Optimized U-MOT for experiments with ultracold atoms near surfaces

统计与决策2021年第3期·总第567期管理决策0引言近年来，随着经济社会的发展以及工业化和城镇化进程的加快，空气污染问题日益突显，PM 2.5作为大气污染中的主要污染物之一，对空气质量以及人们的生活造成了很大的影响。

对PM 2.5浓度做出精准的预测，有效地减少和控制大气污染，降低公众健康风险，可以为相关政策制定者提供有意义的参考。

目前PM 2.5等空气污染物浓度的预测，主要可分为三类：确定性模型、经典的统计学模型和机器学习模型。

确定性方法主要包括天气研究预报模式（Weather Researchand Forecasting，WRF ）[1]、多尺度空气质量模型（CommunityMultiscale Air Quality Modeling System,CMAQ）[2]、WRF-CMAQ 气象化学耦合模型[3]等，利用相关的气象数据和污染源数据，模拟污染物复杂的排放、累积、扩散和转移过程，从而对污染物的浓度进行预测和分析。

该方法复杂且和其他模型相比，在预测精度上并没有明显的优势[4]。

统计学模型中自回归移动平均(Autoregressive Moving Inte-grated Average,ARIMA)模型[5]、多元线性回归（MultipleLinear Regression，MLR ）[6]模型等经常被用于PM 2.5浓度的预测。

然而，PM 2.5浓度受到多种可变因素的影响，具有较强的非线性和复杂性，这类模型在处理非线性时间序列数据上，通常不能达到令人满意的效果[7]。

随着数据挖掘技术的兴起，机器学习方法以其优越的预测性能受到了大量的关注，被广泛应用于空气污染物的浓度预测中，主要包括人工神经网络(Artificial Neural Network,ANN)[8—10]、支持向量机[11，12]等。

Park 等（2018）[9]建立了人工神经网络模型，通过室外PM 10浓度、地铁列车的运行数量和通风率的信息，对6个地铁站内的PM 10浓度进行了预测，精度达到了67%~80%。

QMOT STEPPER MOTORSTRINAMIC Motion Control GmbH & Co. KGHamburg, GermanyV 1.06QMOT QSH4218 MANUAL++QSH-4218 -35-10-02742mm1A, 0.27Nm-41-10-03542mm1A, 0.35Nm-51-10-04942mm1A, 0.49Nm++-47-28-04042mm2.8A, 0.40NmTable of Contents1Life support policy (3)2Features (4)3Order codes (5)4Mechanical dimensions (6)4.1Lead wire configuration (6)4.2Dimensions (6)5Torque figures (7)5.1Motor QSH4218-35-10-027 (7)5.2Motor QSH4218-41-10-035 (7)5.3Motor QSH4218-51-10-049 (8)5.4Motor QSH4218-47-28-040 (8)6Considerations for operation (9)6.1Choosing the best fitting motor for an application (9)6.1.1Determining the maximum torque required by your application (9)6.2Motor Current Setting (9)6.2.1Choosing the optimum current setting (10)6.2.2Choosing the standby current (10)6.3Motor Driver Supply Voltage (10)6.3.1Determining if the given driver voltage is sufficient (11)6.4Back EMF (BEMF) (11)6.5Choosing the Commutation Scheme (12)6.5.1Fullstepping (12)6.5.1.1Avoiding motor resonance in fullstep operation (12)7Revision history (13)7.1Document revision (13)List of FiguresFigure 4.1: Lead wire configuration (6)Figure 4.2: Dimensions (all values in mm) (6)Figure 5.1: QSH4218-35-10-027 speed vs. torque characteristics (7)Figure 5.2: QSH4218-41-10-035 speed vs. torque characteristics (7)Figure 5.3: QSH4218-51-10-049 speed vs. torque characteristics (8)Figure 5.4: QSH4218-47-28-040 speed vs. torque characteristics (8)List of TablesTable 2.1: Motor technical data (4)Table 4.1: Lead wire configuration (6)Table 6.1: Motor current settings (9)Table 6.2: Driver supply voltage considerations (10)Table 6.3: Comparing microstepping and fullstepping (12)Table 7.1: Document revision (13)1Life support policyTRINAMIC Motion Control GmbH & Co. KG does notauthorize or warrant any of its products for use in lifesupport systems, without the specific written consentof TRINAMIC Motion Control GmbH & Co. KG.Life support systems are equipment intended tosupport or sustain life, and whose failure to perform,when properly used in accordance with instructionsprovided, can be reasonably expected to result inpersonal injury or death.© TRINAMIC Motion Control GmbH & Co. KG 2011Information given in this data sheet is believed to be accurate and reliable. However neither responsibility is assumed for the consequences of its use nor for any infringement of patents or other rights of third parties, which may result from its use.Specifications are subject to change without notice.2FeaturesThese two phase hybrid stepper motors are optimized for microstepping and give a good fit to the TRINAMIC family of motor controllers and drivers. They are all used in the 42mm PANdrive™ family. The QSH4218-35-10-027, QSH4218-41-10-035 and QSH4218-51-10-049 are motors with 1A RMS coil current. The QSH4218-47-28-40 is different: It is designed for 2.8A RMS coil current to provide maximum torque at high velocities.Characteristics of all QSH4218 motors:∙NEMA 17 mounting configuration∙flange max. 42.3mm * 42.3mm∙step angle: 1.8˚∙optimized for microstep operation∙optimum fit for TMC236/TMC246 based driver circuits∙ 4 wire connection∙CE approvedSpecial characteristics of the QSH4218-47-28-40 motor:∙optimized for 2.8A RMS coil current to provide maximum torque at high velocities∙0.25Nm torque at 1200 RPM with a 24V supply∙holding torque 0.40NmTable 2.1: Motor technical data3Order codesThe length of the motor is specified without the length of the axis. For the total length of the product add 24mm.Tabelle 3.1: Order codes4 Mechanical dimensions4.1 Lead wire configurationTable 4.1: Lead wire configurationFigure 4.1: Lead wire configuration4.2DimensionsFigure 4.2: Dimensions (all values in mm)blackr e db l u e5Torque figuresThe torque figures detail motor torque characteristics for half step operation in order to allow simple comparison. For half step operation there are always a number of resonance points (with less torque) which are not depicted. These will be minimized by microstep operation in most applications.5.1Motor QSH4218-35-10-027Testing conditions: driver supply voltage +24V DC, coil current 1.0A RMS, half step operationFigure 5.1: QSH4218-35-10-027 speed vs. torque characteristics5.2Motor QSH4218-41-10-035Testing conditions: driver supply voltage +24V DC, coil current 1.0A RMS, half step operationFigure 5.2: QSH4218-41-10-035 speed vs. torque characteristics5.3Motor QSH4218-51-10-049Testing conditions: driver supply voltage +24V DC, coil current 1.0A RMS, half step operationFigure 5.3: QSH4218-51-10-049 speed vs. torque characteristics5.4Motor QSH4218-47-28-040Testing conditions: driver supply voltage: +24V DC, coil current: 2.8A RMS, half step operation Figure 5.4: QSH4218-47-28-040 speed vs. torque characteristics6Considerations for operationThe following chapters try to help you to correctly set the key operation parameters in order to get a stable system.6.1Choosing the best fitting motor for an applicationFor an optimum solution it is important to fit the motor to the application and to choose the best mode of operation. The key parameters are the desired motor torque and velocity. While the motor holding torque describes the torque at stand-still, and gives a good indication for comparing different motors, it is not the key parameter for the best fitting motor. The required torque is a result of static load on the motor, dynamic loads which occur during acceleration/deceleration and loads due to friction. In most applications the load at maximum desired motor velocity is most critical, because of the reduction of motor torque at higher velocity. While the required velocity generally is well known, the required torque often is only roughly known. Generally, longer motors and motors with a larger diameter deliver a higher torque. But, using the same driver voltage for the motor, the larger motor earlier looses torque when increasing motor velocity. This means, that for a high torque at a high motor velocity, the smaller motor might be the fitting solution. Please refer to the torque vs. velocity diagram to determine the best fitting motor, which delivers enough torque at the desired velocities.6.1.1Determining the maximum torque required by your application Just try a motor with a torque 30-50% above the application’s maximum requirement. Take into consideration worst case conditions, i.e. minimum driver supply voltage and minimum driver current, maximum or minimum environment temperature (whichever is worse) and maximum friction of mechanics. Now, consider that you want to be on the safe side, and add some 10 percent safety margin to take into account for unknown degradation of mechanics and motor. Therefore try to get a feeling for the motor reliability at slightly increased load, especially at maximum velocity. That is alsoa good test to check the operation at a velocity a little higher than the maximum application velocity.6.2Motor Current SettingBasically, the motor torque is proportional to the motor current, as long as the current stays at a reasonable level. At the same time, the power consumption of the motor (and driver) is proportional to the square of the motor current. Optimally, the motor should be chosen to bring the required performance at the rated motor current. For a short time, the motor current may be raised above this level in order to get increased torque, but care has to be taken in order not to exceed the maximum coil temperature of 130°C respectively a continuous motor operation temperature of 90°C.Table 6.1: Motor current settings6.2.1Choosing the optimum current settingGenerally, you choose the motor in order to give the desired performance at nominal current. For short time operation, you might want to increase the motor current to get a higher torque than specified for the motor. In a hot environment, you might want to work with a reduced motor current in order to reduce motor self heating.The Trinamic drivers allow setting the motor current for up to three conditions:-Stand still (choose a low current)-Nominal operation (nominal current)-High acceleration (if increased torque is required: You may choose a current above the nominal setting, but be aware, that the mean power dissipation shall not exceed the motors nominal rating)6.2.2Choosing the standby currentMost applications do not need much torque during motor standstill. You should always reduce the motor current during standstill. This reduces power dissipation and heat generation. Depending on your application, you typically at least can half power dissipation. There are several aspects why this is possible: In standstill, motor torque is higher than at any other velocity. Thus, you do not need the full current even with a static load! Your application might need no torque at all, but you might need to keep the exact microstep position: Try how low you can go in your application. If the microstep position exactness does not matter for the time of standstill, you might even reduce the motor current to zero, provided that there is no static load on the motor and enough friction in order to avoid complete position loss.6.3Motor Driver Supply VoltageThe driver supply voltage in many applications cannot be chosen freely, because other components have a fixed supply voltage of e.g. 24V DC. If you have the possibility to choose the driver supply voltage, please refer to the driver data sheet and consider that a higher voltage means a higher torque at higher velocity. The motor torque diagrams are measured for a given supply voltage. You typically can scale the velocity axis (steps/sec) proportionally to the supply voltage to adapt the curve, e.g. if the curve is measured for 48V and you consider operation at 24V, half all values on the x-Axis to get an idea of the motor performance.For a chopper driver, consider the following corner values for the driver supply voltage (motor voltage). The table is based on the nominal motor voltage, which normally just has a theoretical background in order to determine the resistive loss in the motor.Comment on the nominal motor voltage: Array(Please refer to motor technical data table.)Table 6.2: Driver supply voltage considerations6.3.1Determining if the given driver voltage is sufficientTry to brake the motor and listen to it at different velocities. Does the sound of the motor get raucous or harsh when exceeding some velocity? Then the motor gets into a resonance area. The reason is that the motor back-EMF voltage reaches the supply voltage. Thus, the driver cannot bring the full current into the motor any more. This is typically a sign, that the motor velocity should not be further increased, because resonances and reduced current affect motor torque.Measure the motor coil current at maximum desired velocityFor microstepping: If the waveform is still basically sinusoidal, the motor driver supply voltage is sufficient.For Fullstepping: If the motor current still reaches a constant plateau, the driver voltage is sufficient.If you determine, that the voltage is not sufficient, you could either increase the voltage or reduce the current (and thus torque).6.4Back EMF (BEMF)Within SI units, the numeric value of the BEMF constant has the same numeric value as the numeric value of the torque constant. For example, a motor with a torque constant of 1 Nm/A would have a BEMF constant of 1V/rad/s. Turning such a motor with 1 rps (1 rps = 1 revolution per second = 6.28 rad/s) generates a BEMF voltage of 6.28V.The Back EMF constant can be calculated as:is multiplied by 2 in this The voltage is valid as RMS voltage per coil, thus the nominal current INOMformula, since the nominal current assumes a full step position, with two coils switched on. The torque is in unit [Nm] where 1Nm = 100cNm = 1000mNm.One can easily measure the BEMF constant of a two phase stepper motor with a (digital) scope. One just has to measure the voltage of one coil (one phase) when turning the axis of the motor manually. With this, one gets a voltage (amplitude) and a frequency of a periodic voltage signal (sine wave).The full step frequency is 4 times the frequency the measured sine wave.6.5Choosing the Commutation SchemeWhile the motor performance curves are depicted for fullstepping and halfstepping, most modern drivers provide a microstepping scheme. Microstepping uses a discrete sine and a cosine wave to drive both coils of the motor, and gives a very smooth motor behavior as well as an increased position resolution. The amplitude of the waves is 1.41 times the nominal motor current, while the RMS values equal the nominal motor current. The stepper motor does not make loud steps any more – it turns smoothly! Therefore, 16 microsteps or more are recommended for a smooth operation and the avoidance of resonances. To operate the motor at fullstepping, some considerations should be taken into account.Table 6.3: Comparing microstepping and fullsteppingMicrostepping gives the best performance for most applications and can be considered as state-of-the art. However, fullstepping allows some ten percent higher motor velocities, when compared to microstepping. A combination of microstepping at low and medium velocities and fullstepping at high velocities gives best performance at all velocities and is most universal. Most Trinamic driver modules support all three modes.6.5.1FullsteppingWhen operating the motor in fullstep, resonances may occur. The resonance frequencies depend on the motor load. When the motor gets into a resonance area, it even might not turn anymore! Thus you should avoid resonance frequencies.6.5.1.1Avoiding motor resonance in fullstep operationDo not operate the motor at resonance velocities for extended periods of time. Use a reasonably high acceleration in order to accelerate to a resonance-free velocity. This avoids the build-up of resonances. When resonances occur at very high velocities, try reducing the current setting.A resonance dampener might be required, if the resonance frequencies cannot be skipped.7Revision history 7.1Document revisionTable 7.1: Document revision。

2An unbeatable solutionThe high-performance HP 8753E RF network analyzer brings together an unbeatable com-bination of speed, performance and ease-of-use features to solve your measurement needs,whether on the production floor or in R&D. With an integrated S-parameter test set that covers 30 kHz to 3 or 6 GHz, up to 110 dB of dynamic range, and both frequency and power sweeps,the HP 8753E accurately and quickly characterizes the linear and nonlinear behavior of active and passive components.Highly evolved for test successThe HP 8753E vector network analyzer provides both magnitude and phaseinformation, as well as, gain compression, group delay and time domain measurements —all with vector error correction to minimize measurement uncertainty.Superb speed, accuracy, and productivity features The HP 8753E extends the performance of the industry-standard HP 8753D with an impressive 300% speedimprovement in measurement sweep, data transfer,instrument-state recall and time-domain analysis. Interface and code compatability with the HP 8753D makes the transition to the HP 8753E fast and easy.Superb measurement accuracy as well as numerous calibration techniques for coaxial, on-wafer and in-fixture measurements enhance product yields. In addition, the world’s most popular RF network analyzer now offers four-parameterdisplay, VGA output, numerous productivity enhancements, and a smaller size.Enhancements formanufacturing includemore calibration techniques,four-parameter display, VGA output for enhanced visability, and smaller size.Impressive 300% speedimprovement in measurement sweep, data transfer,instrument-state recall andtime-domain analysis.1981Feature highlightsFaster measurement speed Exceptional sweep speed, register recall and data-transfer rate.´Faster tuning/testingSimultaneously display all four S-parameters while tuning devices.Quickly record or print all four S-parameters for record keeping. Enhance yourproduct throughputImproved measurement accuracy for noninsertable devices with adapter-removal calibration.Maximize measurement speed with fast sweep list mode.Flexible fitto your environmentVGA output for enhanced pact size to maximize space and ease transport.Reduced support effortMake firmware upgrades quickly using built-in disk drive.Backward compatibility Backward compatibility with HP 8753D makes leveraging your investment in software fast and easy.Built-in data storageConvenient storage of instrument states, calibrations, data, and test sequences via floppy disk and nonvolatile memory.Test sequencingInternally configure and automate measurements easily and quickly with keystroke recording.A winning combination ofperformance and productivity featuresMeasurement speed improvements for the HP 8753EFunction 8753D 8753E % increase in speed 201 points, two-port calibration.510.145352Recall 201 points.740.138536Time domain 201 points.350.046760Trace dump 201 points.061.021290List frequency 201 points.795.104764Powerful options customize your measurementsExtended frequency coverage• Option 006 for characterizing components to 6 GHz.Time-domain analysis• Locate and resolve discontinuities in your test device or fixture versus distance. Gain more insight into the behavior of your device by displaying the step response or by gating to remove unwanted responses such as connector mismatch. Quickly and accurately locate cable faults or resolve multiple discontinuities.Harmonic measurement capability • Swept second and third harmonic levels can be displayed directly or relative to the fundamental carrier (dBc) by using the harmonic-measurement capability. Measure amplifier harmonics as low as –40 dBc with a simple press of a button.Configuration flexibility• Use a dedicated HP test set or delete the built-in test set with Option 011.Enhance performanceFaster measurement speed• New CPU provides exceptional sweep speed, register recall and data transfer rate.•10 Hz to 6 KHz IF filters optimize speed versus accuracy trade-offs.Four-parameter display•Increase your tuning speed with simul-taneous display of all four S-parameters.Now it’s easy and convenient to record and print all four S-parameters.VGA output•Add an external monitor to enhance the visibility of displayed data.Smaller size•More compact and lighter, the HP 8753E maximizes test-station space and is easier to move.Backwards compatable with 8753D •Easy transition from previous model with compatible HP-IB and front-panel features.Greater productivity•Improve the measurement accuracy of noninsertable devices using the new adapter-removal calibration technique.•Reduce measurement time by selecting swept list mode. Choose frequency ranges with independent IF bandwidths and power levels.• Touchstone ©-compatible S2P data format. Flash memory•Use the built-in disk drive to quickly and easily upgrade firmware, and reduce support efforts.Standard features provide solid valueBuilt-in disk drive•Allows convenient storage of instrument states, data, and test sequences.Nonvolatile memory•Allows internal storage of calibration data, as well as up to 31 instrument states.Wide dynamic range•110 dB dynamic range (30 kHz to 3 GHz)and 105 dB dynamic range (3 GHz to 6 GHz).TRL*/LRM* calibration•Allows convenient calibration for in-fixture environments such as microstrip.Serial, parallel and HP-IB interfaces •Supports modern printers and plotters. Make direct color or black-and-white hardcopies of data.Test sequencing•Internally configure and automatemeasurements with an enhanced form of keystroke recording.Real-time clock•Convenient time-stamping of data printouts and files.Fast, accurate measurements are essential in today’s fast paced, competitive environment whether you are in manufacturing or R&D. When you have an HP 8753E vector network analyzer, you have the best measurement tool available for characterizing the performance of high-frequency components and devices. The HP 8753E provides a host of capabilities including magnitude, phase, and group-delay measurements. It also performs power sweeps, providing gain compression and AM-to-PM conversion measurements, shows impedances on a Smith chart, and with the time-domain option, shows the distance from the test port to an impedance mismatch or cable fault. So whether your application is manufacturing components for the wireless market or designing a state-of-the-art RF filter or amplifier,the HP 8753E helps you maximize your measurement throughput and designs.Powerful analysis of RF componentsSuperb speed and performance The HP 8753E provides quick,high-performance measure-ments for a broad range ofintegrated and discrete devicessuch as filters, duplexers,amplifiers, mixers, modulators,cables and antennas. Thesemeasurement capabilities areideally suited for today’s R&Dand high volume manufacturersin the RF communication,aerospace/defence and consumer electronics industries. 1-Hz frequency resolutionallows accurate crystal filter measurements of both passband transmission response andgroup delay.SAW device measurementsCompletely characterize surfaceacoustic wave (SAW) deviceswith high-resolution return loss,insertion loss, and group-delaymeasurements. Measurepassband ripple with 0.001-dBmagnitude resolution and0.01-picosecond delay resolution.Gain additional insight aboutyour device’s unwanted timeresponses, such as RF leakageand triple transit, with optionaltime-domain capabilty. Applygating (a time-domain selectivefilter) to the main lobe response,and view your device’s responseindependently of matchingnetworks and connectors.Simultaneously viewboth the frequency responseand the time responseof your SAW device.Filter measurementsHigh resolution, a stable sourceand wide-dynamic-range recieversare essential for transmissionand reflection measurements ofnarrowband devices such asresonators, filters and duplexers.The HP 8753E synthesizedsource provides 1-Hz resolutionwith uncompromisingmeasurement speed, while thesensitive tuned receivers provideup to 110 dB of dynamic range.Optimize the measurementdynamic range with the 10-Hzbandwith IF filter or choose thenew 6-KHz IF filter for maximummeasure ment speed.Minimize adjustment time withmarker functions that providemultiple marker values simul-taneously and give on-screeninformation about your device’s3-dB point, passband ripple,and maximum or minimumvalue. The marker-trackingfunction provides fast, updatedmagnitude and frequencyinformation during tuning ofyour device.Amplifier measurements Measuring an amplifier that has a high gain and high reverse isolation with a low input signal requires a test system with wide dynamic range and high sensitivity. The HP 8753E is ideal for these measurements with up to –100 dBm sensitivity,110 dB of dynamic range, 0.05 dB dynamic accuracy, and theability to set the test-port power level exactly over a 95 dB range.In addition, the use of swept power to simplify gain-compression measurements and power-meter calibration forimproving source power level as well as receiver-measurement accuracy, make the HP 8753E one of the most productive tools available.Mixer measurementsTraditionally, vector network analyzers have operated over a single stimulus-and-response frequency range, and therefore,could not test the transmission characteristics of mixers. The HP 8753E has the ability to offset its receiver frequency from that of its own internally synthesized source. This enables you to test the transmission characteristics of mixers (conversion loss, amplitude,phase tracking and group delay)by stimulating a device over one frequency range and viewing its response over another.HP 8753E mixermeasurement menuFour parameter displayDesigned for speed, the HP 8753E measures and displays all four S-parameters simultaneously 1.Real-time tuning of your RF communication duplexers and isolators couldn’t be easier.Display any combination of reflection and transmission parameters with magnitude,phase, group delay, Smith chart,polar, SWR, or time-domain formats. Easy-to-use softkeys let you access mesurement functions quickly and viewresults in overlay or split-screen format on a crisp, LCD color display with one, two or four graticules. For enhanced visability, use theVGA-compatible output to drive larger external monitors.1. Available Q298Simultaneously view all four S-parameters for faster tuning/testingHP 8753E productivity featuresspeed and simplify your measurementsPass/fail testingReduce test times by letting the network analyzer determine if measurement results are within user-defined limits. You can easily choose any combination of single-point, horizontal or sloping line limits from thefront-panel. Pass/fail is indicated visually from the display, audibly with a beep, over HP-IB, or from a BNC rear-panel TTL e rear-panel TTL output as an input for automated part handlers.Automate repetitive tasks,without a computerIn test-sequencing mode, you make the measurement once and the network analyzerrecords the keystrokes. Complex measurements can be stored in a sequence and recalled rapidly and consistently with the touchof a button. Since a sequence is created by the same front-panel keystrokes used during manual operation, no additional pro-gramming expertise is required.Display user prompts for tuning and other manual adjustments or make go/no-go decisionsduring sequence execution. You can even control other HP-IB instruments, or use the parallel port to control part handlers.User-definedfrequency testingSpeed up your testing bymeasuring your device at only selected frequencies. You can specify up to 30 arbitrary CW frequencies or frequency sweep segments at which to test your device. Set test-port power levels and IF bandwidth independently for each segment. Reduce test time and increase measurement throughput by optimizing each segment to your specific test requirements.Test sequence flowchartExample of "fast swept list"output.Transfer datato your CAE program Store ASCII disk files inconformance to the CITIFILE standard with the HP 8753E.Touchstone ©-compatible (S2P)format provides component data files describing frequency-dependent two-port S-parameters.Measured data can then be read by electronic design automation applications, including HP EEsof design software, and used for circuit simulation. These pro-grams can also access datadirectly via HP-IB.StartSet stimulus & test limits Connect device Measure parameters Passed test limits Tune devicePlot data & limits Next device StopYNIncrease throughput using automated pass/fail.Change frequenciesand remain calibratedSave time and avoid recalibration when changing frequencies by using the interpolative error-correction mode. Perform a broadband calibration with up to 1,601 points and then adjust your frequency span or number of measurement points for the particular device under test. The HP 8753E will automatically recalculate the error terms based on the new values of either parameter. This allows you the flexibility of investigating your device’s performance over any portion of the calibratedfrequency range while main-taining full display resolution. RF limiterAchieve solid, reliable network analyzer protection for high power applications. The HP 11930RF limiter externally attaches to one or both ports of the analyzer, in essence creating an external fuse. This optional accessory provides an inexpen-sive, quick method of ensuring against potential high-power transients from external devices that can cause hardware failures.RF electroniccalibrationAchieve fast, consistent, high-accuracy calibrations auto-matically with HP’s optionalRF electronic calibration(RF ECal) products. The easy-to-use PC/Windows based solution provides 30 kHz to 6 GHz electronic calibration, which reduces user calibration errors, connector damage and main-tenance. HP ECal provides accurate calibrations for non-insertable and mixed connector devices for improved product yeilds. Designed for high-volume manufacturing environments, HP RF ECal provides the means to optimize your productionmeasurement environment. Reduce measurementcalibration time withHP’s RF ECal.Hewlett-Packard service and supportThe reliability of the HP 8753E has been proven throughextensive environmental testing of shock, vibration, and extreme temperature cycling. Further improvements in reliability have been realized by applying improvements gained fromcareful analysis of the entire HP RF network analyzer family.HP literature Number HP 8753ETechnical Specifications 5966-0054E HP 8753EConfiguration Guide5966-0055EData Subject to Change Copyright © 1997Hewlett-Packard Company Printed in U.S.A. 12/975966-0053EApplications and test expertise at your service Our Systems and Service Division is ready to help you with test process analysis,consulting, and softwaredevelopment. Take advantage of HP’s wide variety of technical training offered around the world, even customized to fit your specific needs.24-hour telephone support HP offers telephone technical and applications support 24hours a day in most countries served by HP .One-year warranty Hewlett-Packard offers astandard one-year, return-to-HP warranty with the HP 8753E.At a low, fixed cost, you can order support options to extend warranty or cover periodic calibrations.Test fixturesFor more information on test fixtures, ask for HP literature number 5091-4254E, or contact:Inter-Continental Microwave 1515 Wyatt DriveSanta Clara, Ca 95054, USA Telephone: (408) 727-1596Fax: (408 727-0105HP on lineFor more information about Hewlett-Packard test and measurement products,applications, services, and for a current sales office listing, visit our web site,/go/tmdir.Quality and reliability by designThe HP 8753E is manufactured in ISO 9002 -registered facilities in concurrence with HP’s commitment to quality.。

NVIDIA Jetson Orin NX Series Orin performance. Nano size.The Most Advanced AI Computer for Smaller, Lower-Power Autonomous MachinesNVIDIA® Jetson Orin™ NX series modules deliver up to 100 TOPS of AI performance in the smallest Jetson form-factor, with power configurable between 10W and25W. This gives you 3X the performance of NVIDIA Jetson AGX Xavier™ and 5X the performance of Jetson Xavier™ NX, making it ideal for small form-factor, low-power products like drones and handheld devices.These system-on-modules support multiple concurrent AI application pipelines with an NVIDIA Ampere architecture GPU, next-generation deep learning and vision accelerators, high-speed IO, and fast memory bandwidth. Now, you can develop solutions using your largest and most complex AI models to solve problems such as natural language understanding, 3D perception, and multi-sensor fusion.Jetson runs the NVIDIA AI software stack and takes advantage of use case-specific application frameworks, including NVIDIA Isaac™ for robotics, DeepStream for vision AI, and Riva for conversational AI. You can also save significant time with NVIDIA Omniverse™ Replicator for synthetic data generation (SDG), and with NVIDIA TAO Toolkit for fine-tuning pretrained AI models from the NGC™ catalog.Jetson ecosystem partners offer additional AI and system software, developer tools, and custom software development. They can also help with cameras and other sensors, as well as carrier boards and design services for your product. Jetson Orin modules are unmatched in performance and efficiency for robots and other autonomous machines, giving you the flexibility to create the next generation of AI solutions with the latest NVIDIA GPU technology. Together with the world-standard NVIDIA AI software stack and an ecosystem of services and products, your road to market has never been faster.DatasheetReady to Get Started?To learn more about the Jetson Orin, visit: /Jetson-Orin© 2023 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, CUDA, Jetson AGX Xavier, NGC, NVIDIA Isaac, NVIDIA Jetson, and NVIDIA Orin are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. ARM, AMBA and ARM Powered are registered trademarks of ARM Limited. Cortex, MPCore and Mali are trademarks of ARM Limited. All other brands or product names are the property of their respective holders. “ARM” is used to represent ARM Holdings plc; its operating company ARM Limited; and the regional subsidiaries ARM Inc.; ARM KK; ARM Korea Limited.; ARM Taiwan Limited; ARM France SAS; ARM Consulting (Shanghai) Co. Ltd.; ARM Germany GmbH; ARM Embedded Technologies Pvt. Ltd.; ARM Norway, AS and ARM Sweden AB. Other company and product names may be trademarks of the respective companies withwhich they are associated. 2605318. JAN23* Virtual channel-related camera information for Jetson Orin NX is not final and subject to change.Refer to the Software Features section of the latest NVIDIA Jetson Linux Developer Guide for a list of supported features.NVIDIA Jetson Orin NX Series Modules。

《基于NSGA-Ⅱ遗传算法的M100甲醇发动机多目标性能优化》篇一一、引言随着能源危机和环保意识的日益增强，甲醇发动机作为一种清洁、可再生的动力源，受到了广泛关注。

M100甲醇发动机作为其中的一种重要类型，其性能优化对于提高能源利用效率、降低排放具有重要意义。

本文旨在探讨基于NSGA-Ⅱ遗传算法的M100甲醇发动机多目标性能优化方法，以期为相关研究提供参考。

二、NSGA-Ⅱ遗传算法概述NSGA-Ⅱ（非支配排序遗传算法II）是一种多目标优化算法，通过模拟自然选择和遗传学机制，实现多目标决策问题的优化。

该算法具有全局搜索能力强、收敛速度快、解集分布均匀等优点，适用于复杂系统的多目标优化问题。

在M100甲醇发动机性能优化中，NSGA-Ⅱ遗传算法可以有效地处理多目标冲突，找到帕累托最优解。

三、M100甲醇发动机性能优化目标M100甲醇发动机的性能优化主要涉及以下几个方面：动力性、经济性、排放性能和可靠性。

动力性是指发动机输出的功率和扭矩；经济性指发动机的燃油消耗率；排放性能主要关注尾气中有害物质的排放量；可靠性则涉及发动机的寿命和故障率。

这些目标之间往往存在冲突，需要通过多目标优化方法进行平衡。

四、基于NSGA-Ⅱ遗传算法的M100甲醇发动机多目标性能优化1. 问题建模：将M100甲醇发动机性能优化问题转化为多目标优化问题，明确决策变量、目标函数和约束条件。

2. 初始化种群：根据问题的特点，生成一定规模的初始种群。

3. 遗传操作：通过选择、交叉和变异等操作，产生新一代种群。

其中，选择操作根据个体的非支配排序和拥挤度进行；交叉和变异操作则模拟生物遗传过程中的基因重组和突变现象。

4. 评估与选择：对新产生的种群进行评估，计算每个个体的目标函数值，并根据评估结果进行选择，保留优秀个体。

5. 迭代优化：重复上述遗传操作和评估选择过程，直至满足终止条件（如迭代次数、解集质量等）。

五、实验结果与分析通过实验，我们发现基于NSGA-Ⅱ遗传算法的M100甲醇发动机多目标性能优化方法能够有效地平衡各目标之间的冲突，找到帕累托最优解。

440C-CR30COMLOCK 00 01 02 03 04 05 06 07 08 0 10 1112 13 14 15 16 17 18 1 20 21+DC -DC 24 24Note: Configuration is what we tested,not the max capacityMicro800 PTO Motion PopularConfiguration Drawing• Optimized for compatibility with MicroLogix™ , SLC™, and Micro800 controllers.• Features include unicode language switching, alarm messages and history, and basic recipe capability• User Interface through Web browser•Panelview Component Configuration Publication CCSIMP-QR002A-EN-P• Supports communication via RS-232 (DH-485), RS-232 (DF1), RS-485 and Ethernet• Two USB ports for transferring files or updating firmware , Supports SD memory cardsPanelview Component Capacity• 4.3, 5.7 ,10.7 inch Display Sizes • Monochrome Transmissive FSTN Options/Color Transmissive TFT, Analog Touch• Automatic motor recognition and gain settings • Cost-effective motion control solution for smaller, low-axis count applications• Flexible command interfaces including digital I/O, analog, pulse train and Modbus-RTU• Configurable with Connected Component Workbench SoftwareKinetix 3 Configuration• Input volt range of 170…264V AC single and three-phase• Continuous power output of 0.05 W…1.5 kW• Modbus-RTU serial communications for changing drive parameters during runtimeKinetix 3 Capacity• CCAT contains Building blocks for indexing up to 3 axisKinetix 3 PerformanceExternal Power Supply• Embedded I/O 28 In, 20 Out,Base Analog I/O channels via plug in or expansion. Maximum Digital I/O :132, Max Expansion I/O Modules:4, 6 High Speed Counters, 3 PTOMicro850 Configuration• Designed for larger standalone machine applications that require more I/O or higher performance analog I/O• Number of I/O points embedded in the base 24,48 points (2085 expansion I/O’s)Micro850 PerformanceMicro 850 Overview 2080-LC50-48xxx Embedded communication USB, Serial, Ethernet Basic I/O size 48 points Max number of I/OUp to 132Max number of Expansion I/O 4Motion axis (Only available in model with transistor output)Up to 3 axis HSCUp to 6 embeddedTo Plant Wide Ethernet network To Computer for ProgrammingStratix 2000PanelView ComponentMicro850Kinetix 3External Encoder Differential / Open collector typeStratix 2000 Unmanaged Switch - 5 Copper PortsPanelview Component Performance 3• Compact DIN rail mount safety relay• 22 embedded Safety I/O • Support for 16 standardI/O via plug-ins440C-CR30 Safety Relay Micro850 Capacity• LC50-48xxx has 10Ksteps of instructions and 20KB of data memory and supports 5 plug in I/O modules & 4 expansion modules.• Supports Ladder Programming,Function Block Diagram, Structured TextC400Micro850 with2080-SERIALISOL (slot 1) & 2080-MOT-HSC (slot 2-4) plug-in modules.• Rotary Motors • TL-Series • Linear Motors• LDC-Series Iron Core • LCL-Series IronlessServo motorsSafety Diagnostics to Micro850 and to PVcSafety I/OConnectionsPulse Train Outputs (24V)High Speed feedback signals from Kinetix 3 Servo drive 5V Differential3High Speed feedback signals from External Encoder . (Up to 5V/24V, Up to 100kHz open collector 250kHz differential)Kinetix 3Kinetix 3Protocol RS-232Micro850: Modbus RTU Master 440C-CR30: SlaveDaisy ChainRS-485 to three Kinetix 3About this ConfigurationThis Micro 850 Controller based low cost system demonstrates the power and scalability of the Component class on an Ethernet I/P based network. This system utilizes standard Ethernet I/P technology to connect to plant wide network. This configuration also highlights Micro850 Capabilities in Connecting Safety relay, Motion control & HMl .A key advantage of this architecture is the ability to use Connected Components Workbench a common integrated environment for programming, configuration, commissioning, and motion tools for the Kinetix 3 ,Panelview Component , and Micro 800 Controller family of products.About the ProductsMicro 850 Controllers• The Micro850 controller is equipped with the same form factor, plug-in support, instruction/data size and motion capabilities as the 24-pt and 48-pt Micro830 controllers• Support up to four Micro850 Expansion I/O modules, Up to a maximum of 132 I/O points (with 48-pt model)• Designed for larger standalone machine applications that require more I/O or higher performance analog I/O than supported by Micro830.• Structured Text, Ladder Diagram and Function Block editors that support • Symbolic addressing• Connected Components Workbench software is used for PLC programming, HMI, drives and safety relay configurationMicro 850 Plugin Module2080-MOT-HSC• The new Allen-Bradley Micro800 Motion Feedback Axis plug-in enhances high speed counter function to Micro800 controllers.• With input frequency to 250 kHz, it supports both 5V differential line driver input and 24VDC input with 1 physical digital output and 15 virtual digital outputs.• Its programmable limit switch function makes it easy for user to turn ON and OFF output based on the counter value.• Through its motion feedback axis mode, virtual axis mode and touch probe function, the new plug-in addition provides an easy and intuitive way for users to obtain motion positioning information..440C-CR30 Safety Relay• Programmable Safety Relay designed for flexibility in a simple safety system• Predefined Safety Functions provide an easy to use system that is programed and configured using CCW software• Compact DIN rail mount designed with 22 embedded Safety I/O and support for 16 standard I/O via plug-ins in a 110mm wide package • Built in communications for programming and HMI diagnostic displayPanelView Component• A full line of displays ranging from 2” to 10”,• Communicate to MicroLogix, Micro800 and SLC controllers via serial (RS232 or RS422/RS485) networks on all terminals, and Ethernet on the C400, C600 and C1000 displays.• Features include unicode language switching, alarm messages and history, and basic recipe capability• PanelView Component software (DesignStation) is an offline programming software offering better user experience with significant DesignTime performance improvementKinetix 3 Servo DrivesSingle-axis solution for low-complexity motion applications, with or without a PLC • Digital I/O, analog, preset velocity, and pulse-train command interfaces• Performs indexing on up to 64 points through serial communication or over digital I/O • 170…264V AC, (200V-class) single-phase or three-phase1783 Stratix 2000™ Unmanaged Switches• Multiple port count and fiber options are available • Copper ports are 10/100 Mbps, full- or half-duplex• Fiber ports are 100 Mbps, full-duplex, with an LC fiber optic connector • Ideal Ethernet switch for small, isolated networks • Auto negotiates speed and duplex settings • Operates on 20V AC or 24V DC power • Automatic cable cross-over detectionFor More Information and HelpFor more information contact your local distributor or Rockwell Automation sales representative.• • Publication Library • My Support• A – Z Product DirectoryBill of MaterialsQty Catalog #DescriptionSystem: Controller Hardware 12080-LC50-48QBB Micro 850 Controller12080- SERIALISOL RS232/485 isolated serial port Plug in Module 32080-MOT-HSC Micro800 Motion Feedback Axis plug-in ModuleHMI Hardware12711C T4T PanelView C400, touch terminalSafety Relay Module1440C-CR30-22BBB Compact DIN rail mount safety relay designed with 22 embedded Safety I/O and support for 16 standard I/O via plug-ins 11761-CBL-HM02Communication cableMotion Hardware22071-AP1Kinetix 3 , 2.38Amps 2TL-A130P-BJ32AA Servo Motor22090-CFBM6DD-CCAA03Cable,Feedback,M-Circ.Plastic.D-DB15,INCR CC ,Std,3M 22090-CPWM6DF-16AA03Cable,Power,M-Circ.Plastic.D-Flying Lead,16AWG ,Std,3M12090-CCMDSDS-48AA01Serial Communication cable between K3 drive & 2080-SERIALISOL plug-in 1Serial Communication cable between K3 drivesNote: Catalog numbers consist of characters, each of which identifies a specific version or option for that component. Reference Publication GMC-SG001-EN-E Kinetix Motion Control Selection Guide for additional informationSystem: Communication Hardware11783-US05T Stratix 2000 Unmanaged Switch - 5 Copper Ports 41585J-M8TBJM-1M Shielded Ethernet CableConfiguration Tool Required19328-SO001-EN-CConnected Components Workbench software。

高通量计算集成机器学习催化描述符设计新型二维MXenes析氢催化剂摘要：二维MXenes作为一种具有优异催化性能的材料，其析氢性能的研究显得尤为重要。

然而，传统的试错方法耗费时间和资源，难以大规模筛选出性能优异的MXenes。

因此，我们提出了一种基于高通量计算和机器学习的催化描述符设计方法，以加速和优化MXenes的析氢性能预测和发现过程。

本文首先通过大量密度泛函理论计算筛选出112种可能的析氢MXenes，并通过Fe原子掺杂进一步优化其析氢性能，得到7种性能优异的Fe doped MXenes。

接着，我们基于多项式回归、随机森林和支持向量回归等机器学习算法构建了基于17种物理和化学性质的催化描述符，并通过训练集和测试集的误差分析，选择了随机森林作为最佳预测模型。

最后，我们使用该模型预测了所有112种MXenes的析氢性能，并发现了15种前所未有的性能优异MXenes，其中析氢活性高于Ni和Pd催化剂，且可能具有实际应用价值。

关键词：MXenes；催化描述符；高通量计算；机器学习；析氢。

Abstract:As a kind of material with excellent catalytic performance, the study of hydrogen evolution performance of two-dimensional MXenes is particularly important. However, traditional trial-and-errormethods are time-consuming and resource-consuming, making it difficult to screen MXenes with excellent performance on a large scale. Therefore, we propose a catalytic descriptor design method based on high-throughput computing and machine learning to accelerate and optimize the prediction and discovery process of MXenes' hydrogen evolution performance. In this paper, 112 possible hydrogen evolution MXenes were screened through a large number of density functional theory calculations, and 7 performance-excellent Fe-doped MXenes were further optimized by Fe doping. Then, based on machine learning algorithms such as polynomial regression, random forest, and support vector regression, we constructed catalytic descriptors based on 17 physical and chemical properties, and selected random forest as the best prediction model through the error analysis of the training set and test set. Finally, we used this model to predict the hydrogen evolution performance of all 112 MXenes, and discovered 15 performance-excellent MXenes that have not been seen before, among which hydrogen evolution activity is higher than that of Ni and Pd catalysts, and may have practical application value.Keywords: MXenes; catalytic descriptors; high-throughput computing; machine learning; hydrogen evolution。

光学 精密工程Optics and Precision Engineering第 29 卷 第 5 期2021年5月Vol. 29 No. 5May 2021文章编号 1004-924X( 2021)05-1127-09联合训练生成对抗网络的半监督分类方法徐哲，耿杰*，蒋雯，张卓，曾庆捷(西北工业大学电子信息学院，西安710072)摘要：深度神经网络需要大量数据进行监督训练学习，而实际应用中往往难以获取大量标签数据°半监督学习可以减小深度网络对标签数据的依赖，基于半监督学习的生成对抗网络可以提升分类效果,旦仍存在训练不稳定的问题°为进一步提高网络的分类精度并解决网络训练不稳定的问题，本文提出一种基于联合训练生成对抗网络的半监督分类方法，通 过两个判别器的联合训练来消除单个判别器的分布误差，同时选取无标签数据中置信度高的样本来扩充标签数据集，提高半监督分类精度并提升网络模型的泛化能力°在CIFAR -10和SVHN 数据集上的实验结果表明，本文方法在不同数量的标签数据下都获得更好的分类精度°当标签数量为2 000时，在CIFAR -10数据集上分类精度可达80.36% ;当标签 数量为10时，相比于现有的半监督方法，分类精度提升了约5%°在一定程度上解决了 GAN 网络在小样本条件下的过拟合问题°关键词：生成对抗网络；半监督学习；图像分类；深度学习中图分类号:TP391文献标识码：Adoi ：10. 37188/OPE. 20212905.1127Co -training generative adversarial networks forsemi -supervised classification methodXU Zhe , GENG Jie * , JIANG Wen , ZHANG Zhuo , ZENG Qing -jie(School of E lectronics and Information , Northwestern Polytechnical University , Xian 710072, China )* Corresponding author , E -mail ： gengjie@nwpu. edu. cnAbstract ： Deep neural networks require a large amount of data for supervised learning ； however , it is dif ­ficult to obtain enough labeled data in practical applications. Semi -supervised learning can train deep neuralnetworks with limited samples. Semi -supervised generative adversarial networks can yield superior classifi ­cation performance ； however ， they are unstable during training in classical networks. To further improve the classification accuracy and solve the problem of training instability for networks ， we propose a semi -su ­pervised classification model called co -training generative adversarial networks ( CT -GAN ) for image clas ­sification. In the proposed model ， co -training of two discriminators is applied to eliminate the distribution error of a single discriminator and unlabeled samples with higher confidence are selected to expand thetraining set , which can be utilized for semi -supervised classification and enhance the generalization of deep networks. Experimental results on the CIFAR -10 dataset and the SVHN dataset showed that the pro ­posed method achieved better classification accuracies with different numbers of labeled data. The classifi ­cation accuracy was 80. 36% with 2000 labeled data on the CIFAR -10 dataset , whereas it improved by收稿日期：2020-11-04；修订日期:2021-01-04.基金项目:装备预研领域基金资助项目(No. 61400010304);国家自然科学基金资助项目(No. 61901376)1128光学精密工程第29卷about5%compared with the existing semi-supervised method with10labeled data.To a certain extent, the problem of GAN overfitting under a few sample conditions is solved.Key words：generative adversarial networks；semi-supervised learning；image classification；deep learn­ing1引言图像分类作为计算机视觉领域最基础的任务之一，主要通过提取原始图像的特征并根据特征学习进行分类［11o传统的特征提取方法主要是对图像的颜色、纹理、局部特征等图像表层特征进行处理实现的，例如尺度不变特征变换法［21,方向梯度法［31以及局部二值法［41等。