基于免疫优化的电力系统电压安全非线性预测控制  

NONLINEAR PREDICTIVE CONTROL SCHEME WITH IMMUNE OPTIMIZATION FOR VOLTAGE SECURITY CONTROL OF POWER SYSTEM

作　　者：李艳君[1] David J.Hill 吴铁军[1] 

机构地区：[1]浙江大学工业控制技术国家重点实验室,浙江省杭州市310027 [2]香港城市大学电子工程系

出　　处：《电力系统自动化》2004年第16期25-31,共7页Automation of Electric Power Systems

基　　金：国家重点基础研究专项经费资助项目(2002CB312200);香港城市大学研究经费资助项目(9380026)

摘　　要：提出了一种基于免疫优化的非线性预测控制方法来求解电力系统电压安全控制问题。采用基于非线性微分一代数方程的电力系统模型来预测系统行为。提出一种目标分段协调方法,以便在不同的预测时段根据系统响应情况调整各子目标的重要性,通过Pareto意义下的目标加权将这些子目标集成为一个集总目标函数。提出一种免疫算法,用具有多基因链结构的抗体来表达复杂优化问题的候选解特征;采用模式识别技术提取优良抗体的基因模式,并利用算法的学习和记忆能力识别各预测时段内已求解的优化问题类型,为未来预测时段内的最优解搜索过程估计出较好的初始解,加快最优解搜索速度。将此方法和基于树搜索算法的非线性预测控制方法通过一个6母线电力系统实例进行了仿真研究,性能比较的结果表明,本文提出的算法具有更强的优化搜索能力和更好的实时性。An immune algorithm based nonlinear predictive control scheme is proposed to solve power system voltage security control problems. A nonlinear differential-algebraic model is used to predict system behavior. A gradational targeting method is developed to decompose global horizon control targets into sub-objectives in receding prediction intervals via Pareto-type weighting functions. A novel immune algorithm is presented, using a multiple gene chain structure of antibodies to represent the solution candidates of the complicated optimization problem. A pattern recognition technique is employed to extract gene patterns of better antibodies. Similar antigen patterns are identified via learning, and memorized to create a better initial guess of solutions in order to accelerate the convergence of the optima searching procedure. System performance comparative results are reported based on the emergency voltage control of a six-bus example power system. The results indicate the promising application potential of the method proposed in this paper.

关 键 词：模型预测控制 非线性系统 免疫算法 电压安全控制 电力系统控制 

分 类 号：TM712[电气工程—电力系统及自动化] TP18[自动化与计算机技术—控制理论与控制工程]