The Research Of Monkey Optimization Algorithm And Its Application

Monkey optimization algorithm is a new type of population-based intelligentalgorithm which derives from the simulation of the mountain-climbing processesof monkeys. It consists of three processes: climb process, watch-jump process andsomersault process. The algorithm has simple structure, strong robustness and isnot easy to fall into local optimum solutions and “dimension disaster”. Therefore,monkey algorithm has been successfully applied to solve various complicatedoptimization problems. But the algorithm also has some shortcomings, namely lowaccuracy, long running time and being easy to escape the solution interval. In viewof these deficiencies, this paper improved the monkey algorithm to perfect thetheoretical basis of the algorithm and widen the application range of the algorithm.This paper mainly obtains following research results:(1) In view of the problem of the low accuracy and spending long running time,this paper introduces the inertial step and reduces climb number to decreaserunning time. The searching strategy of traditional simple method is introduced tomake the monkey who has poor position to reach a new location with reflex,extend, compress and shrink, so as to speed up the convergence speed and improvethe precision of the solution.(2) Aiming at the problem of monkey algorithm easily falling into the localoptimal solutions in solving the0-1knapsack problem and poor populationdiversity, this paper discretized MA, improved the somersault process andintroduced the control parameter to avoid algorithm trapped into local optimum.(3) In view of the problem of low precision of traditional methods in solvingthe clustering analysis problem and easily falling into the local optimal solutions,in this paper, the search operator of artificial bee colony is introduced to strengthenthe local search ability with the climb process and the somersault process isimproved by K-means method to overcome the deficiency of traditional methods.(4) According to the feature of airplane landing problem, this paper adopted aninteger coding method and improved the somersault process to keep the populationdiversity. Experiments show that the improved algorithm has a good performance in solving the airplane landing problem.