Evolutionary Computation Approach Research Articles

Microstructural and mechanical characterization of direct energy deposited 316L stainless steel developed using an evolutionary computational approach

Wire arc additive manufacturing (WAAM) is a technology that has revolutionized the fabrication of large metallic components in various industries. However, previous studies have predominantly relied on trial and error or statistical techniques to determine process parameters, neglecting the potential of evolutionary algorithms for optimizing. In the present work, detailed investigations were performed on the WAAM-based gas tungsten arc welding to predict and optimize the process parameters of the 316L austenitic stainless steel. The input process parameters viz., current, wire feed speed, travel speed, and wire feed direction play a significant role in determining the single bead geometry. The study investigated the influence of these parameters on bead characteristics (such as height to width ratio and percentage dilution) and surface roughness. Multiobjective optimization was carried out using both the response surface methodology-desirability approach (RSM-DA) and the non-dominated sorting genetic algorithm-II (NSGA-II). In the case of NSGA-II, the optimal parameters were determined through a fuzzy decision method and it was validated experimentally. The NSGA-II approach showed superior performance over the traditional RSM-DA model in predicting the responses. In addition, the thin-walled structure printed for the parameters suggested using RSM-DA and NSGA-II approach. In a comprehensive property-based comparative analysis, the NSGA-II optimized parameters demonstrated improved mechanical properties attributed to increased low-angle grain boundaries and kernel average misorientation values. This study highlights the effectiveness of evolutionary computation methods in achieving superior performance and mechanical characteristics in the fabrication of SS316L components.

Portia spider algorithm: an evolutionary computation approach for engineering application

The Portia spider, a notable member of the jumping spider family (Salticidae), is widely recognized for its intricate hunting strategies and remarkable problem-solving prowess. Several species fall under the “Portia” genus, with habitats spanning regions in Africa, Asia, and Australia. Demonstrating the ability to tackle new challenges, these spiders can learn and adapt their strategies based on prior experiences. This study introduces the Portia Spider Algorithm (PSA), a swarm-based technique inspired by the unique predatory strategies of the Portia spider. We conducted rigorous assessments of PSA performance against 23 classical test functions, 29 CEC2017 test cases, and 5 engineering optimization tasks. To demonstrate the effectiveness of the PSA, outcomes were juxtaposed with those of renowned algorithms. This paper explores the mechanics, advantages, and potential applications of PSA within the vast domain of computational optimization.Graphical

PoMeLo: a systematic computational approach to predicting metabolic loss in pathogen genomes

BackgroundGenome streamlining, the process by which genomes become smaller and encode fewer genes over time, is a common phenomenon among pathogenic bacteria. This reduction is driven by selection for minimized energy expenditure in a nutrient-rich environment. As pathogens evolve to become more reliant on the host, metabolic genes and resulting capabilities are lost in favor of siphoning metabolites from the host. Characterizing genome streamlining, gene loss, and metabolic pathway degradation can be useful in assessing pathogen dependency on host metabolism and identifying potential targets for host-directed therapeutics.ResultsPoMeLo (Predictor of Metabolic Loss) is a novel evolutionary genomics-guided computational approach for identifying metabolic gaps in the genomes of pathogenic bacteria. PoMeLo leverages a centralized public database of high-quality genomes and annotations and allows the user to compare an unlimited number of genomes across individual genes and pathways. PoMeLo runs locally using user-friendly prompts in a matter of minutes and generates tabular and visual outputs for users to compare predicted metabolic capacity between groups of bacteria and individual species. Each pathway is assigned a Predicted Metabolic Loss (PML) score to assess the magnitude of genome streamlining. Optionally, PoMeLo places the results in an evolutionary context by including phylogenetic relationships in visual outputs. It can also initially compute phylogenetically-weighted mean genome sizes to identify genome streamlining events. Here, we describe PoMeLo and demonstrate its use in identifying metabolic gaps in genomes of pathogenic Treponema species.ConclusionsPoMeLo represents an advance over existing methods for identifying metabolic gaps in genomic data, allowing comparison across large numbers of genomes and placing the resulting data in a phylogenetic context. PoMeLo is freely available for academic and non-academic use at https://github.com/czbiohub-sf/pomelo.

A Self-Adaptive Meta-Heuristic Algorithm Based on Success Rate and Differential Evolution for Improving the Performance of Ridesharing Systems with a Discount Guarantee

One of the most significant financial benefits of a shared mobility mode such as ridesharing is cost savings. For this reason, a lot of studies focus on the maximization of cost savings in shared mobility systems. Cost savings provide an incentive for riders to adopt ridesharing. However, if cost savings are not properly allocated to riders or the financial benefit of cost savings is not sufficient to attract riders to use a ridesharing mode, riders will not accept a ridesharing mode even if the overall cost savings is significant. In a recent study, the concept of discount-guaranteed ridesharing has been proposed to provide an incentive for riders to accept ridesharing services through ensuring a minimal discount for drivers and passengers. In this study, an algorithm is proposed to improve the performance of the discount-guaranteed ridesharing systems. Our approach combines a success rate-based self-adaptation scheme with an evolutionary computation approach. We propose a new self-adaptive metaheuristic algorithm based on success rate and differential evolution for the Discount-Guaranteed Ridesharing Problem (DGRP). We illustrate effectiveness of the proposed algorithm by comparing the results obtained using our proposed algorithm with other competitive algorithms developed for this problem. Preliminary results indicate that the proposed algorithm outperforms other competitive algorithms in terms of performance and convergence rate. The results of this study are consistent with the empirical experience that two people working together are more likely to come to a correct decision than they would if working alone.

Particle swarm optimization based leader-follower cooperative control in multi-agent systems

Multi-agent systems (MAS) have attracted significant attention in recent years due to their wide applications in cooperative control, formation control, synchronization of complex networks, and distributed coordination. A fundamental problem in MAS is the leader-follower consensus or cooperative tracking, where the followers are required to track the state trajectory of the leader agent. To solve the leader-follower consensus problem, we propose a novel evolutionary computation approach to design the optimal distributed control protocols for leader-follower MAS. First, we formulate the design of distributed control gains for leader-follower consensus as an optimization problem to minimize tracking errors. Then, we leverage particle swarm optimization as an efficient evolutionary technique for distributed gain optimization in multi-agent networks. Finally, we guarantee stability for the closed-loop dynamics under directed communication topologies based on algebraic graph theory. The simulation results indicate that the proposed method yields a diminished tracking error, expedites the convergence process, and minimizes the requisite control effort while enhancing computational efficiency. Furthermore, these results exemplify the method's versatility when applied to nonlinear dynamic scenarios, directed network topologies, fluctuating disturbances, and optimization across multiple domains.

Handling Large-Scale Document Collections using Information Retrieval in the Age of Big Data

This paper's primary goal is to present an overview of big data and its analysis utilizing various methodologies, particularly evolutionary computing techniques, which improve information retrieval over standard search methods. is obtainable. Big data is defined as a huge, diverse collection of data that is difficult to handle with conventional computational methods and necessitates the use of more sophisticated statistical approaches in order to extract pertinent information from the data. Along with providing an overview of evolutionary computational approaches, this study also discusses some of the main models used for information retrieval.

Bio‐Inspired Morphological Evolution of Metastructures with New Operation Modalities

Bio-Inspired Metastructure Evolution In article number 2300019, Qianyun Zhang, Amir H. Alavi, and co-workers explore the principles of natural evolution for design and discovery of thousands of metastructures with hitherto unknown structures and new modalities of operation. The evolutionary computational approach uses representative unit cells with simple shapes to launch the evolutionary process and morphologically evolves complicated microstructures over generations. The framework is used to explore a series of 2D and 3D mechanical metamaterial structures with maximum bulk modulus, maximum shear modulus and minimum Poisson’s ratio.

A Survey on Evolutionary Computation for Computer Vision and Image Analysis: Past, Present, and Future Trends

Computer vision (CV) is a big and important field in artificial intelligence covering a wide range of applications. Image analysis is a major task in CV aiming to extract, analyse and understand the visual content of images. However, image-related tasks are very challenging due to many factors, e.g., high variations across images, high dimensionality, domain expertise requirement, and image distortions. Evolutionary computation (EC) approaches have been widely used for image analysis with significant achievement. However, there is no comprehensive survey of existing EC approaches to image analysis. To fill this gap, this paper provides a comprehensive survey covering all essential EC approaches to important image analysis tasks including edge detection, image segmentation, image feature analysis, image classification, object detection, and others. This survey aims to provide a better understanding of evolutionary computer vision (ECV) by discussing the contributions of different approaches and exploring how and why EC is used for CV and image analysis. The applications, challenges, issues, and trends associated to this research field are also discussed and summarised to provide further guidelines and opportunities for future research.

Multi-objective distributed Web service composition—A link-dominance driven evolutionary approach

Service-oriented architecture (SOA) encourages the creation of modular applications involving Web services as the reusable components. Data-intensive Web services manipulate and deal with the massive data that emerged by technological advances and their various applications. Distributed Data-intensive Web Service Composition (DWSC) includes the selection of data-intensive Web services from diverse locations on the network and composing them into a service to accomplish a complex task. Optimising Quality of Service (QoS) while satisfying the functional requirements is a fundamental challenge for service developers. The multi-objective and distributed nature of the DWSC problem demands satisfying algorithms to automatically produce the Pareto optimal set of composite services. In this paper, we propose a new Evolutionary Computation (EC) approach based on the Non-dominated Sorting Genetic Algorithm (NSGA-II) and a novel local search technique to effectively solve the multi-objective distributed DWSC problems. Our local search technique employs an innovative concept, communication link dominance, to incorporate domain-specific knowledge. The performance of our proposed link-dominance driven EC approach is evaluated on benchmark datasets. The results show that our proposed method has the highest quality with acceptable execution time for most of the composition tasks among competing algorithms.

Development and Comparison of Ten Differential-Evolution and Particle Swarm-Optimization Based Algorithms for Discount-Guaranteed Ridesharing Systems

Savings on transportation costs provide an important incentive for shared mobility models in smart cities. Therefore, the problem of maximizing cost savings has been extensively studied in the ridesharing literature. Most studies on ridesharing focus on the maximization of the overall savings on transportation costs. However, the maximization of the overall savings on transportation costs may satisfy users’ expectations for cost savings. For people to adopt ridesharing as a means to reduce costs, a minimal expected cost savings discount must be offered. There is obviously a gap between the existing studies and the real problems faced by service providers. This calls for the development of a study to formulate a ridesharing model that guarantees the satisfaction of a minimal expected cost savings discount. In this paper, we considered a discount-guaranteed ridesharing model that ensures the provision of a minimal expected cost savings discount to ridesharing participants to improve users’ satisfaction with the ridesharing service in terms of cost savings. The goal was to maximize the overall cost savings under certain capacity, spatial, and time constraints and the constraint that the discount offered to ridesharing participants could be no lower than the minimal expected cost savings discount. Due to the complexity of the optimization problem, we adopted two evolutionary computation approaches, differential evolution and particle swarm optimization, to develop ten algorithms for solving the problem. We illustrated the proposed method by an example. The results indicated that the proposed method could guarantee that the discount offered to ridesharing participants was greater than or equal to the minimal expected cost savings discount. We also conducted two series of experiments to assess the performance and efficiency of the different solution algorithms. We analyzed the results to provide suggestions for selecting the appropriate solution algorithm based on its performance and efficiency.

Evolutionary natural-language coreference resolution for sentiment analysis

Communicating messages on social media usually conveys much implicit linguistic knowledge, which makes it difficult to process texts for further analysis. One of the major problems, the linguistic coreference resolution task involves detecting coreference chains of entities and pronouns that coreference them. It has mostly been addressed for formal and full-sized text in which a relatively clear discourse structure can be discovered, using Natural-Language Processing techniques. However, texts in social media are short, informal and lack a lot of underlying linguistic information to make decisions so traditional methods can not be applied. Furthermore, this may significantly impact the performance of several tasks on social media applications such as opinion mining, network analysis, sentiment analysis, text categorization. In order to deal with these issues, this research address the task of linguistic co-referencing using an evolutionary computation approach. It combines discourse coreference analysis techniques, domain-based heuristics (i.e., syntactic, semantic and world knowledge), graph representation methods, and evolutionary computation algorithms to resolving implicit co-referencing within informal opinion texts. Experiments were conducted to assess the ability of the model to find implicit referents on informal messages, showing the promise of our approach when compared to related methods.

Adaptive dynamic surrogate-assisted evolutionary computation for high-fidelity optimization in engineering

Surrogate-assisted evolutionary computation have received much more attention in the field of optimization because of its ability to reduce effectively the CPU cost. However, as the dimension of the search space increases, the number of samples required to construct the global accurate surrogate model will increase exponentially. In order to improve the computational efficiency for achieving high-precision optimization in high-dimensional search space, an adaptive dynamic surrogate-assisted evolutionary computation approach based on variable search region is presented in this paper. The basic idea of the present method is to abandon the high accurate approximation of surrogate model on the global search space which requires a large number of samples, but focus on the search of the smaller local region where the optimal solution is located. Then refine the samples to construct a higher-precision local surrogate model on the smaller local search region, which moves with the movement of the current optimal solution. Numerical experiments show that the highly accurate optimal solution can be obtained by less than five times of adaptation. It is also applied successfully to the aerodynamic shape design optimization of transonic airfoil and wing, and the results show that present adaptive approach greatly improves the computational efficiency by about ten times compared with the traditional static global approximation surrogate model.

Optimization of gas cyclone design using evolutionary computing approach

The concept of Evolutionary Computing method covers the process of searching for an optimal solution inspired by natural evolution. It can also be viewed as a family of trial and error problem solvers which can be considered as global optimization methods with a metaheuristic or stochastic optimization concept, characterized by the use of a population of candidate solutions. Such methods include Genetic Algorithm, Particle Swarm Intelligence and Differential Evolution among others. The conventional approach adopted in the design process is to use mathematical models and sensitivity approach to obtain relevant optimal design parameters. Accurate computation and optimization of the design parameters of new equipment is one of the main concerns of design engineers. The goal here is to apply evolutionary computing methods to design a gas cyclone with optimum design parameters taking into cognisance that the optimization process is complicated which requires an extensive search of a very large input space. The motivation of this research effort is the avoidance of complex mathematical models and sensitivity approach for gas cyclone design. The result shows that a hybrid Differential Evolution based Particle Swarm Optimization outperformed standard Genetic Algorithm, Particle Swarm Intelligence and Differential Evolution.

Genetic Algorithm Approach for Image Fusion: A Simple Method and Block Method

The sensors available nowadays are not generating images of all objects in a scene with the same clarity at various distances. The progress in sensor technology improved the quality of images over recent years. However, the target data generated by a single image is limited. For merging information from multiple input images, image fusion is used. The basis of image fusion is on the image acquisition as well as on the level of processing and under this many image fusion techniques are available. Several input image acquisition techniques are available such as multisensor, multifocus, and multitemporal. Also, image fusion is performed in four different stages. These levels are the level of the signal, pixel level, level of feature, and level of decision-making. Further, the fusion methods are divided into two domains i.e spatial and frequency domains. The fusion in spatial domain images uses inputs directly to work on pixels, while the transition refers to frequency domain image fusion on input images before fusion. The limitation of spatial domain image fusion is spectral degradation. To overcome this limitation, the fusion of transform domain images is preferred which uses several transforms. The results generated by transform methods are superior to spatial domain methods. But there is a scope to improve the results or to find the optimized results. Optimization can be achieved by using evolutionary approaches. The evolutionary computation approach is an effective way of finding the required solution for a complex problem. An evolutionary algorithm is a guided random search used for optimization. The biological model of evolution and natural selection inspires it. The different types of evolutionary computing algorithms include Genetic algorithm, Genetic Programming, Evolutionary programming, Learning Classifier System, Ant Colony Optimization, Artificial Bee Colony Optimization, Particle Swarm Optimization, Evolution strategy, Swarm intelligence, Tabu Search, Cuckoo Search, etc. Three genetic algorithm-based image fusion techniques are proposed: a genetic algorithm with one population, a genetic algorithm with separate populations, and a block method. In the block method, an array of numbers in one chromosome is generated. The result obtained by the proposed techniques are compared with existing methods and observed that the results are improved. The graphical representation of performance parameters reflects that the block method is better.

Regularization-free multicriteria optimization of polymer viscoelasticity model

This paper introduces a multiobjective optimization (MOP) method for nonlinear regression analysis which is capable of simultaneously minimizing the model order and estimating parameter values without the need of exogenous regularization constraints. The method is introduced through a case study in polymer rheology modeling. Prevailing approaches in this field tackle conflicting optimization goals as a monobjective problem by aggregating individual regression errors on each dependent variable into a single weighted scalarization function. In addition, their supporting deterministic numerical methods often rely on assumptions which are extrinsic to the problem, such as regularization constants and restrictions on parameter distribution, thereby introducing methodology inherent biases into the model. Our proposed non-deterministic MOP strategy, on the other hand, aims at finding the Pareto-front of all optimal solutions with respect not only to individual regression errors, but also to the number of parameters needed to fit the data, automatically reducing the model order. The evolutionary computation approach does not require arbitrary constraints on objective weights, regularization parameters or other exogenous assumptions to handle the ill-posed inverse problem. The article discusses the method rationales, implementation, simulation experiments, and comparison with other methods, with experimental evidences that it can outperform state-of-art techniques. While the discussion focuses on the study case, the introduced method is general and immediately applicable to other problem domains.

A multi-objective worker selection scheme in crowdsourced platforms using NSGA-II

Crowdsourcing has led to a paradigm shift in how commercial houses execute projects by lowering the production cost. A crucial aspect of crowdsourcing is selecting the best set of workers to perform a task. The environment envisaged in this work is an independent pool of workers, each equipped with a pre-defined set of skills. We assume that these skills do not follow any priority order over each other. Given a task with a set of required skills, we aim to select a team of workers who can collectively fulfil the task’s requirements, while maximizing the collective expertise and minimizing the team cost. We propose a nondominated sorting genetic algorithm II (NSGA II) based algorithm to find the best set of workers that can perform the task. The basic operators of the evolutionary computation approach are tuned in accordance with our problem objectives. We perform a detailed analysis to show that the solution is well distributed along the Pareto optimal, converges exponentially apropos the number of generations and is cost-efficient. The proposed approach is compared with an optimal strategy, a greedy-based method, and another evolutionary-based algorithm to establish its effectiveness. Extensive simulation results using real data set and a synthetic data set were presented to validate our claim.

Early prediction of cycle life for lithium-ion batteries based on evolutionary computation and machine learning

Accurate early cycle life prediction of lithium-ion batteries is critical for efficient and rational battery energy distribution and saving the technology development period. However, relatively little research has been carried out on the early prediction based on evolutionary computation approaches. The principal purpose of this study is to explore the accurate early prediction of cycle life for lithium-ion batteries based on evolutionary computation techniques and machine learning approaches. Firstly, twenty features related to the capacity degradation are extracted using data from only the first 100 cycles. Next, four filter methods, four wrapper feature selection methods based on evolutionary computational strategies, and fusion feature selection methods are used to determine their feature subsets combined with the elastic net, respectively. Afterward, seven machine learning methods are selected to conduct comparative performance studies combined with the optimal feature set chosen. Finally, the research results show that the fusion feature selection method that combines Pearson correlation coefficient and differential evolution-based wrapper approach performs the best cycle life early prediction performance. Among, the prediction and average percentage errors are as small as 43.38 cycles and 5.21%, respectively; the coefficient of determination is as high as 0.98.

Trust-Based Recommendation for Shared Mobility Systems Based on a Discrete Self-Adaptive Neighborhood Search Differential Evolution Algorithm

Safety is one concern that hinders the acceptance of ridesharing in the general public. Several studies have been conducted on the trust issue in recent years to relieve this concern. The introduction of trust in ridesharing systems provides a pragmatic approach to solving this problem. In this study, we will develop a trust-aware ridesharing recommender system decision model to generate recommendations for drivers and passengers. The requirements of trust for both sides, drivers and passengers, are taken into consideration in the decision model proposed in this paper. The decision model considers the factors in typical ridesharing systems, including vehicle capacities, timing, location and trust requirements, etc. The decision model aims to determine the shared rides that minimize cost while respecting the trust and relevant constraints. As the decision problem is a nonlinear integer programming problem, we combine a self-adaptive neighborhood search with Differential Evolution to develop an algorithm to solve it. To assess the effectiveness of the proposed algorithm, several other evolutionary computation approaches are also applied to solve the same problem. The effectiveness assessment is done based on the performance of applying different algorithms to find solutions for test cases, to provide a guideline for selecting a proper solution approach.

EvoDCNN: An evolutionary deep convolutional neural network for image classification

Developing Deep Convolutional Neural Networks (DCNNs) for image classification is a complicated task that needs considerable effort and knowledge. By employing an evolutionary computation approach, one can automatically generate the network models. However, the Neuroevolution is computationally expensive, and in some cases it needs hundreds of GPU days for training. Therefore, there is a need to find optimum Neuroevolutionary models with minimum computation to deal with this problem. In this paper, by utilising a Genetic Algorithm (GA), we introduce EvoDCNN, as a block-based evolutionary model for developing an evolutionary deep convolutional network for image classification. Such that by using the proposed fixed-length encoding model, we can generate variable-length networks with high accuracy while using less computation. The proposed model by utilising a straightforward evolutionary framework is able to establish small networks with high classification accuracy. Eight datasets: CIFAR10, MNIST, and six versions of EMNIST, that include balanced and unbalanced datasets, are used for evaluation of the proposed model. We did a comprehensive evaluation where we compared the results with many previous works, and outperformed the previous state-of-the-art accuracy for classification of five of the datasets.

Consequence of natural gas injection in blast furnace: a critical appraisal using a thermodynamic and evolutionary computation approach

ABSTRACT The data generated from a model-based simulator are used in this study for evolutionary multi-objective Optimisation, correlating input parameters like natural gas and pulverised coal injection with coke rate, carbon rate and fuel cost. Evolutionary optimisers are used to minimise coke input and total fuel cost and corresponding CO2 emissions were analysed. Co-injection of gas with coal in a blast furnace generally lowers furnace CO2 emissions by increasing hydrogen in the input fuel. This study reveals that major emission reduction possibilities at the lowest optimised coke rate and total fuel cost do not exist when coal and gas co-injection is attempted. A tri-objective Optimisation study was also conducted for simultaneous minimisation of coke rate, carbon rate and fuel cost with co-injection of gas and coal and several trends emerged. Although significant reductions of coke rate and fuel costs were predicted, no marked reduction of carbon rate, hence emissions, was foreseen.