Temperature Prediction for Subway Station Based on LSTM-SVR Algorithm with Multi Period Characteristics

Abstract By analyzing the timing characteristics of historical temperature data subway sensors, aiming at problem poor accuracy a single prediction model, combined with long-term trend, multi period and irregular change data, this paper proposes model based on Long Short-Term Memory network (LSTM) Support Vector Regression (SVR) theory, mean error LSTM-SVR is lower than which can predict station high accuracy, so as to provide basis for controlling air-conditioning ventilation equipment, also saving energy reducing consumption.