基于改进BP神经网络的中央空调冷负荷预测研究

Prediction of Cooling Load of HVAC System Based on Improved BP Model 

摘要:

针对中央空调系统冷负荷预测中BP神经网络预测收敛慢,易陷入局部最优,精度相对低的缺点,采用了收敛速度快,全部搜索能力强的粒子群优化算法进行改进,同时对BP结构中的输入参数,添加控制误差反馈参数,形成了基于粒子群与控制误差回馈的BP神经网络预测技术;其预测精度较BP神经网络和粒子群BP神经网络分别提高5.94%和0.82%。

Abstract:

Since the shortcomings of the BP neural network prediction method in predicting the building cooling load, such as slow convergence, easy to fall into local optimum and relatively low accuracy, the particle swarm optimization algorithm which has faster convergence and stronger global search capability is introduced to improve the prediction. By adding control error feedback parameters to the input parameters of the BP neural network structure, the BP neural network forecasting techniques based on particle swarm optimization and control error feedback is formed, applied in the central air conditioning system, and enhance the prediction accuracy.