Gravitational search algorithm based on multiple adaptive constraint strategy
- PDF / 1,735,581 Bytes
- 41 Pages / 439.37 x 666.142 pts Page_size
- 56 Downloads / 238 Views
Gravitational search algorithm based on multiple adaptive constraint strategy Jingsen Liu1 · Yuhao Xing2 · Yixiang Ma2 · Yu Li3 Received: 8 September 2019 / Accepted: 16 June 2020 © Springer-Verlag GmbH Austria, part of Springer Nature 2020
Abstract In order to improve the convergence speed and optimization accuracy of gravitational search algorithm, the improved gravitational algorithm with dynamically adjusting inertia weight and trend factors of speed and position is proposed. This kind of algorithm with dynamic inertia weight improves the updating way of particle mass. Moreover, the mass change has a nonlinear decreasing trend and improves the algorithm’s optimization accuracy and convergence speed. At the same time, the speed trend factor and location adaptive factor is introduced, which can dynamically constrain the moving step of each generation of particles according to the number of iterations of the current population. So the algorithm is multi-adaptive. Through classical test function and the CEC2017 benchmark function, the improved algorithm is compared and tested. The theoretical analysis proves the convergence and time complexity of the improved algorithm. Simulation results show that the improved algorithm has a remarkable improvement in terms of optimal performance, high convergence speed and optimization precision. Keywords Gravitational search algorithm · Dynamic inertia weight · Velocity trend factor · Position adaptive factor · Adaptivity Mathematics Subject Classification 90C59
* Yu Li [email protected] Jingsen Liu [email protected] Yuhao Xing [email protected] 1
Institute of Intelligent Network System, and College of Software, Henan University, Kaifeng 475004, Henan, China
2
College of Software, Henan University, Kaifeng 475004, Henan, China
3
Institute of Management Science and Engineering, Henan University, Kaifeng 475004, Henan, China
13
Vol.:(0123456789)
J. Liu et al.
1 Introduction In recent years, with the progress of science and technology, many complex combinatorial optimization problems have emerged. As for optimization problems, these traditional numerical methods cannot meet the demands of speed and convenience. In order to solve these problems, researchers have commenced seeking a new kind of algorithm. They have put forward a series of heuristic algorithms, such as ant colony algorithm [1] for simulating ant collective behaviors, particle swarm optimization [2, 3] for simulating the behaviors of bird population, water circulation algorithm [4] for simulating natural water circulation process, tabu search algorithm [5] for simulating the memory process of human intelligence, differential evolution algorithm [6] based on cooperation and competition among groups to optimize search and biogeography optimization algorithm [7] originated from genetic algorithm and particle swarm optimization. These heuristic algorithms are better than traditional numerical methods in solving various complex optimization problems [8]. Gravitational search algorithm [9] (GSA) is a new
Data Loading...
