Search

A design engineer has to deal with increasingly complex design tasks on a daily basis, for which the available design time is shrinking. Market competitiveness can be improved by using optimization if the design process can be automated. If there is limited information about the behavior of the objective function, global search methods such as simulated annealing (SA) should be used. This algorithm requires the selection of a number of parameters based on the task. A procedure for reducing the time spent on tuning the SA algorithm for computationally expensive, simulation-driven optimization tasks was developed. The applicability of the method was demonstrated by solving a shape optimization problem of a rubber bumper built into air spring structures of lorries. Due to the time-consuming objective function call, a support vector regression (SVR) surrogate model was used to test the performance of the optimization algorithm. To perform the SVR training, samples were taken using the maximin Latin hypercube design. The SA algorithm with an adaptive search space and different cooling schedules was implemented. Subsequently, the SA parameters were fine-tuned using the trained SVR surrogate model. An optimal design was found using the adapted SA algorithm with negligible error from a technical aspect.

One of the most critical issues during polymer finite element simulations is the selection of the proper material models. The widely used and accepted multilinear material models require load case-specific material tests, which are time and cost demanding. Data for these characteristics must be acquired by standardized measurements. On the other hand, the parameters required to create a linear elastic material model in most cases are easy to obtain, and the establishment of the model is a shorter process. This research is aimed to provide information to engineers about the possibility of modeling the nonlinear elastic materials by using linear elastic material models and about the limits of such models. To create the most accurate material models, laboratory measurements were performed on polyamide (PA6) material, which is a widely used raw material in the industry. Test specimens were manufactured to obtain material constants according to the ISO 527-2 standard, and for validating the effectiveness of the applied material models, three different tensile specimens were created, which were tested under quasi-static loading in the elastic region. A comprehensive finite element investigation was performed, and the numerical results were then compared to laboratory measurements using the GOM Aramis digital image correlation (DIC) system. By comparing the optically measured strain data to the numerical results, it was determined that the nonlinear elastic materials can be modeled using linear elastic models in a well identifiable strain range with sufficient accuracy.

In rubber bumper design, the most important mechanical property of the product is the force–displacement curve under compression and its fulfillment requires an iterative design method. Design engineers can handle this task with the modification of the product shape, which can be solved with several optimization methods if the parameterization of the design process is determined. The numerical method is a good way to evaluate the working characteristics of the rubber product; furthermore, automation of the whole process is feasible with the use of Visual Basic for Application. An axisymmetric finite element model of a rubber bumper was built with the use of a calibrated two-term Mooney–Rivlin material model. A two-dimensional shape optimization problem was introduced where the objective function was determined as the difference between the initial and the optimum characteristics. Our goal was to integrate a surrogate model-based parameter selection of local search algorithms for the optimization process. As a metamodeling technique, cubic support vector regression was selected and seemed to be suitable to accurately predict the nonlinear objective function. The novel optimization procedure which applied the support vector regression model in the parameter selection process of the stochastic search algorithm proved to be an efficient method to find the global optimum of the investigated problem.