Prediction of energy consumption in hotel buildings via support vector machines

Abstract This paper studies and analyzes the energy consumption of hotel buildings by establishing a support vector machine energy consumption prediction model. The support vector machine model takes the weather parameters and operating parameters of the hotel air-conditioning system as input variables, and determines the critical value of the input parameters by determining the normal-distribution interval, so as to avoid the influence of the outliers on the model prediction stability. The RBF kernel function is selected as the kernel function of the support vector machine, and the accuracy of the model prediction is improved by optimizing the kernel parameters. The MSE value of the final model prediction was 2.22% and R2 was 0.94. By predicting the results, you can visually assess the actual energy usage of the hotel and suggest timely improvements to the hotel's operations to reduce the hotel's energy consumption.