Investigation of Suitable Perturbation Rate Scheme for Spider Monkey Optimization Algorithm
Spider Monkey Optimization (SMO) is a new metaheuristic whose strengths and limitations are yet to be explored by the research community. In this paper, we make a small but hopefully significant effort in this direction by studying the behaviour of SMO un
- PDF / 563,760 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 70 Downloads / 183 Views
Abstract Spider Monkey Optimization (SMO) is a new metaheuristic whose strengths and limitations are yet to be explored by the research community. In this paper, we make a small but hopefully significant effort in this direction by studying the behaviour of SMO under varying perturbation rate schemes. Four versions of SMO are proposed corresponding to constant, random, linearly increasing and linearly decreasing perturbation rate variation strategies. This paper aims at studying the behaviour of SMO technique by incorporating these different perturbation rate variation schemes and to examine which scheme is preferable to others on the benchmark set of problems considered in this paper. A benchmark set of 15 unconstrained scalable problems of different complexities including unimodal, multimodal, discontinuous, etc., serves the purpose of studying this behaviour. Not only numerical results of four proposed versions have been presented, but also the significance in the difference of their results has been verified by a statistical test. Keywords Metaheuristics Perturbation rate
Spider monkey optimization Control parameters
List of symbols Dim R(a, b) NG Pr G[k] I[k][0] I[k][1]
No. of dimensions Uniformly generated random number between a and b Number of groups in current swarm Perturbation rate Number of members in the kth group Index of first member of the kth group in the swarm Index of last member of the kth group in the swarm
Kavita Gupta (&) K. Deep Department of Mathematics, Indian Institute of Technology Roorkee, Roorkee 247667, Uttarakhand, India e-mail: [email protected] K. Deep e-mail: [email protected] © Springer Science+Business Media Singapore 2016 M. Pant et al. (eds.), Proceedings of Fifth International Conference on Soft Computing for Problem Solving, Advances in Intelligent Systems and Computing 437, DOI 10.1007/978-981-10-0451-3_75
839
840
Si sij Snew Sr llkj glj sminj smaxj fitness (Si) LLCk
Kavita Gupta and Kusum Deep
Position vector of ith spider monkey in the swarm jth coordinate of the position of ith spider monkey A trial vector for creating a new position of a spider monkey Position vector of randomly selected member of group jth co-ordinate of the local leader of the kth group jth co-ordinate of the global leader of the swarm Lower bound on the jth decision variable Upper bound on the jth decision variable Fitness of the position of ith spider monkey Limit count of the local leader of the kth group
1 Introduction Last few decades witness the successful application of metaheuristics in solving real-world optimization problems [1]. Genetic Algorithms (GA) [2], Ant Colony Optimization (ACO) [3], Particle Swarm Optimization (PSO) [4], Differential Evolution (DE) [5], Artificial Bee Colony (ABC) [6], etc., are just a few names in the emerging list of metaheuristics which have been successfully applied to find solutions of complex real-world optimization problems. Metaheuristics are placed in the category of modern optimization techniques. The main reason of increasing popul
Data Loading...
