Load Distribution Optimization for Multiple Chillers Using Improved Tunicate Swarm Algorithm
In order to reduce the energy consumption of central air-conditioning systems,and aimed at the problem of load distri-bution optimization in multiple chillers,this paper proposes a performance prediction model for chills based on random forest feature optimization combined with kernel function extreme learning machine,the prediction accuracy of the model is improved by eliminating redundant features.Then an improved tunicate swarm algorithm(ITSA)based on hybrid strategy is proposed.A whale spiral search strategy is integrated to improve individual update methods,introducing non-linear dynamic weight balancing for global exploration and local development,a flip disturbance strategy is used to avoid falling into partial optimum.Finally,based on the energy-efficiency model,the ITSA is used to optimize the load allocation of multiple chillers.The experimental results show that the random forest fea-ture optimization can effectively improve the accuracy of the energy efficiency prediction model.The ITSA can effectively save the po-tential energy of the system by optimizing the on-off status and load ratio of the chillers.Compared with the original method,the en-ergy consumption can be reduced by about 6％,which shows that the method is appropriate for optimal load allocation of multiple chillers.
万方数据
维普
