基于优化BP网络的电网短期负荷预测研究

如何提升电网负荷短期预测水平是电网企业亟待解决的问题。本文针对传统的BP神经网络算法所存在的学习过程收敛速度慢、算法易陷入局部极小点和鲁棒性差等缺陷,引入粒子群优化算法对其进行优化和改进,使之具备更加完善的性能。通过实际电网负荷预测的实验与比较,证明了所构建的符合预测系统的准确度。

Particle Swarm Optimization-based Neural Network Model for Short-term Load Forecasting

ANN is a method of shortterm load forecasting which has been researched the most recently. This paper proposes a new load forecasting model based on Particle Swarm Optimization（PSO）. PSO is a novel random optimization method which has extensive capability of global optimization. PSO is used to optimize the weighting factor of Radial Basis Function（RBF）neural network and the optimal model is applied to forecast load. The application of ANN in combination forecasting has been summarized.