Multiple-Output Fracture Characteristics Optimization of Bi-material Interfaces for Composite Pipe Repair Using Swarm Intelligence Technique

For gas and oil pipelines undergoing corroding metal damage, a standard alternative method for reinforcing them is by using bi-material composite repair. Internal pressure at interface of the repair forms a blister and the stresses induced due to pressure. Various practical models are studied for evaluating and designing such repairs against breakdown. This study examined various optimum levels of factors affecting fracture mechanics characteristics (FMC) of failure through crack propagation at the interface using particle swarm optimization (PSO). The significant factors from crack length, composite repair material, testing models and FMC’s regression models found out using ANOVA to predict outputs without experimentations. Response surface technique discussed the effect of square and interaction levels of factors on outputs. Multiple-outputs synchronization together provided multiple set of optimum factor levels were obtained using PSO technique integrated with a genetic algorithm. It prolongs convergence to obtain multiple solutions, and PSO also used an optimal Pareto front to convert into a non-dominated useful set of solutions. Thus, Multi-output PSO analysis improved the accuracy of composite repair characteristics by having minimum thickness of composite repair while withstanding maximum pressure. Vast data on the combination of the optimal solution have been generated for every possible combination of factors. This generated data may be used as a reference manual for researchers in the field of composite repair.