Reliability-based weight reduction optimization of forearm of bucket-wheel stacker reclaimer considering multiple uncertainties

This paper focuses on the weight reduction optimization of a forearm of a bucket-wheel stacker reclaimer considering uncertainties of structural parameters, material properties, loads, and surrogate model. However, the optimization problem is a high-dimensional problem, with dozens of independent variables, which has negative effects on the optimization efficiency. Considering that millions of iterations are required for the reliability-based optimization, the finite element model can cause overwhelming computational cost. In addition, due to its complex structure and working conditions, multiple uncertainties exist in practical applications and affect the reliability of a design, especially the uncertainty of the surrogate model. To address these challenges, the sensitivity analysis is performed to improve the optimization efficiency by selecting main factors. The Kriging model with high accuracy is constructed to reduce the computational cost. In order to improve the optimization efficiency further, the deterministic optimization is performed firstly, and the optimal design is used as the initial point of the reliability-based optimization algorithm. For estimating the reliability, the multiple uncertainty models are constructed. Finally, according to the design requirements and taking the multiple uncertainties into account, the reliability-based optimization is proposed and carried out. The result proves that the weight is reduced greatly and the reliability is kept at a high level.