Batch Information Sharing Cuckoo Algorithm And Its Rolling Bearing Optimization Application

Rolling bearings are necessary components of all rotating machinery and are widely used in all walks of life.Although they may seem trivial,their failure is costly.Therefore,the rolling bearing needs to be optimally designed,and its structural parameters are designed to make it run smoothly and increase its service life.The optimal design of rolling bearing is an involved optimization problem with multi-dimension.The process of finding the best value is complicated,it needs massive complex calculations.In recent years,the emergence of meta-heuristic algorithm has brought a new idea to solve complex optimal value problems.By simulating the habits of some creatures in the nature,some individual positions are randomly generated in the complex problem search domain,and the fitness values of these individuals are calculated and the better solution is still retained.Finally,the value of the best solution of the target problem are found through a certain number of iterations.This kind of random searching method greatly reduces the computation amount of solving the high-dimensional complex optimal value problem.Cuckoo algorithm(CS)is an excellent meta-heuristic algorithm,which shows good performance because of its special Levy flight search strategy.This paper proposes a novel algorithm: group information sharing cuckoo algorithm(GSCS).An information sharing strategy is used to increase the relationship between individuals in the population,and the diversity capability of GSCS is increased at the early stage of iteration,make it easier for the algorithm to jump out of the neighborhood of the local minimum and find the neighborhood of the global minimum.In the later stage of the iteration,the information sharing strategy makes full use of the location information between individuals,frequently searches in the neighborhood of the optimal value,increases the local intensive search ability of the algorithm,and improves the accuracy of the final result.The batching strategy can well reduce the calculation amount of the entire algorithm and reduce the time required for algorithm optimization.The batch information cuckoo algorithm retains the excellent random search performance of the original cuckoo algorithm Levy flight search strategy,and also makes the batch information sharing strategy increase the algorithm's global search and local search performance.The batch information sharing cuckoo algorithm not only has strong convergence,but also has high solution accuracy and good stability.In order to verify the performance of the cuckoo algorithm for batch information sharing,experiments and comparisons were carried out on 16 benchmark functions with other algorithms,and comparisons were made in 10 dimensions,30 dimensions,50 dimensions and 100 dimensions,which proved The performance of batch information sharing cuckoo algorithm in dealing with difficult and complex maximum and minimum problems.And as the problem dimension increases,this excellent optimization performance can still remain unchanged.Finally,using the batch information sharing cuckoo algorithm to optimize the structure of the deep groove ball bearing,the performance of the rolling bearing is optimized from the three objective functions of basic dynamic load rating,elastohydrodynamic lubrication and minimum film thickness friction power loss.After the optimized design of the cuckoo algorithm for batch information sharing,the rolling bearing has a greater fatigue life and can accept a longer period of stable operation.