仿真功能，实现了飞控系统的建模、控制律设计和仿真，并建⽴了全机仿真分析模型，实现了全机仿真模型的快速配置以及飞⾏仿真。
4）对MATLAB的RTW⼯具以及引擎技术进⾏了研究，实现了仿真模型的编译处理以及外部调⽤。
在此基础上，将带仿真模型的分析程序进⾏了编译处理；采⽤iSIGHT软件对编译⽣成的可执⾏⽂件进⾏了集成，完成了操稳特性快速评估系统的构建；系统可以单独运⾏，也可以在飞机多学科优化设计系统中作为⼦系统应⽤。
5）在飞机总体设计中应⽤操稳特性快速评估系统，研究了飞翼布局⽆⼈机的操稳特性，以及⼤型民⽤飞机放宽静稳定度技术。
应⽤研究充分表明，本⽂所提出的⽅法和建⽴的操稳特性快速评估系统特别适合飞机总体设计阶段对操稳特性的评估，为飞机多学科优化设计以及应⽤主动控制技术进⾏飞机总体设计提供了有效的操稳分析⽅法和⼯具。
关键词：飞机总体设计；操稳特性；主动控制技术；飞⾏控制系统；控制律；飞⾏仿真操稳特性快速评估及其在飞机设计中的应⽤AbstractIn the conceptual design of aircraft, the S&C (stability and control) of the aircraft need to be evaluated rapidly. When ACT (Active Control Technology) is used in the conceptual design，the rapid evaluation of the S&C need to include the FCS (Flight Control System). The Methods and tool were developed in this thesis to meet this need. The research work in this thesis is presented as the followings:1) In terms of computing aerodynamic coefficient and derivative, some useful methods for input file modeling was improved for DATCOM, some interface code for input and output of DATCOM was developed. The interface code make the DATCOM easier to use and more convenient to be integrated with other codes.2) In Simulink software environment, aerodynamic data interpolation has been completed and non-linear aerodynamic model was set up. Linear model could be extracted from non-linear model by functions in MATLAB. And then the inherent S&C of the aircraft could be analyzed.3) By use of modeling and simulation function in Simulink software, the modeling, control law designing and simulation of the FCS was completed. Then the flight simulation model of the whole aircraft with FCS could be set up.4) By use of the RTW tools and ENGINE technolony in MATLAB, flight simulation model could be compiled. Therefore, all ofthe analysis codes have been compiled to executable files. These files have been integrated by iSIGHT software and built into an integrated rapid evaluation system of the S&C. This system could be used alone or as a sub-system in MDO (Multidisciplinary Design Optimization).5) The rapid evaluation system have been verified by two examples. An unmanned air vehicle with fly-wing configuration was analyzed by this system, and a civil jet transport with RSS (Relaxed Static Stability) technolony was analyzed by this system.The applications demonstrate that the methods and rapid evaluation system developed in this thesis can meet the need of S&C evaluation in the conceptual design. It can be used as an effective tool in the conceptual design of aircraft with ACT technology, as well as a subsystem of MDO.Keywords:Conceptual Design of Aircraft; S&C; ACT; FCS; Control Law; Flight Simulation南京航空航天⼤学硕⼠学位论⽂图表清单图1.1 主动控制技术（下）与传统飞机设计⽅法（上）⽐较 (3)图1.2 飞机多学科优化设计系统框架 (4)图1.3 操稳特性快速评估系统框图 (8)图2.1 本⽂所采⽤的飞机坐标系 (9)图2.2 DATCOM输⼊⽂件的构成⽅式 (12)图2.3 DATCOM输⼊⽂件⽰例 (13)图2.4 导⼊到MATLAB中的DATCOM输出数据 (15)图2.5 输⼊接⼝程序⽣成的DATCOM输⼊⽂件及构成关系 (17)图2.6 利⽤插值处理模块对阻⼒系数有关数据插值处理 (23)图2.7 以Simulink封装模块建⽴的飞机本体⾮线性模型 (27)图2.8 以封装模块与S-function运动⽅程建⽴的飞机本体⾮线性模型 (27)图3.1 飞控系统的⼀般构成 (34)图3.2 飞机纵向线性系统仿真分析模型⽰例 (35)图3.3 ⼤⽓扰动仿真模型 (36)图3.4 爬升指令仿真模型 (37)图3.5 传感器与测量模型 (38)图3.6 ⼤⽓数据计算机仿真模型中的离散采样环节 (39)图3.7 陀螺仪和线加速度计模型 (39)图3.8 舵机仿真模型 (40)图3.9 飞机系统仿真模型 (40)图3.10 全机飞⾏仿真分析模型 (41)图3.11 Simulink与FlightGear视景仿真的接⼝模型 (43)图4.1 系统集成⽅案原理框图 (45)图4.2 采⽤iSIGHT完成的系统集成 (47)图5.1 飞翼⽆⼈机外形及舵⾯布置⽰意图 (48)图5.2 飞控系统的Simulink仿真模型 (49)图5.3 增稳后的扰动运动模态曲线（纵向） (50)图5.4 飞⾏⾼度变化仿真曲线 (50)图5.5 飞机俯仰⾓变化仿真曲线 (51)图5.6 放宽静稳定度对超声速运输机构型的影响 (52)图5.7 飞机优化设计前的基本构型 (53)操稳特性快速评估及其在飞机设计中的应⽤图5.8 优化后的平尾平⾯形状与初始设计形状的对⽐ (55)表2.1 DATCOM中常⽤参数表及控制参数的功能 (11)表2.2 常⽤的DATCOM计算输出结果 (14)表2.3 DATCOM+输出⽂件的类型及其功能 (16)表2.4 DATCOM+中各程序的功能 (16)表3.1 常⽤飞控系统的复杂度分类 (34)表3.2 FlightGear中常⽤的配置参数 (42)表4.1 操稳特性快速评估系统中的各部分程序 (44)表4.2 MATLAB引擎与VC的接⼝函数 (46)表5.1 ⽆⼈机本体扰动运动模态（纵向） (48)表5.2 增稳后的全机扰动运动模态（纵向） (49)表5.3 飞机主要的总体设计参数 (53)表5.4 设计变量、约束、⽬标及其优化结果 (54)表5.5 优化后主要外形特征参数及升阻特性的变化 (55)南京航空航天⼤学硕⼠学位论⽂注释表A展弦⽐ H , h ⾼度 a声速；主轴⽅位⾓ I 转动惯量 b展长 K n 纵向静稳定裕度 c A平均⽓动弦长 L M N 总⼒矩在机体轴系上的分量 A C轴向⼒系数 A A A L M N ⽓动⼒矩在机体轴系上的分量D C阻⼒系数 M a 飞⾏马赫数 L C升⼒系数 P 发动机推⼒或拉⼒ m C俯仰⼒矩系数 Q 动压，0.5ρV 2 N C法向⼒系数 p q r 滚转，俯仰，偏航⾓速度 L C α升⼒系数对攻⾓的导数 T 发动机作⽤⼒在机体x 轴的分量m C α俯仰静稳定性导数 u v w 空速在机体轴上的分量 Y C β侧⼒系数对侧滑⾓的导数 V 空速 n C β偏航静稳定性导数 W 飞机重量 l C β滚转静稳定性导数 cp x 压⼒中⼼的相对位置 Lq C升⼒系数对俯仰⾓速度的导数 ac x 焦点（⽓动中⼼）的相对位置mq C 俯仰⼒矩系数对俯仰⾓速度的导数cg x 重⼼的相对位置L C α升⼒系数对攻⾓变化率的导数 Y 侧⼒ m C α俯仰⼒矩系数对攻⾓变化率的导数α攻⾓（迎⾓） Yp C侧⼒系数对滚转⾓速度的导数β铡滑⾓ lp C滚转阻尼导数 e δ升降舵偏转⾓ lr C滚转交叉导数 f δ襟翼偏转⾓ np C航向交叉导数 a δ副翼偏转⾓ nr C航向阻尼导数φ滚转⾓ C P螺旋桨拉⼒系数θ俯仰⾓ D阻⼒ψ偏航⾓ F x F y F z空⽓动⼒在机体轴系上的分⼒ρ空⽓密度 G重⼒ξ阻尼⽐ g重⼒加速度 n ω固有频率承诺书本⼈声明所呈交的硕⼠学位论⽂是本⼈在导师指导下进⾏的研究⼯作及取得的研究成果。
外文文献翻译：原文+译文文献出处：Gunasekaran A. The study of purchasing process management information system[J]. European Journal of Operational Research, 2016, 1(2): 29-45.原文The study of purchasing process management information systemGunasekaran AAbstractThe definition of process management is a set of creating value for customers, interrelated, the activities of the relationship between input and output combinations. Specification of the operation of the enterprise's process is on the business can be constantly summarize and curing excellent experience in business operations. Made the enterprise work efficiency is higher, the cost is lower, create a product or service quality more good, to gain more customer satisfaction. Combining with the latest process management theory and ideas, design the procurement process architecture and process management system, the application of industry in the process of benchmarking management tool, to solidify and auxiliary process management, process management and IT tools were discussed by the process of ascension in helping the business management level, and at the same time comparative analysis of several commonly used industry process management tools, for the enterprise procurement process change management to provide reference and basis. Keywords: purchasing management; process management; the information system1 IntroductionTo customer demand as the guidance of the build process type organization; Carding the business process in a process management way, according to the hierarchical classification process management ideas, building the business process monitoring management, performance evaluation and the closed loop management system, internal control auditing protect health of the whole process running, high efficiency and low cost. In guaranteed, on the basis of organization and Process system, and then based on the integration (P2A, Process To the Application ProcessTo the Application system), the mechanism of using ARIS tools To build information integrated Process management platform, To effectively support the business Process of efficient operation and management, at the same time can adapt To the continuous improvement of business Process optimization and IT system on the synergy. In the real sense, the research significance of this article embodied in two aspects: first, the implementation of business process management for the enterprise to provide reference and guidance for the construction of process management system. From the process management system, rules, the design, the pilot, push, monitoring, evaluation and improvement of the whole life cycle process management and process management and department in the process of the organization and management of the combination of introduction and discussion, lets the enterprise process management ideas and methods to get more detailed understanding. At the same time, the analysis of the problems in the process of the whole also can make the enterprise before the implementation of process management ready to more fully. Second, in the process management system to provide the reference and guidance on building, enterprise to obtain competitive process to adapt to the market, the integration of supply chain enterprise, enterprise's process must continuous improvement. In this way, the process becomes dynamic, across the organization, and full of flexibility. Business process management system using the unified visual process modeling tool to business process of abstracting, can the enterprise complicated business flow synchronization application logic separation, improves the flow process of flexibility. Such rapid construction of system, can satisfy the continuous improvement of the enterprise, and thus gain a competitive edge for the enterprise. Thirdly, based on the research of the purchasing process management and architecture, for domestic enterprises how to establish purchasing management system and purchasing process architecture construction for reference, at the same time from the research problems in the process of discussion, help enterprise thinking and avoid problems in the process of building, better service to the development of the enterprise.2 Literature reviewIf there is a suitable leader and the appropriate business management tools, and aclear transparent industry value chain model, and the model is able to adjust for differences in some condition, the enterprise can be based on the global market competition. Enterprises in the implementation of its operating strategy, need to set up a flexible business process, at the same time also need to have a can turn process optimization and process innovation management concept in the information system. The best business solutions are the process configuration to the application system, build process-oriented, suitable organization, and reduce the interface between business processes and the external. Business process management covers the entire value chain of enterprise management, is across organizational process management and control of a concept and method. Business Process Management (BPM, Business Process Management) which is very wide, from the Process analysis and optimization, transformation, and to such as software system implementation, so that the Process of automatic determination of the execution, control and evaluation index, all belong to the category of BPM. Business flow management is, as it were, a cycle management, only to keep the loop unobstructed, can make the enterprise constantly adapt to market changes. Can be found through the analysis of BPM and make full use of enterprise and the customer (and the customer's customer) and suppliers (and suppliers) of processes across the enterprise boundaries between the potential to save cost and increase efficiency. Business process management is to improve the foundation of enterprise competitiveness and ability to innovate, it directly affects the production process, product innovation, quickly enter the market, etc.), manufacture and service process (customer orientation, marginal benefit/profit, quality, etc.), the support process (less management costs, higher employee satisfaction, etc.) as well as the management and control process (change management, strategy, etc.).From the perspective of process management concept is: there is thought that create customer value and enterprise, the enterprise process is creating value for customers, excellent business process, can achieve the success of the enterprise, excellent process operation depends on excellent process management. To sum up, the meaning of the process management of enterprise, is to simplify and improve the company's business process system, and make it more agile response to customer demand, expand theroutine management, reduce the exception management, improve efficiency, plugging loopholes, enables the enterprise to gradually into the operational excellence.Earlier with the gradually development of IT technology, EAI (enterprise application integration), process modeling, process optimization, workflow technology evolved into the business process management. Business process management technology lies in process modeling, process analysis, process simulation to the development of three aspects. Around the business process analysis technology as the center of the research, the main representative is: a person thousands of activity-based cost analysis method in the activities of business process reengineering model and analysis of the application of auxiliary activities, as well as li-hua huang from the change of management thinking, based on the process of information technology and optimization based on analysis of information flow in the process of hyper graph three aspects proposed the six rules of process optimization. Around the business process modeling technology and the development of a lot of modeling technology.Business process management system is usually refers to the technical implementation of business process management, business process management system for process management of enterprise value choice must implement the following functions: independently managing resources, information and the ability to process, and can realize the function allocation and combination; Have the ability to measure the impact of business changes and process management system can provide better than before the change of business process information; The ability to achieve business rules and business change quickly, process management system can help business managers to react more quickly to deal with the business of change; Keep the consistency of the integrity of business processes and information; So the process is not simply a technical management system, through the process management system components research, we can more clearly understand.3 Purchasing management researchProcurement refers to the way in addition to the purchase to obtain goods or services, also can make items in the following ways to use, to achieve the goal ofmeet the demand. General procurement main ways are: exchange, lending and leasing. The function of purchasing in large enterprises is not only purchasing department internal or buyer; also composition is an important part of enterprise supply chain as a whole. Procurement in accompanied by the emergence and spread of the integration of logistics management, its operation process itself is no longer a simple transaction oriented, its core concept should be extended to the supplier to build strategic alliances, sharing information system effectively, and promote global local procurement.Basic types of procurement organization consisted of a mixed type purchasing organization and cross-functional procurement team, centralized purchasing organization, dispersible purchasing organization. Hybrid purchasing organization, in some of the major manufacturing enterprises, at the company level there is a company purchasing department, however, independent business unit of strategic and tactical purchasing activities. Cross-functional team is a relatively new form of organization, purchasing a new purchasing organization USES a single point of contact with suppliers (commodity group), is provided by the team for the entire organization on the requirements of the integration of all components. Centralized procurement organization structure, the company has a central purchasing department level, including work mainly include: the company's procurement experts in strategic and tactical level of operation; Product specifications set centrally; Supplier selection strategy; to prepare and negotiate the contract with suppliers. Dispersible purchasing organization is an important characteristic of each business unit responsible persons responsible for own financial consequences.With the rapid development of information technology, IT, and changes in today's global economy, the status of purchasing management in the business management activity is more and more important. Simple is summarized and the trend of the development of procurement management, procurement will be from simple pure trading buy and sell to "reasonable purchasing" development, namely choosing the right product, at the right price, in the right time, with the right quality and through the appropriate suppliers. Enterprise organization structure also gradually to thedevelopment of functional and ability, which can accurately locate on the market value of procurement, to obtain products and services effectively. Mainly embodies in the following three trends: one, strategic sourcing, and supplier to build strategic alliance, a long-term common development with technology, diverse and complex market environment in the future for enterprises in the first stand firm foundation, can have exceptional value. Second, e-commerce, purchasing will increasingly rely on the Internet, rapid, real-time access to various global suppliers, market information such as price, quantity, at the same time also can get more purchasing data, tries to study the value of the data itself, on the purchasing management also has far-reaching significance. Three, strategic cost management: as technology, equipment and other fields gradually reduce, reduce the cost of space and procurement in the position to reduce the cost of the also more and more obvious, and the enterprise's strategic cost management, the entire supply chain of each link only together to reduce costs, to achieve a "win-win".译文采购流程管理信息系统研究Gunasekaran A摘要流程管理的定义就是一组为客户创造价值的，有相互关联的、有输入输出关系的活动组合。
随着时代的飞速发展，高度自主化的机器人在人类社会中的地位与作用越来越大。
而机械臂作为机器人的一个最主要操作部件，其运动规划问题，例如准确抓取物体，在运动中躲避障碍物等，是现在研究的热点，对其运动规划的不断深入研究是非常必要的。
机械臂的运动规划主要在高维空间中进行。
RRT （Rapidly-exploring Random Tree）算法[1]基于随机采样的规划方式，无需对构型空间的障碍物进行精确描述，同时不需要预处理，因此在高维空间被广为使用。
近些年人们对于RRT算法的研究很多，2000年Kuffner等提出RRT-connect算法[2]，通过在起点与终点同时生成两棵随机树，加快了算法的收敛速度，但存在搜索路径步长较长的情况。
2002年Bruce等提出了ERRT（Extend RRT）算法[3]。
2006年Ferguson等提出DRRT （Dynamic RRT）算法[4]。
2011年Karaman和Frazzoli提出改进的RRT*算法[5]，在继承传统RRT算法概率完备性的基础上，同时具备了渐进最优性，保证路径较优，但是会增加搜索时间。
2012年Islam等提出快速收敛的RRT*-smart算法[6]，利用智能采样和路径优化来迫近最优解，但是路径采样点较少，使得路径棱角较大，不利于实际运用。
2013年Jordan等通过将RRT*算法进行双向搜索，提出B-RRT*算法[7]，加快了搜索速度。
同年Salzman等提出在下界树LBT-RRT中连续插值的渐进优化算法[8]。
2015年Qureshi等提出在B-RRT*算法中插入智能函数提高搜索速度的IB-RRT*算法[9]。
同年Klemm等结合RRT*的渐进最优和RRT-connect的双向搜基于改进的RRT*-connect算法机械臂路径规划刘建宇，范平清上海工程技术大学机械与汽车工程学院，上海201620摘要：基于双向渐进最优的RRT*-connect算法，对高维的机械臂运动规划进行分析，从而使规划过程中的搜索路径更短，效率更高。
本科毕业论文（设计）题目：五子棋博弈系统研究以及单机版网络版的实现姓名：方杰学号：***********院（系）：信息工程学院专业：软件工程系指导教师：杨林权职称：副教授评阅人：职称：2011 年6 月学位论文原创性声明本人郑重声明：所呈交的论文是本人在导师的指导下独立进行研究所取得的研究成果。
除了文中特别加以标注引用的内容外，本论文不包含任何其他个人或集体已经发表或撰写的成果作品。
本人完全意识到本声明的法律后果由本人承担。
作者签名：年月日学位论文版权使用授权书本学位论文作者完全了解学校有关保障、使用学位论文的规定，同意学校保留并向有关学位论文管理部门或机构送交论文的复印件和电子版，允许论文被查阅和借阅。
本人授权省级优秀学士学位论文评选机构将本学位论文的全部或部分内容编入有关数据库进行检索，可以采用影印、缩印或扫描等复制手段保存和汇编本学位论文。
本学位论文属于1．保密□，在_________年解密后适用本授权书。
2．不保密□√。
（请在以上相应方框内打“√”）作者签名：年月日导师签名：年月日摘要人工智能是近年来很活跃的研究领域之一。
计算机博弈是人工智能研究的一个重要分支，它的研究为人工智能带来了很多重要的方法和理论，产生了广泛的社会影响和学术影响。
国内外对博弈的研究已经较为广泛，特别是IBM的国际象棋程序“深蓝”，已经达到了人类的世界冠军水平。
“深蓝”的研究成果，特别是基于剪枝的极大极小树搜索技术为设计其它的计算机棋类博奕系统提供了良好的参照。
但是不同的棋类博奕，其规则的千差万别赋予了每一种棋类博奕特殊的专业知识。
这就必然要求设计一个具体的棋类博奕系统时应该深入研究它的基本原理和内在规律。
随着网络的发展，简单的单机版已满足不了人们的需要，将单机版晋升成为网络版已是必然的。
C++语言是一种面向对象语言，尽管在当前，可视化语言发展迅速，晋级很快，但c++语言作为一种基础的语言，它还是有它的存在价值，甚至有时它是不可替代的，特别是在和硬件接口技术相联系的软件方面。
第 51 卷 第 5 期石 油 钻 探 技 术Vol. 51 No.5 2023 年 9 月PETROLEUM DRILLING TECHNIQUES Sep., 2023doi:10.11911/syztjs.2023087引用格式：曾凡辉，胡大淦，张宇，等. 数据驱动的页岩油水平井压裂施工参数智能优化研究[J]. 石油钻探技术，2023, 51（5）：78-87.ZENG Fanhui, HU Dagan, ZHANG Yu, et al. Research on data-driven intelligent optimization of fracturing treatment parameters for shale oil horizontal wells [J]. Petroleum Drilling Techniques，2023, 51(5)：78-87.数据驱动的页岩油水平井压裂施工参数智能优化研究曾凡辉1， 胡大淦1， 张 宇1， 郭建春1， 田福春2， 郑彬涛3（1. 油气藏地质及开发工程全国重点实验室（西南石油大学）， 四川成都 610500；2. 中国石油大港油田分公司石油工程研究院，天津 300280；3. 中国石化胜利油田分公司石油工程技术研究院，山东东营 257000）摘 要: 针对目前数智化压裂施工参数设计针对性不足、流程不畅通等问题，建立了基于数据驱动的压裂施工参数智能优化方法。
以CD区块32口页岩油井为研究对象，采用主成分分析法处理代表储层地质特征、工程品质和施工参数的15项产量影响因素，使之降低维度，引入高斯隶属函数和熵权法进行储层压裂非均质性模糊综合评价，结合支持向量回归和粒子群优化算法，以产量最高为目标，推荐射孔位置、段长、簇间距、单位长度液量、单位长度砂量和排量。
研究结果表明，渗透率、孔隙度、热解游离烃含量、单位长度液量和单位长度砂量为研究区块的产量主控因素。
未来科技的对话英文作文The Dialogue of Future TechnologyIn the not-so-distant future, the relationship between humans and technology will evolve in profound and unprecedented ways. As artificial intelligence (AI) and other cutting-edge innovations continue to advance at a rapid pace, the lines between the digital and physical worlds will blur, giving rise to a new era of human-machine interaction. This dialogue between humanity and our technological creations will be marked by increased interdependence, seamless integration, and perhaps even a degree of symbiosis.One key aspect of this future dialogue will be the integration of AI into nearly every facet of our lives. Intelligent digital assistants, already present in the form of Siri, Alexa, and Google Assistant, will become increasingly sophisticated, capable of understanding natural language, anticipating our needs, and providing personalized guidance and support. These AI companions will serve as our digital concierges, scheduling appointments, managing our calendars, and even offering emotional support and companionship.Moreover, AI will be deeply embedded within the smart devices and connected systems that comprise the Internet of Things (IoT). Our homes, workplaces, and even our modes of transportation will be imbued with AI-powered automation and optimization, adjusting to our preferences and patterns to enhance our comfort, efficiency, and productivity. Imagine a future where your smart home can automatically adjust the temperature, lighting, and music based on your presence and activity, or where your self-driving car can navigate you to your destination while you use the commute time to work or relax.As AI becomes more advanced, it will also play a crucial role in fields such as healthcare, scientific research, and decision-making. Intelligent diagnostic systems will be able to analyze medical data, identify patterns, and provide early detection of diseases, while AI-powered research tools will accelerate the pace of scientific discovery by rapidly processing vast amounts of information and generating novel hypotheses. In the realm of decision-making, AI-powered algorithms will assist policymakers, business leaders, and individuals in making more informed and data-driven choices, drawing upon a wealth of data and predictive modeling capabilities.However, the dialogue between humans and technology will not be without its challenges. The increasing reliance on AI and automationraises important questions about privacy, security, and the potential displacement of human labor. As our personal data becomes more closely intertwined with these intelligent systems, we must grapple with issues of data protection, algorithm transparency, and the ethical use of technology. Moreover, the displacement of certain jobs by AI-powered automation will require a reevaluation of educational systems, job training, and social safety nets to ensure a smooth transition and prevent widespread economic disruption.Additionally, the integration of AI and other emerging technologies, such as brain-computer interfaces and genetic engineering, will raise profound questions about the nature of human identity, consciousness, and our relationship with technology. As we venture into the realm of human enhancement and the blurring of the line between biological and artificial intelligence, we will be confronted with complex moral and philosophical dilemmas that challenge our traditional conceptions of what it means to be human.Despite these challenges, the dialogue between humans and technology holds immense potential to improve our lives and unlock new realms of human potential. By embracing the symbiotic relationship between humanity and our technological creations, we can harness the power of AI and other innovations to tackle some of the world's most pressing challenges, from climate change and disease to poverty and social inequality. Through this dialogue, wemay even uncover new ways of understanding ourselves, our place in the universe, and the very nature of consciousness.As we look to the future, it is clear that the relationship between humans and technology will be one of the defining narratives of the 21st century. The dialogue that unfolds will shape the course of human civilization, redefining the way we live, work, and interact with the world around us. By navigating this uncharted territory with foresight, empathy, and a commitment to ethical and responsible innovation, we can ensure that the dialogue between humans and technology leads to a future that is both prosperous and profoundly meaningful.。
Rapid evaluation and optimization of machine tools with position-dependent stabilityMohit Law,Yusuf Altintas n ,A.Srikantha PhaniDepartment of Mechanical Engineering,The University of British Columbia,2054-6250Applied Science Lane,Vancouver,BC,Canada,V6T 1Z4a r t i c l e i n f oArticle history:Received 14November 2012Received in revised form 4February 2013Accepted 6February 2013Available online 21February 2013Keywords:Machine tool Model reduction StabilityOptimizationa b s t r a c tMachine tool’s productivity is a function of the dynamic response between the spindle nose and table,which varies as a function of drive positions within the machine work volume.The position-dependent structural dynamics results in varying stability of the machine.This paper presents a computationally efficient methodology to evaluate and improve dynamic performance of a machine tool at the design stage.An efficient position-dependent multibody dynamic model of a machine tool is developed based on reduced model substructural synthesis.The experimentally validated reduced machine model simulates position-dependent behavior with significantly less computational effort than commonly used full order Finite Element models.The proposed modeling strategy is used to identify weak components of an experimental machine,which limit the productivity due to chatter.The identified weak machine component is modified and the complete dynamics are rapidly analyzed by virtually re-assembling the machine using reduced order models.Optimal design modifications are shown to increase productivity by $25%.The proposed method can be used for efficient simulation of structural dynamics,stability assessment as well as interactions of the CNC and cutting process with the machine tool structure in a virtual environment.&2013Elsevier Ltd.All rights reserved.1.IntroductionThe present machine tool industry requires rapid analysis of structural dynamics,machine-cutting process and controller interac-tions in a virtual environment.The process–machine interactions are influenced by the dynamic stiffness between the tool center point (TCP)and workpiece which determines the maximum depth of cut due to chatter constraints.The dynamic stiffness and the maximum stable depth of cut are further influenced by the changing structural dynamics of the machine as the tool moves along the tool path in the machine work volume.The position-varying structural vibrations between the TCP and servo drives further limit the positioning speed and accuracy.The objective of the machine tool designer is to maximize the dynamic stiffness between TCP —workpiece and TCP —drive motors while keeping the machine mass light for high speed positioning and high-productivity machining.Evaluation and optimization of machine tool’s performance to deliver improved dynamic behavior over the whole work volume necessitates rapid assessment of several design alternatives in the virtual environment before eventual physical prototyping [1].Typically,machine tool design and response analyses are carried out in virtual environments using a finite element (FE)model of the machine.These models are efficient for subsystem level design analyses such as modeling of ball-screw feed-drive systems [2],and spindles [3,4].FE models for full machine analyses have also proven to be useful for structural modification based on process–machine interactions [5–7],as well as for modeling control structure interactions [8].However,response analyses of full machine models which are typically on the order of 1,000,000degrees of freedom (DOFs)or more,is computation-ally costly,and can take up significant portion of the total computational effort required for the design and analyses of machine tools [9].Moreover,modeling position-dependency in such large order FE models requires cumbersome adaptive/re-meshing strategies,making it time consuming in practice.Position-dependency has been modeled using co-simulation techniques in which FE solvers are coupled to multibody simula-tion codes [10–12].These methods have been shown to be effective for rigid-flexible body motion analyses and are less suited to the flexible bodies of a machine undergoing relative motion.Changing structural dynamics have also been modeled with simplifying assumptions of modeling the multiple flexible bodies in contact as rigid-flexible [13,14].Position-dependency for multiple flexible bodies in contact was also modeled in [15]for a simplified single set nodal compatibility condition.Increas-ing modularity in the machine tool development process often requires substructures to be designed and modeled separately,resulting in different mesh resolutions at the contacting interface.Contents lists available at SciVerse ScienceDirectjournal homepage:/locate/ijmactoolInternational Journal of Machine Tools &Manufacture0890-6955/$-see front matter &2013Elsevier Ltd.All rights reserved./10.1016/j.ijmachtools.2013.02.003nCorresponding author.Tel.:þ16048225622;fax:þ16048222403.E-mail addresses:altintas@mech.ubc.ca,altintas@mail.ubc.ca (Y.Altintas).International Journal of Machine Tools &Manufacture 68(2013)81–90Ensuring mesh compatibility during synthesis for such models which are simultaneously in contact over multiple nodes is difficult to guarantee,and if this condition was not necessary —modeling time could be halved,as estimated in [9].To facilitate rapid assessment of design alternatives,this paper offers an integrated virtual approach based on a computationally efficient position-dependent multibody dynamic model of the machine.A reduced model substructural synthesis approach proposed earlier by the authors in [16]is extended here to model the entire machine tool system.Position-dependency is modeled by synthesizing reduced substructures whose response is position-independent.This approach allows for modularity in the design process by tolerating mesh incompatibility at substructural interfaces by having displacement compatibility between adjacent substructures through sets of algebraic constraint equations.Position-dependent dynamic behavior of a representative three-axis vertical machining center —FADAL 2216is modeled and assessed using the generalized integrated virtual design scheme shown in Fig.1.Position and feed-direction-dependent-process-stability is proposed as a performance criterion to evaluate the productivity of the machine tool in its entire work volume.This criterion also identifies parameters limiting the target productivity levels.Improve-ments to meet design targets,if necessary,are made possible with the iterative approach as shown in Fig.1.The paper is organized as follows:a generalized substructural synthesis formulation to model position-dependency is presented in Section 2.The virtual machine model is compared with experimental results in Section 3.For a defined set of cutting conditions and productivity levels,the position and feed-direction-dependent stabi-lity is evaluated in Section 4.Mechanical parameters limiting productivity are identified and modified to meet target productivity in Section 5;followed by conclusions in Section 6.2.Generalized substructural formulation for position-dependencyPosition-dependency at the TCP is modeled based on a twostage substructural assembly approach.At first,each of the majorsubstructures of the machine under consideration,namely:spindle–spindle-housing,column,base,cross-slide,and table are reduced independently and synthesized subsequently together with the three individual feed drive models as shown in Fig.2.A second stage substructural assembly involves coupling the tool–tool-holder response (Substructure I)to the response obtained at the spindle nose from the first stage (Substructure II)using a receptance coupling approach.This allows different tool–tool-holder combinations to be modeled independently and coupled subsequently without having to regenerate machine (spindle)models [17].FE models for the structural substructures are generated from their respective CAD models using 10noded solid tetrahedron elements with material properties assigned as:modulus of elasticity of 89GPa;density of 7250kg/m 3;and,Poisson’s ratio of 0.25.The spindle,three ball-screw drive models and the tool–tool-holder are modeled with Timoshenko beam elements.The spindle assembly including the spindle shaft,cartridge,bearings,drive pulley,and other accessories such as nuts and rotary couplings are modeled based on work reported by Cao and Altintas [3].This spindle assembly is integrated as a separate substructure coupled rigidly to the spindle housing.Each of the substructural models is exported,after convergence checks,to the MATLAB environment for reduction and synthesis.2.1.Substructure model reductionEach of the main substructural machine components is reduced independently as discussed here.For any substructure under consideration,the undamped equations of motion are represented as:M €uþKu ¼f ð1Þwhere {M ,K }F ÂF are the mass and stiffness matrices for the total DOFs F ;and f F Â1is the force vector.Reduced substructures are synthesized at their interface DOFs.Hence the displacement vector u is partitioned into the DOFs to be retained,i.e.the exterior/interface (E )DOFs,u E ,that are in physical contact with the other substructure(s),and,the DOFs which are to be elimi-nated,i.e.the interior (I )DOFs,u I .Develop machine concept(CAD model)Define requirementsSubstructurally synthesized reduced order FE modelSimulate TCP FRF Identify modes limiting productivityDesign improvement: Modify (stiffen) substructures Productivity goals fulfilled?Detailed machine designUpdate position within work-volumeYesDefine representative milling operation and productivity goals formachineSimulate process-stabilityPosition-dependency checked?NoYesCheck feasibility to reduce massDesign improvement: Topology optimization Not feasibleFeasibleNoFig.1.Flow chart of a virtual integrated position-dependent process–machine interaction approach for designing milling machines ensuring targeted productivity.w et al./International Journal of Machine Tools &Manufacture 68(2013)81–9082The displacement vector u is expressed in terms of a reduced set u R R Â1by a transformation matrix T F ÂR (R 5F )as:u F Â1¼T F ÂR u R R Â1ð2ÞwhereT ¼I EE 0EPT ICMS IE U IP"#ð3Þis a modified form of the standard component mode synthesis (CMS)transformation matrix as previously presented by the authors in [16].I EE is a unit matrix,and,T ICMS is an improvement over the equivalent quasi-static transformation obtained by including the inertial terms [16].The effect of the interior DOFs is modeled by complementing the reduced set of exterior DOFs by a set of generalized (modal)coordinates represented by selecting modes belonging to the interior DOFs.This is done by selecting a subset of significant P modes,i.e.U IP D U II in Eq.(3);wherein the eigenvector U II is obtained by solving the eigenvalue problem corresponding to the interior DOFs.The selected mode sets are able to represent higher order dynamics of the substructure while keeping the order of the reduced model to a minimum by spanning a much widerfrequency range with fewer modes than would be required with standard CMS methods.For further details on reduction proce-dure and significant mode selection criteria,see [16].Using the transformation matrix of Eq.(3),the reduced substructural matrices {M R ,K R }R ÂR are represented as:M R ¼T T MT ;K R ¼T T KT ;f R ¼T T f :ð4ÞThe size of each of the reduced models which consists of the exterior/interface DOFs u E ,complemented by a set of P significant modes is listed in Table 1.2.2.Substructural synthesisEach of the reduced substructures are synthesized at the contacting interfaces as shown in detail ‘‘A’’in Fig.2,where the substructures S (1)and S (2)may represent any of the machine tool substructural components with/without relative motion between them.Each substructure has been modeled separately resulting in different mesh resolutions at their contacting interfaces.To synthesize such incompatible substructures,an approximate model of surface interaction is obtained by defining a virtual condensation node placed at the center of each of the interface surfaces in contact,as shown in the detail ‘‘A’’in Fig.2.Moving interfaces ColumnSpindle Hsg. + SpindleZ feed drivedisplacement compatibilityMachine BaseCross-slideTableY feed driveMoving interfacesFixed interfacesSpindle noseSubstructure IISubstructure IAuuS uuS uurΔX feed drive 2Fig.2.Two stage substructural synthesis of the machine tool.Tool–tool-holder response (Substructure I)is coupled to the position-dependent response of the synthesized substructural machine model (Substructure II).Table 1Division of DOFs for individual substructures.Spindle–spindle housingColumn Base Cross-slide Table Total Reduced modelFull order model 21,99245,98397,21450,20115,623231,013Interface DOFs (u E )10371533194525981908Significant component modes (P )27284627249173w et al./International Journal of Machine Tools &Manufacture 68(2013)81–9083Displacement compatibility between the virtual nodes for substructures in rigid contact is represented as [16]:u ð1ÞC Àu ð2ÞC ¼0ð5ÞThe displacement of the virtual node u ðn ÞCfor substructure n (n ¼1,2)consisting of a set of translational and rotational DOFs u C *x C ,y C ,z C ,y x C ,y y C ,y z C ÀÁis linked to the interface DOFs with a displacement operator such that:u ðn ÞC ¼C ðn Þuðn Þð6ÞCoefficients of the displacement operator C (n )are extracted by linking the DOFs of the condensation node to the interface nodal DOFs it is meant to represent using a multipoint constraint (MPC)equation formulation.The MPCs represent the DOFs of the condensation node as a linear combination of all the DOFs of the nodes in contact at the interface.The motion of the condensation node,i.e.displacements (u C )and rotations (a C )are fully described as the weighted average of the motion of the interface nodes in contact as [18]:u C ¼P mk ¼1w E k u E kP mk ¼1w E k a C ¼P mk ¼1w E k r C k Âu E kP mk ¼1w E k 9r C k 9ð7Þwhere w E k represents the weight factors for each of the exteriorDOFs corresponding to the node k in contact at a particular position.r C k is the vector from the condensation node to the node corresponding to the interface node k.Eq.(7)results in a set of m constraint equations.The set of nodal DOFs for each node k for volumetric elements in this work is:u *x k ,y k ,z k .To ensure that the condensation node represents the average motion of the contacting interface surface it is meant to represent,the weight factors w E k for each DOF are chosen proportional to the part of the interface surface its node represents;and are assigned as the coordinates of the nodes being coupled in this study.For addi-tional details on selection of weight factors,see [16].Synthesizing each substructural interface with a set of m constraint equations,the undamped equation of motion for the synthesized reduced model at a particular position is:M R S 000000M R C000000M R B000000M R CS 000000M R T 000000026666666643777777775€u R S €u R C €u R B €u R CS €u R T €k8>>>>>>>>><>>>>>>>>>:9>>>>>>>>>=>>>>>>>>>;þK R S 0000C T 0K R C 000C T 00K R B 00C T 000K R CS 0C T 0000K R T C T C C C C C 0266666666664377777777775u R S u R C u R Bu R CS u R T k 8>>>>>>>><>>>>>>>>:9>>>>>>>>=>>>>>>>>;¼f R S f R C f R B f R CS f R T 08>>>>>>>><>>>>>>>>:9>>>>>>>>=>>>>>>>>;ð8Þwhere {M R ,K R }are the reduced substructural mass and stiffnessmatrices and the subscripts S ,C ,B ,CS ,and T correspond to the spindle–spindle housing combine,column,base,cross-slide,and table respectively.C ðu R S ,u R C ,u R B ,u R CS ,u R T Þis the displacement operator coupling these substructures,whose coefficients are obtained from Eqs.(5)–(7).k represents a discrete set of m Lagrange multipliers corresponding to the number of constraint equations [19].The synthesized model enables prediction of dynamic response at the spindle nose as one component changes its position relative to another by solving the eigenvalue problem form of Eq.(8),byvarying the tool position in the work volume by adjusting D X ,D Y ,or D Z (see Fig.2).For each new position,since a new set of nodes come into contact,while others fall out of contact,the displacement operator in Eq.(6)is updated by instantaneously coupling/de-coupling the corresponding nodes on the interfaces.Since the substructurally synthesized reduced machine model size is $1/25th the size of the full model (Table 1),it allows for very efficient position-dependent dynamic modeling of the machine.Further modularity in the design process is facilitated by synthesizing the frequency response function (FRF)at the spindle nose with that of separately modeled tool–tool-holder model response as discussed in the next section.2.2.1.Tool point FRF with receptance couplingSince both the spindle and the tool–tool-holder are modeled with Timoshenko beam elements,each of the coupling nodes has six DOFs,three translational and three rotational;for which the respective component receptances in compact matrix form are:u x u y u zy x y y y z8>>>><>>>>:9>>>>=>>>>;i¼h xx h xy h xz l xm x l xm y l xm z h yxh yy h yz l ym x l ym y l ym z h zx h zyh zz l zm x l zm y l zm z t y x fxt y x f y t y x f z p y x m x p y x m y p y x m z t y y fxt y y f y t y y f z p y y m x p y y m y p y y m z t y z f xt y z f yt y z f zp y z m xp y z m yp y z m z26666643777775ijf x f y f z m x m y m z8>>>><>>>>:9>>>>=>>>>;jð9Þwhere h represents the displacement-to-force receptance;l representsthe displacement-to-couple receptance;t represents the rotations-to-force receptance;and p represents the rotation-to-couple receptance;and i and j are the respective measurement and excitation locations.Eq.(9)may be rewritten in its generalized form as:x i ¼R ij q jð10Þwhere R ij is the generalized receptance matrix that describes both translational and rotational component behavior,and,x i and q j are the corresponding generalized displacement/rotation and force/couple vectors.The direct receptances at,and cross receptances between the free-end (i.e.location 1in Fig.2)and the coupling end (location 10in Fig.2)of the tool–tool-holder model are:x 1¼R 11q 1ð11Þx 10¼R 1010q 10ð12Þx 1¼R 110q 10ð13ÞDue to symmetry and Maxwell’s reciprocity,R 110¼R 101.The direct receptances at the spindle nose (location 2in Fig.2)are similarly represented as:x 2¼R 22q 2ð14ÞEnsuring displacement compatibility,i.e.x 10Àx 2¼0,and force equilibrium,i.e.q 10þq 2¼0,at the rigid connection between the tool–tool-holder and the spindle nose,the assembled receptances,G 11,at the TCP in the generalized form are given as [20]:G 11¼R 11ÀR 110ðR 1010þR 22ÞÀ1R 101ð15Þwhere G 11has the same structure as R ij ,and its constituent receptances are obtained from solutions in Eqs.(11)–(14).The two stage substructural synthesis methodology yields the position-dependent dynamic response at the TCP.A face-mill cutter of 50mm diameter with an overhang of 70mm from the spindle nose with a CAT40type tool-holder is modeled and its response is obtained from Eqs.(11)–(13).Response at the spindle nose obtained from Eqs.(8)and (14)is synthesized with the tool–tool-holder response,which leads to the response at the free-end of the tool using Eq.(15).w et al./International Journal of Machine Tools &Manufacture 68(2013)81–90842.3.Position-dependent dynamic response at TCP for machine model TCP FRFs are simulated with the substructurally synthesized reduced order machine model at three different positions of the machine.The top position is the configuration shown in Fig.2;the mid and bottom positions are when the tool has moved in the Z -direction by an amount of À0.2m;and À0.4m respectively.A uniform damping ratio of z ¼0.02has been assumed in simulating FRFs,which is updated by correlating with experimental modal damping from measurements on a similar available machine as presented in the following section.The stability of the machining process is primarily influenced by the low-frequency structural modes [6]such as column,table,spindle housing,and the spindle shaft;hence X and Y directional direct TCP FRFs are compared up until 350Hz in Fig.3.Higher frequency tool and tool-holder modes are usually not related to the design of the machine tool structure since they are machining application specific and are more local in nature,i.e.they do not exhibit strong position-dependency.As evident in Fig.3,the global modes corresponding to the column (30–100Hz)exhibit stronger position-dependency as compared to the spindle housing modes (100–350Hz).The low-frequency column mode in the X direction at $40Hz varies by $50%in dynamic stiffness,whereas the second X directional column bending mode at $70Hz varies by up to 8%in natural frequencies.The dominant low-frequency column mode in the Y direction at $45Hz varies by as much as 15%in natural frequencies and $135%in dynamic stiffness over the full Z stroke of the machine.Dominant low-frequency mode shapes for the reduced model when the headstock is at the top position are shown in Fig.4.The full model shown in Fig.4(a)is represented by only the interface DOFs in Fig.4(b).Mode shapes are shown by overlaying the deformed configuration over the un-deformed configuration.The first mode at $41Hz corresponds to a global column bending mode in the YZ plane,in Fig.4(c),while the second dominantmode at $42Hz corresponds to the global column bending mode in the XZ plane,in Fig.4(d).Position-dependent results for the dominant low frequency modes obtained with the reduced model are compared with the full order model results in Table 2,which also includes error estimates rounded off to the nearest integer.Full order results are obtained by ‘gluing’substructures at the interfaces in the ANSYS s environment [21].As the comparison shows,the synthesized reduced model is able to reasonably capture full order model behavior with errors in natural frequencies ranging from 2%to 10%for the dominant X directional modes;and by at most 18%for the dominant Y directional mode.Dynamic stiffness too is approximated reasonably well with errors ranging from 0%to 9%;with the exception of the X -directional mode at $42Hz,for which error in approximation is as much as $50%.Errors are mainly thought to be due to the interpolation constraint formula-tion underestimating contact stiffness at the interfaces [18],leading to underestimation of stiffness of overall assembled machine.Other sources of errors may be attributed to modeling simplifications,and to a lesser extent due to the reduction process [16].Overall however,even though the synthesized reduced model is $1/25th the size of the full model,yet it gives reason-ably close results;and importantly,the reduced model leads to considerable simulation time savings for the designer;taking $1min/position as compared to $12h/position for the full model (Intel s i7 2.67GHz processor with 9GB RAM)thereby facilitating position-dependent analyses.The verified reduced model is validated against measurements by considering joint characteristics in the next section.3.Validation of the substructurally synthesized reduced machine modelDynamic response is measured at the tool tip with a 50mm diameter face mill on the three axis machine —FADAL2216,shown in Fig.5.Measurement including data acquisition,modal analysis and FRF curve fitting is carried out with CUTPRO s [22].Measured modal damping ratio is used in the synthesized reduced machine model and the full machine model.Joints at the contacting interfaces are idealized as two translational springs perpendicular to the direction of motion.The connections between the tool–tool holder-spindle and between the spindle–spindle housing are assumed to be rigid.The joint stiffness at bolted interfaces are taken as the corresponding bolt’s stiffness.Detailed modeling of the contact stiffness at the rolling interfaces is beyond the present scope of this paper;in which interfaces are idealized as being connected by linear spring elements for which the equivalent contact stiffness values were obtained from manufacturers 0catalogs [23].Equivalent contact stiffness for each of the three axes for the guide-block and guide-rail interface is assigned as 187N/m m (THK SVR series);and as 280N/m m for the ball-screw–nut interface (THK SBN series)[23].Although the joint characteristics are a function of contacting surface conditions which vary over the entire contacting inter-face,for simplicity,joint parameters are assumed to be position-independent.Measured fitted FRFs and simulated (full,reduced)TCP FRFs are compared in Fig.6in both the X and Y machine directions for the representative position of the tool being at the bottom of the work volume.Though the model predicted response is able to capture the trend of the measured response as evident in Fig.6;the dynamic stiffness of the low-frequency column modes is underestimated,particularly in the X -direction.The model predicted response also overestimates the first bending mode in the YZ plane —predicted at 47Hz,and measured at 33Hz.These anomalies are mainly00.55Frequency [Hz]0Frequency [Hz]YY FRF Comparisons0.40.20.10.20.3M a g n i t u d e [μm /N ]M a g n i t u d e [μm /N ]parison of reduced model TCP FRFs at three different tool positions:top,mid,and bottom.w et al./International Journal of Machine Tools &Manufacture 68(2013)81–9085thought to be due to the difficulty in modeling the base-mounting springs whose contact stiffness if overestimated,leads to errors in prediction of the low-frequency modes.Order mismatch between the number of simulated and measured modes between 50and 100Hz and at $250Hz may be attributed to modeling simplifi-cations in representing the machine accessories like the auto-matic tool changer and cabinets by lumped mass elements.The model response is updated at other positions as well,and is used to evaluate the achievable productivity levels of the machine tool in the next section.4.Material removal rate and position-dependent stability When the structural dynamics of the machine vary within themachine’s work space,the chatter stability and the resulting limits on the material removal rates vary as well.The variation of stability is demonstrated here by considering face milling ofAISI 4340steel with 80%engagement,which is treated as the target application for the envisaged machine.A minimum speed and feed-direction independent stable depth of cut of 4mm is targeted in the whole working range of the machine.Effects of position-dependent directional compliances on machining stabi-lity are investigated by generating feed-direction-dependent absolute machining stability charts.4.1.Oriented FRFs and feed-direction-dependent stabilityA multi-degree-of-freedom machine tool may chatter in any of its dominant modes.The effect of each mode is determined by its dynamic characteristics;whether or not the mode is aligned with the principal machine directions (X MT Y MT );and,the direction of feed.Consider a machine tool with horizontal (x )and normal (y )axes;and let the tool be traveling in the feed direction (u )with an angular orientation (y )as shown in Fig.7.Additionally,if the structural mode ðm xy iÞis not aligned with the principal machine Table 2Comparison of dominant modes for full order model (F )and reduced order model (R ),and the error estimates (E )in the X and Y directions.Natural frequencies (f n )[Hz];dynamic stiffness (K d )[N/m m];error [%].Bottom position Mid position Top position f nK d f n K d f n K d Mode no.F R %E F R %E F R %E F R %E F R %E F R %E X direction 142432 6.7 3.2À5241435 3.5 2.6À2640425 3.6 2.1À4228176À6 1.8 1.968073À9 1.9 1.9À7971À10 1.925Y direction15647À169.99.2À75245À135.95.7À35041À184.33.9À9ball -slide +X axisscrew + nut Y axis-screw + nut Cross -slide +Z YZXfreq. = 42 Hzfreq. = 41 HzFig.4.(a)Full order model,(b)reduced model with only interface DOFs,(c)mode shape of column bending mode in YZ plane,and (d)mode shape of column bending mode in XZ plane.w et al./International Journal of Machine Tools &Manufacture 68(2013)81–9086。
