Research on the Construction of Energy Consumption Prediction Model of Chiller Based on Data-Driven

Le résumé fourni par la source

A model for predicting the energy consumption of chiller based on data-driven approach is put forward, which can enhance the efficiency of the chiller and optimize energy conservation. Initially, the research collected a large quantity of historical data during the chiller's operation, such as environmental temperature, cooling load, and cooling water flow. Relying on the random forest algorithm, this paper analyzes and constructs models for these data, and utilizes its powerful nonlinear fitting capacity to predict the energy consumption of the chiller. To confirm the validity of the model, a simulation system was devised to simulate the energy consumption of the chiller under different operating conditions. Through the precise data analysis of the simulation results, the model can effectively predict the energy consumption, and the prediction error is less than 5%. Furthermore, the random forest algorithm is more precise and stable compared with the traditional one. It is demonstrated that the data-driven random forest model can be utilized to optimize the energy efficiency of chillers.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.