APO Algorithm Based On Elite Learning And Its Application In Rolling Shear Shear Mechanism

Optimization problems exist in many application fields,such as scientific research,industrial technology and so on.However,the reality of optimization is becoming more and more complex.People's demands for efficient optimization technology and optimization algorithm become increasingly urgent.The artificial physics optimization algorithm(APO)has emerged as a new intelligent optimization algorithm.Artificial physics is inspired by the laws of physics,and it is based on the Newton's second law by simulating the virtual force acting on the object and the motion of the object.By simulating the force of virtual force among objects,the adaptive value of individuals can be attracted by the better individuals,and the adaptive value of the individual to exclude the better individual,and the optimal individual to attract all the other individuals so as to achieve the optimization of the population.It has better global search ability and search ability.However,for some difficult optimization problems,it is easy to be attracted by the single global optimal solution and the local optimal solution occurs.Due to the limitation,this paper,inspired by the theory of machine learning,introduces elite learning strategies and gives a framework for the optimization algorithm of physical physics based on elite learning,allowing the individual to learn from a number of elite individuals in order to improve the diversity of individual learning.Different schemes are designed for the framework.The algorithm consists of three stages,groups,group learning,and inter group learning.This paper designs the three stages respectively,and the simulation experiments compare the different design schemes,and verify the feasibility and effectiveness of the algorithm.Then,on the basis of the framework,the algorithm is further improved by the combination of the opposition-based learning and the population diversity,and the elite learning APO optimization algorithm based on reverse learning,population diversity and adaptive iterative probability selection is proposed respectively.Through theoretical analysis and simulation experiments,it is proved that the improved algorithm is superior to other algorithms.The optimization results prove that the search precision and speed have been greatly improved,which reflects the strong optimization ability.Finally,the APO algorithm and the improved GEAPO_div algorithm are applied to the optimization design of the rolling shear mechanism respectively.On the basis of the establishment of the mechanism model,the two algorithms are used to optimize the solution.The GEAPO_div algorithm obtains the ideal trajectory curve and the rod length parameters of the mechanism and equipment,and verifies the model.Effectiveness.In order to provide convenience for users with visual interface,the design and development of simulation platform for rolling shear mechanism is realized by using MATLAB/GUI graphic development tool.In the rolling shear mechanism.