Evolutionary algorithms and their applications to engineering problems

Słowik, Adam; Kwaśnicka, Halina

doi:10.1007/s00521-020-04832-8

Cited by 377 publications

(172 citation statements)

References 105 publications

Supporting

Mentioning

170

Contrasting

Unclassified

Order By: Relevance

“…The main advantage of the GA compared to the other optimization search methods, is that the GA is more robust in finding the global optimal solutions, particularly in real-world multi-objective optimization problems. 12,38 NSGA-II is a stochastic evolutionary multi-objective algorithm. It generates for the users, in one single execution, a complete set of Pareto solutions.…”

Section: The Genetic Algorithm Nsga-iimentioning

confidence: 99%

“…The genetic algorithm (GA) is an advanced search and optimization algorithm. The main advantage of the GA compared to the other optimization search methods, is that the GA is more robust in finding the global optimal solutions, particularly in real‐world multi‐objective optimization problems 12,38 . NSGA‐II is a stochastic evolutionary multi‐objective algorithm.…”

Section: Optimization Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Smart system management and techno‐environmental optimal sizing of a desalination plant powered by renewables with energy storage

Affi

Cherif

2021

Int J Energy Res

View full text Add to dashboard Cite

The present study assessed the performances of a brackish water reverse osmosis desalination unit powered by (PV/Wind/Battery) system with hydraulic storage. The hybrid renewable energy system and the reverse osmosis desalination plant were modeled for a specific site in southern Tunisia using real meteorological and water consumption data for 1 year. A total of 8760 simulations were treated under a dynamic simulator coupled to a smart energy management strategy to reach the hourly water load demand. Although the recognized reliability of the hybrid renewable sources compared to one-power source, their design may still critical. Therefore, in this paper, the authors suggest applying the methodology of optimizing the presented unit using an integrated systemic meta-modeling which is built through the design of experiments tool and defined by an equation involving the decision variables of the optimization as its parameters. This equation outlines the objects of the optimum sizing. The selected optimal design of the proposed unit is predicted by the genetic algorithm NSGA-II based on minimizing two objectives in conflict: Embodied energy and hydraulic loss of power supply. The CPU time of the NSGA-II optimization via the meta-model reached only 15 minutes of treatment (with one day to build the design of experiments) compared to the classic optimization via dynamic simulation. The results of the optimum sizing show that it is possible to consider a low water shortage rate (0% ⩽ LPSP-H < 5%) for a marginal variation of embodied energy rate which does not exceed (7%). K E Y W O R D S design of experiments, hybrid renewable energy, meta-model, multi-objective genetic algorithm, optimization, water desalination, water-energy nexus 1 | INTRODUCTION Climate change affects many sides of the environment and severely emphasizes the issue of water scarcity. Indeed, global warming attacks the planet's humidity and the water's underground basins, and if the water management strategies don't change on the behalf of "rescuing water sources", by 2050 the number of environmental refugees will be doubled five times than the actual number. Currently, the world population is about 7.6 billion, reaching 8.6 billion by 2030 and exceeding 9 billion in 2050 whom will be in increasing need of freshwater. 1,2 Desalting, as it was known during World War II while the freshwater supplies were restricted, is likely to be the

show abstract

Section: The Genetic Algorithm Nsga-iimentioning

confidence: 99%

Section: Optimization Algorithmmentioning

confidence: 99%

Smart system management and techno‐environmental optimal sizing of a desalination plant powered by renewables with energy storage

Affi

Cherif

2021

Int J Energy Res

View full text Add to dashboard Cite

show abstract

“…For the experiments, we chose five metaheuristics, among them: GA [5,42], (µ + λ)EA [46,47,48], ABC [54,55,56], ACO [57,58], PSO [59,60,47,44,61]. These metaheuristics were based on successful applications for feature selection and other areas of applications, such as: Engineering, Computer Science, Manufacturing, and so on.…”

Section: G Metaheuristicsmentioning

confidence: 99%

“…However, the (µ + λ)EA has an extra component (a) that represents the interval width of the mutation, where the modification has a uniform probability [−a, a]. Thus, for each individual, the parameter is adjusted adaptively through random mutation events [47,48].…”

Section: G Metaheuristicsmentioning

confidence: 99%

“…Several studies have shown robust results when using metaheuristics for feature selection and other applications, such as Genetic Algorithm (GA) [5,42,43,44,45], Evolutionary Algorithm ((µ + λ)EA) [46,47,48], Bat algorithm (BA) [49,50,51,52], Artificial Bee Colony (ABC) [53,54,55,56], Ant Colony Optimization (ACO) [57,58], and Particle Swarm Optimization (PSO) [59,60,47,44,61]. Furthermore, the literature presents several other representational EA-based feature selection methods, such as variablesize cooperative co-evolutionary PSO for feature selection on high-dimensional data [62], multi-objective PSO approach for cost-based feature selection in classification [63], binary differential evolution with self-learning for multi-objective feature selection [64], cost-sensitive feature selection using two-archive multi-objective ABC algorithm [65], and returncost-based binary firefly algorithm for feature selection [66].…”

mentioning

confidence: 99%

See 1 more Smart Citation

A Novel Decomposing Model With Evolutionary Algorithms for Feature Selection in Long Non-Coding RNAs

et al. 2020

View full text Add to dashboard Cite

Machine learning algorithms have been applied to numerous transcript datasets to identify Long noncoding RNAs (lncRNAs). Nevertheless, before these algorithms are applied to RNA data, features must be extracted from the original sequences. As many of these features can be redundant or irrelevant, the predictive performance of the algorithms can be improved by performing feature selection. However, the most current approaches usually select features independently, ignoring possible relations. In this paper, we propose a new model, which identifies the best subsets, removing unnecessary, irrelevant, and redundant predictive features, taking the importance of their co-occurrence into account. The proposed model is based on decomposing solutions and is called k-rounds of decomposition features. In this model, the least relevant features are suppressed according to their contribution to a classification task. To evaluate our proposal, we extract from 5 plant species datasets, a set of features based on sequence structures, using GC content, kmer (1-6), sequence length, and Open Reading Frame. Next, we apply 5 metaheuristics approaches (Genetic Algorithm, (µ + λ) Evolutionary Algorithm, Artificial Bee Colony, Ant Colony Optimization, and Particle Swarm Optimization) to select the best feature subsets. The main contribution of this work was to include in each metaheuristic a decomposition model that uses round and voting scheme. To investigate its relevance, we select the REPTree classifier to assess the predictive capacity of each subset of features selected in 8 plant species. We identified that the inclusion of the proposed decomposition model significantly reduces the dimensions of the datasets and improves predictive performance, regardless of the metaheuristic. Furthermore, the resulting pipeline has been compared with five approaches in the literature, for lncRNA, when it also showed superior predictive performance. Finally, this study generated a new pipeline to find a minimum number of features in lncRNAs and biological sequences.

show abstract

Fermentation Process Control and Optimization

et al. 2022

View full text Add to dashboard Cite

Process control and optimization is important in ensuring a process in sustaining profitable income while maintaining the required quality. The fermentation process has attracted attention from food processing, pharmaceutical energy, and waste treatment industries due to its lower impact on the environment and lower operating cost than the conventional chemical process. However, its control and monitoring are relatively complex as living cell activity is not easily comprehended and highly nonlinear. Various control techniques applied in the fermentation process are reviewed, including a summary of the control objective, manipulated and controlled variables chosen, and their strengths and limitations in determining the optimal fermentation process.

show abstract

Evolutionary algorithms and their applications to engineering problems

Cited by 377 publications

References 105 publications

Smart system management and techno‐environmental optimal sizing of a desalination plant powered by renewables with energy storage

Smart system management and techno‐environmental optimal sizing of a desalination plant powered by renewables with energy storage

A Novel Decomposing Model With Evolutionary Algorithms for Feature Selection in Long Non-Coding RNAs

Fermentation Process Control and Optimization

Contact Info

Product

Resources

About