基于遗传算法优化的动态冰蓄冷负荷预测  

Prediction model of GA-BP on air conditioning load

作　　者：闫华光[1] 石坤[1] 许高杰[1] 李德志[1] 张艳辉[2] 林仕立[2] 黄冲[2] 

机构地区：[1]中国电力科学研究院,北京100190 [2]中国科学院广州能源研究所,广州510640

出　　处：《电子测试》2014年第7期25-27,共3页Electronic Test

基　　金：国家重点基础研究发展计划(973计划)项目(2012CB724400)

摘　　要：本文中利用时间、大气干球温度、环境温度、太阳辐射强度、t-1时刻的系统冷负荷和t-24时刻的系统冷负荷作为输入变量进行建模预测。本文充分利用遗传算法的全局搜索的优势以及BP神经网络精于局部精确搜索的特性,采用遗传算法(Genetic Algorithm,GA)优化神经网络算法各因子的初始权重,充分达到了两种智能算法有机结合,达到了优势互补的目的。结果表明,遗传算法和神经网络的有效结合显著提高了预测精度,证明了这种方法的有效性和可靠性,为指导动态冰蓄冷空调系统负荷预测和提高预测精度提供了新途径。The prediction model of air conditioning load was established that selected environment factors, according to the magnitude of the effect of time, wet bulb temperature, relative atmospheric temperature, Solar radiation intensity, and system cooling load at t-land t-24 as the input variable. The model combines genetic algorithm（GA）based on global optimization with back propagation（BP）based on gradient descent in the paper make the linking weights of networks self-adaptive evolution in constantly iterative process. The results show that GA-BP could significantly increased model computational speed and accuracy, proved the effectiveness and reliability of this method provides a new way for guiding the prediction of air conditioning load and improve model accuracy.

关 键 词：冰蓄冷系统 负荷预测 遗传算法 神经网络 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]