Analysis of Underhood Temperature Fields using Linear Superposition 2017-01-0138
The analysis of thermal fields in the underhood region is complicated by the complex geometry and the influence of a multitude of different heat sources. This complexity means that running full CFD analyses to predict the thermal field in this region is both computationally expensive and time consuming. A method of predicting the thermal field using linear superposition has been developed in order to analyse the underhood region of a simplified Formula One race car, though the technique is applicable to all vehicles. The use of linear superposition allows accurate predictions of the thermal field within a complex geometry for varying boundary conditions with negligible computational costs once the initial characterisation CFD has been run. A quarter scale, rear end model of a Formula One race car with a simplified internal assembly is considered for analysis, though the technique can also be applied to commercial and industrial vehicles. Different parts of the internal assembly are used as heat sources in order to simulate the thermal boundary conditions of a real-world underhood zone. The method utilises a combination of initial CFD analyses and subsequent post-processing within MATLAB in order to rapidly predict the thermal field under varying boundary conditions. Predictions of the thermal field using linear superposition are generated in under a second for a six million node domain. These predictions are compared to CFD analyses with the same explicit boundary conditions, which represent the current industry design methodology for thermal management strategy. Comparisons between the methods are presented for varying thermal boundary conditions and are extended to varying inlet boundary conditions using additional scaling factors. Current limitations of this approach and recommendations for future development are also discussed.