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

 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.