【摘要】 准确的短期负荷预测能够减少发电机组停机备用和旋转备用,其预测效果直接影响电网的安全稳定和经济效益。针对BP神经网络初值敏感、易陷入局部最优的缺点,提出了一种改进人工蜂群算法优化BP神经网络的负荷预测方法。首先融合负荷数据与温度、湿度等天气数据并进行高斯滤波处理,再采用搜索位置更新实现人工蜂群算法的改进,利用其算法完成BP网络权值和阈值的优化,最后建立用于短期负荷预测的优化模型,并通过实例进行仿真验证。结果表明:该改进预测模型与传统BP算法相比预测精度及收敛速度均有大幅提高,具备工程实用价值。更多还原

【Abstract】 Accurate short-term load forecasting can reduce generator set shutdown and spin reserve,and its prediction effect directly affects the safety and stability of the grid and economic benefits.Aiming at the shortcomings of BP neural network initial value sensitivity and easy to fall into local optimum,a load forecasting method based on artificial bee colony algorithm to optimize BP neural network is proposed.Firstly,the weather data of load data and temperature,humidity and so on are combined and Gaussian filtering is processed.Then the search location update is used to improve the artificial bee colony algorithm.The algorithm is used to optimize the BP network weight and threshold,and finally used for short-term load forecasting.The optimization model is verified by simulation.The results show that the improved prediction model has a significant improvement in prediction accuracy and convergence speed compared with the traditional BP algorithm,and has engineering practical value.更多还原