【作者】 王婧；

【导师】 王学伟；

【作者基本信息】 北京化工大学 ， 控制科学与工程， 2020， 博士

【摘要】 进入21世纪,为解决能源与环境间的矛盾,能源的供给侧与需求侧发生了重大变革,我国《十三五规划纲要》中明确提出“深入推进能源革命,着力推动能源生产利用方式变革”。经过多年的技术创新与应用,落实习近平总书记提出的“创新、协调、绿色、开放、共享”五大发展理念,我国能源生产与利用方式在发生重大变化的同时,也为电能的准确计量带来了挑战。电网供给侧可再生新型能源大规模发电,其输出功率具有较强的不确定性、间歇性和随机波动性。需求侧大功率非线性动态负荷的广泛应用,导致负荷电流表现出复杂的快速随机动态波动特性,进而引起电能表电能计量严重超差。根据国家能源局统计数据,2019年,我国以非线性动态负荷使用为主的工业用电量占全社会用电量的67.1%,因而,由动态负荷信号快速随机波动所导致的电能计量1%的误差就可能造成几十亿元的经济损失。目前,国内外缺少对快速随机波动条件下智能电能表动态误差的测试理论与技术。本文以上述国家战略实施中存在的问题为导向,发现并提炼出智能电能表动态误差测试的科学问题,研究探索电能表动态误差测试的理论和方法,形成了原创性的研究成果,主要包括:(1)研究分析电网中实际动态负荷信号的典型本质特性,在此基础上,针对现有的电能表误差测试信号模型无法反映实际动态负荷信号快速随机波动特性的问题,建立了一种新的畸变波形m序列伪随机动态测试信号结构化参数模型,并研究了此类信号的产生方法,所提出的测试信号模型满足电能表动态误差测试信号建模的要求,为开展智能电能表动态误差测试提供了有效的解决方法。(2)为提高智能电能表动态误差的测试效率,根据压缩感知理论中的测量矩阵线性编码调制理论,采用结构化方法,构建正交伪随机测量矩阵,通过矩阵映射产生正交伪随机幅度调制函数,建立畸变波形正交伪随机动态测试信号模型。使其在反映实际电网中动态负荷典型本质特性的同时具备紧凑性,提高了电能表动态误差的测试效率。解决了压缩感知理论在工程领域应用的难题。(3)针对国内外电参量测量领域广泛使用的窗函数卷积算法在快速随机动态条件下测量准确度明显降低的问题。基于压缩检测信号处理理论,分析离散畸变波形伪随机动态瞬时功率测试信号的频域稀疏性,通过构建最小误差有功功率检测滤波器,提出了动态电能量值准确测量的非交叠移动压缩检测(Nonoverlapping moving compressive measurement,NOLM-CM)算法,在仿真与实验条件下,验证了 NOLM-CM算法具有更高的准确度。为智能电能表在快速随机动态条件下的电能量值准确测量提供指导。(4)针对智能电能表动态误差测试,在所提出的两类畸变波形伪随机动态测试信号模型的基础上,定义测试信号的游程似然函数,建立智能电能表动态误差的似然函数间接测试方法,解决了从动态参考电能量值到稳态参考电能量值的溯源问题。其次,搭建智能电能表动态误差测试系统,实验验证了本文所建立的畸变波形伪随机动态测试信号和电能表动态误差似然函数间接测试方法的有效性,且测量不确定度显著降低。本文从理论研究到仿真分析,再到实验验证,形成了智能电能表动态误差测试的完整理论体系,解决了测试关键技术,研究成果对保证快速随机动态波动条件下电能的准确计量与公平交易,促进电能替代绿色发展与创新发展,具有重要意义和广阔的应用前景。更多还原

【Abstract】 In the twenty-first century,in order to solve the contradiction between energy and environment,great changes have taken place in the supply side and demand side of energy grid.In 13th five year plan outline of the China,it is clearly proposed that "further promote the energy revolution and strive to promote the transformation of energy production and utilization mode".After years of technological innovation and application,by implementing the five development concepts of "innovation,coordination,green,openness and sharing" put forward by general secretary Xi Jinping,our energy production and utilization patterns have changed significantly,at the same time,it has also brought challenges to the accurate measurement of electrical energy.The output power of large scale renewable energy generation on the supply side of energy grid has strong uncertainty,intermittent and random volatility.The wide application of large power nonlinear dynamic loads on the demand side leads to the complex and rapid dynamic fluctuation of load current,which leads to the serious over tolerance of electrical energy measurement for smart electricity meters.According to the statistics of nation energy administration,in 2019,the industrial power consumption dominated by nonlinear dynamic loads accounts for 67.1%of the total social electrical energy consumption Therefore,the 1%error of electrical energy measurement caused by the rapid random fluctuation of dynamic loads may cause several billion yuan of economic loss.However,at present,there is no corresponding theory and technology for the dynamic error testing of smart electricity meter under the condition of rapid fluctuation at home and abroadGuided by the above problems in the implementation of the national strategy,the scientific problems for the dynamic error testing of smart electricity meter are found and refined in this paper,theory and method of the smart electricity meter dynamic error test are researched and explored.The original research results are formed,which mainly including(1)The typical intrinsic characteristics of actual dynamic load singals in the grid are studied and analyzed.On this basis,aiming.at the problem that the existing test signals for electricity meter error testing can not reflect the fast and random fluctuation characteristics of actual dynamic load signals,a new structual parameter model of distorted waveform m-sequence pseudorandom dynamic test signal is established,and the signal generation method is proposed.The test signal model meets the requirements of the dynamic error test signal modeling for smart electricity meter,and provides an effiective solution for the dynamic error testing of smart electricity meter.(2)In order to improve the efficiency of the smart electricity meter dynamic error testing,according to the linear code modulation theory of measurement matrix in the compressive sensing theory,an orthogonal pseudorandom measurement matrix is constructed by using the structural method,the orthogonal pseudorandom amplitude modulation function is generated by matrix mapping,and the distorted waveform orthogonal pseudorandom dynamic test signal model is proposed.It can reflect the typical intrinsic characteristics of actual dynamic load signals,at the same time,it has compactness,which improves the efficiency of smart electricity meter dynamic error test.Meanwhile,the problem of the application using compressive sensing theory in engineering field is solved.(3)The windowed algorithm,which is widely used in the field of electrical parameter measurment at home and abroad,has the problem that the measurement accuracy is obviously reduced under fast random dynamic conditions.Based on the compressive measurement signal processing theory,the frequency-domain spareness of pseudorandom dynamic instantaneous power test signal is analyzed.By constructing a minimum error active power measurment filter,a nonoverlapping moving compressive measurement(NOLM-CM)algorithm is proposed for the dynamic electrical energy accurate measurement.In the simulation and experimental conditions,the NOLM-CM algorithm is verified to have higher accuracy.It can provide guidance for the accurate measurement of dynamic electrical energy measurement under fast random dynamic conditions.(4)Aiming at the smart electricity meter dynamic error test,based on the two kinds of distorted waveform pseudorandom dynamic test signals,the run length likelihood functions of test signals are defined,and a dynamic error likelihood function indirect test method of smart electricity meter is establised,which solves the traceability from dynamic reference electrical energy to steady-state reference electrical energy.Meanwhile,the dynamic error test system of smart electricity meter is builted.Experimental results verify the validity of the distorted waveform pseudorandom dynamic test signal models and dynamic error likelihood function indirect test method.The measurement uncertainty is significantly reduced.In this paper,from theoretical research to simulation analysis,then to experimental verifications,a complete theorectical system of smart electricity meter dynamic error test has been formed,and the key technologies have been solved.The research results have broad application propects to ensure the accurate measurment and fair trading of electrical energy under the condition of rapid random dynamic fluctuations,and promote the green development and innovation development of electrical energy substitution.更多还原