Multi variable collaborative optimization control of magnetic levitation chiller based on PSO-SVM model

A multi variable collaborative optimization control method based on PSO-SVM model is proposed to address the problems of low energy efficiency and poor stability caused by the neglect of the coupling relationship between multiple variables in traditional magnetic levitation chiller optimization control methods. Firstly, with the help of multivariate identification, clarify the coupling relationship between the input, output, and state variables of the unit; Construct an optimization function with energy efficiency ratio and operational stability as the objectives, and use PSO algorithm to automatically optimize the parameters of SVM model to improve the accuracy of state prediction; On this basis, a collaborative control mechanism based on PSO-SVM is designed to achieve multivariable online rolling optimization. The experimental results show that the proposed method has significantly higher energy efficiency than traditional methods under six typical operating conditions. The variance of cooling capacity during dynamic processes is reduced by 51.9% to 65.8%, significantly improving the energy efficiency and operational stability of the unit under all operating conditions. This provides an effective approach for intelligent control of complex electromechanical systems.