Improvement And Application Of Spider Monkey Optimization Algorithm

Spider monkey optimization(SMO)algorithm is a new swarm intelligence optimization algorithm proposed by J.C.Bansal and other scholars in 2014.It is inspired by the foraging activities of spider monkey with fission-fusion social structure.SMO mainly updates the position through the experience of global leader,local leaders and individual group members,and divides the group into smaller groups or fuses all groups at an appropriate time.It has the characteristics of simple principle and few parameters.However,like other optimization algorithms,SMO has some problems,such as low solution accuracy,slow convergence speed and easy to fall into local optimum.And SMO is used to solve unconstrained optimization problems,while most of the problems in reality are constrained optimization problems.Therefore,the research on SMO has great potential.For unconstrained optimization problems,SMO is improved by using self-adaptive inertia weight and crisscross strategy.And two improved algorithms are proposed.Firstly,aiming at the low solution accuracy of SMO,the adaptive inertia weight based on individual function value is introduced in the local leader phase to improve the self-learning ability of spider monkey,so as to enhance the global search ability and local search ability of individuals.A modified spider monkey optimization algorithm based on self-adaptive inertia weight(SAWSMO)is proposed.Secondly,in view of the shortage that SMO is easy to fall into local optimization and the reduction of population diversity in the later stage,adding individual horizontal crossing and vertical crossing between the global leader phase and local leader learning phase of spider monkey optimization algorithm to increase population diversity.At the same time,it enhances the ability of global search and the ability to jump out of local optimization of the algorithm.A spider monkey optimization algorithm with crisscross optimization(CSMO)is proposed.The above two improved algorithms are compared with other swarm intelligent optimization algorithms in different aspects through numerical experiments.The results show that the solution accuracy of these two algorithms is higher than that of SMO,and the amount of calculation is also reduced compared with SMO.For the constrained optimization problem,the constrained optimization problem is transformed into a series of bounded constrained optimization problems by the augmented Lagrange multiplier method,and then the spider monkey optimization algorithm with crisscross is used to solve the bounded optimization problems.An improved spider monkey optimization algorithm for solving constrained optimization problems(LCSMO)is proposed.Numerical experiments are carried out on the algorithm by using the constrained optimization benchmark problem.The experimental results show that the LCSMO can effectively solve the constrained optimization problem,and the effect is better than other algorithms.And LCSMO is more stable.LCSMO is used to solve the classical pressure vessel design problem and the tension/compression spring design problem.The numerical comparison results with other algorithms shows that LCSMO can provide a better design scheme.