A novel hybrid adaptive Kriging and water cycle algorithm for reliability-based design and optimization strategy: Application in offshore wind turbine monopile

Abstract

Metaheuristic algorithms have been widely concerned by scholars because of their global optimization ability that does not depend on gradient information. In this study, Water Cycle Algorithm (WCA) as metaheuristic optimization is applied for a decoupling strategy to improve the efficiency of the original dual-loop Reliability-Based Design and Optimization (RBDO) algorithm. Furthermore, an adaptive Kriging-model-assisted RBDO method considering hybrid uncertainty is proposed based the decoupling WCA for improving the accuracy of RBDO results. In this method, in order to improve the computational burden of optimization design under uncertainties with highly accuracy, an adaptive Kriging model is developed to approximate the actual performance function. A novel updated strategy for Kriging model is proposed to achieve the construction of the adaptive Kriging model. Two cases studies, including a numerical example and a crank-slider mechanism optimization example is used to illustrate the superiorities of the proposed method. Finally, the proposed hybrid adaptive RBDO method is applied for optimization design of offshore wind turbine monopile under hybrid uncertainties. Compared with the original design, the weight of the optimized scheme decreased by 12.78%.