摘要
对电力负荷进行预测是电网的一个传统问题。有效的电力负荷峰值预测,可以对发电机组进行经济合理的起停、保持电力系统的安全稳定运行,在保证社会的正常生产和生活的前提下,又能降低发电的成本,提高社会及经济效益。文章重点对电力负荷预测的特点进行研究,提出动态蚂蚁遗传神经网络算法(DAAGA-BP)预测模型,并对多种预测算法进行深入的对比分析。通过仿真计算,对比拟合数据与实际负荷值,预测值曲线与实际值曲线较为接近。证明该模型应用于电网负荷预测分析具有适应性,通常是可行且有效的。
Predicting power load is a traditional problem in the power grid.Effective peak load forecasting can economically and reasonably arrange the start and stop of generator units,maintain the safety and stability of the power system,under the premise of ensuring the normal production and life of society,reduce the cost of electricity generation,and effectively improve social and economic benefits.The characteristics of power load forecasting have been studied in detail,We proposed a DAAGA-BP neural network prediction model and conducted in-depth comparative analysis of multiple prediction algorithms.Through simulation calculations,comparing the fitted data with the actual load values,the predicted value curve is closer to the actual value curve.It has been proven that the model has adaptability and is generally feasible and effective when applied to power grid load forecasting analysis.
作者
邱金鹏
QIU Jinpeng(Southwest Electric Power Design Institute Co.,Ltd.of China Power Engineering Consulting Group,Chengdu 610021,China)
出处
《电力勘测设计》
2025年第7期35-41,共7页
Electric Power Survey & Design
关键词
负荷预测
神经网络模型
遗传算法
load forecasting
neural network model
genetic algorithm
