Swarm Intelligence Algorithms Research Articles

Long, short, and mediumterms wind speed prediction model based on LSTM optimized by improved moth flame optimization algorithm.

Time series prediction of wind speed has been widely used in wind power generation. The volatility and instability of wind speed have a large negative impact on wind turbines and power systems, which can lead to grid collapse in severe cases. Therefore, accurate wind speed prediction is crucial for wind power generation. In this paper, considering the influence of different parameters on algorithm training and prediction, an improved moth flame optimization algorithm is constructed to optimize the LSTM wind energy prediction system to obtain better performance. The system consists of three modules: data preprocessing, optimization, and prediction. The data preprocessing module uses fuzzy information granulation to blur the input data. On this basis, the combination of swarm intelligent optimization algorithm and prediction model can effectively predict wind speed time series. Taking the California wind farm as an example, the MAPE of the experiment in the short-term forecast is 3.15%, the MAPE of the medium-term forecast is 4.38%, and the MAPE of the long-term forecast is 18.28%. The experimental results show that the proposed model has obvious advantages over the previous model.

Multi-strategy modified sparrow search algorithm for hyperparameter optimization in arbitrage prediction models.

Deep learning models struggle to effectively capture data features and make accurate predictions because of the strong non-linear characteristics of arbitrage data. Therefore, to fully exploit the model performance, researchers have focused on network structure and hyperparameter selection using various swarm intelligence algorithms for optimization. Sparrow Search Algorithm (SSA), a classic heuristic method that simulates the sparrows' foraging and anti-predatory behavior, has demonstrated excellent performance in various optimization problems. Hence, in this study, the Multi-Strategy Modified Sparrow Search Algorithm (MSMSSA) is applied to the Long Short-Term Memory (LSTM) network to construct an arbitrage spread prediction model (MSMSSA-LSTM). In the modified algorithm, the good point set theory, the proportion-adaptive strategy, and the improved location update method are introduced to further enhance the spatial exploration capability of the sparrow. The proposed model was evaluated using the real spread data of rebar and hot coil futures in the Chinese futures market. The obtained results showed that the mean absolute percentage error, root mean square error, and mean absolute error of the proposed model had decreased by a maximum of 58.5%, 65.2%, and 67.6% compared to several classical models. The model has high accuracy in predicting arbitrage spreads, which can provide some reference for investors.

Automatic calibration of a discrete element model of a masonry arch by swarm intelligence methods

This paper addresses the automatic calibration of constitutive parameters of the Discrete Element Method (DEM) models of masonry structures. Three swarm intelligence algorithms were applied, which are efficient in problems characterised by the small number of unknown parameters but high computational demand for the evaluation of the target function. Trust-Based Particle Swarm Optimisation (TBPSO) was chosen due to its excellent convergence rate. It was already tested that TBPSO converged faster than the original Genetic Algorithm and Particle Swarm Optimisation Methods. Therefore, TBPSO was compared with two recent and promising methods, the Marine Predator Algorithm (MPA) and the Golden Eagle Optimiser (GEO). The paper also presents a novel two-step approach, which enhances the performance of the tested optimisation strategies. The methodology is capable to account for and precisely find the mode of failure using Convolutional Neural Networks (CNN). The three optimisation methods and the two-step approach were applied on the DEM model of a quasi-static masonry arch collapse experiment from the literature. Performance of the three algorithms in identifying the appropriate constitutive parameters values of the DEM model were compared. Clear evidence on the advantage of the two-step approach with TBPSO is demonstrated. Remarks about different modelling issues are also included.

A cyber defense system against phishing attacks with deep learning game theory and LSTM-CNN with African vulture optimization algorithm (AVOA)

AbstractPhishing attacks pose a significant threat to online security, utilizing fake websites to steal sensitive user information. Deep learning techniques, particularly convolutional neural networks (CNNs), have emerged as promising tools for detecting phishing attacks. However, traditional CNN-based image classification methods face limitations in effectively identifying fake pages. To address this challenge, we propose an image-based coding approach for detecting phishing attacks using a CNN-LSTM hybrid model. This approach combines SMOTE, an enhanced GAN based on the Autoencoder network, and swarm intelligence algorithms to balance the dataset, select informative features, and generate grayscale images. Experiments on three benchmark datasets demonstrate that the proposed method achieves superior accuracy, precision, and sensitivity compared to other techniques, effectively identifying phishing attacks and enhancing online security.

Red-billed blue magpie optimizer: a novel metaheuristic algorithm for 2D/3D UAV path planning and engineering design problems

Numerical optimization, Unmanned Aerial Vehicle (UAV) path planning, and engineering design problems are fundamental to the development of artificial intelligence. Traditional methods show limitations in dealing with these complex nonlinear models. To address these challenges, the swarm intelligence algorithm is introduced as a metaheuristic method and effectively implemented. However, existing technology exhibits drawbacks such as slow convergence speed, low precision, and poor robustness. In this paper, we propose a novel metaheuristic approach called the Red-billed Blue Magpie Optimizer (RBMO), inspired by the cooperative and efficient predation behaviors of red-billed blue magpies. The mathematical model of RBMO was established by simulating the searching, chasing, attacking prey, and food storage behaviors of the red-billed blue magpie. To demonstrate RBMO’s performance, we first conduct qualitative analyses through convergence behavior experiments. Next, RBMO’s numerical optimization capabilities are substantiated using CEC2014 (Dim = 10, 30, 50, and 100) and CEC2017 (Dim = 10, 30, 50, and 100) suites, consistently achieving the best Friedman mean rank. In UAV path planning applications (two-dimensional and three − dimensional), RBMO obtains preferable solutions, demonstrating its effectiveness in solving NP-hard problems. Additionally, in five engineering design problems, RBMO consistently yields the minimum cost, showcasing its advantage in practical problem-solving. We compare our experimental results with three categories of widely recognized algorithms: (1) advanced variants, (2) recently proposed algorithms, and (3) high-performance optimizers, including CEC winners.

An analog circuit fault diagnosis method using improved sparrow search algorithm and support vector machine.

In analog circuits, component tolerances and circuit nonlinearity pose obstacles to fault diagnosis. To solve this problem, a soft fault diagnosis method based on Sparrow Search Algorithm (SSA) and Support Vector Machine (SVM) is used. In this study, ISSA is obtained by optimization using four strategies for SSA deficiency. Twenty-three benchmark functions are used for optimization experiments, and ISSA converges faster, more accurately, and with better robustness than other swarm intelligence algorithms. Finally, ISSA is used to optimize the SVM parameters and establish the ISSA-SVM fault diagnosis model. In the Sallen-key test circuit diagnosis experiments, the correct fault diagnosis rates of SSA-SVM and ISSA-SVM are 97.41% and 98.15%, respectively. The results show that the optimized ISSA-SVM model has a good analog circuit fault diagnosis with an increase in diagnostic accuracy.

Improving and comparing performance of machine learning classifiers optimized by swarm intelligent algorithms for code smell detection

In complex systems, the maintenance phase engenders the emergence of code smells due to incessant shifts in requirements and designs, stringent timelines, and the developer's relative inexperience. While not conventionally classified as errors, code smells inherently signify flawed design structures that lead to future bugs and errors. It increases the software budget and eventually makes the system hard to maintain or completely obsolete. To mitigate these challenges, practitioners must detect and refactor code smells. However, the theoretical interpretation of smell definitions and intelligent establishment of threshold values pose a significant conundrum. Supervised machine learning emerges as a potent strategy to address these problems and alleviate the dependence on expert intervention. The learning mechanism of these algorithms can be refined through data pre-processing and hyperparameter tuning. Selecting the best values for hyperparameters can be tedious and requires an expert. This study introduces an innovative paradigm that fuses twelve swarm-based, meta-heuristic algorithms with two machine learning classifiers, optimizing their hyperparameters, eliminating the need for an expert, and automating the entire code smell detection process. Through this synergistic approach, the highest post-optimization accuracy, precision, recall, F-measure, and ROC-AUC values are 99.09%, 99.20%, 99.09%, 98.06%, and 100%, respectively. The most remarkable upsurge is 35.9% in accuracy, 53.79% in precision, 35.90% in recall, 44.73% in F-measure, and 36.28% in ROC-AUC. Artificial Bee Colony, Grey Wolf, and Salp Swarm Optimizer are the top-performing swarm-intelligent algorithms. God and Data Class are the most readily detectable smells with optimized classifiers. Statistical tests underscore the profound impact of employing swarm-based algorithms to optimize machine learning classifiers, corroborated by statistical tests. This seamless integration enhances classifier performance, automates code smell detection, and offers a robust solution to a persistent software engineering challenge.

АНАЛІЗ АЛГОРИТМІВ РОЙОВОГО ІНТЕЛЕКТУ

This paper conducts a comprehensive review of swarm intelligence algorithms, highlighting the significant potential and development prospects of multi-agent systems and swarm intelligence. It underscores the ongoing research activity in this field and the continuous expansion of application areas. By examining various studies and publications, the paper concludes the importance of integrating approaches from different scientific disciplines to tackle diverse and complex problems using swarm intelligence. Future research is aimed at providing a more detailed analysis and comparison of various swarm intelligence algorithms across different application domains, as well as exploring their integration with other artificial intelligence methods. This work points towards the growing relevance of swarm intelligence in solving real-world problems, showcasing its versatility and effectiveness across multiple sectors, including military, agriculture, search and rescue, and environmental monitoring.

Design and measurement of three-dimensional uniform-field coils based on swarm intelligence algorithms under ferromagnetic boundaries

Combination systems of high permeability magnetic shields and coils are widely used in fields such as the magnetic field standard metrology and the zero magnetic room construction. To solve the problem that the magnetic field of uniform-field coils under the ferromagnetic boundary of magnetic shields is difficult to design precisely and the uniformity is difficult to improve, three-dimensional uniform field coils are designed based on the magnetic field model of coils under the ferromagnetic boundary using the target-field method (TFM), and the weight coefficient vectors determining the coils’ structure are optimized using intelligent optimization algorithms, which improves the uniformity of the magnetic field in the target region by 2 orders of magnitude through whale optimization algorithm (WOA). Then, the radial WOA-TFMUFY coil and the axial WOA-TFMUFZ coil are designed considering sidewall optical holes, and their magnetic field distribution is analyzed and verified under different permeabilities and axial opening sizes of the shield. Compared to the traditional saddle coil and Lee-whiting coil, the maximum magnetic field non-uniformity in the target uniform field region is reduced by 82.2 % and 92.8 % for the WOA-TFMUFY coil and WOA-TFMUFZ coil. The maximum relative errors between the measured magnetic field generated by the designed coils and the finite element method (FEM) simulation results are no higher than 2.2%. This novel method can provide more precise and uniform magnetic field metrology standard elements.

Antenna modeling based on meta-heuristic intelligent algorithms and neural networks

As wireless communication technology continues to advance, the antenna, as an essential front-end device in radio communication system, is surrounded by more and more complex electromagnetic wave environments with increasing variety, resulting in greater demand for antennas and higher design requirements. While the traditional antenna design methods suffered from the disadvantage of low design efficiency, a powerful tool for accelerating antenna design is the modelling of antennas with neural networks. Aiming to enhance the modeling accuracy of neural network, multiple novel meta-heuristic swarm intelligent algorithms are introduced and part of them are modified for the purpose of applying to optimizing network’s weights and biases so as to raise the antenna model`s prediction precision on the basis of neural network. Specifically, the intelligent algorithms and their improvement directions include the strategy of optimizing weights and biases for neural networks with seagull optimization algorithm, optimizing the weights and biases of neural network with the improved butterfly algorithm fused with reverse learning, and the artificial rabbit algorithm optimizing the neural network weights and biases. In addition, two intelligent optimization algorithms that are already more mature: the particle swarm algorithm and the genetic algorithm are added to compare with the above three algorithms. The accuracy of neural network prediction before and after the optimisation of neural network by seagull algorithm, the butterfly algorithm incorporating reverse learning, the artificial rabbit algorithm, the particle swarm algorithm, and the genetic algorithm are got through the results respectively. The results of the experiments displayed that the neural network optimized of the improved butterfly algorithm incorporating reverse learning has a prediction accuracy of 99.69 % with stable results, the optimised neural network prediction accuracy of the seagull algorithm reaches 99.51%, and the optimised neural network prediction accuracy of the artificial rabbit algorithm is 99.49 %. The remaining two traditional algorithms optimized neural network accuracy is 83.1 % and 99.43 % respectively. Therefore, the improved butterfly algorithm incorporating reverse learning is the most effective among these three new algorithms applied to the field of antenna prediction. Moreover, the running time of the network optimized by different algorithms is quite distinct, among which the neural network optimized by the improved butterfly algorithm incorporating reverse learning takes the shortest time, which increases the prediction efficiency of the network by more than 70%. In summary, the application of the fused reverse learning improved butterfly algorithm in optimizing neural network predictions yields the shortest processing time and highest accuracy. This not only enables faster and more precise antenna design but also holds greater significance for the field of antenna design and analysis.

On the Initialization of Swarm Intelligence Algorithms for Vector Quantization Codebook Design.

Vector Quantization (VQ) is a technique with a wide range of applications. For example, it can be used for image compression. The codebook design for VQ has great significance in the quality of the quantized signals and can benefit from the use of swarm intelligence. Initialization of the Linde-Buzo-Gray (LBG) algorithm, which is the most popular VQ codebook design algorithm, is a step that directly influences VQ performance, as the convergence speed and codebook quality depend on the initial codebook. A widely used initialization alternative is random initialization, in which the initial set of codevectors is drawn randomly from the training set. Other initialization methods can lead to a better quality of the designed codebooks. The present work evaluates the impacts of initialization strategies on swarm intelligence algorithms for codebook design in terms of the quality of the designed codebooks, assessed by the quality of the reconstructed images, and in terms of the convergence speed, evaluated by the number of iterations. Initialization strategies consist of a combination of codebooks obtained by initialization algorithms from the literature with codebooks composed of vectors randomly selected from the training set. The possibility of combining different initialization techniques provides new perspectives in the search for the quality of the VQ codebooks. Nine initialization strategies are presented, which are compared with random initialization. Initialization strategies are evaluated on the following algorithms for codebook design based on swarm clustering: modified firefly algorithm-Linde-Buzo-Gray (M-FA-LBG), modified particle swarm optimization-Linde-Buzo-Gray (M-PSO-LBG), modified fish school search-Linde-Buzo-Gray (M-FSS-LBG) and their accelerated versions (M-FA-LBGa, M-PSO-LBGa and M-FSS-LBGa) which are obtained by replacing the LBG with the accelerated LBG algorithm. The simulation results point out to the benefits of the proposed initialization strategies. The results show gains up to 4.43 dB in terms of PSNR for image Clock with M-PSO-LBG codebooks of size 512 and codebook design time savings up to 67.05% for image Clock, with M-FF-LBGa codebooks with size N=512, by using initialization strategies in substitution to Random initialization.

Analysis of swarm intelligence algorithms

Analysis of swarm intelligence algorithms

A multi-direction guided mutation-driven stable swarm intelligence algorithm with translation and rotation invariance for global optimization

In recent years, there have been notable advancements in the theoretical research of existing meta-heuristic algorithms and in the development of novel algorithms. However, the research on the definition and construction methods of the so-called “mutation” operator, a concept widely used in existing meta-heuristic algorithms, remains notably insufficient. In addition, how to directly construct a meta-heuristic algorithm under the guidance of stability and invariance analysis continues to pose a challenge that has yet to resolve. This paper carefully investigates and rigorously defines the concepts of “mutation operator” and “mutation-driven meta-heuristic algorithm” as found in the literature. According to these definitions, a novel meta-heuristic algorithm, called Multi-direction guided Mutation-driven Stable swarm Intelligence algorithm with Translation and Rotation invariance (M2SITRI) is proposed, and the corresponding mutation operator is carefully designed based on the analysis of translation invariance, rotation invariance, order-1 stability, and order-2 stability. The stability and invariance properties of the M2SITRI algorithm are validated through numerical simulations, and its superior solving performance is demonstrated by benchmarking it against several other well-known existing meta-heuristic algorithms across a variety of test problems.

ENHANCED SURFACE QUALITY AND STRENGTH OF FDMed SPECIMENS USING BBD AND BIO-INSPIRED ALGORITHMS

This research investigated and optimized the parameters of the FDM process by employing bio-inspired algorithms for determining the optimal parameter settings in terms of surface quality and mechanical performance. Four important process parameters including layer thickness (0.11–0.33[Formula: see text]mm), part orientation (0–90∘), raster width (0.2–0.56[Formula: see text]mm), and the raster angle (0–60∘) at three variation levels were selected for fabricating the specimens (ABS material P430) using the statistical Box–Behnken design. ANOVA analysis and multiple regression analysis were used to fit the experimental data to a second-order polynomial equation. Through, the RSM analysis, the layer thickness is the key important factor that accounts for all of the responses. The fracture behavior of specimens was examined using a scanning electron microscope (SEM). From the SEM analysis, a substantial amount of plastic deformation on the fracture surface indicative of craze cracking is visible from a 0∘ orientation, indicating a totally ductile fracture mechanism. Then, three swarm intelligence algorithms such as Tasmanian Devil Optimization (TDO), Remora Optimization Algorithm (ROA), Tuna Swarm Optimization (TSO) were implemented to optimize the input parameters that would lead to minimum surface roughness and maximum tensile strength. Experimental data and predicted values varied between 1.64% and 1.84%, as shown by verification experiments.

Dynamic Bayesian network structure learning based on an improved bacterial foraging optimization algorithm

With the rapid development of artificial intelligence and data science, Dynamic Bayesian Network (DBN), as an effective probabilistic graphical model, has been widely used in many engineering fields. And swarm intelligence algorithm is an optimization algorithm based on natural selection with the characteristics of distributed, self-organization and robustness. By applying the high-performance swarm intelligence algorithm to DBN structure learning, we can fully utilize the algorithm's global search capability to effectively process time-based data, improve the efficiency of network generation and the accuracy of network structure. This study proposes an improved bacterial foraging optimization algorithm (IBFO-A) to solve the problems of random step size, limited group communication, and the inability to maintain a balance between global and local searching. The IBFO-A algorithm framework comprises four layers. First, population initialization is achieved using a logistics-sine chaotic mapping strategy as the basis for global optimization. Second, the activity strategy of a colony foraging trend is constructed by combining the exploration phase of the Osprey optimization algorithm. Subsequently, the strategy of bacterial colony propagation is improved using a "genetic" approach and the Multi-point crossover operator. Finally, the elimination-dispersal activity strategy is employed to escape the local optimal solution. To solve the problem of complex DBN learning structures due to the introduction of time information, a DBN structure learning method called IBFO-D, which is based on the IBFO-A algorithm framework, is proposed. IBFO-D determines the edge direction of the structure by combining the dynamic K2 scoring function, the designed V-structure orientation rule, and the trend activity strategy. Then, according to the improved reproductive activity strategy, the concept of "survival of the fittest" is applied to the network candidate solution while maintaining species diversity. Finally, the global optimal network structure with the highest score is obtained based on the elimination-dispersal activity strategy. Multiple tests and comparison experiments were conducted on 10 sets of benchmark test functions, two non-temporal and temporal data types, and six data samples of two benchmark 2T-BN networks to evaluate and analyze the optimization performance and structure learning ability of the proposed algorithm under various data types. The experimental results demonstrated that IBFO-A exhibits good convergence, stability, and accuracy, whereas IBFO-D is an effective approach for learning DBN structures from data and has practical value for engineering applications.

Fractional-Order Water Flow Optimizer

The water flow optimizer (WFO) is the latest swarm intelligence algorithm inspired by the shape of water flow. Its advantages of simplicity, efficiency, and robust performance have motivated us to further enhance it. In this paper, we introduce fractional-order (FO) technology with memory properties into the WFO, called fractional-order water flow optimizer (FOWFO). To verify the superior performance and practicality of FOWFO, we conducted comparisons with nine state-of-the-art algorithms on benchmark functions from the IEEE Congress on Evolutionary Computation 2017 (CEC2017) and four real-world optimization problems with large dimensions. Additionally, tuning adjustments were made for two crucial parameters within the fractional-order framework. Finally, an analysis was performed on the balance between exploration and exploitation within FOWFO and its algorithm complexity.

OBABC: A one-dimensional binary artificial bee colony algorithm for binary optimization

Artificial bee colony (ABC) algorithm is a widely utilized swarm intelligence (SI) algorithm for addressing continuous optimization problems. However, most binary variants of ABC (BABC) algorithms may suffer from issues such as invalid searches and high complexity when applied to binary problems. To address these challenges, we first establish a set of criteria for developing a BABC algorithm. Following these criteria, we propose a novel BABC algorithm, denoted as oBABC, which not only adheres to the defined criteria but also successfully inherits the advantages of original ABC algorithm. To evaluate the performance of oBABC and verify its effectiveness, experiments are conducted on two typical binary problems: uncapacitated facility location problem (UFLP) and maximum cut problem (Max-Cut). The experimental results reveal the following findings: 1) The validity of the criteria and the accuracy of the theoretical analysis are confirmed. oBABC exhibits high search efficiency with an invalid learning rate (ILR) of 0 %, while the ILRs of other BABC algorithms almost exceeds 20 %. 2) In terms of search efficiency and capability, oBABC exhibits a significant improvement in search efficiency and consistently ranks at the top in terms of optimization capability. These results suggest that oBABC may be a highly efficient and effective tool for solving binary problems.

High-pressure and high-temperature modulation of one-dimensional infinite chain in SeO2

The structural evolution of lone-pair compounds under high-pressure and high-temperature conditions has been a subject of fundamental interest in revealing modulated polymorphs. As one of the archetypal lone-pair compounds, selenium dioxide (SeO2) has attracted much attention due to the pressure modulation of its one-dimensional infinite W-shaped chain arrangement. Here, through swarm intelligence algorithm in conjunction with the first-principles simulation, we propose the existence of an orthorhombic Pnma-SeO2 structure, characterized by V-shaped chains interconnected via vertex-sharing SeO3 pyramids. These V-shaped chains demonstrate reduced compressibility along their chain direction compared to the W-shaped chains. Calculations indicate that Pnma-SeO2 is a semiconductor with a large indirect bandgap of 2.39 eV. Remarkably, we synthesized the predicted Pnma-SeO2 in a laser-heated diamond anvil cell at a pressure of 48.5 or 87 GPa as identified by in situ synchrotron x-ray diffraction data. Our findings lead to a significant extension of the phase diagram and transition path of SeO2 and provide key insights into understanding the pressure modulation in lone-pair compounds.

A parallel chimp optimization algorithm based on tracking-learning and fuzzy opposition-learning behaviors for data classification

Chimp optimization algorithm (ChOA), which simulates the social behaviors of chimps, is a novel swarm intelligence algorithm for solving global optimization problems. ChOA has the advantages of fast convergence and avoiding falling into local optimum. However, the global search capability is weakened and the time overhead is too large when solving complex optimization problems. In order to improve the overall performance of ChOA, a parallel chimp optimization algorithm based on tracking-learning and fuzzy opposition-learning behaviors (PChOA) is proposed in this paper. First, a tracking-learning behavior is designed to improve the search accuracy. Second, a fuzzy opposition-learning behavior is adopted to enhance the global search capability. Third, a parallel computing architecture is developed to accelerate computational speed. Moreover, the convergence of our proposed PChOA has been analyzed theoretically. To validate the effectiveness of PChOA, it is applied to solve classification problem. The experimental results demonstrate that the classification performance of our proposed algorithm outperforms six other state of the art algorithms on most used datasets. Meanwhile, the time overhead of PChOA is significantly reduced in the environment of parallel computing. When the number of processors is increased to 16, PChOA costs less time than NBTree which is the fastest comparison algorithm in the experiment.

Intelligent Scheduling Technology of Swarm Intelligence Algorithm for Drone Path Planning

Different kinds of swarm intelligence algorithm obtain superior performances in solving complex optimization problems and have been widely used in path planning of drones. Due to their own characteristics, the optimization results may vary greatly in different dynamic environments. In this paper, a scheduling technology for swarm intelligence algorithms based on deep Q-learning is proposed to intelligently select algorithms to realize 3D path planning. It builds a unique path point database and two basic principles are proposed to guide model training. Path planning and network learning are separated by the proposed separation principle and the optimal selection principle ensures convergence of the model. Aiming at the problem of reward sparsity, the comprehensive cost of each path point in the whole track sequence is regarded as a dynamic reward. Through the investigation of dynamic environment conditions such as different distances and threats, the effectiveness of the proposed method is validated.