Directed Self-learning Hybrid Swarm Intelligence Algorithm And Its Application In Hydraulic Reliability Optimization

The hydraulic system is a complex system of electro-hydraulic fluid coupling,and its reliability is the key to the operation of the main engine.Combining advanced optimization techniques to optimize system reliability can achieve high reliability,low cost and other goals,but hydraulic system optimization problems usually have nonlinear and uncertain characteristics,which cannot be effectively solved by traditional methods.Therefore,the optimization technology based on swarm intelligence algorithm provides a better solution to this problem.The existing single biomimetic swarm intelligent algorithm has optimization and application limitations.For this reason,focusing on the optimization issues related to hydraulic systems,the hybrid swarm intelligence algorithm mixed by standard particle swarm optimization algorithm,two-stage particle swarm optimization algorithm and ant colony algorithm based on the static and the dynamic topology is studied.Firstly,the intelligent algorithm for single bionic behavior has the defects of local optimality,poor robustness and slow convergence in the optimization process.Standard particle swarm optimization algorithm,two-stage particle swarm optimization algorithm and ant colony algorithm are co-evolved.And a static topology is used to establish the relationship among the three algorithms in the iterative process.The static topology hybrid swarm intelligent algorithm is proposed,to search optimal solution based on information migration and knowledge sharing.And the population diversity and optimization ability of the proposed algorithm are tested and analyzed.The proposed algorithm is used to optimize the reliability of the bridge system.By comparing with other single bionic search algorithms,better optimization performance of the proposed algorithm is verified.Secondly,in order to make full use of the advantages of sparse topology and dense topology algorithms,standard particle swarm optimization algorithm,two-stage particle swarm optimization algorithm and ant colony algorithm are combined with a directed self-learning topology whose density changes with iterative changes.A directed self-learning hybrid swarm intelligence algorithm is proposed and tested.The optimization performance of the proposed algorithm is tested,and the connection status between sub-algorithms is analyzed in the case of changes in topology metric parameters.The proposed algorithm is used to optimize the reliability of the bridge system and the scheduling of the hydraulic valve block shop to ensure that the best optimization targets are obtained under the constraint conditions,and the comparison shows that the directed self-learning topology can more effectively combine the advantages of different algorithms.Finally,aiming at the optimization of the reliability distribution of hydraulic working system and the polymorphic system,based on the analysis of the TS fault tree and the general generator function,an optimization model with different optimization objectives is established,and a directed self-learning hybrid group intelligent algorithm is applied to solve it.By comparing with single bionic search algorithm,static topology hybrid swarm intelligent algorithm and algorithm optimization results in the literature,the superiority of the proposed algorithm for solving complex optimization problems is verified.