电网可靠性评估的PSO-SVR评估模型

摘要

The complicated structure and big data size of city power systems are the problems of power systems reliability evaluation, which leads to the poor evaluation results of traditional power systems reliability evaluation method.Thus, particle swarm support vector regression algorithm (PSO-SVR) is presented to evaluate the power systems reliability in the paper to solve the problems of power systems reliability evaluation due to complicated structure of city power systems and big data size of city power systems, and power supply reliability rate is employed as evaluation index.Particle swarm support vector regression algorithm can solve the local extremum problem of traditional artificial neural network evaluation algorithm.Firstly, the feature parameters and evaluation index of power systems reliability evaluation are determined, and the structure of evaluation model is determined.Particle swarm optimization algorithm is used to optimize training parameters of support vector regression model.The experimental simulation results show that the evaluation accuracy of particle swarm support vector regression evaluation algorithm is higher than artificial neural networks.Therefore, power systems reliability evaluation algorithm based on particle swarm support vector regression has better application value.%城市电网结构复杂,数据量大是电网可靠性评估的难点,导致了传统的电网可靠性评估方法难以有效评估.为提高评估的精度和效率,提出一种基于粒子群支持向量回归法的电网可靠性评估的新方法解决电网可靠性评估的问题,采用供电可靠率作为评估指标,粒子群支持向量回归法能克服传统的人工神经网络可靠性评估方法易陷人局部极值.采用电网可靠性评估特征参数与评估指标,确定评估模型结构,再用粒子群优化算法优化支持向量回归模型参数.仿真结果表明,粒子群支持向量回归法可靠性评估精度高于人工神经网络.证明粒子群支持向量回归的电网可靠性评估方法具有更好的应用价值.