Research
MONSDA: – A Novel Multi-objective Non-Dominated Sorting Dragonfly Algorithm
This novel article presents the multi-objective version of the recently proposed Dragonfly Algorithm (DA) known as Non-Dominated Sorting Dragonfly Algorithm (NSDA). This proposed NSDA algorithm works in such a manner that it first collects all non-dominated Pareto optimal solutions in achieve till the evolution of last iteration limit. The best solutions are then chosen from the collection of all Pareto optimal solutions using a crowding distance mechanism based on the coverage of solutions and swarming strategy to guide dragonflies towards the dominated regions of multi-objective search spaces. For validate the efficiency and effectiveness of proposed NSDA algorithm is applied to a set of standard unconstrained, constrained and engineering design problems. The results are verified by comparing NSDA algorithm against Multi objective Colliding Bodies Optimizer (MOCBO), Multi objective Particle Swarm Optimizer (MOPSO), non-dominated sorting genetic algorithm II (NSGA-II) and Multi objective Symbiotic Organism Search (MOSOS).The results of proposed NSDA algorithm validates its efficiency in terms of Execution Time (ET) and effectiveness in terms of Generalized Distance (GD), Diversity Metric (DM) on standard unconstraint, constraint and engineering design problem in terms of high coverage and faster convergence.
Optimal Power Flow using a Hybrid Particle Swarm Optimizer with Moth Flame Optimizer
In this work, the most common problem of the modern power system named optimal power flow (OPF) is optimized using the novel hybrid meta-heuristic optimization algorithm Particle Swarm Optimization-Moth Flame Optimizer (HPSO-MFO) method. Hybrid PSO-MFO is a combination of PSO used for exploitation phase and MFO for exploration phase in an uncertain environment. Position and Speed of particle are reorganized according to Moth and flame location in each iteration. The hybrid PSO-MFO method has a fast convergence rate due to the use of roulette wheel selection method. For the OPF solution, standard IEEE-30 bus test system is used. The hybrid PSO-MFO method is implemented to solve the proposed problem. The problems considered in the OPF are fuel cost reduction, Voltage profile improvement, Voltage stability enhancement, Active power loss minimization and Reactive power loss minimization. The results obtained with hybrid PSO-MFO method is compared with other techniques such as Particle Swarm Optimization (PSO) and Moth Flame Optimizer (MFO). Results show that hybrid PSO-MFO gives better optimization values as compared with PSO and MFO which verifies the effectiveness of the suggested algorithm.
A Novel Quasi Opposition Based Passing Vehicle Search Algorithm Approach for Large Scale Unit commitment problem
This paper presents a novel approach population based metaheuristics algorithm known as Quasi Oppositional Passing Vehicle Search (QOPVS) algorithm for solve the Unit commitment problem (UCP) of thermal units in an electrical power system. Passing vehicle search (PVS) algorithm is a population based algorithm which mechanism is inspired by passing vehicles on two-lane rural highways. As algorithms are population based so enables to provide improved solution with integration of powerful techniques. In this article, such a powerful technique named Opposite based learning techniques (OBLT) is integrated with proposed PVS algorithm. OBLT provides enough strength to proposed PVS algorithm to gain a better approximation for both current and opposite population at the same time, as it provide a solution which is more nearer solution from optimal based from starting by checking both solutions. Thermal unit scheduling problem is a nonlinear, non convex, discrete, complex and constrained optimisation problem. To verify the effectiveness of the proposed QOPVS algorithm is applied to some standard benchmark test function and various IEEE test systems with the number of thermal units 5-, 6-, 10-, 20-, and 40-unit in a 24-hour load scheduling horizon. The results show an improvement in the quality of solutions obtained compared with other methods results in the literature. The proposed algorithm is considerably fast and provides feasible nearoptimal solutions. Simulations results have proved the performance of the proposed QOPVS algorithm to solving large UC problems within a faster convergence and reasonable execution time.
Non-Dominated Sorting Whale Optimization Algorithm (NSWOA): A Multi-Objective Optimization algorithm for Solving Engineering Design Problems
This novel article presents the multi-objective version of the recently proposed the Whale Optimization Algorithm (WOA) known as Non-Dominated Sorting Whale Optimization Algorithm (NSWOA). This proposed NSWOA algorithm works in such a manner that it first collects all non-dominated Pareto optimal solutions in achieve till the evolution of last iteration limit. The best solutions are then chosen from the collection of all Pareto optimal solutions using a crowding distance mechanism based on the coverage of solutions and bubble-net hunting strategy to guide humpback whales towards the dominated regions of multi-objective search spaces. For validate the efficiency and effectiveness of proposed NSWOA algorithm is applied to a set of standard unconstrained, constrained and engineering design problems. The results are verified by comparing NSWOA algorithm against Multi objective Colliding Bodies Optimizer (MOCBO), Multi objective Particle Swarm Optimizer (MOPSO), non-dominated sorting genetic algorithm II (NSGA-II) and Multi objective Symbiotic Organism Search (MOSOS). The results of proposed NSWOA algorithm validates its efficiency in terms of Execution Time (ET) and effectiveness in terms of Generalized Distance (GD), Diversity Metric (DM) on standard unconstraint, constraint and engineering design problem in terms of high coverage and faster convergence.
