Optimum Design of Four Mechanical Elements Using Cohort Intelligence Algorithm
In this study, Cohort Intelligence (CI) algorithm is implemented for solving four mechanical engineering problems such as design of closed coil helical spring, belt pulley drive, hollow shaft, and helical spring. As these problems are constrained in natur
- PDF / 734,490 Bytes
- 25 Pages / 439.37 x 666.142 pts Page_size
- 79 Downloads / 211 Views
Abstract In this study, Cohort Intelligence (CI) algorithm is implemented for solving four mechanical engineering problems such as design of closed coil helical spring, belt pulley drive, hollow shaft, and helical spring. As these problems are constrained in nature, a penalty function approach is incorporated. The performance of the constrained CI is compared with other contemporary algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization, Artificial Bee Colony (ABC), Teaching–Learning-Based Optimization (TLBO), and TLBO with Differential Operator (DTLBO). The performance of the constrained CI was better than other algorithms in terms of objective function. The computational cost was quite reasonable, and the algorithm exhibited robustness solving these problems. Keywords Design of mechanical elements · Single-objective optimization · Cohort intelligence algorithm · Penalty function approach
K. Marde · A. J. Kulkarni (B) Symbiosis Institute of Technology, Symbiosis International (Deemed University), Pune 412115, Maharashtra, India e-mail: [email protected]; [email protected] K. Marde e-mail: [email protected] A. J. Kulkarni Odette School of Business, University of Windsor, 401 Sunset Avenue, Windsor, ON N9B3P4, Canada P. K. Singh ABV - Indian Institute of Information Technology and Management Gwalior, Gwalior 474015, Madhya Pradesh, India e-mail: [email protected]; [email protected] © Springer Nature Singapore Pte Ltd. 2019 A. J. Kulkarni et al. (eds.), Socio-cultural Inspired Metaheuristics, Studies in Computational Intelligence 828, https://doi.org/10.1007/978-981-13-6569-0_1
1
2
K. Marde et al.
1 Introduction Several nature-/bioinspired metaheuristic techniques have been proposed so far, such as Swarm Intelligence (SI) methods and Evolutionary Algorithms (EAs). The major SI methods include Ant Colony Optimization (ACO) [1], PSO [2], ABC [3], bacterial foraging optimization algorithm [4], bat algorithm [5], cuckoo search algorithm [6], glowworm swarm optimization [7], firefly optimization [8], predator–prey algorithm [9], etc. The EA-based methods include Genetic Algorithm (GA) [10, 11], evolutionary strategies [12], biogeography-based optimization algorithm [13], differential evolution [14], artificial immune system [15], memetic algorithms [16], learning classifier systems [17], etc. There are few other optimization algorithms that are also available such as backtracking search algorithm [18], harmony search algorithm [19], random optimization algorithm [20], random search algorithm [21], scatter search algorithm [22], tabu search algorithm [23], Teaching–Learning-Based Optimization (TLBO) [24], etc. Cohort Intelligence (CI) algorithm is an AI-based optimization methodology [25]. It is a socio-inspired optimization method in which cohort candidates learn from one another through interaction and competition to achieve a goal which is common to all. In every learning attempt, every candidate chooses certain candidate in the cohort from which it can learn certain qua
Data Loading...
