Research On Improvement Of Colliding Bodies Optimization Algorithm And Its Application

The colliding bodies optimization algorithm is a meta-heuristic algorithm.Its principle is the law of collision between one-dimensional objects in physics.The basic idea is that after two objects with a certain mass and speed collide,they are separated and generate a new speed,move to a new position,and continuously loop the process to search for the global optimal solution.Because colliding bodies optimization algorithm has the advantages of simple structure,easy implementation,and not dependent on any internal parameters,it is applied in the field of scientific design and production practice by researchers at home and abroad.However,with the in-depth investigation and research of scientists,it is found that the colliding bodies optimization algorithm still needs to be strengthened in search efficiency and convergence quality.In this regard,after an in-depth study of the algorithm,this article proposes some improvement measures for the shortcomings of the algorithm,and uses the function test set to test the performance of the improved algorithm.At the same time,it is used to solve complex and multi-constrained problems and test them performance on real problems.The main research contents of this paper are as follows:(1)Propose a rotation learning-based colliding bodies optimization algorithm.First of all,the proposed new algorithm integrates the rotation learning strategy into the colliding bodies optimization algorithm.By establishing a specific circle in a two-dimensional space,adjusting different angles to search for any point in the specific circle,expand the search space of the algorithm,and increase the possibility of obtaining the optimal solution.Secondly,the mirror image processing strategy is adopted for the out-of-bounds particles,to avoid a large number of out-of-bounds objects accumulating on the search boundary and affecting the optimization accuracy of the algorithm.Finally,the function test set is used to simulate the improved algorithm and other several intelligent optimization algorithm,to test the feasibility of the improved algorithm.(2)Proposed a colliding bodies optimization algorithm based on adaptive parameter adjustment strategy and good point set strategy.The algorithm uses the best point set strategy to replace the method of randomly generating the initial population in the colliding bodies optimization algorithm.By evenly taking points in the search space to ensure the richness of the population,improve the accuracy of the algorithm,and enhance the stability of the algorithm.In order to effectively balance the exploration and development of the algorithm,it uses the logarithmic function and fitness value to redesign an adaptive parameter adjustment strategy,so that the change of the control parameter is closer to the change trend of the problem itself.Finally,a random differential mutation mechanism is introduced to colliding bodies optimization algorithm,performing mutation operations on the position update of the colliding bodies,which increase the disturbance of the algorithm,and avoid the algorithm from falling into a stagnant state in the later stage.The improved new algorithm and other comparison algorithms are tested on 23 benchmark functions.The experimental results prove that the performance of the newly proposed algorithm is better than that of other experimental algorithms.(3)Apply the improved two colliding bodies optimization algorithms to the optimal power flow calculation problem in the power system,verify the performance of the improved algorithm in solving complex and multi-constrained problems.Experiments prove that the improved colliding bodies optimization algorithm has achieved good results on the optimal power flow problem,reducing the running time and improving the optimization accuracy.