Sea Test Location Problem Based On Hybrid Particle Swarm Optimization

Sea test is one of the most important processes in the research and development of underwater weapons. A large-scale sea test relates to many resources throughout the country such as personnel/ platform/ equipment and the product being tested. Because of the long period and big consumption, how to reducing the test expenses on the premise of quality is an urgent problem to research and solve. To solve this kind of complex optimization problem, the traditional algorithms can't satisfy with the request for actual application very good for limitation of performance. Therefore, aimed at sea test location problem, how to design high-effective optimization algorithm has very good significance in theory and practice.Based on the deep research on location and job-shop scheduling problem, the characteristic of sea test location optimization problem has been analyzed. Transporting all these resources to an appropriate sea area belongs to the transportation problem. The choice of test areas may affect that of test resources and consequently test scheduling, which is a job-shop scheduling problem. Meanwhile, test scheduling will affect the choice of resources and test area. Therefore, the choice of sea area is the coupling of the transportation problem and job-shop scheduling problem.In order to solve sea test location problem, new hybrid PSO algorithm is put forward based on deep research on PSO algorithm. The algorithm adopts particle swarm optimization and queuing theory to deal with respectively precedence constraints and resource constraints of test scheduling. During the evaluation of the fitness function, a heuristic algorithm is used for resource assignment and transportation before evaluation of the total test cost, including transportation, construction cost and operational cost of the resources.The mathematic model is build for sea test location problem and the experimental simulation is carried out using the new hybrid PSO algorithm then. The experimental result shows that the algorithm can solve sea test location problem very good. The research has high reference value for research in relevant domain such as algorithm and application.