关键词: 空调表冷系统, 斐波拉契搜索, 极限学习机, 预测控制, 中心平衡优化算法

Abstract: According to the environment variables of outdoor temperature and relative humidity, this paper uses the Extreme Learning Machine (ELM) to build an indoor temperature and humidity prediction model of the air conditioning surface cooling system to solve its coordinated valve control problem.The number of hidden layer nodes of ELM is optimized by using the Fibonacci search algorithm. The Fibonacci extreme learning machine (FELM) is proposed to improve the accuracy of the prediction model.The traditional equilibrium optimization (EO) algorithm has a slow convergence speed and is easy to fall into local minima.The KMedoids clustering algorithm is embedded into the equilibrium optimization algorithm to improve the performance of the optimization algorithm.Then, the KMedoids equilibrium optimization (KEO) algorithm is used to conduct rolling optimization of the control quantity of the three valves of the surface cooling system, such as the opening of the main surface cooling valve, the auxiliary surface cooling valve and the electric three-way valve. The simulation results show that, compared with the traditional ELM predictive control algorithm, KEO-FELM predictive control has higher stability and tracking abilities with better energy-saving effect.