Optimal Configuration of Voltage Sag Monitoring Points Based on Ridge Regression and Improved Genetic Algorithm

The distribution network is directly facing the vast number of power users. Once the voltage sag occurs, it is liable to cause problems such as equipment outage, factory shutdown, residential power interruption, etc., and even seriously affect the safe and normal operation of the power system. To understand voltage sag situations in a network, a series of voltage sag monitoring devices need to be configured first. However, installing monitoring devices in buses may generate unnessary redundant information. Additionally, configuring a large number of monitors may result in high costs of investment and maintenance. To this end, this paper proposes a method for optimising the configuration of system monitoring points based on multiple linear regression and an improved genetic algorithm for determining the optimal number and location of monitors. Firstly, in order to reduce the redundancy of monitoring devices, Mallow's Cp is introduced as the evaluation criterion of the sublinear regression model and set as the fitness function of the improved genetic algorithm. Then, the obtained optimal configuration strategy for the whole system is further filtered by minimising the sum of squared errors for the voltages of unmonitored buses. Finally, based on the Matlab/Simulink simulation platform, the correctness and effectiveness of the proposed method is veried in the IEEE 33-bus system.