Inverse Design for Thermal Environment and Energy Consumption of Vehicular Cabins with PSO–CFD Method
To create a comfortable thermal environment in the passenger compartment, designers usually redesign the parameters of the HVAC system and simulate a large number of flow fields with different boundary conditions including the inlet velocity and temperature. However, the traditional method is often inefficient and ineffective. This paper presents an inversed design method based on back propagation neural network (BPNN) and particle swarm optimization (PSO). The BPNN is used to obtain more computational fluid dynamics (CFD) cases by fitting some known case results. The PSO method is used to identify the best flow control conditions. Besides, this paper used the predicted mean vote (PMV) and energy consumption as the objective function to optimize the inlet boundary conditions. The results show that the proposed method can reduce CFD calculation time and achieve optimized design of the HVAC system in-vehicle cabins.
KeywordsCabin thermal environment Back propagation neural network Particle swarm optimization CFD PMV
We would like to acknowledge the National Natural Science Foundation of China (51708493), National Research Program for Key Issues in Air Pollution Control (DQGG0207), Zhejiang Provincial Natural Science Foundation (LR19E050002), Zhejiang Province Key Science and Technology Project (2018C01020, 2018C01060, 2019C01057), and the Youth Funds of State Key Laboratory of Fluid Power & Mechatronic Systems (SKLoFP_QN_1804).