基于多重遗传算法的中央空调制冷系统节能优化方法OA

In response to the problems of complex coupling of operating parameters in central air conditioning refrigeration systems,strong reliance on traditional manual parameter adjustment experience,and the tendency of single intelligent algorithms to fall into local optima,resulting in poor energy efficiency optimization effects,a new energy-saving optimization method based on multiple genetic algorithms(MGA)is proposed.Through grey correlation analysis(GRA),three key influencing parameters,namely the outlet water temperature of the main unit,the frequency of the chilled water pump,and the frequency of the cooling water pump,are selected from 10 parameters.The backpropagation(BP)neural network is used to construct an energy efficiency ratio(EER)prediction model with a prediction error of less than 0.4%as the fitness function.The parameter association rules mined by the association rule algorithm(Apriori)are introduced as constraints.The MGA is used for global optimization.The results show that under a load rate of 80%to 90%in a commercial building,the method increases the system energy efficiency ratio by an average of 5.75%and up to 7.93%,effectively improving the system operation efficiency.