基于灰色关联分析和IPSO-LSSVM的线损预测模型研究

Research on Line Loss Prediction Model Based on Grey Correlation Analysis and IPSO-LSSVM

为了准确、全面评估10 kV配电网线损情况,给出了一种灰色关联分析和改进粒子群算法优化最小二乘支持向量机的线损预测方法。通过灰色关联度分析定量分析了电气指标和10 kV配电网线损之间的关联性,改进标准粒子群算法学习因子的变化规律,使用IPSO优化LSSVM的惩罚因子,建立IPSO-LSSVM预测模型。通过某地区10 kV配电网线路实际计算,对比不同方法,验证IPSO-LSSVM模型具有更好的精度和收敛能力。

In order to evaluate accurately and comprehensively the line loss situation of 10 kV distribution network,the grey loss analysis and the improved particle swarm algorithm to optimize the least squares support vector machine for the line loss prediction method are presented.The grey correlation analysis is used to analyze quantitatively the correlation between the electrical indicators and the line loss of the 10 kV distribution network.The inertia weight of the standard particle swarm optimization algorithm and the change law of the learning factor are improved.The penalty factor of the LSSVM is optimized using IPSO,and the IPSO-LSSVM prediction is established.Through the actual calculation of 10 kV distribution network lines in a certain area and comparing different methods,it is verified that IPSO-LSSVM has better accuracy and convergence ability.