Additional fuel system optimization for CNG vehicle base on MDO method

Usually CNG vehicle is modified based on conventional vehicle in structure, the additional fuel system is very heavy due to its gas and structure, which increases the danger of potential gas leakage and explosion. Therefore the safety and reliability of the CNG vehicle should be highly emphasized. In this paper, we analyze the characteristic fuel structure of CNG vehicle and emphatically study on rear impact and NVH performance. A finite element model of CNG vehicle for rear impact and NVH analysis was created. Using OPTIMUS software with multidisciplinary feasible algorithm which integrated with Pam-Crash and MD Nastran software for collaborative optimization, we made the additional fuel system lighter and fit all requirements. From sensitivity analysis of initial design space we found sensitive ingredient of structure, and built response surface model for these sensitive parts as design parameters base on RBF function. We apply an adaptive evolutionary algorithm with discrete variables to find global optimal solution on response surface. Calculated results show that the response surfaces have high accuracy, and the optimization calculation converges very fast. Finally the mass of additional fuel system is significantly reduced, and the model and MDO method is proved to be effective and correct.