基于近似动态规划的建筑能量管理系统日内优化调度方法

In order to maintain the internal temperature of the building and cope with the fluctuation of random amount，an intraday optimization scheduling method for building energy management system is proposed. The thermal balance relationship of the building envelope is comprehensively considered，the human comfort function related to room temperature is obtained，and the scheduling optimization model of building energy management system is established with the goal of economic optimization. Then，a solution method combining approximate dynamic programming（ADP） and model predictive control（MPC） is proposed，using historical data as the training set to cope with a large number of uncertainties such as new energy output，temperature and radiation，and the segmented linear function obtained offline training is used as the future value function of intraday scheduling. The numerical analysis shows that the proposed MPC-ADP algorithm can obtain an approximately optimal real-time scheduling scheme through continuous iteration，and greatly improve human comfort.