非平稳畸变信号条件下有功电能计量新方法研究

In modern grid, besides steady-state fundamental component and harmonic signals, there are unstable signals with complex quasi-periodic and time-variable characteristics. This has caused complex power flow change in the grid, and exposed serious inaccuracy and unreasonableness in the existing measurement methods. It is a topic with great theoretical significance and application values that researching on a novel measuring method of electrical energy under the conditions of non-stationary distortion signals. This project solves the problems of the accurate and reasonable measurement of active electrical energy under the conditions of non-stationary distortion signals using mathematic tools such as random signal analysis, functional analysis and wavelet analysis. First, functional series models of distortion signal in actual grid are established. Second, active power flow is analyzed using the methods of relative functions. Factors affecting the reasonable measurement of the active electrical energy are determined. A novel measuring method is presented for active electrical enery under the conditions of non-stationary distortion signals. Furthermore, frequency tracking algorithm of grid signals is studied . The algorithm provides accurate fundamental frequency to the construction of tight supported bi-orthogonal wavelet bases. Finally, tight supported bi-orthogonal wavelet bases are constructed according to the different characteristics of the non-stationary distortion signals. The adaptive wavelet algorithm which meets accuracy and real-time requirements is studied in order to measure the power and electrical energy of distortion signals accurately and effectively. A device is developmented to test and verify the correctness and the real-time performance of the novel methods presented.

现代电网中，除了平稳的基波和谐波信号之外，还出现了准周期、时变的非平稳畸变信号，由此引发了复杂电网功率潮流的变化，突显了现有电能计量方式的不准确、不合理。研究非平稳畸变信号条件下电能计量新方法是电能计量领域一个极具理论意义及应用价值的课题。本申请项目采用随机信号分析、泛函分析及小波分析等方法解决非平稳畸变信号条件下有功电能准确合理计量的问题。建立与实际情况相符的电网信号的泛函级数模型；采用相关函数法进行有功功率潮流分析，确定影响有功电能计量合理性的因素，提出非平稳畸变信号条件下有功电能计量的新方法；研究信号频率跟踪算法，为构造紧支撑双正交小波基提供准确的基波频率；基于非平稳畸变信号的不同特征，构造相应的紧支撑双正交小波基，并研究满足准确性和实时性要求的自适应小波算法，实现畸变功率及电能的准确高效计量。试制非平稳畸变信号条件下电能计量新方法验证装置，验证所提方法的正确性与实时性。

随着电网中非线性负载的迅速增加，电能质量日趋恶化，这不仅严重影响电网安全高效的运行，而且对经典的电能计量理论、方法和仪表的设计都提出了新的挑战。现代电网中，除了平稳的基波和谐波信号之外，还出现了准周期、时变、非平稳的畸变信号，进而引发了复杂电网功率潮流的变化，深刻暴露了现有电能计量方式的不准确、不合理性。在国内，经济越发达地区电网畸变越严重，畸变导致了电能计量中大量电量流失。计量技术的提高是合理计费的基础，研究畸变信号条件下有功电能计量新方法不仅能促进电能计量技术的发展，为研发计量更准确合理的电能表提供理论基础，而且还能为解决电力部门的重大技术、经济问题，制定国家用电政策、法规提供理论依据。因此，该项研究不仅具有理论意义，而且具有现实意义和应用价值。.本项目应用wiener-G泛函级数建立了非平稳畸变信号条件下电网信号的数学模型，以电弧炉、半导体整流器件和电力机车三种典型畸变源为例，通过Matlab仿真验证了所建模型的正确性。采用相关函数法进行了有功功率潮流分析，研究了各功率潮流的物理意义及影响有功功率及电能计量合理性的因素，提出了非平稳畸变信号条件下有功电能计量新方法。鉴于电网中三相不平衡等因素导致的频率波动及有功电能计量误差，研究了基于unscented卡尔曼滤波器与加权相位平滑差分法的电网基波信号频率跟踪算法，为小波分解算法提供准确的基波频率。在50Hz存在白噪声干扰，频率、幅值及相位同时跳变，频率缓变三种情况下，频率跟踪误差为10-4 - 10-3 Hz。应用基于信息熵评价指标的多小波变换对谐波性负载和冲击性负载进行有功功率分解测量，测量准确度比应用Db40小波高一个数量级。试制非畸变信号条件下电能计量方法验证装置，针对不同畸变信号的小波分解算法测量准确度为10-4 - 10-3，实时性约2ms，验证了所提方法的准确性与实时性。