Learning Particle Swarm Optimization Research Articles

Estimating the optimal parameters of solid oxide fuel cell‐based circuit using parasitism‐predation algorithm

The process of constructing a reliable mathematical model of solid oxide fuel cell (SOFC) is a challenge due to its complex nature. This paper proposes a new methodology incorporated a recent meta-heuristic algorithm named parasitism-predation algorithm (PPA) to estimate the optimal parameters of SOFC equivalent circuit. Two experiments are conducted in this work; the first one comprises four measured datasets for a commercial enhanced cylindrical SOFC manufactured by Siemen Energy. While the second series consists of five measured datasets for a theoretical dynamic SOFC stack with 96 connected cells. The collected datasets are measured at different operating conditions. An excessive comparative study is presented with other optimizers of comprehensive learning particle swarm optimization (CLPSO), improved PSO with difference mean with perturbation (DMP_PSO), heterogeneous CLPSO (HCLPSO), locally informed PSO (LIPS), modified CSO with tri-competitive mechanism (MCSO), opposition-based learning competitive PSO (OBLCPSO), ranking-based biased learning swarm optimizer (RBLSO), competitive swarm optimizer (CSO), hybrid Jaya with DE (JayaDE), and social learning PSO (SLPSO). Furthermore, statistical analyses of the ranking tests, multiple sign tests, Friedman tests, and ANOVA are performed. The obtained results confirmed the proposed PPA's competence in constructing a reliable model of SOFC as it provides the least mean square error (MSE) between the measured and estimated characteristics of 2.164e−6 in the first series of experiments at 1073 K, in contrast, the most peer (CLPSO) provides 5.57e−6. Similarly, in the second series of experiments, PPA achieves lease MSE of 7.17e−2 at 973 K; meanwhile, the most peer (CLPSO) attains 5.44e−1.

Chaotic theory incorporated with PSO algorithm for solving optimal reactive power dispatch problem of power system

In this paper, the chaotic particle swarm optimization (CPSO) algorithm is combined with MATPOWER toolbox and used as an optimization tool for attaining solving the optimal reactive power dispatch (RPD) problem, by finding the optimal adjustment of reactive power control variables like a voltage of generator buses (VG), capacitor banks (QC) and transformer taps (Tap) while satisfying some of equality and inequality constraints at the same time. CPSO and Simple PSO algorithms will be checked in a large system such as IEEE node -118. CPSO and Simple PSO algorithms have been implemented and simulated in the MATLAB program, version (R2013b/m-file). Then compassion these results with the results obtained in the other algorithms in the literature like the comprehensive learning particle swarm optimization (CLPSO) algorithm. The simulation results confirm that the CPSO algorithm has high efficiency and ability in terms of decrease real power losses (P loss), and improve voltage profile compared with the obtained by using the simple (PSO) algorithm and (CLPSO) at light load.

A hybrid model for short-term dissolved oxygen content prediction

Dissolved oxygen (DO) content is a significant indicator of water quality in intensive aquaculture ponds and is profoundly related to healthy fish growth. Accurately forecasting DO and analysing its change trends in aquaculture ponds is crucial for fish survival. A novel forecasting model based on a combination of complete ensemble empirical mode decomposition with an adaptive noise Lempel-Ziv complex (CEEMDAN-LZC) and a gated recurrent unit (GRU) with a generalized opposition-based learning particle swarm optimization algorithm (GOBLPSO) has been proposed to improve the prediction precision of DO. The DO content numerical sequence was decomposed and reconstructed into several new features by the CEEMDAN-LZC method in the modelling process. Independent models were structured to fit the components obtained above using GRUs, and the prediction value was superimposed to obtain the ultimate result. Furthermore, the time-step parameter t and the unit number parameter u, which significantly influence the performance of the GRU, were selected by the GOBLPSO algorithm. The simulation results based on DO content data in river crab culture ponds show that the proposed model has excellent essential feature extraction capability and is very powerful and reliable for DO content prediction in aquaculture.

Nonlinear rotor side converter control of DFIG based wind energy system

The present study proposes an alternative approach following command-filtered integral backstepping principle to control the rotor side converter (RSC) of a doubly-fed induction generator (DFIG) based wind energy conversion system (WECS). This controller provides better benefit compared to the classical backstepping controller; since the errors introduced by the input signal differentiation is eliminated. This controller essentially stabilises the rotor speed by controlling the electromagnetic torque. The global stability of the overall closed loop system is established using Lyapunov criteria. Moreover, the parameters of the controller are optimally selected using comprehensive learning particle swarm optimisation (CLPSO) technique. The performance of the controller is compared with two other controllers, which include a classical proportional integral derivative (PID) and ΔΣ–based PID. The results of comparative investigation demonstrate that the proposed command-filtered integral backstepping controller is superior and could effectively control the RSC under various fault conditions.

Trajectory optimisation in electrical discharge machining of three-dimensional curved and twisted channels

Numerical control electrical discharge machining (NC EDM) is one of the most widely used machining technologies for manufacturing a closed blisk flow path, particularly for three-dimensional (3D) curved and twisted flow channels. In this process, tool electrode design and machining trajectory planning are the key factors affecting machining accessibility and efficiency. Herein, to reduce the difficulty in designing the electrode and its motion path in the closed curved and twisted channels, a heuristic search hybrid optimisation strategy based on channel grids is adopted to realise the initial electrode trajectory design search and optimised size reduction. By transferring the trajectory optimisation constraints from the complex free-form surface to numbered grids, the search is found to be more orderly and accurate. The two trajectory indicators, namely argument angle and minimum distance, are analysed separately for the optimised results of the adaptive learning particle swarm optimisation algorithm, demonstrating that they can meet the actual processing requirements. Experimental results of NC EDM indicate that the motion path generated by this design method can meet the machining requirements of 3D curved and twisted flow channels.

Enhanced Marine Predators Algorithm for identifying static and dynamic Photovoltaic models parameters

Providing an accurate and precise photovoltaic model is a vital stage prior to the system design, therefore, this paper proposes a novel algorithm, enhanced marine predators algorithm (EMPA), to identify the unknown parameters for different photovoltaic (PV) models including the static PV models (single-diode and double-diode) and dynamic PV model. In the proposed EMPA, the differential evolution operator (DE) is incorporated into the original marine predators algorithm (MPA) to achieve stable, and reliable performance while handling that nonlinear optimization problem of PV modeling. Three different real datasets are used to show the effectiveness of the proposed algorithm. In the first case study, the proposed algorithm is used to identify the unknown parameters of a single-diode and double-diode PV models. The root-mean-square error (RMSE) and standard deviation (STD) values for a single-diode are 7.7301e-04 and 5.9135e-07. Similarly for double diode are 7.4396e-04 and 3.1849e-05, respectively. In addition, the second case study is used to test the proposed model in identifying the unknown parameters of a double-diode PV model. Here, the proposed algorithm is compared with classical MPA in five scenarios at different operating conditions. In this case study, the RMSE and STD of the proposed algorithm are less than that obtained by the MPA algorithm. Moreover, the third case study is utilized to test the ability of the proposed model in identifying the parameters of a dynamic PV model. In this case study, the performance of the proposed algorithm is compared with the one obtained by MAP and heterogeneous comprehensive learning particle swarm optimization (HCLPSO) algorithms in terms of RMSE ± STD. The obtained value of RMSE ± STD by the proposed algorithm is 0.0084505±1.0971e-17, which is too small compared with that obtained by MPA and HCLPSO algorithms (0.0084505±9.6235e-14 and 0.0084505±2.5235e-9). The results show the proposed model’s superiority over the MPA and other recent proposed algorithms in data fitting, convergence rate, stability, and consistency. Therefore, the proposed algorithm can be considered as a fast, feasible, and a reliable optimization algorithm to identify the unknown parameters in static and dynamic PV models. The code of the dynamic PV models is available via this link: https://github.com/DAyousri/Identifying-the-parameters-of-the-integer-and-fractional-order-dynamic-PV-models?_ga=2.104793926.732834951.1616028563-1268395487.1616028563.

RETRACTED ARTICLE: Reputation based clustering system in vehicular adhoc networks

Vehicular ad hoc networks (VANETs) are among the various magnificent innovations for a clever transportation framework that diminishes vehicle–vehicle to raise multi-jump specially appointed organizations. Directing is one of the motivations in VANETs in view of its geography creation’s different alterations. Yet, because of the high flexibility and uncommon appearance of the vehicles out and about, it is a moving undertaking to course the correspondences to their last objective in VANETs. To this end, bunching has been widely utilized in various existing plans in the writing. In this methodology archive, we recommend a reputation-based weighted clustering protocol (RBWCP) for VANET. The proposed RBWCP is planned with vehicle way, territory and speed determined, path assurance, number of close by vehicles, and every hub’s unwavering quality in pay for keeping up the VANET geography. Additionally, the administration of different RBWCP the executives boundaries is actualized to enhance the difficult cycle. This board applies the multi-target trouble wanted by RBWCP boundaries as it goes in and endeavors to actualize progressed block life, lessen block load, and progressed bundle move proportion.The Firefly Multipurpose Algorithm (MOFA), a developmental methodology, is utilized to streamline RBWCP boundaries. Reenactments run utilizing MATLAB emulator and MOEA improvement structure. Results are assessed utilizing similar formative improvement methodologies. Investigations are performed with sensible guides from Open Street Maps. The paper results are assessed utilizing other multi-target improvement procedures. Multi-objective particle swarm optimization (MO-PSO) and comprehensive learning particle swarm optimization (CL-PSO). The proposed RBWCP technology is simulated using MATLAB, simulation results are detected on three parameters of packet delivery ratio measures, Cluster packet overhead obtained to be more than 90%, and an average number of clusters used in this research is about 50 nodes.

Multi Algorithmic Approach to Galactic Swarm Optimization (MAGSO)

Multi population algorithms have noted advantages over single population algorithms when optimizing over multimodal function space. The existence of multiple attractors allows exploration on multiple optimality regions which in turn provide more exploration compared with single population algorithms. This paper is a novel utilization of Galactic Swarm Optimization (GSO) framework in which the subpopulations are evolved with multiple variants of Particle Swarm Optimizers (PSOs) rather than single PSO. The proposed algorithm is called Multi Algorithmic Galactic Swarm Optimization (MAGSO). MAGSO utilizing multivariant PSOs in subswarm level combined with the unique superswarm tailored for exploitation has shown significant improvement of performance over GSO. Experiments done on Congress on Evolutionary Computation 2013 (CEC 2013) and other standard general benchmark suite indicate that the proposed MAGSO algorithm outperforms the previously proposed state-of-the-art swarm algorithms such as Bollinger bands approach on boosting ABC (ABCBB) and Dynamic Neighbourhood Learning Particle Swarm Optimizer (DNLPSO) besides GSO. Detailed statistical evidence over 51 independent trials on the benchmark functions indicate the best performance of MAGSO. The main research goal is to propose a superior performing algorithm by exploiting the GSO framework. In order to show its efficacy, the algorithm is compared with state-of-the-art swarm algorithms such as ABCBB, DNLPSO and unmodified GSO. The method proposed here is quite general and demonstrates that more ability and extension work possible from the basic GSO framework.

Source mask optimization for extreme-ultraviolet lithography based on thick mask model and social learning particle swarm optimization algorithm.

Extreme ultraviolet (EUV) lithography plays a vital role in the advanced technology nodes of integrated circuits manufacturing. Source mask optimization (SMO) is a critical resolution enhancement technique (RET) or EUV lithography. In this paper, an SMO method for EUV lithography based on the thick mask model and social learning particle swarm optimization (SL-PSO) algorithm is proposed to improve the imaging quality. The thick mask model's parameters are pre-calculated and stored, then SL-PSO is utilized to optimize the source and mask. Rigorous electromagnetic simulation is then carried out to validate the optimization results. Besides, an initialization parameter of the mask optimization (MO) stage is tuned to increase the optimization efficiency and the optimized mask's manufacturability. Optimization is carried out with three target patterns. Results show that the pattern errors (PE) between the print image and target pattern are reduced by 94.7%, 76.9%, 80.6%, respectively.

Bee-foraging learning particle swarm optimization

Numerous particle swarm optimization (PSO) algorithms have been developed for solving numerical optimization problems in recent years. However, most of existing PSO algorithms have only one search phase. There is no strengthened search phase for the well-performed particles, and also no re-initialization phase for the exhausted particles. These issues may still restrict the performance of PSO for complex optimization problems. In this paper, inspired by the bee-foraging search mechanism of artificial bee colony algorithm, a novel bee-foraging learning PSO (BFL-PSO) algorithm is proposed. Different from existing PSO algorithms, the proposed BFL-PSO has three different search phases, namely employed learning, onlooker learning and scout learning. The employed learning phase works like traditional one-phase-based PSO, while the onlooker learning phase performs strengthened search around those well-performed particles to exploit promising solutions, and the scout learning phase re-initializes those exhausted particles to introduce new diversity. The proposed BFL-PSO is comprehensively evaluated on CEC2014 benchmark functions, and compared with state-of-the-art PSO algorithms as well as artificial bee colony algorithms. The experimental results show that BFL-PSO achieves very competitive performance in terms of solution accuracy. In addition, the effectiveness of the newly introduced onlooker learning and scout learning phases in BFL-PSO is verified.

Probability-optimal leader comprehensive learning particle swarm optimization with Bayesian iteration

In this paper, a novel comprehensive learning particle swarm optimization algorithm, which is based on the Bayesian iteration method and named as Bayesian comprehensive learning particle swarm optimization (BCLPSO), is proposed. In the original PSO, the flying direction of each particle is based on its own historical best position and global optimum. This updating mechanism, however, easily falls into the local optimum, and the potential optimum solution may be ignored in the iteration and update process. Therefore, the BCLPSO is designed to facilitate discovering potential solution and avoid the problem of premature convergence. In the BCLPSO algorithm, the exemplar of the swarm is not the global best position but the particle location with the largest posterior probability based on the Bayesian formula. The posterior probability is developed by historical prior information. This means that the posterior probability can inherit the historical information of particles that may be exploited. In this way, the swarm diversity can be preserved to prevent premature convergence. The BCLPSO is experimentally validated on the CEC2017 benchmark functions and compared with other state-of-the-art particle swarm optimization algorithms. The results show that BCLPSO outperforms other comparative PSO variants on the CEC 2017 test suite. Furthermore, the algorithm is applied to the quality control process of an automated welding production line for the automobile body and is found to exhibit superior performance.

Gait Optimization Method for Humanoid Robots Based on Parallel Comprehensive Learning Particle Swarm Optimizer Algorithm.

To improve the fast and stable walking ability of a humanoid robot, this paper proposes a gait optimization method based on a parallel comprehensive learning particle swarm optimizer (PCLPSO). Firstly, the key parameters affecting the walking gait of the humanoid robot are selected based on the natural zero-moment point trajectory planning method. Secondly, by changing the slave group structure of the PCLPSO algorithm, the gait training task is decomposed, and a parallel distributed multi-robot gait training environment based on RoboCup3D is built to automatically optimize the speed and stability of bipedal robot walking. Finally, a layered learning approach is used to optimize the turning ability of the humanoid robot. The experimental results show that the PCLPSO algorithm achieves a quickly optimal solution, and the humanoid robot optimized possesses a fast and steady gait and flexible steering ability.

Optimal allocation of distributed energy storage in active distribution network via hybrid teaching learning and multi-objective particle swarm optimization algorithm

This work aims at solving complex problems of the optimal scheduling model of active distribution network, teaching strategies are proposed to improve the global search ability of particle swarm optimization. Moreover, based on the improved Euclidean distance cyclic crowding sorting strategy, the convergence ability of Li Zhiquan algorithm is improved. With the cost and voltage indexes of the energy storage system of the distribution network as the goal, different optimized configuration schemes are constructed, and the improved HTL-MOPSO algorithm is adopted to find the solution. The results show that compared with the traditional TV-MOPSO algorithm, the proposed algorithm has better convergence performance and optimization ability, and has a lower economic cost. In short, the algorithm proposed can provide a basis for improving the optimization of active distribution network scheduling strategies.

Prediction of Blood Glucose Concentration Based on CEEMD and Improved Particle Swarm Optimization LSSVM.

Aiming at the difficulty of accurate prediction due to the randomness and nonstationary nature of blood glucose concentration series, a blood glucose concentration prediction model based on complementary ensemble empirical mode decomposition (CEEMD) and least squares support vector machine (LSSVM) is proposed. Firstly, CEEMD is used to convert the blood glucose concentration sequence into a series of intrinsic mode functions (IMFs) to reduce the impact of randomness and nonstationary signals on prediction performance. Then, a LSSVM prediction model is established for each mode IMF. The comprehensive learning particle swarm optimization (CLPSO) algorithm is used to optimize the kernel parameters of LSSVM. Finally, the prediction results of all IMFs are superimposed to yield the final blood glucose concentration prediction value. The experimental results show that the proposed prediction model has higher prediction accuracy in short-term blood glucose concentration values.

Data-Driven Bidding Strategy for DER Aggregator Based on Gated Recurrent Unit–Enhanced Learning Particle Swarm Optimization

Distributed energy resources (DERs) such as wind turbines (WTs), photovoltaics (PVs), energy storage systems (ESSs), local loads, and demand response (DR) are highly valued for environmental protection. However, their volatility poses several risks to the DER aggregator while formulating a profitable strategy for bidding in the day-ahead power market. This study proposes a data-driven bidding strategy framework for a DER aggregator confronted with various uncertainties. First, a data-driven forecasting model involving gated recurrent unit–enhanced learning particle swarm optimization (GRU-ELPSO) with improved mutual information (IMI) is employed to model renewables and local loads. It is critical for a DER aggregator to accurately estimate these components before bidding in the day-ahead power market. This aids in reducing the penalty costs of forecasting errors. Second, an optimal bidding strategy that is based on the information gap decision theory (IGDT) is formulated to address market price uncertainty. The DER aggregator is assumed to be risk-averse (RA) or risk-seeker (RS), and the corresponding bidding strategies are formulated according to the risk preferences thereof. Then, an hourly bidding profile is created for the DER aggregator to bid successfully in the day-ahead power market. The proposed data-driven bidding framework is evaluated using an illustrative system wherein a dataset is obtained from the PJM market. The results reveal the effectiveness of handling uncertainty by providing accurate forecasting results. In addition, the DER aggregator can bid effectively in the day-ahead power market according to its preference for robustness or high profit, with a suitable bidding profile.

Enhanced Comprehensive Learning Particle Swarm Optimization with Dimensional Independent and Adaptive Parameters.

Comprehensive learning particle swarm optimization (CLPSO) and enhanced CLPSO (ECLPSO) are two literature metaheuristics for global optimization. ECLPSO significantly improves the exploitation and convergence performance of CLPSO by perturbation-based exploitation and adaptive learning probabilities. However, ECLPSO still cannot locate the global optimum or find a near-optimum solution for a number of problems. In this paper, we study further bettering the exploration performance of ECLPSO. We propose to assign an independent inertia weight and an independent acceleration coefficient corresponding to each dimension of the search space, as well as an independent learning probability for each particle on each dimension. Like ECLPSO, a normative interval bounded by the minimum and maximum personal best positions is determined with respect to each dimension in each generation. The dimensional independent maximum velocities, inertia weights, acceleration coefficients, and learning probabilities are proposed to be adaptively updated based on the dimensional normative intervals in order to facilitate exploration, exploitation, and convergence, particularly exploration. Our proposed metaheuristic, called adaptive CLPSO (ACLPSO), is evaluated on various benchmark functions. Experimental results demonstrate that the dimensional independent and adaptive maximum velocities, inertia weights, acceleration coefficients, and learning probabilities help to significantly mend ECLPSO's exploration performance, and ACLPSO is able to derive the global optimum or a near-optimum solution on all the benchmark functions for all the runs with parameters appropriately set.

Intelligent Modeling and Design of a Novel Temperature Control System for a Cantilever-Based Gas-Sensitive Material Analyzer

Devices used to set and control the environmental temperature are critical to the performance of gas-sensitive material analyzers, which use silicon microcantilevers to characterize the gas-sensitive materials. This paper describes a novel microtemperature-control device that uses a double Peltier structure to replace the traditional refrigerant temperature control system. A proportional-integral-derivative (PID) algorithm is used to achieve accurate and fast temperature control, with a long short-term memory (LSTM) network trained to identify the nonlinear dynamics of the Peltier system. A neighbor hybrid mean center opposition-based learning particle swarm optimization (NHCOPSO) algorithm is proposed to optimize the PID controller. The LSTM network identification is obviously better than that of previous Peltier system identification methods, and the NHCOPSO algorithm is found to be superior to other improved PSO and evolutionary algorithms on benchmark functions and in PID parameter optimization. Experimental results show that the proposed temperature control device greatly improves the accuracy and efficiency of gas-sensitive material analysis with a temperature control range of −40 to 180°C, a temperature control tolerance within ±0.05°C, a maximum heating rate of 20°C/min, and a maximum cooling rate of −10°C/min.

An Optimal Deep Learning based Computer-aided Diagnosis System for Diabetic Retinopathy

Diabetic Retinopathy (DR) is a significant blinding disease that poses serious threat to human vision rapidly. Classification and severity grading of DR are difficult processes to accomplish. Traditionally, it depends on ophthalmoscopically-visible symptoms of growing severity, which is then ranked in a stepwise scale from no retinopathy to various levels of DR severity. This paper presents an ensemble of Orthogonal Learning Particle Swarm Optimization (OPSO) algorithm-based Convolutional Neural Network (CNN) Model EOPSO-CNN in order to perform DR detection and grading. The proposed EOPSO-CNN model involves three main processes such as preprocessing, feature extraction, and classification. The proposed model initially involves preprocessing stage which removes the presence of noise in the input image. Then, the watershed algorithm is applied to segment the preprocessed images. Followed by, feature extraction takes place by leveraging EOPSO-CNN model. Finally, the extracted feature vectors are provided to a Decision Tree (DT) classifier to classify the DR images. The study experiments were carried out using Messidor DR Dataset and the results showed an extraordinary performance by the proposed method over compared methods in a considerable way. The simulation outcome offered the maximum classification with accuracy, sensitivity, and specificity values being 98.47%, 96.43%, and 99.02% respectively.

Hybrid Particle Swarm and Grey Wolf Optimizer and its application to clustering optimization

Grey Wolf Optimizer (GWO) and Particle Swarm Optimization (PSO) algorithm are two popular swarm intelligence optimization algorithms and these two algorithms have their own search mechanisms. Based on their unique search mechanisms and their advantages after the improvements on them, this paper proposes a novel hybrid algorithm based on PSO and GWO (Hybrid GWO with PSO, HGWOP). Firstly, GWO is simplified and a novel differential perturbation strategy is embedded in the search process of the simplified GWO to form a Simplified GWO with Differential Perturbation (SDPGWO) so that it can improve the global search ability while retaining the strong exploitation ability of GWO. Secondly, a stochastic mean example learning strategy is applied to PSO to create a Mean Example Learning PSO (MELPSO) to enhance the global search ability of PSO and prevent the algorithm from falling into local optima. Finally, a poor-for-change strategy is proposed to organically integrate SDPGWO and MELPSO to obtain an efficient hybrid algorithm of GWO and PSO. HGWOP can give full play to the advantages of these two improved algorithms, overcome the shortcomings of GWO and PSO and maximize the whole performance. A large number of experiments on the complex functions from CEC2013 and CEC2015 test sets reveal that HGWOP has better optimization performance and stronger universality compared with quite a few state-of-the-art algorithms. Experimental results on K-means clustering optimization show that HGWOP has obvious advantages over the comparison algorithms.

AN OPERANT CONDITIONING APPROACH FOR LARGE SCALE SOCIAL OPTIMIZATION ALGORITHMS

The changes that positive or negative results cause in an individual's behavior are called Operant Conditioning. This paper introduces an operant conditioning approach (OCA) for large scale swarm optimization models. The proposed approach has been applied to social learning particle swarm optimization (SL-PSO), a variant of the PSO algorithm. In SL-PSO, the swarm particles are sorted according to the objective function and all particles are updated with learning from the others. In this study, each particle's learning rate is determined by the mathematical functions that are inspired by the operant conditioning. The proposed approach adjusts the learning rate for each particle. By using the learning rate, a particle close to the optimum solution is aimed to learn less. Thanks to the learning rate, a particle is prevented from being affected by particles close to the optimum point and particles far from the optimum point at the same rate. The proposed OCA-SL-PSO is compared with SL- PSO and pure PSO on CEC 13 functions. Also, the proposed OCA-SL-PSO is tested for large-scale optimization (100-D, 500-D, and 1000-D) benchmark functions. This paper has a novel contribution which is the usage of OCA on Social Optimization Algorithms. The results clearly indicate that the OCA is increasing the results of large-scale SL-PSO.