Article Preview
TopIntroduction
Engineering optimization comprises of complex types of algorithms with various application in the domain of power system networks. In the early 1950s, some scientific techniques or deterministic methods are applied like Linear Programming, Hill climbing, search techniques (gradient based), etc. for solving various engineering problems (Song et al., 2020). However, those deterministic techniques take rather long time to obtain a solution and failed to assure the optimum solution as well (Yang et al., 2016). Moreover, there is a higher chance for the optimization techniques to be trapped in a local minimum or maximum rather than achieving the optimal global solution (Kaluri & Pradeep, 2017). In the case of second generation of algorithms, most techniques are problem-specific and their mechanism mostly rely on the early assumption of the solution. Hence, these techniques such as Genetic Algorithm (GA) and Differential Evolution (DE) also require domain specific design of computational study (Bhattacharya, 2020; Kaluri & Pradeep, 2018).
The latest edition called third generation optimization techniques are known as improved heuristic techniques, evolutionary algorithms or meta-heuristic (Reddy et al., 2020). The key strength of meta-heuristic techniques is founded intensely on random initial solutions (Rahman & Mohamad-Saleh, 2018). Two of such algorithms dedicated for underwater are the Artificial Fish Swarm Algorithm (AFSA) (Li, 2013) and Whale Optimization Algorithm (WOA) (Mirjalili & Lewis, 2016). The discussed optimization techniques are inspired by the performance of supportive like following other fishes to reach the food sources and shielding the swarm against pressures during the hunting of food (Mirjalili & Lewis, 2016). In order to study the interactions among fish swarm for hunting, in the year 2003, Artificial Fish Swarm Algorithm (AFSA) had been articulated mimicking the cooperative social hunting capability among the swarm of fishes (Yan et al., 2020). Meanwhile, AFSA passed through few alternations for the enhancement of overall performances of optimization techniques. Whale Optimization Algorithm (WOA) has been introduced a decade later using a distinct hunting performance of humpback whales in the early 2016. The performances of AFSA and WOA have been compared with few well developed and very prominent bio-inspired techniques like Particle Swarm Optimization (PSO), DE, etc.
However, lately WOA has established less accurate solution, biased reliability and higher computational complexity to obtain optimal solution (Gharehchopogh & Gholizadeh, 2019). To improve the overall performance of WOA, some research studies have been applied. However, the variants are either computationally complex or poor global and local search capabilities (Ray et al., 2020). The foundation of this study is to blend the exploitation capability of WOA with the exploration of AFSA to exploit both techniques’ power. Authors further extends the previous study published in (Rahman et al., 2019) by applying the WOA variant into a power system optimization problems with related figures and Tables.
The rest of the paper is organized as follows: the following section, BACKGROUND discuss the general topics related to the work. The NEW ENHANCED WOA section explains about the proposed enhanced WOA, involving the modifications done onto it as well as presenting the performance results of the proposed enhanced WOA. The next section, APPLICATION IN RPO serves to demonstrate the practicality of the proposed enhanced WOA at solving a real-world problem. The CONCLUSION section is the final section with conclusive remarks and future recommendations.