To accomplish the better control of heating, ventilation, and air conditioning (HVAC) with the energy saving target, the accurate indoor temperature prediction plays a fundamental role. However, existing temperature prediction approaches often fails in extracting sufficient features in a single model, which results in subsequent unideal control effects. Considering this defect, this paper developed a novel hybrid artificial neural network (ANN) and long short term memory (LSTM) model (HAL) to predict indoor short and long-term temperature. The hybrid HAL can model the temperature pattern and learn the relations among other variables simultaneously. The proposed HAL is implemented on a real-world indoor environmental dataset and extensive experimental assessments including single and multiple step prediction are conducted. The results show that the proposed HAL outperforms other data driven methods with strong robustness in not only single but multiple step prediction.
Hybrid ANN-LSTM based temperature prediction for HVAC control Lianjie Jiang, Xinli Wang, Lei Wang, Mingjun Shao, Liping Zhuang