RBF surrogate model and EN1317 collision safety-based optimization of two guardrails

Abstract

The purpose of using roadside safety equipment is to protect vehicle occupants during an accident by reducing the severity of impact. Using poorly designed safety equipment can have serious consequences. While roadside safety elements are designed primarily for safety, cost-effectiveness cannot be overlooked. This study is aimed to optimize H1W4 and H2W4 performance level guardrails manufactured from S235JR, S275JR, and S355JR grade steel materials in terms of safety and economy. For this purpose, surrogate model-based optimization is utilized. In this context, the design variables are selected as post width (x1) and cross-sectional thickness of rail (x2), while objective functions are selected as working width (w) and vehicle exit angle (α). The design variables (x1, x2) are derived by full factorial design (FFD) method and crash test simulations are utilized to construct objective and constraint functions. Consequently, radial basis function (RBF)-based metamodels are created with the help of the obtained data and later the accuracy of the models is validated. Finally, the RBF metamodels are optimized using the multi-objective genetic algorithm (MOGA). As a result of the simulation-based design optimization (SBDO), the optimum designs of H1W4 and H2W4 guardrail systems for different steel material grades are achieved. It is concluded that the final guardrail designs meet the safety criteria as well as provide an economic advantage of up to 23% compared to existing guardrail designs.