Short-Term Load Forecasting Based on Particle Swarm Algorithm Optimised Neural Network
Neural networks based on particle swarm algorithm(PSO)show excellent results in short-term load forecasting.Through PSO,the weights and biases of the neural network are effectively optimised so that it learns and adapts to the load changes of the power system faster and more accurately.The method achieves global search and optimisation of the neural network model by randomly initialising a group of"particles",each representing a set of parameters of the neural network,and then iteratively updating the particle positions.The results show that the PSO-optimised neural network exhibits a higher degree of fit in short-term load forecasting,which is closer to the actual load profile,and its prediction error is significantly reduced compared with the traditional neural network.This method provides a reliable tool for power systems and effectively improves the accuracy and reliability of short-term load forecasting.
NETL
NSTL
万方数据
