基于PSO-LSSVM模型空调负荷预测研究

Research on Air Conditioning Load Prediction Based on PSO-LSSVM Model

提出利用粒子群优化(PSO)算法优化的最小二乘支持向量机(LSSVM)预测模型(PSO-LSSVM模型)。以厦门某公共建筑作为研究对象,将平均绝对误差绝对值、平均相对误差绝对值作为评价指标,评价LSSVM模型、PSO-LSSVM模型对空调负荷的预测效果。LSSVM模型、PSO-LSSVM模型的空调负荷预测值与实测值变化趋势基本一致。与LSSVM模型相比,PSO-LSSVM模型的预测值平均绝对误差绝对值、平均相对误差绝对值更小,PSO-LSSVM模型的预测准确性更高。

The least squares support vector machine(LSSVM)prediction model optimized by particle swarm optimization(PSO)(PSO-LSSVM model)is proposed.Taking a public building in Xiamen as the research object,the absolute value of average absolute error and the absolute value of average relative error are used as evaluation indexes to evaluate the prediction effect of LSSVM model and PSO-LSSVM model on air conditioning load.The predicted values of air conditioning load of the LSSVM model and the PSO-LSSVM model are basically consistent with the measured values.Compared with the LSSVM model,the absolute value of average absolute error and the absolute value of average relative error of the predicted value of the PSO-LSSVM model are smaller,indicating that the prediction accuracy of the PSO-LSSVM model is higher.