面向微网复杂源荷环境的电能质量动态跟踪及实时检测方法研究

The real-time and accurate dynamic tracking of parameters such as power quality magnitude, phase, and frequency, real-time detection of power quality disturbance start and end times, and identification of sources of disturbance are necessary requirements for micro-grid power quality control and optimization of power electronic control systems. In the micro-grid, there is a complex source and load environment in which nonlinear interactions occur between sources or between the source and the load through the electrical coupling of the system impedance. The complex source and load environment causes the traditional dynamic tracking and detection algorithms to have problems such as poor accuracy, missed detection, delay, and state mutation causing the algorithm to fail. The project uses the time-frequency domain method to analyze disturbance characteristics of the typical source and load output power and its change rate. The failure reason of the traditional Kalman dynamic tracking algorithm under the influence of complex disturbances revealed. AC change rate was introduced to establish a unified state space model. The optimal estimation theory is adopted to accurately describe the power AC signal and propose a pure sine wave tracking model. The fundamental frequency adaptive Kalman dynamic tracking algorithm based on the pure sine wave tracking model is proposed. The aircraft target maneuver tracking, detection and identification technology will be introduced. a disturbance feature value is extracted by detecting the parameter change of the tracking model. The detection and identification methods of source-load disturbance sources based on the change rate of AC power are proposed and discussed. The research on the project will overcome the influence of the time-frequency domain interference caused by the complex micro-grid source and load environment, and provide real-time accurate power quality parameter estimates and disturbance source information.

实时准确动态跟踪电能质量幅值、相位和频率等参数，实时检测电能质量扰动起止时刻及识别扰动来源，是微网电能质量治理和优化电力电子控制系统的必要需求。针对微网中各源之间、源荷之间通过系统阻抗电气耦合作用而存在非线性交互影响的复杂源荷环境，使传统动态跟踪和检测算法存在精度差、漏检、时延和状态突变造成算法失效等问题。课题采用时频域法分析典型源荷输出电量及其变化率扰动特征，揭示复杂扰动影响下传统卡尔曼动态跟踪算法失效原因，引入交流电量变化率建立统一状态空间模型，采用最优估计理论准确描述电力交流信号提出纯正弦波跟踪模型，提出基于该模型的基波频率自适应卡尔曼动态跟踪算法；引入飞行器目标机动跟踪、检测与识别技术，通过检测跟踪模型的参数变化提取扰动特征量，提出并探讨基于交流电量变化率特征的源荷扰动源检测识别方法。课题研究将克服复杂微网源荷环境带来的时频域干扰影响，提供实时准确的电能质量参数估计值和扰动源信息。

项目组完成工作为：(1)搭建了复杂微网环境MATLAB/Simulink仿真模型，设置短路故障、变压器投运和感应电机启动等情况采集了大量数据，分析了典型电压暂降源输出幅值及其变化率的电能质量扰动特征；(2)在分析传统KF算法时延、失效的原因基础上，验证了引入电量导数建立状态空间模型的有效性；(3)结合锁频环（FLL）和锁相环（PLL），基于所建动态跟踪模型，提出克服复杂电能质量扰动影响的动态跟踪算法；(4)将所提算法应用于电网动态谐波估计、有源滤波器基波正序分量提取、电压暂降特征提取及辨识、电网行波测距、基于电能质量优化指标的逆变器和储能双向变流器控制参数整定等领域；(5)仿真和实验结果表明，本项目研究内容和所提方法是有效可行的。. 通过四年的理论研究、仿真和实验工作，本项目组发表论文、申请专利及研究生培养数量已保质保量完成了项目计划书预定目标。预定目标为：(1)在国内外SCI/EI期刊录用论文4～6篇；(2)时机成熟时将申请国家发明专利1项；(3)培养硕士研究生4人以上。完成情况为：(1)发表期刊论文10篇（其中，SCI期刊论文5篇，中文核心期刊5篇）；通过努力完成了申请书中个人承诺，即在SCI的TOP期刊《IEEE Transactions on Industrial Electronics》独著发表1篇学术论文；(2)授权发明专利8项；(3)培养毕业硕士研究生9人，其中2人获国家研究生奖学金，1人获江西省政府研究生奖学金，12人次获省级竞赛奖励。另外，参加国内外学术合作交流会议15次。. 针对微网中各源之间、源荷之间通过系统阻抗电气耦合作用而存在非线性交互影响的复杂源荷环境，造成传统跟踪检测算法具有精度差、漏检、时延和失效等问题，本项目采用时频域法分析典型源荷输出电量及其变化率扰动特征，揭示复杂扰动影响下传统算法失效原因，引入交流电量变化率建立统一状态空间模型，采用最优估计理论准确描述电力交流信号提出正弦波动态跟踪模型，进而提出基于动态跟踪算法的电能质量扰动源实时检测方法。仿真和实验结果验证了所提方法的可行性和有效性，其克服了复杂微网源荷环境带来的时频域干扰影响，能够提供实时准确的电能质量参数估计和扰动源信息。