Research of the Unity Theory Between Three-level Space Vector and Carrier-based PWM Modulation Strategy

两电平空间矢量脉宽调制（space vector pulse widthmodulation,SVPWM）与正弦脉宽调制（sinusoidal pulse widthmodulation,SPWM）存有一定的本质关系,对于三电平,在常见的8段式SVPWM序列中可以证明,SVPWM与SPWM本质上也存在某种统一。实际的SVPWM调制中,由于要考虑中点电压平衡、减少谐波等其它性能问题,一般会利用开关冗余状态得到8段以上的开关调制序列。目前的文献中还未曾有对这些情况进行统一性地验证,因此三电平SVPWM与SPWM的统一理论存在一定的不足。研究发现,可通过严格地理论推导将传统意义上的调制波分解为2个子调制波,再进行SPWM调制,可以得到与SVPWM完全吻合的序列。该文在此方法基础上深入研究了8段以上（10、12、14段）的SVPWM调制序列与SPWM的统一。经过严格地理论推导,成功实现了含任意段数的SVPWM序列与SPWM的统一。仿真结果证明了该理论的正确性。

Indirect Vector Control of Linear Induction Motors Using Space Vector Pulse Width Modulation

Vector control schemes have recently been used to drive linear induction motors(LIM)in high-performance applications.This trend promotes the development of precise and efficient control schemes for individual motors.This research aims to present a novel framework for speed and thrust force control of LIM using space vector pulse width modulation(SVPWM)inverters.The framework under consideration is developed in four stages.To begin,MATLAB Simulink was used to develop a detailed mathematical and electromechanical dynamicmodel.The research presents a modified SVPWM inverter control scheme.By tuning the proportional-integral(PI)controller with a transfer function,optimized values for the PI controller are derived.All the subsystems mentioned above are integrated to create a robust simulation of the LIM’s precise speed and thrust force control scheme.The reference speed values were chosen to evaluate the performance of the respective system,and the developed system’s response was verified using various data sets.For the low-speed range,a reference value of 10m/s is used,while a reference value of 100 m/s is used for the high-speed range.The speed output response indicates that themotor reached reference speed in amatter of seconds,as the delay time is between 8 and 10 s.The maximum amplitude of thrust achieved is less than 400N,demonstrating the controller’s capability to control a high-speed LIM with minimal thrust ripple.Due to the controlled speed range,the developed system is highly recommended for low-speed and high-speed and heavy-duty traction applications.

Realization of Space Vector Pulse Width Modulation Method Based on DSP

In order to compensate for the deficiency of Sine Pulse Width Modulation（SPWM）, on the base of analyzing the principle of space w tot pulse width modulation and being compared with SPWM, the method of solving working time of adjacent vector and the method of generate space voltage vector were introduced. The experiment to the inverter which consists of IGBT proves that SVPWM centrol algorithm can reduce harmonic effectively, it is beneficial to enhancing the utilization rate of voltage source inverter direct current power supply.

Research on Parallel Interleaved Inverters with Discontinuous Space-Vector Modulation

Parallel converter can significantly increase the capacity of the converter and improve the power quality of AC side, but the circulation which can lead to high switching loss and even damage the devices will easily exist in the direct parallel converters. In this paper, the average model of parallel interleaved inverters system to analyze the circulation current is shown, and the cross current is relevant to DC-bus voltage and the overlap time of zero vectors in the switching period. Based on this observation, a discontinuous space vector modulation without using zero vectors (000) is eliminate and suppress the zero-sequence current to entire system. Finally, the effectiveness of modulation strategy is verified by the simulations in this paper.

ANFIS Based Space Vector Modulation-DTC for Switched Reluctance Motor Drive

Direct torque control (DTC) of Switched reluctance motor is known straightforward control structure with similar execution to that of field situated control strategies. In any case, the part of ideal determination of the voltage space vector is one of the weakest focuses in a routine DTC drive because of adjustable switching frequency and high torque ripple. In this paper, ideal choice of voltage space vectors is accomplished utilizing ANFIS (Adaptive Neuro Fuzzy Inference System) with space vector Modulation. SVM-DTC gives consistent switching frequency and the proposed ANFIS controller’s structure manages the torque and stator flux error signals through the fuzzy deduction to get a yield that takes the type of space voltage vector. Simulation results accept the proposed evolutionary system with quick torque and flux reaction with minimized torque ripple and flux ripple.

基于CBAM-CGRU-SVM的Android恶意软件检测方法

随着Android恶意软件的种类和数量不断增多,检测恶意软件以保护系统安全和用户隐私变得越来越重要。针对传统的恶意软件检测模型分类准确率较低的问题,提出一种基于卷积神经网络(CNN)、门控循环单元(GRU)和支持向量机(SVM)的模型CBAM-CGRU-SVM。首先,在CNN中添加卷积块注意力模块(CBAM)以学习更多恶意软件的关键特征;其次,利用GRU进一步提取特征;最后,为了解决图像分类时模型泛化能力不足的问题,使用SVM代替softmax激活函数作为模型的分类函数。实验使用了Malimg公开数据集,该数据集将恶意软件数据图像化作为模型输入。实验结果表明,CBAM-CGRU-SVM模型分类准确率达到94.73%,能够更有效地对恶意软件家族进行分类。

Classification of power quality combined disturbances based on phase space reconstruction and support vector machines

Power Quality (PQ) combined disturbances become common along with ubiquity of voltage flickers and harmonics. This paper presents a novel approach to classify the different patterns of PQ combined disturbances. The classification system consists of two parts, namely the feature extraction and the automatic recognition. In the feature extraction stage, Phase Space Reconstruction (PSR), a time series analysis tool, is utilized to construct disturbance signal trajectories. For these trajectories, several indices are proposed to form the feature vectors. Support Vector Machines (SVMs) are then implemented to recognize the different patterns and to evaluate the efficiencies. The types of disturbances discussed include a combination of short-term dis-turbances (voltage sags, swells) and long-term disturbances (flickers, harmonics), as well as their homologous single ones. The feasibilities of the proposed approach are verified by simulation with thousands of PQ events. Comparison studies based on Wavelet Transform (WT) and Artificial Neural Network (ANN) are also reported to show its advantages.

Harmonic Domain Modelling of Space Vector Based STATCOM

This paper presents frequency domain method for harmonic analysis of space vector based STATCOM. Space Vector Pulse Width Modulation (SVPWM) method is an advanced PWM method. It is a best method among all the PWM techniques. It provides a freedom in a switching cycle for placement space vector. In this paper, the SVPWM is used for switching of STATCOM. The harmonic (or frequency) domain is a steady-state form of harmonic analysis method, which represents converters to their harmonic spectra. This paper presents harmonic analysis by means of harmonic domain for space vector based Static shunt converter (STATCOM). Performance of the STATCOM is evaluated in harmonic domain simulation studies in MATLAB environment.

Dynamic Harmonic Domain Modelling of Space Vector Based SSSC

This paper presents analytical frequency domain method for harmonic modeling and evaluation of Space Vector Pulse Width Modulation (SVPWM) based static synchronous series converter (SSSC). SVPWM is the best among all the PWM techniques. It gives a degree of freedom of space vector placement in a switching cycle. Dynamic modeling technique is used for space vector modulation (SVM) based voltage source converter that is adapted as a static synchronous series converter (SSSC) for harmonic analysis using dynamic harmonic domain. Performance of the SSSC is evaluated in dynamic harmonic domain simulation studies in MATLAB environment. The switching function spectra are necessary for harmonic transfer matrix which is calculated using Fourier series. This paper presents the analysis of harmonics for space vector based SSSC during steady state and dynamic condition.

Improved VSVPWM Modulation Method for Three-level NPC Inverter in Rural Power Grid

With the acceleration of agricultural electrification,a lot of nonlinear and shock loads appear in the rural power grid,and the resulting harmonic and reactive currents pollute the rural power grid more and more seriously.To solve the above problem,three-level neutral point clamped(NPC)inverters have been widely used,but their development is greatly restricted by the defect of neutral point voltage imbalance.In this paper,an improved virtual space vector pulse width modulation(VSVPWM)was proposed.Firstly,the mathematical models of various space vectors were established,and the influence of various space vectors on neutral point voltage was analyzed.The sum of the vector current at the neutral point was zero and the voltage control at the neutral point was completed by.introducing the time offset into different switching times of the redundant small vector.This method was simple in design and avoided the redundant calculation of the traditional VSVPWM method and tedious switch sequence design.This balancing control strategy could greatly reduce the influence of virtual vectors on neutral point voltage and effectively control the low-frequency oscillation of neutral point voltage.The validity of the method was verified by establishing a matlab simulation model.

基于混合采样和SE_ResNet_SVM的不平衡多分类研究

针对结构化多分类算法中不平衡数据集类别分布不均导致分类难度增加的问题,本文提出了一种基于混合采样、压缩与激励(squeeze and excitation, SE)模块、改进深度残差网络和支持向量机(support vector machines, SVM)的网络模型SNSMRS (SMOTEENNmixed residual networks-SVM network)。首先,通过合成少数过采样和编辑最近邻技术来改善数据分布;然后,构建融合SE模块与通过融合批次归一化和群组归一化的深度残差网络来提取特征;最后,通过SVM进行输出网络模型。其中,SE模块增强了模型对特征的区分能力,提升了模型的鲁棒性;基于融合归一化的残差网络受批次大小的影响较小,并且避免了传统神经网络梯度消失和精度退化等问题,增强了网络的稳定性与准确度;SVM可以根据特征向量在空间上的分布进行全部特征的分割,特征利用率高,提高了模型的分类精度。在7个不同规模和领域的非平衡公开数据集上进行了对比和消融实验,结果表明,本文所提的网络模型SNSMRS不仅优于其他深度学习模型,而且相对于未改良的ResNet,Macro-F1和G-mean值分别提升了约3%和4%,同时在4个数据集上的Macro-F1和G-mean值均超过了95%。

An Improved Hybrid Space Vector PWM Technique for IM Drives

In this paper, an improved hybrid space vector pulse width modulation (HSVPWM) technique is proposed for IM (induction motor) drives. The basic principle involved in the proposed random pulse width modulation (RPWM) cuddled SVPWM is amalgamating the pre-calculated switching timings for various sections of hexagonal space vector boundary and the random selection of carrier between two triangular signals, in order to disband acoustic switching noise spectrum with improved fundamental component. The arbitrary selection between triangular carriers, which is decided by digital signal states (Low or High) of the linear feedback shift register (LFSR) based pseudo random binary sequence (PRBS) generator. The SVPWM offers a control degree of freedom in terms of positioning of vectors inside every sampling interval and hence it has six possible variants of the voltage vectors arrangements in each sector. The developed HSVPWM is thoroughly analyzed in using the MATLAB? based simulation for all SVPWM variants. From the simulation and experimental results viz. harmonic spectrum, harmonic spread factor (HSF), total harmonic distortion (THD) etc., and the superiority of the proposed scheme such as better utilization of DC bus and the randomization of the harmonic power are evidenced. For the practical implementation, Xilinx XC3S500E FPGA device has been used.

基于反向传播神经网络的SVM技术在电压型变流器中的应用研究

在分析三相电压型变流器空间矢量调制(SVM)技术基本原理的基础上,提出了一种基于反向传播神经网络结构的 SVM 技术(CPN-SVM)的实现方法。该方法采用CPN 竞争层来计算 SVM 中各个矢量的具体作用时间,避免了计算正弦函数这一非线性运算,缩短了计算时间,采样周期的可进一步缩短。仿真和实验表明:CPN-SVM 除了具备 SVM 的固有优点外,还有下述几个显著优点:①大大降低了整个控制系统的软硬件成本,提高了对开关瞬态位置判断的准确性;②随着采用周期的缩短,最大开关频率相应增大,从而提高了整个系统的传输带宽③避免了由于计算误差给 SVM 波形中所带来的附带谐波。

基于支持向量机(SVM)的回采工作面瓦斯涌出混沌预测方法研究

针对瓦斯涌出传统的线性预测方法存在的问题,根据瓦斯涌出时间序列固有的确定性和非线性,利用混沌动力系统的相空间延迟坐标重构理论,结合基于机器学习理论的支持向量机(SVM),建立基于SVM理论的瓦斯涌出混沌时间序列预测模型。经Ⅱ1024回采工作面瓦斯涌出时间序列仿真计算,仿真结果显示该预测模型具有比传统的回归方法更好的泛化能力,预测方法具有很高的预测精度。同时,该模型具有以往传统机器学习的瓦斯涌出预测模型建立简便、训练速度快等优点。由于充分考虑瓦斯涌出时间序列的混沌性,并利用SVM预测的优良特性,使得预测更科学。

改进的SVM决策树分类算法

为解决多类分类问题,在分析SVM决策树分类器及存在问题的基础上,通过引入类间可分离性测度,并将其扩展到核空间,提出一种改进的SVM决策树分类器。实验表明了该分类算法对提高分类正确率的有效性。

中点电压偏移对SVM控制的三电平逆变器的影响及补偿措施

简要介绍了三电平逆变器的传统SVM控制算法。针对逆变器负载运行时,负载动态特性对于三电平逆变器的中点电压的影响,以及由中点电压偏移造成的空间矢量图的变化,引入中点电压前馈控制方法,进行新的矢量分区和作用时间计算,从而改进逆变器带负载时的输出电流波形。在Matlab软件平台上进行仿真,结果表明改进的SVM得到的输出电流波形较传统的SVM控制方法下的输出电流波形有明显的改善。

无轴承异步电机SVM-DTC系统研究

针对无轴承异步电机这一多变量、非线性、强耦合的系统,采用空间电压矢量调制技术(SVM)与直接转矩控制(DTC)相结合的方法来控制无轴承异步电动机。阐述了无轴承异步电机的工作原理,给出无轴承异步电机的数学模型,将整个控制系统分为旋转控制系统模块和悬浮控制系统模块,旋转模块引入SVM方法,并详细给出了基于定子磁链矢量偏差法的无轴承异步电动机SVM-DTC的分析与实现过程;悬浮模块采用电流追踪型PWM逆变器控制的方法。实验和仿真结果表明:转矩波动显著减少,运行更加平稳,系统具有良好的动态和静态性能。

基于商空间粒度理论的大规模SVM分类算法

利用商空间粒度理论对已有的SVM分类算法进行改进,给出了一种新的SVM分类算法——SVM-G。该算法将SVM分类问题划分成两个或多个子问题,从而降低了SVM分类复杂度。实验表明,改进的算法适用于处理大数据量的样本,能在保持分类精度的情况下有效地提高支持向量机的学习和分类速度。