基于PSO-ELM组合算法的热力站负荷预测研究  

Heat substation load prediction based on PSO-ELM combined algorithm

作　　者：马文菁 郭晓杰 曹姗姗 孙春华[1] 夏国强[1] 齐承英[1] Ma Wenjing;Guo Xiaojie;Cao Shanshan;Sun Chunhua;Xia Guoqiang;Qi Chengying(Hebei University of Technology,Tianjin)

机构地区：[1]河北工业大学,天津300401

出　　处：《暖通空调》2024年第3期157-162,共6页Heating Ventilating & Air Conditioning

基　　金：河北省高等学校科学技术研究项目“基于多元时序数据挖掘的供热系统用能模式识别与诊断研究”(编号:QN2021212)。

摘　　要：提出了一种粒子群优化极限学习机(PSO-ELM)算法用于热力站负荷预测,应用粒子群(PSO)算法优化极限学习机(ELM)的输入权值和隐含层阈值。将提出的组合算法应用于天津市某小区热力站的负荷预测中,并与ELM、支持向量回归(SVR)和粒子群优化支持向量回归(PSO-SVR)算法在同等条件下进行比较。结果表明,PSO-ELM在预测精度上优于其他算法;在热负荷波动较大时,表现优于PSO-SVR;在一定范围内样本容量对预测结果影响不大,PSO-ELM可遗忘更多的数据。In this paper,a particle swarm optimization extreme learning machine(PSO-ELM)algorithm is proposed to predict the load of the heat substation,and the input weight and hidden layer threshold of the extreme learning machine(ELM)are optimized by the particle swarm optimization(PSO)algorithm.The combined algorithm is applied to the load prediction of a residential heat substation in Tianjin,and compared with ELM,support vector regression(SVR)and particle swarm optimization support vector regression(PSO-SVR)under the same conditions.The results show that the PSO-ELM has better prediction accuracy than other algorithms.When the heating load fluctuation is large,its performance is better than the PSO-SVR.In a certain range,the sample size has little effect on the prediction results,and the PSO-ELM can forget more data.

关 键 词：热力站 热负荷预测 极限学习机 粒子群优化 负荷波动 训练集样本容量 

分 类 号：TM621[电气工程—电力系统及自动化] TP181[自动化与计算机技术—控制理论与控制工程]