A review of applications of animal‐inspired evolutionary algorithms in reservoir operation modelling

Abstract

Successful operation of reservoir systems to guarantee the optimal use of available water resources has been the subject of many studies. The advent and applications of evolutionary algorithms (EAs) in the field of reservoir operation have led to significant advances in our capacity to improve the planning and management of complex reservoir systems. This study reports a review of the applications of animal‐inspired EAs to reservoir operation optimization selected among a large number of available papers in this area of research. The animal‐inspired EAs herein identified concern algorithms that mimic biologic traits of animal (wild) species. Among the animal‐inspired EAs ant colony optimization (ACO), particle swarm optimization (PSO), shuffled frog leaping algorithm (SFLA), artificial bee colony (ABC), honey bee mating optimization (HBMO), firefly algorithm (FA), cuckoo search (CS) and the bat algorithm (BA) are the best‐known ones selected for this review. This paper presents a brief description of the algorithmic characteristics and various employed improved versions or varieties thereof of each of the stated EAs. Furthermore, the differences between the proposed animal‐inspired EAs and their improved versions are identified by comparing the performance of the implemented animal‐inspired EAs in the reviewed literature. PSO and its varieties have the largest number of reported applications. Our comparison results revealed that constrained, discrete and randomized varieties of the animal‐inspired EAs outperformed unconstrained, continuous and deterministic varieties, respectively because of larger feasible search space, better solution quality and shorter computational time. Moreover, all the animal‐inspired EAs outperformed traditional methods of reservoir optimization, such as nonlinear programming (NLP) and dynamic programming (DP).

Multi-Criteria Analysis of the “Lake Baikal—Irkutsk Reservoir” Operating Modes in a Changing Climate: Reliability, Resilience, Vulnerability

“…It should be noted that although we used GA to find the optimal configuration of release rules (dispatch schedules) [1][2][3], however, classical optimization methods give the best results, both in terms of computation time and the value of the objective function. In general, the authors, in contrast to [11,12], prefer to use nonlinear programming methods (the method of generalized reduced gradient), both implemented in standard programs such as Solver (Solver Excel) and developed by the authors to solve specific optimizing problems of reservoir cascades operation [13,14]. This is since OF and constraints, as a rule, are continuous, piecewise differentiable functions (unfortunately, strongly nonconvex), for which the conjugate gradient method is applicable (see Section 3 of this article).…”

“…In almost all studies [8][9][10][11][12], optimization methods were used for a comparative assessment of the existing management rules' reliability, or for the promising measures formation to improve the water management situation in the face of climate change (strategic planning).…”

“…Study [11] reported a review on applications of animal inspired EAs to reservoir operation optimization. Among them are ant colony optimization, particle swarm optimization, artificial bee colony, firefly algorithm, cuckoo search etc.…”

“…In this article (in contrast to [8][9][10][11][12]), the authors formulated a mathematical model, algorithm and computational technology, which allow, on the basis of optimization methods, to form in real time (for the current time interval) releases from the reservoir. To implement the optimization approach, a proprietary method was developed that allows the solving of large tasks in a reasonable time.…”

Optimal Reservoir of Small Reservoirs by Optimization Techniques on Reservoir Simulation Model

“…In general, water allocation from a reservoir can be considered a complex and nonlinear problem due to a large number of conditions and variables. However, from the past ten years, improvements have been made to reservoir rule curves to address complex problems using the optimization technique, which is widely studied [5,6], because it provides quantitative results consistent with the hydrological change situation, including its application in water allocation to reduce economic losses caused by pollution [7,8]. It can also be used as a tool for decision support systems (DSS) in finding the optimal land use in relation to groundwater vulnerability [9].…”