Swarm Intelligence-based Technique Research Articles

A Comparative Evaluation of Initialization Strategies for K-Means Clustering with Swarm Intelligence Algorithms

Clustering is a fundamental data analysis task that presents challenges. Choosing proper initialization centroid techniques is critical to the success of clustering algorithms, such as k-means. The current work investigates six established methods (random, Forgy, k-means++, PCA, hierarchical clustering, and naive sharding) and three innovative swarm intelligence-based approaches—Spider Monkey Optimization (SMO), Whale Optimization Algorithm (WOA) and Grey Wolf Optimizer (GWO)—for k-means clustering (SMOKM, WOAKM, and GWOKM). The results on ten well-known datasets strongly favor swarm intelligence-based techniques, with SMOKM consistently outperforming WOAKM and GWOKM. This finding provides critical insights into selecting and evaluating centroid techniques in k-means clustering. The current work is valuable because it provides guidance for those seeking optimal solutions for clustering diverse datasets. Swarm intelligence, especially SMOKM, effectively generates distinct and well-separated clusters, which is valuable in resource-constrained settings. The research also sheds light on the performance of traditional methods such as hierarchical clustering, PCA, and k-means++, which, while promising for specific datasets, consistently underperform swarm intelligence-based alternatives. In conclusion, the current work contributes essential insights into selecting and evaluating initialization centroid techniques for k-means clustering. It highlights the superiority of swarm intelligence, particularly SMOKM, and provides actionable guidance for addressing various clustering challenges.

An Intelligent Hybrid Ensemble Gene Selection Model for Autism Using DNN

Autism Spectrum Disorder (ASD) is a complicated neurodevelopmental disorder that is often identified in toddlers. The microarray data is used as a diagnostic tool to identify the genetics of the disorder. However, microarray data is large and has a high volume. Consequently, it suffers from the problem of dimensionality. In microarray data, the sample size and variance of the gene expression will lead to overfitting and misclassification. Identifying the autism gene (feature) subset from microarray data is an important and challenging research area. It has to be efficiently addressed to improve gene feature selection and classification. To overcome the challenges, a novel Intelligent Hybrid Ensemble Gene Selection (IHEGS) model is proposed in this paper. The proposed model integrates the intelligence of different feature selection techniques over the data partitions. In this model, the initial gene selection is carried out by data perturbation, and the final autism gene subset is obtained by functional perturbation, which reduces the problem of dimensionality in microarray data. The functional perturbation module employs three meta-heuristic swarm intelligence-based techniques for gene selection. The obtained gene subset is validated by the Deep Neural Network (DNN) model. The proposed model is implemented using python with six National Center for Biotechnology Information (NCBI) gene expression datasets. From the comparative study with other existing state-of-the-art systems, the proposed model provides stable results in terms of feature selection and classification accuracy.

Soft Hunting Algorithm for Auto-Tuning Software Reliability Growth Models

Nature-inspired algorithms are popular for auto-tuning software reliability growth models in recent decades due to their derivative-free natural tendency to circumvent the local optima problem. These methods have indeed exhibited enormous effectiveness in estimating software dependability. The goal of this study is to present a new nature-inspired approach for parameter improvement of system reliability predictions based on the hunting abilities of smoother-coated otters. The otters’ most notable characteristic is their ability to hunt in groups. In this study, clever otter hunting behavior is used to enhance reliability engineering parameters of the model. Otters work well together and have a strong sense of teamwork. Matte finish coated otters’ smart fish scavenging capacity distinguishes them from many other swarm intelligence-based techniques. Three stages of otter searching are accomplished: traveling in a V formation in the direction of the prey’s movements, advancing forcefully via the stream, and then assaulting the prey on the beach. Three software dependability models are utilized to validate the applicability of the suggested approach. The study’s findings demonstrate that the suggested algorithm outperformed the ABC, GA, and PSO algorithms by 75% and 50%, respectively, in terms of reduced SSE and lower MSE. The smooth-coated otter’s cognitive foraging behavior gives great gain capabilities in parameters software cost estimation system reliability analysis. The outcomes are encouraging for auto-tuning software reliability growth models. Smooth-covered otter optimization can also be used to solve other efficiency challenges.

Swarm intelligence-based technique to enhance performance of ANN in structural damage detection

Artificial neural network (ANN), a powerful technique, has been used widely over the last decades in many scientific fields including engineering problems. However, the backpropagation algorithm in ANN is based on a gradient descent approach. Therefore, ANN shows high potential in local stagnancy. Besides, choosing the right architecture of ANN for a specific issue is not an easy task to deal with. This paper introduces a simple, effective hybrid approach between an optimization algorithm and a traditional ANN for damage detection. The global search-ability of a heuristic optimization algorithm, namely grey wolf optimizer (GWO), can solve the drawbacks of ANN and also improve the performance of ANN. Firstly, the grey wolf optimizer is used to update the finite element (FE) model of a laboratory steel beam based on the vibration measurement. The updated FE model of the tested beam then is used to generate data for network training. For an effective training process, GWO is utilized to identify the optimal parameters for ANN, such as the number of the hidden nodes, the proportion of dataset for training, validation, test, and the training function. The optimization process provides an optimal structure of ANN that can be used to predict the damages in the beam. The obtained results confirm the accuracy, effectiveness, and reliability of the proposed approach in (1) alleviating the differences between measurement and simulation and (2) damage identification including damage location and severity, in the tested beam considering noise effects. For both applications, dynamic characteristics like natural frequencies and mode shapes of the beam derived from the updated FE model, are collected to calculate the objective function

A multi-objective optimization model for robust tuning of wide-area PSSs for enhancement and control of power system angular stability

A multi-objective optimization model for robust tuning of wide-area power system stabilizers (WAPSSs) aiming the improvement and control of angular stability of electric power system (EPS) is presented in this paper. In this work the WAPSSs are referred to power system stabilizers (PSSs) of fixed-structure with insertion of remote signals, from phasor measurement units (PMUs), in the control loops, that are used in a wide-area control system (WACS). The proposed model is composed of objective functions to maximize both the small-disturbance angular stability (SDAS) through the damping ratio and the large-disturbance angular stability (LDAS) using an index derived from transient energy function (TEF). Moreover, three objectives functions that contribute to the previous ones are taken into account: maximization of power system robustness for several operating conditions, minimization of eigenvalues sensitivity to small disturbances, and minimization of eigenvalues sensitivity with respect to time delay variations of remote signals. Swarm intelligence-based techniques are used to solve the multi-objective optimization model. Simulation results, with the benchmark interconnected system New England test system – New York power system (NETS-NYPS) of 16-generator and 68-bus, corroborate the applicability of the proposal.

An Enhanced Artificial Bee Colony: Naïve Bayes Technique for Optimizing Software Testing

Software driven technology has become a part of life and the quality of software largely depends on the extent of effective testing performed during various phases of development. A wide range of nature inspired searching techniques are employed over years to automate the testing process and provide promising solutions to elude the infeasibility of exhaustive testing. These techniques use metaheuristics and work by converting the problem space into search space. A subset of optimized solutions is searched that reduces overall time by shortening the testing time. Objective: An enhanced Artificial Bee Colony- Naïve Bayes optimizer for test case selection is proposed in this paper. This article also aims to provide brief insights into the emergence of hybrid swarm-inspired techniques over the last two decades. Method: The modified Artificial Bee colony is applied after component selection and further optimization is achieved using Naïve Bayes classifier. The proposed technique is implemented and evaluated taking three benchmark programs into consideration. The proposed technique is also compared to other competitive swarm intelligence-based techniques of its class. Results: The experimental results show that the proposed technique outperforms other swarm-inspired techniques in terms of execution time in a given scenario and capable of higher detection of faults with minimal test case selection. Conclusion: The proposed approach is an improvement over existing techniques and helps in huge time and cost saving. It will contribute to the testing society and enhance the overall quality of the software.

Optimal Sizing of Recycling Folded Cascode Amplifier for Low Frequency Applications Using New Hybrid Swarm Intelligence-Based Technique

ABSTRACT A new efficient design approach for sizing a high performance analog amplifier circuit namely the Recycling Folded Cascode (RFC) amplifier is presented. A RFC amplifier is an enhanced version of the conventional folded cascode amplifier and achieves better slew rate, gain, bandwidth, offset etc. for same area and power budget. Low frequency amplifiers such as biomedical or neural have a demanding requirement of low area, low power and low noise apart from meeting other optimal design specifications which have inherent trade-off amongst themselves. As a result, manual sizing becomes a computationally inefficient approach. Thus, swarm based optimization techniques have been employed to efficiently determine the optimal sizing for the RFC amplifier such that the area is minimized while meeting all the optimal design specifications considering the constraints. A new hybrid whale particle swarm optimization (HWPSO) algorithm is employed which takes advantage of the good qualities of both the whale algorithm and the PSO algorithm to optimize the area with less computational complexity. Simulations and statistical analysis have been performed and comparisons with other state of art algorithms reveals that HWPSO-based approach achieves a minimum circuit area of 21 µm2 with a mean Friedman’s statistical rank of 2.05 while meeting optimal design specifications for low frequency systems. Finally, validation with circuit design tool Cadence Virtuoso is done and pre as well as post layout analysis have been performed which further illustrated a close agreement with algorithmic results.

Optimal Sizing of Recycling Folded Cascode Amplifier for Low-Frequency Applications Using New Hybrid Swarm Intelligence-Based Technique

ABSTRACT A new efficient design approach for sizing a high-performance analog amplifier circuit namely the recycling folded cascode (RFC) amplifier is presented. An RFC amplifier is an enhanced version of the conventional folded cascode amplifier and achieves better slew rate, gain, bandwidth, offset, etc. for same area and power budget. Low-frequency amplifiers such as biomedical or neural have a demanding requirement of low area, low power, and low noise apart from meeting other optimal design specifications which have inherent trade-off among themselves. As a result, manual sizing becomes a computationally inefficient approach. Thus, swarm-based optimization techniques have been employed to efficiently determine the optimal sizing for the RFC amplifier such that the area is minimized while meeting all the optimal design specifications considering the constraints. A new hybrid whale particle swarm optimization (HWPSO) algorithm is employed which takes advantage of the good qualities of both the whale algorithm and the PSO algorithm to optimize the area with less computational complexity. Simulations and statistical analysis have been performed and comparisons with other state of art algorithms reveals that HWPSO-based approach achieves a minimum circuit area of 21 µm2 with a mean Friedman’s statistical rank of 2.05 while meeting optimal design specifications for low-frequency systems. Finally, validation with circuit design tool Cadence Virtuoso is done and pre- as well as postlayout analysis have been performed which further illustrated a close agreement with algorithmic results.

Purification of naphthalene by zone refining: Mathematical modelling and optimization by swarm intelligence-based techniques

Naphthalene based ionic liquids have been recently reported in applications of photoluminescence at room temperature. Further usages of Naphthalene cover phosphorescence aspects and liquid-phase exfoliation of graphite for graphene sheets production. Naphthalene is also known for its use in resonance studies and scintillation counters. All the aforementioned applications are sensitive to trace contaminants. One of the purification routes consists of zone refining, a large family of techniques based on the solidification/segregation theory. Separation of solute (impurity) and solvent is gradually increased through subsequent passes of a molten zone travelling along a bar. The impurities accumulate at its ends resulting into a high purity material at the intermediate region. Zone refining is a time costly batch technique since subsequent passes are needed and the movement of the molten zone is inherently slow to avoid impurity entrapment when the liquid solidifies. Since the length of the molten zone influences the impurity distribution, the challenge to achieve refining efficiency consists on establishing the best set of these lengths with a view to minimizing the molten zone passes. This work aims to fulfill this challenge using a semi-analytical/numerical model combined with two swarm artificial intelligence approaches. Experimental impurity profiles of Rhodamine after a number of molten zone passes in a Naphthalene bar are compared with theoretical predictions of the mathematical model for validation purposes. The two swarm artificial intelligence approaches applied are bio-inspired algorithms: Particle Swarm Optimization, which mimics the bird flock flying behavior - velocity and coordinate for each bird; and the Cuckoo Search, which is based on the brood parasitism of some cuckoo species by laying their eggs in the nests of other host birds. Each algorithm is shown to successfully interact with the mathematical model permitting the purification of Naphthalene by zone refining to be optimized. Both algorithms have demonstrated that better purification effect can be achieved by using larger zone lengths for the initial zone passes. It is shown that fewer iterations are needed by Cuckoo Search to achieve convergence as compared to Particle Swarm Optimization.

Generation scheduling with large-scale integration of renewable energy sources using grey wolf optimization

PurposeGeneration scheduling (GS) is the most prominent and hard-hitting problem in the electrical power industry especially in an integrated power system. Countless techniques have been used so far to solve this GS problem for proper functioning of the units in the power system to dispatch the load economically to consumers at once. Therefore, this work aims to study for the best possible function of integrated power plants to obtain the most favourable solution to the GS problem.Design/methodology/approachAn appropriate method works in a proper way and assures to give the best solution to the GS problem. The finest function of incorporated power plants should be mathematically devised as a problem and via that the aim of the GS problem to minimize the total fuel cost subject to different constraints will be achieved. In this research work, the latest meta-heuristic and swarm intelligence-based technique called grey wolf optimization (GWO) technique is used as an optimization tool that will work along with the formulated problem for correct scheduling of generating units and thus achieve the objective function.FindingsThe recommended GWO technique provides the best feasible solution which is optimal in its performance for different test cases in the GS problem of integrated power plant. It is further found that the obtained solutions using GWO method are better than the former reports of other traditional methods in terms of solution excellence. The GWO method is found to be unique in its performance and having superior computational efficiency.Practical implicationsDecision making is significant for effective operation of integrated power plants in an electrical power system. The recommended tactic implements a modern meta-heuristic procedure that is applied to diverse test systems. The method that is proposed is efficient in providing the best solutions of solving GS problems. The suggested method surpasses the early techniques by offering the most excellent feasible solutions. Thus, it is obvious that the proposed method may be the appropriate substitute to attain the optimal operation of GS problem.Social implicationsRenewable energy sources are discontinuous and infrequent in nature, and it is tough to predict them in general. Further, integrating renewable energy source-based plants with the conventional plant is extremely difficult to operate and maintain. Operation of integrated power system is full of challenges and complications. To handle those complications and challenges, the GWO algorithm is suggested for solving the GS problem and thus obtain the optimal solution in integrated power systems by considering the reserve requirement, load balance, equality and inequality constraints.Originality/valueThe proposed system should be further tested on diverse test systems to evaluate its performance in solving a GS problem and the results should be compared. Computation results reveal that the proposed GWO method is efficient in attaining best solution in GS problem. Further, its performance is effectively established by comparing the result obtained by GWO with other traditional methods.

A new hybrid strategy for data clustering using cuckoo search based on Mantegna levy distribution, PSO and k-means

Data clustering is one of the data mining techniques which is widely used in some applications, such as pattern recognition, machine learning, image processing, etc. Swarm intelligence-based algorithms are extensively used in data clustering in recent years. Cuckoo Search (CS) is one of the recently proposed algorithms in the category of swarm intelligence-based techniques. In this paper, a new hybrid algorithm which utilises CS, Particle Swarm Optimisation (PSO) and k-means has been proposed (HCSPSO). The proposed algorithm employed PSO and k-means to produce new nests in standard CS to obtain better results. It also benefits from Mantegna levy distribution to obtain higher convergence speed as well as local search. To eliminate the problem of the high number of functional evaluations in standard CS, a fraction of nests has assigned to every section of the algorithm. The proposed algorithm's performance was evaluated by ten standard benchmark datasets. Evaluation results show that the proposed algorithm is an efficient method for data clustering and produces more optimised results in comparison with standard CS, PSO, Elephant Search Algorithm (ESA), Enhanced Bat Algorithm (EBA), Bird Flock Gravitational Search Algorithm (BFGSA), Improved Cuckoo Search (ICS) and k-means.

A new hybrid strategy for data clustering using cuckoo search based on Mantegna levy distribution, PSO and k-means

Data clustering is one of the data mining techniques which is widely used in some applications, such as pattern recognition, machine learning, image processing, etc. Swarm intelligence-based algorithms are extensively used in data clustering in recent years. Cuckoo Search (CS) is one of the recently proposed algorithms in the category of swarm intelligence-based techniques. In this paper, a new hybrid algorithm which utilises CS, Particle Swarm Optimisation (PSO) and k-means has been proposed (HCSPSO). The proposed algorithm employed PSO and k-means to produce new nests in standard CS to obtain better results. It also benefits from Mantegna levy distribution to obtain higher convergence speed as well as local search. To eliminate the problem of the high number of functional evaluations in standard CS, a fraction of nests has assigned to every section of the algorithm. The proposed algorithm's performance was evaluated by ten standard benchmark datasets. Evaluation results show that the proposed algorithm is an efficient method for data clustering and produces more optimised results in comparison with standard CS, PSO, Elephant Search Algorithm (ESA), Enhanced Bat Algorithm (EBA), Bird Flock Gravitational Search Algorithm (BFGSA), Improved Cuckoo Search (ICS) and k-means.

An exploration of the literature on the use of 'swarm intelligence-based techniques' for public service problems

The importance of studying public service systems and finding robust solutions to the problems encountered in public service management has increased considerably over the past decade. One of the main objectives is to find acceptable solutions to Public Service Problems (PSPs) within an affordable period of time. However, many PSPs remain difficult to solve within a reasonable time due to their complexity and dynamic nature. This requires solving PSPs with techniques which provide efficient algorithmic solutions. There has been increasing attention in the literature to solving PSPs through the use of Swarm Intelligence-Based Techniques (SIBTs) like ant colony optimisation, particle swarm optimisation, Bee(s) Algorithm (BA), etc. This paper presents a review of Swarm Intelligence (SI) applications in public services (including PSPs in specific application areas), as well as the models and SI algorithms that have been reported in the literature.