Short-term Prediction Model for Thermal Comfort Based on ISSA-LSTM
During short-term prediction on test days,to solve the influence of random factors such as building inertia and person-nel in human thermal comfort,resulting in the low prediction accuracy in rural and urban areas,an optimized long short-term memory neural network(LSTM)based on improvement sparrow search algorithm(ISSA)is proposed to build the novel short-term prediction model for the thermal comfort of residential air conditioners.Firstly,this paper analyzes the dynamic data of weather on test days,verifies the validity of the data,and constructs various thermal comfort prediction models;Then,the new household thermal comfort short-term prediction model(ISSA-LSTM)is selected to predict thermal comfort.The results show that compared with the sparrow search algorithm(SSA)and Dung beetle optimizer(DBO)optimized LSTM,the proposed method increases the highest prediction mean squared error(MSE)of 0.022 96 and 0.108 27,respectively.The ISSA-LSTM method is adopted to improve the accuracy of short-term thermal comfort prediction,and the performance of split air conditioners to control temperature through thermal comfort.
residential air conditionersthermal comfortISSAneural networkshort-term prediction
万方数据
维普
