Electric vehicle air conditioning system performance prediction based on artificial neural network. (5th October 2015)

• Record Type: Journal Article
• Authors: Tian, Z.; Qian, Ch.; Gu, B.; Yang, L.; Liu, F.
• Is Part Of: Applied thermal engineering. Volume 89(2015:Oct.)
• Page Start: 101
• Page End: 114
• Publication Date: 2015-10-05
• Subjects: Air conditioning system -- Artificial neural network -- Electric vehicle -- Electronic expansion valve -- Performance prediction; Heat engineering -- Periodicals; Heating -- Equipment and supplies -- Periodicals; Periodicals; 621.40205
• DOI: 10.1016/j.applthermaleng.2015.06.002
• Languages: English
• ISSNs: 1359-4311

Abstract

Abstract: In this study, electric vehicle air conditioning system (EVACS) performances with scroll compressor and electronic expansion valve (EEV) were experimentally investigated by varying scroll compressor speed, EEV opening and environment temperature. An artificial neural network (ANN) model for EVACS performances (such as refrigerant mass flow rate, condenser heat rejection, refrigeration capacity and compressor power consumption) prediction was developed based on experimental data. The ANN model was tested with two transfer functions (logsig and tansig) and different hidden neurons (3–13) using Levernberg-Marquardt algorithm. The optimized ANN was determined as a configuration of 4-13-4 with logsig transfer function, which demonstrated the best capability with mean relative errors, root mean square errors and correlation coefficients in the range of 0.92–2.71%, 0.0044–0.0141 and 0.9975–0.9998, respectively. Highlights: EEV opening influences on EVACS performance were experimentally studied. ANN used for EVACS performance prediction was trained with two transfer functions. Parametric study and hidden neurons test were performed to determine ANN structure. ANN was defined as a configuration of 4-13-4. ANN showed MRE, RMSE and R 2 in the range of 0.92–2.71%, 0.0044–0.0141 and 0.9975–0.9998, respectively.