摘要
综合能源系统多元负荷预测是有效提升能源利用效率、降低用能成本的主要途径之一。针对综合能源系统数据繁杂、不易预测的问题,首先引入动态FOA算法优化RBF神经网络,帮助RBF神经网络寻优;其次运用Lasso原理对气象因素进行选择,将负荷数据及气象因素输入到动态FOA优化后的RBF神经网络;最后对北方某园区进行综合能源系统负荷预测,并与BP神经网络进行对比验证。预测结果表明,采用该方法进行负荷预测能有效改善预测效果,保障了区域综合能源系统的优化运行。
Multivariate load forecasting of integrated energy system is one of the main ways to effectively improve energy utilization efficiency and reduce energy costs.In this paper,in response to the problem of complex data and difficulty in predicting and planning in integrated energy system,the dynamic FOA algorithm is introduced to optimize RBF neural network and assist in its optimization.Secondly meteorological factors is selected by using the Lasso principle,and the load data as well as meteorological factors are inputted into the dynamic FOA optimized RBF neural network.Finally,a comprehensive energy system load prediction is conducted for a certain park in the north,and compare as well as verify with BP neural network.The prediction results indicate that using this method for load forecasting can effectively improve the prediction effect and ensure the optimal operation of integrated energy system.
作者
黄文静
吴蔚
HUANG Wenjing;WU Wei(Hebei Vocational University of Industry and Technology,Shijiazhuang 050091,China;Shijiazhuang Power Supply Branch of State Grid Hebei Electric Power Co.,Ltd.,Shijiazhuang 050004,China)
出处
《河北电力技术》
2024年第2期8-11,17,共5页
Hebei Electric Power
基金
河北省教育厅科技项目(zc2022042)。
关键词
综合能源
负荷预测
RBF神经网络
FOA算法
integrated energy system
load forecasting
RBF neural network
FOA algorithm
