Dew point evaporative cooling(DPEC)air conditioning achieves cooling through the water evaporation and heat absorption.Due to climatic variations,the performance of DPEC air conditioning differs across different regions.With two machine learning methods,namely the Back Propagation(BP)neural network and the Random Forest algorithm.This study utilizes seven features,including environmental temperature,relative humidity,air velocity,working air ratio,height of the heat exchanger,spacing,and number of layers,as independent variables.The dependent variables include the refrigeration capacity,Coefficient of Performance,dew point efficiency,and wet bulb efficiency of the DPEC air conditioning.Both machine learning methods were employed to predict the performance of the DPEC air conditioning.The results indicate that both machine learning methods can effectively predict the performance,with the prediction error of the Random Forest algorithm being less than 10%and the BP neural network below 2%.The BP neural network method demonstrates superior prediction performance.
维普
万方数据