基于温度多元线性回归模型的空调制冷站在线预测控制方法研究

Online predictive control method of air conditioning chiller plants based on temperature multiple linear regression model

从实际应用角度出发,以制冷站全局COP为优化目标,综合考虑制冷站整体特性及各组件相互耦合特性,提出了一种基于多元线性回归模型的制冷站预测控制方法。该模型的输入变量为冷水进出口温差和冷却水进出口温差,分别对应冷水侧和冷却水侧的运行特性,输出变量为制冷站全局COP,能够反映制冷站系统的综合特性。该模型结构简单、输入参数易获取、可复制性高,可在线应用于大多数制冷站系统。试验结果表明,该预测控制方法可有效提升制冷站整体运行能效。

From the perspective of practical application,the global COP of the chiller plant is taken as the optimization objective,and a predictive control method of the chiller plant based on the multiple linear regression model is proposed by considering the overall characteristics of the chiller plant and the coupling properties among components.The input variables of the model are the inlet and outlet temperature difference of chilled water and the inlet and outlet temperature difference of cooling water,corresponding to the operation characteristics of the chilled water side and the cooling water side respectively,and the output variable is the global COP of the chiller plant,which can reflect the comprehensive characteristics of the chiller plant system.The model has the advantages of simple structure,easily obtained input parameters and high replicability,which renders it able to be implemented online in most chiller plant systems.The experimental results show that the proposed predictive control method can effectively improve the overall energy efficiency of the chiller plant.