Research And Application About Particle Swarm Optimization Algorithms

As a burgeoning evolutionary computation technology, swarm intelligence is now becoming a new research hotspot. All researches of theories and applications related to swarm intelligence proves that it is a kind of effective method to solve most global optimal problems. It is more important that its potential parallel and distributed features provide technical assurance to process data in large database. Compares with the traditional optimized algorithm, the particle swarm algorithm in the multi-dimensional function optimization, the dynamic goal seeks the excellent aspect to have the convergence rate to be quick, the solution quality is high, robustness good and so on merits, especially qualify project application. For the defects of particle swarm optimization algorithm such as lower search velocity, being prone to getting into local best position in later evolution phase, longer search time and lower precision, it is essential to make some researches about improvement of PSO.The main two research parts in the paper are following:Two kinds of improvement particle swarm algorithm is presented to solve the problem that the linearly decreasing weight of the particle swarm algorithm cannot adapt to the complex and nonlinear optimization process. The evolution speed factor and aggregation degree factor of the swarm are introduced in DIPSO. Regarding asks the minimum value the optimized question: The evolutionary speed factor is smaller, indicated that the evolutionary speed rate is quicker, the algorithm may search continually in the big search space, may reduce the inertia weight the value, causes particle swarm algorithm in the small scope the spatial search, so that quicker found the optimal value. If the swarm is scattered, the swarm is not easy to fall into the local optimal value, along with the aggregation degree's enhancement of swarm, the algorithm easy to fall into the local optimal solution, this time, should increase the inertia weight, like this increased the swarm search space, enhances gobal optimization ability of the swarm. Then, the improvement algorithm's inertia weight may is formulated as a function of the evolutionary speed factor and the aggregation degree factor. In each iteration process, the weight is changed dynamically based on the current evolutionary speed factor and the aggregation degree factor, which provides the algorithm with effective dynamic adaptability. The community sufficiency takes the inertia weight's controlled variable in ARIWPSO, causes the inertia weight along with the community sufficiency change, proposed the adaptive inertia weight Particle Swarm Optimization.In view of proposed that a improvement and constrained single–objective particle swarm optimization, to its three project example (second-level helical gear reduction gear, column screw compression spring, column screw tension spring) has carried on the optimization design. The result indicated: it is practical and feasible that based on two kinds of improvement and constrained single–objective particle swarm optimization, has provided the new mentality and the method for the complex machinery optimization design.