基于多元线性回归和改进粒子群算法的输电网监测点优化配置策略

MLR-MPSO-based Arrangement Optimization of Monitoring Points in Transmission Networks

输电网的安全稳定运行非常重要,一旦发生电压暂降就可能严重影响电力系统的安全和正常运行。为获取输电网的运行情况,需配置一系列数据采集设备。但是,配置大量监视器可能会导致高昂的投资和维护成本,且可能会产生不必要的冗余信息。为此,提出了一种基于多元线性回归和改进粒子群算法的输电网监测点优化配置策略,以确定监测器的最佳数量和位置。首先,为了减少监测设备的冗余,引入Mallow的Cp作为线性回归模型的评价标准,并对粒子群算法进行改进。然后,通过最小化未监控母线电压的误差平方和,对获得的整个系统的最优配置策略进行进一步筛选。最后,基于MATLAB/Simulink仿真平台,在IEEE 30-BUS系统中对所提出方法的正确性和有效性进行了验证。

The safe and stable operation of transmission networks is very important.Once a voltage sag occurs,it may seriously impact the operation of power system.Therefore a series of monitoring devices are mandatory in a transmission network in order to obtain the operation status.However the deployment of these monitors should consider efficiency since number-excessive planning results in not only high investment and maintenance costs,but also unnecessary redundant information.The present work made a preliminary attempt to propose and analyze the utility of a MLR-MPSO-based strategy of determining the optimal number and locations of monitoring points in a transmission network.First,in order to reduce the redundancy of monitors,Mallow′s Cp was introduced as the evaluation criterion for the linear regression model,and the particle swarm optimization algorithm was modified.Then taking minimization of the sum of squared errors in the unmonitored bus voltage as the objective,the candidate arrangement strategies for the entire system were further selected.The correctness and the effectiveness of the proposed strategy was verified by a MATLAB/Simulink simulation using the IEEE 30-BUS system.