多级滤波的永磁同步电动机无位置传感器控制

基于滑模观测器的永磁同步电动机无位置传感器控制存在抖振问题。提出通过多级滤波器得到平滑反电势曲线,采用三级相同的低通滤波器串联滤波。该方法避免了传统单级滤波器带来的滤波不干净或过大的问题。然后对滤波后的反电势用锁相环原理提取转速转角信号,避免了传统估算转速转角的复杂性与误差性。本文做了以下分析:观测电流、观测反电势、估算转速转角、动态过程和负载突变运行,结果显示了该改进算法的准确性和有效性。     

永磁同步电机改进型滑模观测器无传感器控制

传统滑模观测器存在抖振和高次谐波的问题,难以满足系统高性能调速要求.为改善这一问题,设计并实现了一种将反电势估算值反馈引入到定子电流观测计算中,并采用扩展卡尔曼滤波器和饱和函数的改进型滑模观测器转子位置估算方法.通过实验结果表明,该改进型滑模观测器能在很大程度上削弱抖振和高次谐波带来的负面影响,并具有良好的鲁棒性及动态跟踪性能.Maflab仿真及DSP实验验证了该改进型滑模观测器的有效性与实用性.     

基于微分进化和SMO算法的烟气含氧量软测量

火电厂中烟气含氧量是一个难以测量的量,采用支持向量机中的序列最小优化(SMO)算法对其进行软测量,并采用改进的SMO算法提高建模速度.SMO算法的性能很大程度上依赖于其学习参数,选择合适的SMO参数是一个亟待解决的问题.而微分进化算法(DE)具有很强的全局搜索能力,在多峰函数的寻优问题上已表现出优异的性能.为此,采用DE算法选择SMO的参数,提出了基于DE算法的SMO参数选择方法.仿真表明,该方法能够准确预测烟气含氧量的变化,比用遗传(GA)算法和粒子群(PSO)算法优化SMO参数具有更高的精度和更快的速度.     

Using Sequential Unconstrained Minimization Techniques to simplify SVM solvers

In this paper, we apply Sequential Unconstrained Minimization Techniques (SUMTs) to the classical formulations of both the classical L1 norm SVM and the least squares SVM. We show that each can be solved as a sequence of unconstrained optimization problems with only box constraints. We propose relaxed SVM and relaxed LSSVM formulations that correspond to a single problem in the corresponding SUMT sequence. We also propose a SMO like algorithm to solve the relaxed formulations that works by updating individual Lagrange multipliers. The methods yield comparable or better results on large benchmark datasets than classical SVM and LSSVM formulations, at substantially higher speeds.     

Hot Working Characteristic of Superaustenitic Stainless Steel 254SMO

Hot deformation characteristic of superaustenitic stainless steel 254SMO has been studied by isothermal compression testing in the temperature range of 950–1,200 °C and strain rate range of 0.01–10 s-1.The activation energy of 496 kJ/mol was calculated by a hyperbolic-sine type equation over the entire range of strain rates and temperatures.In order to obtain optimum hot working conditions,processing maps consisting of power dissipation map and instability map were constructed at different strains.The power dissipation map exhibits two domains with relatively high efficiencies of power dissipation.The first domain occurs in the temperature range of 990–1,070 °C and the strain rate range of0.01–0.1 s-1.Microstructure observation in this domain indicates the partial dynamic recrystallization(DRX) accompanied with precipitation of tetragonal sigma phase.The second domain occurs in the temperature range of 1,140–1,200 °C and the strain rate range of 0.01–1 s-1with a peak efficiency of power dissipation of 39%,and in this domain,the microstructure observation reveals the full DRX.The instability map shows that flow instability occurs at the temperatures below 1,140 °C and the strain rates above 0.1 s-1.     

Fuzzy sliding mode observer with dual SOGI-FLL in sensorless control of PMSM drives

To improve the performance of permanent-magnet synchronous motor (PMSM) drives, a sensorless control scheme based on a sliding mode observer (SMO) with a fuzzy logic controller (FLC) and a dual second-order generalized integrator-frequency locked loop (DSOGI-FLL) is proposed in this paper. The major drawbacks of the conventional SMO, namely chattering phenomena, high-order harmonics and external noise, are discussed. These drawbacks affect the estimated accuracy of the SMO and reduce the control reliability of the system. To eliminate these drawbacks, an FLC is designed and integrated into the SMO to adjust the observer gain in a self-adaptive manner and to reduce the chattering; an existing dual synchronous frequency extract filter-phase locked loop (DSFF-PLL) is used to filter out the main components of high-order harmonics and to calculate the rotor position and speed precisely. Furthermore, to obtain an accurate fundamental frequency for the phase locked loop (PLL) and filter out the remaining harmonics and external noise signals, DSOGI-FLL processing is developed and incorporated into the DSFF-PLL. An overall PMSM sensorless control system based on the proposed SMO is designed, and an experimental platform using the TMS320F28335 DSP controller is built. Comparative experiments using the proposed SMO and the conventional SMO are performed to validate the effectiveness of the proposed FLC and the DSFF-DSOGI-FLL-PLL structures. Performance experiments of the overall proposed SMO-based sensorless control scheme are performed to verify the robustness and control reliability of the system. The results show that the proposed SMO has satisfactory performances and can be used in practical engineering.     

基于LSSVM与SMO稀疏化算法的短期负荷预测

针对最小二乘支持向量机(LSSVM)仅考虑等式约束,失去了支持向量机（SVM）解的稀疏性,影响了二次学习效率的问题,把最小二乘支持向量机与序列最小优化算法(SMO)相结合,后者应用改进的剪枝算法对解进行稀疏化处理,解决了上述问题;算法与最小二乘支持向量机相比降低了计算的复杂度,提高了计算精度。将此方法应用到电力系统短期负荷预测,与SVM和LSSVM算法比较表明该方法运行速度快,精度高,取得了较好的预测结果。     

支持向量机改进序列最小优化学习算法

为提高支持向量机序列最小优化学习算法的学习性能，提出了一种支持向量机改进序列最小优化学习算法，对传统SMO学习方法进行了多方面改进，从优化变量的选择和2个变量的优化方法分别提出具体可行的改进方法．改进后的SMO学习算法提高了学习速度，加快了网络收敛速度．基于改进SMO算法的仿真结果验证了改进SMO算法的有效性和优越性，并通过仿真，与原始算法进行了比较，显示了改进SMO算法的快速性．     

基于SMO的多层次文本分类法研究

在以往的自动文本分类研究中,大多比较流行的分类技术都是在一个层次上将文本分成几个类别。但随着信息检索的量越来越大,文本的种类将越来越多,仅仅通过一层对海量信息进行组织分类越来越不适合海量信息的检索工作,这种平坦式的分类组织难以进一步提高信息检索的速度。论文将SMO分类算法结合到文本分类研究中,通过构建多层支持向量机文本分类树,实现了基于SMO的多层次文本分类系统。     

高压变频器在电励磁同步电机上的应用

电励磁同步电机具有调速范围宽、过载能力强、功率因数可调等优点,广泛应用于大功率传动领域。为此,介绍了一种中性点钳位型(NPC)三电平高压变频器的系统结构及中性点电压平衡控制策略。采用电流频率(IF)控制和滑模观测器(SMO)相结合的无速度传感器矢量控制策略,控制电励磁同步电机变频起动与运行。MATLAB仿真结果验证了采用所提控制策略的变频器具有良好的起动和运行性能,为高压大功率传动应用领域提供了技术保障。