Figure - available from: Soft Computing
This content is subject to copyright. Terms and conditions apply.
Source publication
The aim of this research is to estimate the parameters and performance of a permanent magnet generator (PMG). A 90 W, 50 Hz, 220 V, 1500 rpm single-phase induction motor which is having a squirrel cage rotor has been modified into the PMG by replacing the squirrel cage rotor with the surface mounted permanent magnet rotor. The four ceramic magnets...
Citations
... Thus, there is a significant interest among the scientific community in developing strategies to characterize IMs and thus improve their performance in terms of torque (El Ouanjli et al. 2019;Chasiotis et al. 2022;Younas et al. 2022). For such characterization, an equivalent circuit is often used and modeled mathematically to determine the parameters of the corresponding circuit, such as stator resistance and reactance, rotor resistance and reactance, and magnetizing reactance (Al-Jufout et al. 2018;Carbonieri et al. 2019;Gülbahçe and Karaaslan 2022;Kharlamov et al. 2018;Puri and Chauhan 2022). Estimating the parameters of an IM is important because these parameters can be used to design an appropriate control strategy for speed regulation (Khamehchi et al. 2006;Anagreh and Al-Ibbini 2023). ...
In this paper, we present a new optimization method to estimate the parameters and torques of an induction motor (IM). The proposed method is known as the Vortex Search Algorithm (VSA), whose performance is based on the behavior of the vortices created by stirred fluids. This algorithm was compared with other four optimization methods reported in the specialized literature (grasshopper optimization algorithm, particle swarm optimization, salp swarm algorithm and sine cosine algorithm), and its solution quality, precision, and robustness were validated using two test motors. During the tests, we analyzed the minimum error between the estimated values and the values provided by the manufacturer, as well as the global error of each method and their required processing time. The results show that the VSA is an excellent alternative to estimate the parameters of an IM, as it exhibited the best performance when compared to the other optimization methods.
... It is relevant to highlight that the estimation of parameters in these systems through optimization approaches and metaheuristic techniques may reduce the negative effect of the parametric uncertainties. Some examples of cyber-physical systems that could be considered for parameter estimation by implementing AI techniques with edge-IoT architectures are electric machines, such as ANNs trained with electrical and virtual rotor stages in Brushless DC (BLDC) motors [8] or the Gravitational Search Algorithm (GSA) hybridised with Particle Swarm Optimization (PSO) in Induction motors [9], Black Widow Optimization (BWO) in membrane fuel cells to meet the electrical energy needs with reliability [10], and an Equilibrium Optimizer (EO) improved with an ANN in photovoltaic cells that allows predict more output data of these cells to obtain maximum output power harvest [11]. ...
The vast increase in connected Internet of Things (IoT) devices have revolutionised how data are processed. This fact, coupled with the current trend from cloud to edge computing paradigms, has resulted in the need for efficient and reliable data processing near to data sources using resource-constrained devices. In this article, low-cost edge-IoT architectures are implemented to deploy lightweight Machine Learning (ML) models for air quality estimation, such as Polynomial Regression and Artificial Neural Networks (ANN). ML models are deployed in wireless centralised and distributed parallel architectures with common modules such as sensor fusion for luminosity, temperature, humidity, CO2, and other gases. The centralised architecture uses a Graphic Processing Unit (GPU) and the Message Queuing Telemetry Transport (MQTT) protocol, but low-performance processing devices and the Message Passing Interface (MPI) protocol are used in the distributed one. The training and testing of models are attained with appropriate datasets obtained from multiple peak, step, and transient test cases for each air quality parameter. The results for temperature forecasting, and similar ones for other parameters, supports that the distributed parallel architecture could achieve a slightly better estimation metrics and a better performance in power consumption compared to the centralised architecture despite using low-cost general purpose devices.
... Moslehi et al. (2020) proposed a novel hybrid framework by combining Genetic Algorithm (GA) with PSO (GA-PSO) for mining quantitative association rules. Combining PSO with Gravitational Search Algorithm (GSA), Puri and Chauhan (2022) proposed GSA-PSO to estimate the parameters and performance of a Permanent Magnet Generator (PMG). Recently several meta-heuristic methods are being applied to solve problems in various domains. ...
Particle Swarm Optimization (PSO) has a limitation of early convergence and needs to be improved to find the global optima. The main objective here is to improve its exploration capability without deteriorating the exploitation capability. For this purpose, a modified version of PSO, namely Elitist Random Swapped Particle Swarm Optimization (ERSPSO), has been proposed. The elitist (fittest) particles in the swarm guide the other particles to improve their position. To enhance exploration in the search process a swapping of the randomly selected parts of the elitist particle positions (candidate solution) has been made. Consequently, a perturbation is applied to further improve the exploration. The proposed ERSPSO has been applied to the full benchmark set of 25 functions (CEC 2005) as well as complex real life problems like ‘Gene selection by sample classification’. The new variant ERSPSO has been validated by the statistical metrics, convergence plot, sensitivity analysis using convergence behaviour, p-values using Wilcoxon rank sum test and Friedman rank test. For sample classification in Gene selection, VkNN (a new variant of kNN) is proposed which performs better than kNN in classification accuracy. The combined ERSPSO-VkNN is tested in 6 microarray datasets including 4 diseases. In most of the datasets (5 datasets out of 6) ERSPSO-VkNN performs better than the state-of-the-art methods. In different datasets, the percentage of classification accuracy of ERSPSO-VkNN varies between 89.29 and 100%. Finally, a biological verification is performed to show that many of the selected genes are biologically significant according to the reporting in current literature.
