基于广域多通道量测信号的低频振荡模态参数辨识与安全预警方法研究

项目名称： 基于广域多通道量测信号的低频振荡模态参数辨识与安全预警方法研究

项目编号： No.51507014

项目类型： 青年科学基金项目

立项/批准年度： 2016

项目学科： 电工技术

项目作者： 肖辉

作者单位： 长沙理工大学

项目金额： 20万元

中文摘要： 针对目前基于WAMS量测信号的低频振荡模态参数辨识方法对含噪声、非平稳时变的振荡信号处理效果欠佳，尤其在抗有色高斯噪声及算法计算效率方面仍有待提高，且基于WAMS量测信号的低频振荡监测预警分析主要侧重事故后的告警分析，并未真正意义上起到预警作用的特点，本项目研究内容包括：1）研究一种较强鲁棒性和适应性的改进型TLS-ESPRIT的低频振荡模态参数辨识方法；2）针对现有WAMS系统中低频振荡模态参数辨识是基于SISO信号分析法，存在遗漏关键振荡模式和计算效率偏低的特点，研究一种具有较为出色的并行计算能力，能显著提高主导模式辨识速度和精度的MIMO信号分析法；3）研究一种综合考虑系统安全稳定裕度、系统风险、模式信号能量权重因子的综合预警指标。该项目的研究可望建立一个可靠高效的低频振荡监视与预警体系，为调度运行人员决策提供重要依据与技术支持，也是稳控技术研究的发展趋势。

中文关键词： 电力系统；广域测量系统；低频振荡；模态参数；预警指标

英文摘要： The existing low frequency oscillation modal parameter identification methods using WAMS signals have poor effects on noised, non-stationary and time-varying oscillation signals, especially in the resistance of colored Gaussian noise and algorithm efficiency. In addition, the present low frequency oscillation monitoring and early-warning methods focus mainly on alarm analysis after accidents and don’t really play a role in early-warning. In view of these, this project will study as follows:1) An improved TLS-ESPRIT method will be studied for modal parameter identification, which has the abilities of robustness and flexibility; 2) A MIMO signal analysis method will be studied which has excellent parallel computing ability and be able to quickly and accurately obtain dominant oscillation mode characteristics; 3) A comprehensive early-warning index considering system stability margin, risk and signal energy weight will be established. This project will be good for building a reliable and efficient low frequency oscillation monitoring and early-warning system and can provide technical supports for dispatchers, which is the trend for stability control technologies.

英文关键词： Power system;WAMS;low frequency oscillation;modal parameter;early-warning index