Load forecast and fuzzy control of the air-conditioning systems at the subway stations. (15th May 2022)

• Record Type: Journal Article
• Authors: Bi, Haiquan; Zhou, Yuanlong; Liu, Jin; Wang, Honglin; Yu, Tao
• Is Part Of: Journal of building engineering. Volume 49(2022)
• Publication Date: 2022-05-15
• Subjects: Subway station -- Air-conditioning system -- Energy consumption -- Neural network -- Load forecasting -- Predictive fuzzy control; Building -- Periodicals; 690.05
• DOI: 10.1016/j.jobe.2022.104029
• Languages: English
• ISSNs: 2352-7102

Abstract

Abstract: The average annual electricity consumption of the subway stations in China is 1.8–2.3 million kWh, of which ventilation and air-conditioning systems account for approximately 46%. Optimizing the ventilation and air-conditioning control system is an important energy-saving method in urban rail transit. This study applied the technologies of neural network and fuzzy control to the load forecast and the control of the air-conditioning system in subway stations to reduce the energy consumption of the air-conditioning system. Firstly, the energy consumption of the air-conditioning system was calculated by TRNSYS software. Then, a load forecast model of the air-conditioning system was established using neural network technology, and the accuracy of the load forecast model was verified through comparative analysis. Finally, the predictive fuzzy control model of the air-conditioning system was established. The temperature and the humidity in the subway station with the predictive fuzzy control and the traditional temperature control were studied, as well as the energy consumption of the air-conditioning system. Results showed that the neural network technology could effectively predict the load of the subway station's air-conditioning system. The predictive fuzzy control could offset the delay of control quantity adjustment of the air-conditioning system to a certain extent. Compared with the traditional temperature control method, the temperature fluctuation of the station hall and platform under predictive fuzzy control is smaller, and the total energy consumption of the air-conditioning system in summer is reduced by 7.13%. This study provides a reference for reducing the energy consumption of the air-conditioning system in the urban rail transit stations. Highlights: Optimizing the air-conditioning control system is an important energy-saving method in urban rail transit. This study applied neural network and fuzzy control to the air-conditioning system's load forecast and control. Neural network technology could effectively predict the subway station's air-conditioning system load. Using predictive fuzzy control can reduce the temperature fluctuation in subway stations. The total energy consumption of air-conditioning system in summer is reduced by 7.13% under predictive fuzzy control.