Evolutionary Computation Techniques Research Articles

Expert knowledge for the recognition of leukemic cells.

This work shows the advantage of expert knowledge for leukemic cell recognition. In the medical area, visual analysis of microscopic images has regularly used biological samples to recognize hematological disorders. Nowadays, techniques of image recognition are needed to achieve an adequate identification of blood tissues. This paper presents a procedure to acquire expert knowledge from blood cell images. We apply Gaussian mixtures, evolutionary computing, and standard techniques of image processing to extract knowledge. This information feeds a support vector machine or multilayer perceptron to classify healthy or leukemic cells. Additionally, convolutional neural networks are used as a benchmark to compare our proposed method with the state of the art. We use a public database of 260 healthy and leukemic cell images. Results show that our traditional pattern recognition methodology matches deep learning accuracy since the recognition of blood cells achieves 99.63%, whereas the convolutional neural networks reach 97.74% on average. Moreover, the computational effort of our approach is minimal, while meeting the requirement of being explainable.

Evolutionary correlation filtering based on pseudo-bacterial genetic algorithm for pose estimation of highly occluded targets

An accurate method based on evolutionary correlation filtering to solve pose estimation of highly occluded targets is presented. The proposed method performs multiple correlation operations between an input scene and a bank of filters designed in frequency-domain. Each filter is computed with statistical parameters of a real-world scene and a template that contains information of the target in a single pose parameter configuration. A vast set of templates is generated from multiple views of a three-dimensional model of the target, which are created synthetically with computer graphics. An evolutionary approach in the bank of filter construction for optimizing the pose estimation parameters is implemented. The evolutionary computation technique based on a pseudo-bacterial genetic algorithm yields high estimation accuracy finding the best filter that produces the highest matching score. The proposed evolutionary correlation filtering yields good convergence of the bank of filter optimization, which produces a reduction of the number of computational operations. Experimental results demonstrate the robustness of the proposed method in terms of detection performance and pose estimation of highly occluded targets compared with state-of-the-art methods.

An application of evolutionary computation algorithm in multidisciplinary design optimization of battery packs for electric vehicle

AbstractThe increased concerns of climate change are driving a shift in the transport sector from fossil to green fuels and electric vehicles (EVs) systems capable of delivering long‐term sustainability. However, there are some challenges remaining related to batteries used in EVs. These challenges exist in four different levels, that is, cell, module, pack, and EV level which can be easily solved by evolutionary computation (EC) techniques rather than using conventional modeling or optimization methods. In this study, a framework, using EC is proposed to solve the problems of EV applications. Firstly, a multi‐gene genetic programming is proposed to analyze the data in cell level and formulate a model for lithium‐ion battery capacity. The second case study targets optimization of battery encloser design at pack level using non‐dominated sorting genetic algorithm (NSGA‐II). Validation results manifest the outstanding optimization function of NSGA‐II and improved performance of the enclosure. Moreover, the results show improved performance of NSGA‐II when combined with other artificial intelligence algorithms. The results suggest that EC can be integrated in the EV system for monitoring its performance and ensure its safety.

On evolutionary computation techniques for multi-view triangulation

Multi-view triangulation is an essential step in recovering three-dimensional structure from a set of images. It is a well-studied problem in computer vision with many suboptimal and optimal methods based on different optimality criteria. In this paper, we assess the ability of evolutionary computation (EC) methods in finding highly accurate solutions to this problem. We use an overlaying Luus–Jaakola optimizer to find good parameter configurations and determine appropriate computational budget for the EC methods. Empirical results on synthetic and real data demonstrate the superior performance of EC methods over existing triangulation methods.

SOLVING WIND INTEGRATED DYNAMIC ECONOMIC LOAD DISPATCH USING DYNAMIC NEIGHBORHOOD BASED DIFFERENTIAL EVOLUTION

Economically satisfying the power demand of various consumers in dynamic mode becomes a prime challenge nowadays. Unbalanced load variation makes the power demand and power loss vibrant which leads to the rough load curve. To deal with differential power requirement, dynamic economic (DELD) load dispatch is witnessed as major one in consideration with unauthorized operating zone, economic load dispatch in valve pointed loading, etc. On considering the surplus power demand it is mandatory to find an alternate way to satisfy it economically, hence renewable energy support is crucial. The discontinuous power generation of wind energy with respect to the availability of wind source makes the DELD problem non smooth and non convex in practical. Hence an evolutionary computational technique is prominent to find a better solution. In proposed topology, Neighborhood based Differential Evolution Algorithm has been evaluated for resolving the wind integrated DELD problem including various constraints. To check out values attained from proposed method, 15 unit and 10 unit system test data is optimized for various power demand including the wind support and evaluated with the results from the literature.

Evolutionary Computing Techniques for Resolving Load Dispatch Problem

To have an effective scheduling of the generators in order to achieve a perfect planning and functioning of the electric power generation system so as to satisfy the demands required, the economic load dispatch is important. The economic load dispatch is very essential in the network operations in the market with the derestricted electricity and takes a vital role in the power plant operations. ELD problem scopes to meet the load demands fulfilling the various constraints in the operation and minimizing the cost of the operations. The conventional methods put forth to find the solution were found unsuitable as the cost curves of the units were assume to be uneventfully increasing linear functions and whereas the practical systems are nonlinear. So the evolutionary computing engaging the GA in combination with the PSO to resolve the economic load dispatching problem

Shuffled Frog-Leaping Programming for Solving Regression Problems

There are various automatic programming models inspired by evolutionary computation techniques. Due to the importance of devising an automatic mechanism to explore the complicated search space of mathematical problems where numerical methods fails, evolutionary computations are widely studied and applied to solve real world problems. One of the famous algorithm in optimization problem is shuffled frog leaping algorithm (SFLA) which is inspired by behaviour of frogs to find the highest quantity of available food by searching their environment both locally and globally. The results of SFLA prove that it is competitively effective to solve problems. In this paper, Shuffled Frog Leaping Programming (SFLP) inspired by SFLA is proposed as a novel type of automatic programming model to solve symbolic regression problems based on tree representation. Also, in SFLP, a new mechanism for improving constant numbers in the tree structure is proposed. In this way, different domains of mathematical problems can be addressed with the use of proposed method. To find out about the performance of generated solutions by SFLP, various experiments were conducted using a number of benchmark functions. The results were also compared with other evolutionary programming algorithms like BBP, GSP, GP and many variants of GP.

Prediction of convective clouds formation using evolutionary neural computation techniques

The prediction of convective clouds formation is a very important problem in different areas such as agriculture, natural hazards prevention or transport-related facilities. In this paper, we evaluate the capacity of different types of evolutionary artificial neural networks to predict the formation of convective clouds, tackling the problem as a classification task. We use data from Madrid-Barajas airport, including variables and indices derived from the Madrid-Barajas airport radiosonde station. As objective variable, we use the cloud information contained in the METAR and SPECI meteorological reports from the same airport and we consider a prediction time horizon of 12 h. The performance of different types of evolutionary artificial neural networks has been discussed and analysed, including three types of basis functions (sigmoidal unit, product unit and radial basis function) and two types of models, a mono-objective evolutionary algorithm with two objective functions and a multi-objective evolutionary algorithm optimised by the two objective functions simultaneously. We show that some of the developed neuro-evolutionary models obtain high quality solutions to this problem, due to its high unbalance characteristic.

Evolving deep neural networks using coevolutionary algorithms with multi-population strategy

Deep learning (DL) has achieved state-of-the-art results on benchmark datasets and been most popular and widely accepted for both industrial and research problems. Since the success of deep learning, particularly across multiple domains, and successful implementations by tech giants like Apple, Microsoft, Facebook, etc., the research focus is now diverted towards optimising DL process by using various conventional and unconventional approaches. One such approach is using evolutionary computation techniques to improve the training and learning of DL. This paper, for the first time, proposes a novel multi-population competitive and cooperative neuroevolution approach for evolving optimised deep neural networks (DNNs) to provide a warm start to the DL process. Two separate coevolutionary strategies are used on two different populations that are collectively working towards producing optimised DNNs. Furthermore, three new parameters are proposed to avoid premature convergence and to maintain genetic diversity which is considered to be one of the major problems for population-based approaches. The proposed approach is tested on four different and diversified datasets, i.e. IRIS, MNIST, prostate cancer gene expression and synthetic hierarchical dataset. The experiments are carried out with four different strategies (of two different sizes) using standard DNNs, DNNs evolved with coevolution strategies and the proposed multi-population competitive and cooperative evolutionary (evolutionary deep neural networks or simply EDN) approach. The proposed EDN approach dominated the results by reducing the training and execution times while maintaining the best classification accuracy among all other approaches for all four datasets. The classification accuracy with the proposed approach is 98.3%, 98.7%, 82.2% and 89.1% for IRIS, MNIST, gene expression dataset and synthetic dataset, respectively. With the proposed EDN, the execution time for training and testing (including evolving DNNs) is considerably reduced by over 50% for MNIST and synthetic dataset, whereas it is 15% for IRIS and gene expression datasets when compared to standard deep neural networks with random weights and fixed topology.

Nature-inspired metaheuristic techniques for automatic clustering: a survey and performance study

The application of several swarm intelligence and evolutionary metaheuristic algorithms in data clustering problems has in the past few decades gained wide popularity and acceptance due to their success in solving and finding good quality solutions to a variety of complex real-world optimization problems. Clustering is considered one of the most important data analysis techniques in the domain of data mining. A clustering problem refers to the partitioning of unlabeled data objects into a certain number of clusters based on their attribute values or features, with the objective of maximizing intra-clusters homogeneity and inter-cluster heterogeneity. This paper presents an up-to-date survey of major nature-inspired metaheuristic algorithms that have been employed to solve automatic clustering problems. Further, a comparative study of several modified well-known global metaheuristic algorithms is carried out to solve automatic clustering problems. Also, three hybrid swarm intelligence and evolutionary algorithms, namely, particle swarm differential evolution algorithm, firefly differential evolution algorithm and invasive weed optimization differential evolution algorithm, are proposed to deal with the task of automatic data clustering. In contrast to many of the existing traditional and evolutionary computational clustering techniques, the clustering algorithms presented in this paper do not require any predetermined information or prior-knowledge of the dataset that is to be classified, but rather they are capable of spontaneously identifying the optimal number of partitions of the data points during the course of program execution. Forty-one benchmarked datasets that comprise eleven artificial and thirty real world datasets are collated and utilized to evaluate the performances of the representative nature-inspired clustering algorithms. According to the extensive experimental results, comparisons and statistical significance, the firefly algorithm appeared to be more appropriate for better clustering of both low and high dimensional data objects than were other state-of-the-art algorithms. Further, an experimental study demonstrates the superiority of the three proposed hybrid algorithms over the standard state-of-the-art methods in finding meaningful clustering solutions to the problem at hand.

Traceability Link Recovery between Requirements and Models using an Evolutionary Algorithm Guided by a Learning to Rank Algorithm: Train control and management case

Traceability Link Recovery (TLR) has been a topic of interest for many years within the software engineering community. In recent years, TLR has been attracting more attention, becoming the subject of both fundamental and applied research. However, there still exists a large gap between the actual needs of industry on one hand and the solutions published through academic research on the other.In this work, we propose a novel approach, named Evolutionary Learning to Rank for Traceability Link Recovery (TLR-ELtoR). TLR-ELtoR recovers traceability links between a requirement and a model through the combination of evolutionary computation and machine learning techniques, generating as a result a ranking of model fragments that can realize the requirement.TLR-ELtoR was evaluated in a real-world case study in the railway domain, comparing its outcomes with five TLR approaches (Information Retrieval, Linguistic Rule-based, Feedforward Neural Network, Recurrent Neural Network, and Learning to Rank). The results show that TLR-ELtoR achieved the best results for most performance indicators, providing a mean precision value of 59.91%, a recall value of 78.95%, a combined F-measure of 62.50%, and a MCC value of 0.64. The statistical analysis of the results assesses the magnitude of the improvement, and the discussion presents why TLR-ELtoR achieves better results than the baselines.

Random Walk Particle Tracking Embedded Cellular Automata Model for Predicting Temporospatial Variations of Chlorine in Water Distribution Systems

Cellular Automata (CA) is an evolutionary computing technique that makes discrete idealizations of differential equations and represents the physical system at the mesoscopic scale. A novel CA approach for predicting the temporal and spatial variations of chlorine in Water Distribution Systems (WDSs) is presented in this paper. Random Walk Particle Tracking (RWPT), a stochastic Lagrangian technique, is used to represent the advection and dispersion processes. A one-dimensional CA-based reactive-transport model for chlorine, named as RWPT_CA model, incorporating advective-dispersive transport mechanism is developed and demonstrated. The significance of the cell dimension in the model algorithm is ascertained, and a deterministic approach is formulated for its selection. An indirect numerical solution technique is developed to improve the computational efficiency of the CA algorithm and to minimize the restrictions in the process of discretization of mass into equivalent particles. The numerical accuracy of the proposed RWPT_CA model is verified by applying it on to a benchmark problem. The RWPT_CA model provided excellent representations of the chlorine concentration profiles for low to medium range dispersion in WDSs. The model testing on a benchmark problem from the literature, well tested by researchers, revealed its effectiveness to derive the chlorine concentration patterns under dynamic hydraulic conditions. The dispersion mechanism was found significant in controlling the temporospatial distribution of chlorine at the nodes farther from the source nodes. The models which consider only advective transport mechanism were found over-predicting the chlorine concentrations, and thereby, establishing untrue representations of the quality of the delivered water.

Filter-Based Multi-Objective Feature Selection Using NSGA III and Cuckoo Optimization Algorithm

Feature selection aims to confiscate inappropriate features and yet improve classification performance. These aims are conflicting with one another, and a choice must be made in the presence of the trade-off between them. Numerous researches deal with feature selection problem but, they are mostly single-objective based. Nowadays, multi-objective optimisation approaches are becoming the most suitable approaches to deal with feature selection problems. They can easily create a balance between selected features and classification accuracy or error rate. Evolutionary computation techniques have been applied for multi-objective feature selection. Cuckoo optimisation algorithm is among the most popular technique that is exceptional in solving the problems of feature selection. Based on the binary cuckoo optimisation algorithm, two different multi-objective filter-based feature selection frameworks are presented with the idea of nondominated sorting genetic algorithms NSGAIII (BCNSG3) along with NSGAII (BCNSG2). Thus, four multi-objective filter-based feature selection approaches are proposed by employing mutual information along with gain ratio based-entropy as the respective filter evaluation measures in all the proposed frameworks. The results obtained are examined and analysed against the existing methods and single objective scheme on fourteen (14) datasets of varying degree of difficulties. The outcome of the experiments displays that the proposed multi-objective algorithms successfully derive a set of nondominated solutions that used the least feature size and attained the best error rate than using full-length features. In general, BCNSG2 obtained the best results compared to the existing methods and single-objective algorithm, whereas BCNSG3 outdoes all other approaches.

Multi-Cluster Jumping Particle Swarm Optimization for Fast Convergence

Particle Swarm Optimization (PSO) is a metaheuristic evolutionary computation technique inspired by the social behavior of birds and fish flock. The classical PSO has limitations of slow convergence rate and trapping in local minima, as the dimensions of the data increase. Moreover, majority of improvements made by the researchers in optimization techniques have focused on the accuracy of solution and have overlooked the convergence speed of an algorithm. Keeping in view the need of an optimization algorithm with fast convergence speed, suitable for high dimensional data space, this article proposes a novel concept of Multi-Cluster Jumping PSO. In the proposed method, the particles in the swarm are divided in different clusters to search for the global optimum solution. Each cluster in the swarm has its own cluster best position which is the best position within a cluster and the global best position is located by clusters communication. In order to avoid trapping in the local optima, a jumping strategy is incorporated for stuck particles through relocation of particles to a random new position. Instead of random initialization of the particles, a semi-random initialization is opted by dividing the entire search space and the distribution of particles over a search space is done in independent slots. The proposed approach has the ability to overcome the limitations of classical evolutionary computation methods and is suitable for high dimensional dynamic data. An extensive experimentation is carried out to optimize twelve benchmark functions using the proposed Multi-Cluster Jumping PSO and a significant difference is observed in the convergence speed of the proposed method over the existing state-of-the-art approaches.

Efficient Variable Thickness Radome Design with Insertion Phase Delay Correction

Variable thickness design is capable of yielding excellent electromagnetic (EM) performance for streamlined airborne radomes. The traditional optimization method using evolutionary computation (EC) techniques can yield variable thickness radome (VTR) design with optimal EM performance but uncontrollable thickness profile, leading to manufacturing difficulties. Thickness profile design based on average incident angle is an efficient way for VTR design with smooth thickness profile, whereas the boresight error (BSE) is relatively large. In this work, insertion phase delay (IPD) correction is introduced in the efficient thickness profile design process to realize a better balance between EM performance and thickness profile smoothness of airborne radomes. The thickness profile obtained from average incident angle is adjusted for IPD distribution symmetry with respect to the antenna axis under scan angles with large BSE. Results indicate that the proposed method can effectively improve the BSE characteristics at the expense of slight deterioration of computational efficiency and thickness profile smoothness.

Optimizing earthquake design of reinforced concrete bridge infrastructures based on evolutionary computation techniques

This paper presents a methodology for the earthquake design of reinforced concrete (RC) bridge infrastructures based on the application of multi-objective evolutionary techniques. The purpose of the methodology is to allow better decision-making for the earthquake design of bridges by proposing optimized solutions which offer trade-offs between material quantities, performance/robustness, and cost. For this, two multi-objective problems (MOPs) were defined with two sets of objectives: the first objective set optimizes the amount of material used in the piers; the second set of objectives comprises a cost function and a performance metric. The NSGA-II algorithm was adapted and applied with real coded variables and multiple non-linear dynamic analyses performed in each fitness evaluation. The results of the runs show different Pareto fronts strongly associated with solution schema of pier-deck connections and steel distribution between piers. The results also allow to perceive the influence of ductility through the impact that certain variables in certain piers have on the performance of solutions. The importance of support conditions/connections between infrastructure and deck and of confinement of piers is clear. The results of the two MOPs show that sub-optimal solutions in terms of volume of materials used may be interesting in terms of reliability gain. In the end, the results of applying the methodology present solutions which offer trade-offs and information gain valuable for bridge design.

Imperceptible Steganography Scheme with High Payload Capacity using Genetic Algorithm and Particle Swarm Optimization

Security is the most significant parameter in all type of confidential data transfers. Steganography is used to enhance the security of such confidential communications. Steganography is a method of covert communication in which the existence of secrecy is concealed. In image steganography, achieving high data embedding capacity and simultaneously retaining good visual quality is a very tricky and difficult objective. In this paper, a reversible, secure, extremely imperceptible and high payload capacity steganography technique in the spatial domain is proposed. The proposed method employs evolutionary computation techniques to identify the most optimum locations and arrangements for secret data embedding. The proposed technique uses Particle Swarm Optimization to find the best possible order of data hiding whereas Genetic algorithm is used to identify the best possible arrangements to modify secret data to produce least amount of change in cover-image. The result of the proposed scheme is compared with many steganography techniques and the proposed scheme outperforms the existing schemes in terms of imperceptibility. The proposed technique produces an average PSNR value of 46.40 dB at 2 bit per pixel data embedding rate.

Consolidation assessment using Multi Expression Programming

In this study, new approximate solutions for consolidation have been developed in order to hasten the calculations. These solutions include two groups of equations, one can be used to calculate the average degree of consolidation and the other one for computing the time factor (inverse functions). Considering the complicated nature of consolidation, an evolutionary computation technique called Multi-Expression Programming was applied to generate several non-piecewise models which are accurate and straightforward enough for different purposes for calculating the degree of consolidation for each depth and its average as well for the whole soil layer. The parametric study was also performed to investigate the impact of each input parameter on the predicted consolidation degree of developed models for each depth. Moreover, the results of the consolidation test carried out on four different clays attained from the literature showed the proper performance of the proposed models.

Total Optimization of Energy Networks in a Smart City by Multi-Swarm Differential Evolutionary Particle Swarm Optimization

This paper proposes total optimization of energy networks in a smart city (SC) by multi-swarm differential evolutionary particle swarm optimization (MS-DEEPSO). Efficient utilization of energy is necessary for reduction of CO2 emission, and SC demonstration projects have been conducted around the world for reducing total energies and the amount of CO2 emission. The problem can be formulated as a mixed integer nonlinear programming problem and various evolutionary computation techniques such as particle swarm optimization, differential evolution, and DEEPSO have been applied to the problem. However, there is still room for improving solution quality. Multi-swarm based evolutionary computation methods have been verified to improve solution quality and the approach has a possibility for improving solution quality. Considering these backgrounds, the paper proposes total optimization of energy networks in an SC by MS-DEEPSO with only migration model and with only abest model. The results of MS-DEEPSO with both migration and abest, only migration, and only abest model based methods are compared with those by the original DEEPSO based method with a single swarm. Various migration topologies, migration policies and intervals, and the number of sub-swarms are also investigated. It is verified that MS-DEEPSO with both migration and abest model based method with hyper-cube topology and the W-B policy is the most effective among all of multi-swarm parameters.

Urban Water Demand Forecasting: A Comparative Evaluation of Conventional and Soft Computing Techniques

Previous studies have shown that soft computing models are excellent predictive models for demand management problems. However, their applications in solving water demand forecasting problems have been scantily reported. In this study, feedforward artificial neural networks (ANNs) and a support vector machine (SVM) were used to forecast water consumption. Two ANN models were trained using different algorithms: differential evolution (DE) and conjugate gradient (CG). The performance of these soft computing models was investigated with real-world data sets from the City of Ekurhuleni, South Africa, and compared with conventionally used exponential smoothing (ES) and multiple linear regression (MLR). The results obtained showed that the ANN model that was trained with DE performed better than the CG-trained ANN and other predictive models (SVM, ES and MLR). This observation further demonstrates the robustness of evolutionary computation techniques amongst soft computing techniques.