基于粒子群的电力系统短期负荷预测

Short-term load forecast in Power system based on PSO optimizing algorithm

短期负荷预测是电力系统管理的一项重要方法,准确的负荷预测可以保证用户得到安全、经济、合理的供电。针对负荷预测方法的多样性,在传统的BP网络用于负荷预测的基础上,提出粒子群PSO(Particle Swarm Optimizer)优化神经网络权值的算法,并应用到电力系统短期负荷预测中。仿真结果表明PSO优化算法训练的神经网络不仅收敛速度明显加快,而且其预报精度也明显地得到提高。

Short-term load forecast is very important in power system management, accurate load forecast can ensure customer to get safe, economical and reasonable power supply. Among many optimizing algorithms used for load forecasting, a novel PSO （Particle Swarm Optimizer） algorithm is proposed for optimizing the weight of NN （Neural Network）and is applied to short term load forecast in power system. It is shown that convergence speed and precision of BP neural network are improved.