摘要
适用性强且准确度高的冷热电负荷预测方法能更好地为机组设备选型提供可靠依据,从而整体提升系统综合效率。针对传统BP神经网络局部过优化、预测误差较大等问题,将遗传算法与BP神经网络相结合,提升预测准确性和可靠性。为了验证GA-BP神经网络预测方法的合理性,分别采用两种方法对铁岭某园区冷热电负荷进行预测,结果证明改进后的预测方法误差更小、贴合性更高。
The cooling,heating and power load forecasting method with strong applicability and high accuracy can better provide a reliable basis for unit equipment selection,so as to improve the overall efficiency of the system.Aiming at the problems of local over optimization and large prediction error of traditional BP neural network,the genetic algorithm is combined with BP neural network to improve the prediction accuracy and reliability.In order to verify the proposed GA-BP neural network prediction method,two methods are used to predict the cooling,heating and power load of a park in Tieling.The results show that the improved prediction method has less error and higher fit.
作者
王玥
鞠振河
汤梓涵
WANG Yue;JU Zhenhe;TANG Zihan(Shenyang Institute of Engineering,Shenyang 110136,China)
出处
《电工技术》
2022年第7期40-42,189,共4页
Electric Engineering