Using neural networks to solve testing problems

This paper discusses using Neural Networks for diagnosing circuit faults. As a circuit is tested, the output signals from a Unit Under Test can vary as different functions are invoked by the test. When plotted against time, these signals create a characteristic trace for the test performed. Sensors in the ATS can be used to monitor the output signals during test execution. Using such an approach, defective components can be classified using a Neural Network according to the pattern of variation from that exhibited by a known good card. This provides a means to develop testing strategies for circuits based upon observed performance rather than domain expertise. Such capability is particularly important with systems whose performance, especially under faulty conditions, is not well documented or where suitable domain knowledge and experience does not exist. Thus, neural network solutions may, in some application areas, exhibit better performance     

基于无刷直流电机的电动车控制器研究与设计

传统的电动车用无刷直流电机(BLDCM)控制器功率逆变电路,为实时检测相电流,通过电阻进行相电流采样并反馈给微控制单元(MCU)进行电流环闭环控制,然而,在功率较大的控制器电路应用中,大电流会导致功率损耗增加。为解决这一问题,以STM32系列微型单片机为核心,设计了一套电动车控制器控制电路。重点阐述BLDCM硬件控制。为改进电源短路保护电路,提出了一种新型的MOS管内阻采样方法,并制作铝基板PCB。测试结果表明:该控制电路测控性能稳定,能够应用于工业生产中。     

径向基神经网络在非线性非定常气动力建模中的应用研究

基于南航NH-2风洞中某飞机模型大迎角大振幅单自由度偏航、滚转及偏航-滚转耦合的谐波、阶跃运动实验数据,应用径向基神经网络,研究人工神经网络描述非线性非定常气动力特性的能力.研究结果表明,所建立的径向基神经网络模型的预测结果与训练数据和验证数据都符合得很好,说明神经网络建模方法可以有效地对高度非线性的气动力进行建模.研究还表明,用神经网络建立模型时所需要的风洞实验数据可以减少,从而提高风洞实验效率、减少风洞实验的时间和成本.     

Force measurement using strain-gauge balance in shock tunnel based on deep learning

When a force test is conducted in a shock tunnel, vibration of the Force Measurement System (FMS) is excited under the strong flow impact, and it cannot be attenuated rapidly within the extremely short test duration of milliseconds order. The output signal of the force balance is coupled with the aerodynamic force and the inertial vibration. This interference can result in inaccurate force measurements, which can negatively impact the accuracy of the test results. To eliminate inertial vibration interference from the output signal, proposed here is a dynamic calibration modeling method for an FMS based on deep learning. The signal is processed using an intelligent Recurrent Neural Network (RNN) model in the time domain and an intelligent Convolutional Neural Network (CNN) model in the frequency domain. Results processed with the intelligent models show that the inertial vibration characteristics of the FMS can be identified efficiently and its main frequency is about 380 Hz. After processed by the intelligent models, the inertial vibration is mostly eliminated from the output signal. Also, the data processing results are subjected to error analysis. The relative error of each component is about 1%, which verifies that the modeling method based on deep learning has considerable engineering application value in data processing for pulse-type strain-gauge balances. Overall, the proposed dynamic calibration modeling method has the potential to improve the accuracy and reliability of force measurements in shock tunnel tests, which could have significant implications for the field of aerospace engineering.     

Tapped-inductor filter assisted soft-switching PWM DC-DC power converter

A novel high-frequency transformer linked full-bridge type soft-switching phase-shift pulsewidth modulated (PWM) controlled dc-dc power converter is presented, which can be used as a power conditioner for small-scale photovoltaic and fuel cell power generation systems as well as isolated boost dc-dc power converter for automotive ac power supply. In these applications with low-voltage large-current sources, the full-bridge circuit is the most attractive topology due to the possibility of using low-voltage high-performance metal-oxide-semiconductor field-effect transistor (MOSFET) and achieving high efficiency of the dc-dc power converter. A tapped-inductor filter including the freewheeling diode is newly implemented in the output stage of the full-bridge phase-shift PWM dc-dc converter to achieve soft-switching operation for the wide load variation range. Moreover, in the proposed converter circuit, the circulating current is effectively minimized without using additional resonant circuit and auxiliary power switching devices. The practical effectiveness of the proposed soft-switching dc-dc power converter was verified in laboratory level experiment with 1 kW 100 kHz breadboard setup using power MOSFETs. Actual efficiency of 94-97% was obtained for the wide duty cycle and load variation ranges.     

A Stepping Motor Drive Circuit Using Magnetic Core Logic

A new magnetic core circuit that is used to supply power in the proper phase and time sequence to a bidirectional stepping motor is described. This circuit will power a standard voltage stepping motor from a nonstandard voltage power supply and includes a means for maintaining a constant energy pulse excitation over a wide range of dc power supply input voltage, a means for supplying power to the winding of the motor on pairs of isolated conductors, and a means for conductively isolating the low-level logic circuits from the high-level power circuits.     

Flyback Converter Output Voltage Stabilization

The expression of the flyback converter output voltage (output power) is derived as a function of the supply voltage, load resistance, transformer ratios, transistor current gain, and base-circuit resistor value. Switching period and duty cycle are also calculated. A converter circuit is designed having stabilized output voltage, with respect to supply voltage, at constant load. The transistor base current is controlled by the supply voltage, via a nonlinear circuit. This feedforward circuit approximates with logarithmic characteristics the ideal hyperbolic dependence of the transistor base current as a function of the supply voltage. The converter has high performance and low cost. A cheaper circuit variant is presented, in which the high-voltage control transistor was eliminated.     

飞机二次电源控制器监测电路的分析与计算

对于某些飞机二次电源供电的机载设备,采用两台变流机作为电源,并配备一台控制器。当主变流机发生故障,控制器能使主弯流机停止工作并起动备用变流机向设备供电。文章应用对称分量法分析计算了主变流机在上述两种状态下,监测电路能输出差别甚大的电压信号,从而如实反映主变流机正常工作和故障两种状态。     

音响功放开关电源输出功率检测电路的设计与实现

为了使开关电源很好地应用于音响功放,设计了一种开关电源输出功率检测电路,给出了电路的原理和具体实现方法.经测试证明该电路具有检测准确、动态性能良好的特点.     

Single-phase power-factor-correction AC/DC converters with threePWM control schemes

Three pulse-width modulation (PWM) control schemes for a single-phase power-factor-correction (PFC) AC/DC converter are presented to improve the power quality. A diode bridge with two power switches is employed as a PFC circuit to achieve a high power factor and low line current harmonic distortion. The control schemes are based on look-up tables with hysteresis current controller (HCC) to generate two-level or three-level PWM on the DC side of diode rectifier. Based on the proposed three control schemes, the line current is driven to follow the sinusoidal current command which is in phase with the supply voltage, and two capacitor voltages on the DC bus are controlled to be balanced. The simulation and experimental results of a 1 kW converter with load as well as line voltage variation and shown to verify the proposed control schemes. It is shown that unity PFC is achieved using a simple control circuit and the measured line current harmonics satisfy the IEC 1000-3-2 requirements     

基于神经网络的预膜雾化模型

为了实现对预膜雾化质量的快速预估，基于神经网络建立了以结构参数和工况参数为输入,SMD值为输出的预膜雾化模型。使用试验数据训练和测试网络模型,结果显示,模型在较宽的工作范围内具有较高的精度和较好的泛化能力。SMD预测值随输入参数的变化规律符合试验结论,可作为给定工况参数下预膜式雾化装置的结构参数优化依据。神经网络能够较好地学习预膜雾化过程中的隐含规律。随着供油压差的增大,预膜器逐渐出现阻雾化效应,导致其雾化效果比单喷嘴和直混式雾化均差。     

电压变化时Boost PFC变换器的混沌和分叉现象

利用小信号模型对开关电源控制器进行设计过程中常出现混沌分叉等不稳定现象。以工程设计的角度,建立了平均电流型BoostPFC的精确电路模型,分析了系统中存在的非线性现象,结合非线性分叉和混沌理论对这类非线性现象进行了详细的分析和推导。     

CIECA: Application to Current Programmed Switching Dc-Dc Converters

Sundstrand Advanced Technology Corporation The current injection equivalent circuit approach (CIECA) to modeling switching converter power stages is extended to model the current programmed converter power stages operating in fixed frequency, continuous inductor conduction mode. To demonstrate the method, modeling is carried out for the buck, boost, and buckboost converters to obtain small-signal linear equivalent circuit models which represent both input and output properties. The results of these analyses are presented in the form of linear equivalent circuit models as well as transfer functions. Though current programmed converters exhibit single-pole response, the addition of artificial ramp changes converters to exhibit well damped two-pole response. This has been investigated for the first time using CIECA. The results of these analyses are presented in the form of linear equivalent circuit models as well as transfer functions.     

基于DSP的永磁同步电机控制系统硬件设计

以小功率永磁同步电机(PMSM)为研究对象,结合数字信号处理器TMS320F2812功能特点,给出了一套PMSM驱动控制系统硬件设计方案。详细阐述了功率驱动主电路、反馈信号检测电路以及供电电路的设计,介绍了主要元器件选型和参数计算方法。基于设计的硬件平台,对PMSM调速控制系统进行了测试。试验结果表明,所设计的控制系统硬件设计可靠、性能稳定、控制精度高。     

Implementation of a three-phase high-power-factor rectifier with NPC topology

A three-phase four-wire power factor corrector based on neutral-point-clamped (NPC) topology is adopted to reduce the current harmonics and increase the input power factor. Using the NPC topology, the voltage stress of power switches can be reduced to the half of the dc-link voltage. With the dc-link voltage controller and the phase locked loop circuit, the balanced and sinusoidal line currents can be drawn from the ac supply system under the balance and unbalance mains voltages. The hysteresis current comparators are adopted in the current control loop to track the line current commands. Three voltage levels are generated on the ac terminal to the neutral point. Simulations and experiments are provided to verify the validity and the effectiveness of the proposed control scheme.     

小束斑电子束脉冲轰击偏压电源的研制

为了获得一种性价比较高的电子束表面处理技术,采用定频调压电路,研制出一种新型偏压脉冲电源。这种电源由偏压主电路拓扑、偏压变压器和偏压整流滤波电路组成,其中偏压主电路由偏压基值发生电路和偏压脉冲发生电路组成。偏压基值发生电路控制偏压脉冲基值,偏压脉冲发生电路控制偏压脉冲频率和占空比。脉冲偏压电源的脉冲基值、脉冲频率、占空比均连续可调,相应实现脉冲束流的幅值、束流频率和占空比的调节。检测分析了脉冲偏压电源的输出电压波形,带载时的输出束流波形,并观察分析了电子束轰击后钨合金组织变化规律。结果表明所研制的脉冲偏压电源输出电压稳定、可调性好,能满足小束斑脉冲电子束轰击工艺的要求。     

综合化航电智能电源控制软件设计与实现

综合化航空电子系统已经成为高度复杂的实时系统,新一代航电系统的发展对电源模块的设计提出了新的要求：大电流供电、余度供电、故障隔离等。传统的电源模块已经不能适应这些需求,因此,需要设计具有电源控制及管理功能的智能电源及其控制软件。通过对一种智能控制电源软件设计及其控制策略的分析,详细描述了智能电源在新航电系统中的功能及应用,为智能电源在不同场合下的应用提供了探索。     

基于MPS4021的20W LED驱动电源设计

采用原边反馈方案设计出一款20W LED驱动电源,通过变压器初次级隔离省去了副边反馈电路,提高了驱动电源的安全性和可靠性,并节约了BOM成本。所设计的驱动电源实现了在全电压范围内输出电流精度达±5%、功率因数大于0.9、效率大于85%,使用寿命大于3万小时,电磁干扰符合EN55015限制要求,并且具备有过流保护和过压保护功能,对设计大功率LED驱动电源具有一定的参考价值。     

基于LMD与AO-PNN的中介轴承故障诊断方法

针对航空发动机中介轴承受噪声干扰大、传递路径复杂导致采用传统方法难以进行故障诊断的问题,提出了一种基于局部均值分解（LMD）与相关系数-能量比-峭度准则、结合天鹰座优化算法（AO）优化概率神经网络（PNN）的中介轴承故障诊断方法。使用LMD对传感器采集的振动信号进行分解;利用相关系数-能量比-峭度准则判决筛选分解得到的PF分量,重构筛选后的信号;计算重构信号的多尺度排列熵（MPE）,以构建特征向量;通过AO优化的PNN的平滑因子,将优化后的神经网络用于中介轴承的故障诊断。基于中介轴承故障试验数据对诊断结果进行了分析,结果表明：提出的方法可以有效诊断高背景噪声、复杂路径干扰下的航空发动机中介轴承的典型故障,与粒子群优化的概率神经网络方法（PSO-PNN）和传统的PNN方法相比,其诊断准确率分别提高了3.875%和8.125%,具有较好的全局收敛性和计算鲁棒性。     

电梯驱动系统测试台设计

介绍了采用四象限变频静止电源、ACS880变频器作为电梯驱动系统的试验电源和加载系统电源的电梯驱动系统测试台方案。测试台采用Yokogawa高精度WT1800E六通道电参数测量仪、CT系列宽频电流传感器和Kistler双量程转矩转速传感器,实现了电梯驱动系统的高精度测试;采用能量闭环回馈的方式进行对拖加载试验,能够满足30 kW及以下的电梯驱动系统的综合性能测试。方案试验结果表明:设备运行稳定可靠,试验控制灵活,操作简单方便,具有较高的设计参考价值。