Multi-objective optimization of rolled goods transport pallet based on meta-models and adaptive optimization algorithms
The design and improvement of transport pallets for the railway transportation of rolled goods often rely on engineering experience,utilizing a combination of simulation analysis and actual strength testing.However,this design approach excessively depends on individual experience,resulting in diminished design efficiency,substantial waste of experimental resources,and challenges in achieving the anticipated outcomes.To transform the conventional design approach into a mathematical optimization problem,the introduction of meta-models and adaptive multi-objective optimization algorithms was proposed for the design and enhancement of transport pallets.Establish a three-dimensional model for the pallet,formulate loading reinforcement scheme,and conduct simulation analyses under static and impact loading conditions.Perform strength tests on the pallet,including both static and impact trials,to validate the reliability of simulation analyses.Employ a collaborative simulation approach using SolidWorks and Ansys,establishing a parameter-driven CAE simulation environment.Propose an optimal Latin hypercube sampling method based on the maximum-minimum distance criterion and ESP-induced joint optimization,creating a high-dimensional parameter space matrix.Utilize permissible stress and deformation thresholds for various pallet conditions as constraints to formulate a multi-objective optimization mathematical model.Compare the precision of different types of meta-models,select the GWO-BP neural network,couple it with the adaptive third-generation non-dominated sorting genetic algorithm (A-NSGA-Ⅲ) to obtain the optimal parameter combination,ultimately achieving pallet lightweight,and minimizing stress under two distinct operational conditions.Research findings indicate that,taking the example of the railway RTKD-type transport pallet for rolled goods,multi-objective optimization design is conducted.It is observed that the error between simulated analysis and actual experimental values for the pallet was consistently less than 11％,with the simulated stress distribution closely aligning with the measured values.Post-optimization,the pallet's mass is reduced by 27.97 kg,and the maximum stresses under the two conditions decreases by 23.01 MPa and 37.21 MPa,respectively,aligning with the safety requirements for railway goods transportation.The research results can provide engineering references for the design and improvement of similar transport pallets,playing a crucial role in promoting the development of railway transportation for rolled goods.
railway rolled goods transport palletfinite element analysisstatic load testimpact testmeta-modelsA-NSGA-Ⅲ
万方数据
维普
