Load extraction and load forecasting method of VRF air conditioners based on Fourier transform

The actual cooling capacity of variable refrigerant flow (VRF) air conditioners is usually not equivalent to the real air conditioning load demand of room, resulting in the poor forecasting effect of data-driven load forecasting models. In this paper, a method based on Fourier transform is proposed to extract the real load of air conditioner from the actual cooling capacity. The case forecasting results show that the proposed method can effectively extract the air conditioning load data and improve the accuracy of the ultra-short-term air conditioning load forecasting models based on artificial neural networks by 10.94%.