© 2018 JETIR December 2018, Volume 5, Issue 12 www.jetir.org (ISSN-2349-5162)
JETIR1812A19 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org 139
Review of Metaheuristic Optimization Methods for
Unit Commitment in Power System Operation and
Planning
Abdulhamid Musa
Training Officer
Electrical and Electronic Engineering Department,
Petroleum Training Institute, Effurun, Delta state, Nigeria
Abstract : Unit commitment (UC) is considered one of the essential activities in power system planning and operation. The
complexity and non-linearity nature of the UC problem make metaheuristic optimizations techniques more relevant for their
solutions. Apart from minimizing the operational cost, the increased public awareness regarding the harmful effects of atmospheric
pollutants on the environment and other environmental regulations have led researchers to focus on environmental effect as another
unit commitment objective function. This paper reviews some published research papers based on metaheuristic techniques
considering their single and multi-objective functions. However, the practical, technical and economic importance of the unit
commitment problem is proven by the enormous amount of the available literature for attaining their solutions.
Keywords : Cost of Production, Hybrid Optimization, Metaheuristic Technique, Optimization, Unit Commitment.
I. INTRODUCTION
Unit commitment (UC) problem is an important optimizing task for scheduling the on/off states of generating units with objective
function usually of minimizing cost [1, 2]. The increased public awareness regarding the harmful effects of atmospheric pollutants
on the environment, as well as the tightening of environmental regulations has led to another objective function of UC in minimizing
the environmental effect. The carbon emissions produced by fossil-fueled thermal power plants need also to be minimized. It is
necessary to consider these emissions as another objective [3]. A small improvement in the optimization like 0.5% would bring
millions of dollars cost reduction per year for a large utility grid [4]. The unit commitment problem is considered as one of the key
aspects in power system operation [5]. UC is a traditional mixed-integer non-convex problem and remains a key optimization task
in power system scheduling [4]. The UC problem (UCP) consists of deciding which power generator units must be committed or
decommitted in order to satisfy demand over a planning horizon. A short-term planning is generally split into periods of one hour
each in [6, 7] with varying loads and generations under different generational, environmental and technical constraints [7]. The
production levels at which units operate (pre-dispatch) must also be determined, and the committed units must generally satisfy the
forecasted system load and reserve requirements, as well as a large set of technological constraints [8]. Therefore, UC plays a major
role in the operation planning of power systems [6] including environmental concerns [3].
II. PREVIOUS RESEARCHES
Since increasing the number of generating units makes it difficult to solve in practice, many approaches have been introduced to
solve the UC problem [1]. For the purpose of this writeup, metaheuristic optimization approach is reviewed. The approach of dealing
with UC problems include the Binary Whale Optimization Algorithm (BWOA) [9], Binary Grey Wolf Optimizer (BGWO)
algorithm [10], Grey Wolf Optimization (GWO) Algorithm [11], a Quantum Inspired Binary Grey Wolf Optimizer (QI-BGWO)
[12], HYBRID-BAT search algorithm 10-unit system [13], an improved version of the Binary Quantum-Inspired Gravitational
Search Algorithm (BQIGSA) and proposes a new approach to solve the UC problem based on the improved BQIGSA, called
QGSA-UC. [1], a Hybrid Particle Swarm Optimization Approach With Small Population size (HPSO-SP) in [2], ‘‘local
branching’’ and an hybridization of Particle Swarm Optimization (PSO) with a Mixed Integer Programming Solver [8], Binary
Coded Modified Moth Flame Optimization Algorithm (BMMFOA) [5], Particle Swarm Optimization and Tabu Search algorithm
[14], using Lagrangian Relaxation (LR) and PSO [15], deterministic method named Cut-And-Branch [6], Combination of proposed
Weighted-Improved Crazy Particle Swarm Optimization with a Pseudo Code Based Algorithm and scenario analysis method [16],
Genetic Algorithm (GA) or Dynamic Programming (DP) to solve UC and then Shuffled BAT (BAT) technique as an evolutionary
based approach to solve the constrained Economic Load Dispatch (ELD) problem [17], an improved real-coded genetic algorithm
and an enhanced Mixed Integer Linear Programming (MILP) [18], combinations of three algorithms including Charged Search
System (CSS), PSO and Ants Colony Search (ACS) [19] and hybrid metaheuristic DEEPSO, which is combination of Differential
Evolution (DE), Evolutionary Programming (EP) and PSO [20] metaheuristic approach combined with a non-dominated sorting
known as Biased Random Key Genetic Algorithm [3, 21].