论文部分内容阅读
对空调负荷进行准确预测不仅对优化空调控制的意义重大,也是实现空调经济运行与节能的关键所在。为了提高建筑空调负荷的预测精度,在分析最小二乘支持向量机建模特点的基础上提出了利用PSO-SA优化的一种空调负荷预测算法。该方法利用粒子群—模拟退火方法对最小二乘支持向量机的参数进行优化选择,提高模型的精度和泛化能力。通过空调负荷预测建模的结果表明,该方法具有学习速度快、跟踪性能好以及泛化能力强等优点,为实现空调系统的优化运行奠定了基础。
Accurately forecasting the air conditioning load is not only of great significance to the optimization of air conditioning control, but also the key to realize economical operation and energy saving of air conditioning. In order to improve the prediction accuracy of building air conditioning load, an air conditioning load forecasting algorithm optimized by PSO-SA is proposed based on the analysis of modeling features of least square support vector machines. This method uses particle swarm optimization and simulated annealing to optimize the parameters of least square support vector machine, and improves the precision and generalization ability of the model. The results of modeling with air conditioning load forecasting show that this method has the advantages of fast learning speed, good tracking performance and strong generalization ability, which lays the foundation for the optimal operation of air conditioning system.