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
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。
Abstract：
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.
Keywords:heat substation; heating load prediction; extreme learning machine(ELM); particle swarm optimization (PSO); load fluctuation; training set sample size