Gearbox Fault Detection through PSO Exact Wavelet Analysis and SVM Classifier

The problem with development of the support vector machine (SVM) classifiers using modified particle swarm optimization (PSO) algorithm and their ensembles has been considered. Solving this problem would allow fulfilling the high-precision data classification, especially Big Data classification, with the acceptable time expenditures. The modified PSO algorithm conducts a simultaneous search of the type of kernel functions, the parameters of the kernel function and the value of the regularization parameter for the SVM classifier. The idea of particles' «regeneration» served as the basis for the modified PSO algorithm. In the implementation of this algorithm, some particles change the type of their kernel function to the one which corresponds to the particle with the best value of the classification accuracy. The offered PSO algorithm allows reducing the time expenditures for the developed SVM classifiers, which is very important for Big Data classification problem. In most cases such SVM classifier provides the high quality of data classification. In exceptional cases the SVM ensembles based on the decorrelation maximization algorithm for the different strategies of the decision-making on the data classification and the majority vote rule can be used. Also, the two-level SVM classifier has been offered. This classifier works as the group of the SVM classifiers at the first level and as the SVM classifier on the base of the modified PSO algorithm at the second level. The results of experimental studies confirm the efficiency of the offered approaches for Big Data classification.

In order to improve the performance of Support Vector Machine (SVM) classifier for imbalanced data, this paper proposes dynamic cost-sensitive SVM classifier based on chaos particle swarm optimization (CPDC_SVM). Firstly, this paper introduces dynamic cost-sensitive thought to SVM classifier, and gives the method for structuring dynamic cost and cost-sensitive SVM model. Secondly, we propose the evaluation methodology performance for classifier, and adopts decimal base to code the particles. At last, chaos thought is introduced in particle swarm optimization algorithm, and the Algorithm of the dynamic cost-sensitive SVM classifier is given, which improves convergent speed and accuracy of particle swarm optimization, and can optimize dynamic cost-sensitive SVM well, so CPDC_SVM adds effectively the convergence speed and accuracy for the particle swarm optimization algorithm. Experimental results show CPDC_SVM has higher precision than traditional SVM classifier, and dynamic cost and chaos particle swarm optimization can improve the performance for classifier.

Due to errors, accuracy of Global Positioning System is not so high. Therefore, the Real Time Differential Global Poisoning System (RTDGPS) which is based on the successive transmission message of RTCM protocol, is using in real-time applications. Stability and accuracy of the system, significantly depends to a fast transmission of correction messages. These messages come from the reference station to the user stations and affected by the errors related to each satellite. Receiving correction factors which are transmitted by the reference station are facing with time-lag problem, which can increase the error of the RTDGPS. To overcome this problem, prediction algorithms are used. In this research, support vector machine (SVM) model is used to predict the pseudo range correction. Unfortunately, the practical use of SVM is limited because the quality of SVM models depends on a proper setting of SVM and SVM kernel parameters. Therefore, to determine the main parameters of the SVM, both particle swarm optimization (PSO) and genetic algorithm (GA), as two optimization techniques, are used. The proposed methodology has been implemented by a 6-s predicts time step. Simulations showed that the accuracy of GA–SVM and PSO–SVM are equal to 0.186 and 0.154, respectively.

Collaborative beamforming in wireless sensor networks using a novel particle swarm optimization algorithm variant

Supercapacitors present an attractive energy storage alternative for high‐performance applications due to their compact size and high‐power density. Therefore, the supercapacitors have broad prospects for the development in the field of electric vehicles and renewable energy. To describe the output characteristics of the supercapacitors with high accuracy for the simulation research and practical application, a dynamic modelling method is proposed for the supercapacitors based on support vector machine (SVM) and particle swarm optimisation (PSO) algorithm. In this study, the SVM is used to predict the output voltage of the supercapacitors with the key parameters (temperature, current and initial voltage). The PSO algorithm is adopted to optimise the parameters of the SVM to improve the performance of the dynamic modelling. An experimental platform is established, where an electric machine drive system powered by the supercapacitors is controlled to operate at frequent acceleration and deceleration modes, thus leading to the frequent charging and discharging of the supercapacitors. The experimental data is collected to validate the effectiveness of the proposed method. The results show that the proposed method can effectively predict the output voltage of the supercapacitors.

The particle swarm optimization (PSO) algorithm presents a new way for finding optimal solutions of complex optimization problems. In this paper a modified particle swarm optimization algorithm is presented. We modify the PSO algorithm in some aspects. Firstly, a contractive factor is introduced to the position update equation, and the particles are limited in search region. A new strategy for updating velocity is then adopted, in which the velocity is weakened linearly. Thirdly, using an idea of intersecting two modified PSO algorithms. Finally, adding an item of integral control in the modified algorithm can improve its global search ability. Based on these strategies, we proposed a new PSO algorithm named crossed PSO algorithm. Simulation results show that the crossed PSO is superior to the original PSO algorithm and can get overall promising performance over a wide range of problems.

Neural Network system for SoC is one topic in the world. Recently, SoC began to combine AI technique by Python. When developers consider Intelligent system on SoC, the co-design software and hardware are significant. In this case, small devices and embedded system are target system. For IoT, those techniques will grow up in the world. Those cases, avoiding premature convergence while keeping performance is a challenge in training the neural network, especially in a case of large NNs or a large number of training data. In this case, one problem is a huge memory and processing power. For tiny memory and processing system as SoC, we propose an improved particle swarm optimization (PSO) algorithm called the PSOseed2 algorithm for training NN. The PSOseed2 algorithm solves the premature convergence of the standard PSO (SPSO) algorithm by slightly modifying the velocity update function without adding many computational tasks to the SPSO algorithm. We evaluated this algorithm on field programmable gate array (FPGA)-based NN and software-based NN and trained these NNs with different PSO algorithms that are SPSO, PSOseed, PSOseed2, and dissipative PSO. Experimental results with different datasets confirmed that the NNs trained by the proposed PSOseed2 algorithm had better recognition rates and lower global learning errors than the NN trained by other PSO algorithms. In this talk, I would like to talk those relational research in the world and our results and discussion.

Recently, focus on maintenance strategies has been shifted towards prognostic health management (PHM) and a number of state of the art algorithms based on data-driven prognostics have been developed to predict the health states of degrading components based on sensory data. Amongst these algorithms, Multiclass Support Vector Machines (MC-SVM) has gained popularity due to its relatively high classification accuracy, ability to classify multiple patterns and capability to handle noisy / incomplete data. However, its application is limited by the difficulty in determining the required kernel function and penalty parameters. To address this problem, this paper proposes a hybrid differential evolution – particle swarm optimization (DE-PSO) algorithm to optimize the MC-SVM kernel function and penalty parameters. The differential algorithm (DE) obtains the search limit for the SVM parameters, while the particle swarm optimization algorithm (PSO) determines the global optimum parameters for a given training data set. Since degrading machinery components display several degradation stages in their lifetime, the MC-SVM trained with optimum parameters are used to estimate the health states of a degrading machinery component, from which the remaining useful life (RUL) is predicted. This method improves the classification accuracy of MC-SVM in predicting the health states of a machinery component and consequently increases the accuracy of RUL predictions. The feasibility of the method is validated using bearing prognostic run-to-failure data obtained from NASA public data repository. A comparative study between MC-SVM with parameters obtained using simple grid search with n-fold cross validation and MC- SVM with DE-PSO based on prognostic performance metrics reveals that the proposed method has better performance, with all the cases considered falling within a 10 % error margin. The method also outperforms other soft computing methods proposed in literature.

The Particle swarm optimization (PSO) algorithm is a population-based intelligent stochastic search technique encouraged from the intrinsic manner of bee swarm seeking for their food source. With flexibility for numerical experimentation, the PSO algorithm has been mostly used to resolve diverse kind of optimization problems. The PSO algorithm is frequently captured in local optima meanwhile handling the complex real-world problems. Many authors improved the standard PSO algorithm with different mutation strategies but an exhausted comprehensive overview about mutation strategies is still lacking. This article aims to furnish a concise and comprehensive study of problems and challenges that prevent the performance of the PSO algorithm. It has tried to provide guidelines for the researchers who are active in the area of the PSO algorithm and its mutation strategies. The objective of this study is divided into two sections: primarily to display the improvement of the PSO algorithm with mutation strategies that may enhance the performance of the standard PSO algorithm to great extent and secondly, to motivate researchers and developers to use the PSO algorithm to solve the complex real-world problems. This study presents a comprehensive survey of the various PSO algorithms based on mutation strategies. It is anticipated that this survey would be helpful to study the PSO algorithm in detail for researchers.

The maximum power point tracking (MPPT) system ensures that the solar cell operates at its maximum power point, which is the point at which it produces the most power for a given set of conditions. This study uses an artificial bee colony (ABC) algorithm and particle swarm optimization (PSO) algorithms to optimize the boost converter. The ABC and PSO algorithms help to optimize the boost converter by finding the best possible duty cycle and other parameters that maximize efficiency. The resulting duty cycle of 20% produced the highest efficiency of 78.26% obtained from the boost converter. A system-wide test was conducted between 9:00 and 11:10 WIB. In MPPT systems using boost converters with the PSO algorithm, the average power is 16.80 W, while in MPPT systems using boost converters with the ABC algorithm, the average power is 14.53 W. The novelty of the study discusses a system designed to optimize power by using a boost converter in order to increase power output from solar panels. The algorithms PSO and ABC are used to analyze the data from the PV and adjust the converter parameters to optimize the output in order to increase the efficiency of the system.

In order to solve the problem of interaction between multiple evaluation indexes of seed metering performance under multiple factors of centralized seed feeding device, a multi-objective optimization of structure based on particle swarm optimization (PSO) algorithm was proposed in this paper. The wheat centralized seed feeding device was taken as the research object, and the experimental factors were cone angle of type hole, working speed and seed filling gap. The working process of wheat centralized seed feeding device was simulated by discrete element method (DEM). The average seed number of type hole, the variation coefficient of the average seed number of type hole, and the maximum tangential force between seed and seed feeding mechanism were selected as the evaluation indexes. Through the variance analysis of the evaluation indexes by the experimental factors, the optimization objective function was constructed. Using PSO algorithm, the multi-objective optimization was carried out for the wheat centralized seed feeding device. The optimization results show that the best structural combination parameters of the wheat centralized seed feeding device are the hole cone angle of 31.6° and the seed filling gap of 4.6 mm. The validity of the method was verified by simulation and field test. The results show that the PSO algorithm multi-objective optimization method proposed in this paper can provide a reference for the structural improvement and optimal design of the centralized seed feeding device. Keywords: centralized seed feeding device, multi-objective, optimization, PSO algorithm DOI: 10.25165/j.ijabe.20201306.5665 Citation: Wang Q Q, Li Z D, Wang W W, Zhang C L, Chen L Q, Wan L. Multi-objective optimization design of wheat centralized seed feeding device based on particle swarm optimization (PSO) algorithm. Int J Agric & Biol Eng, 2020; 13(6): 76–84.

This study presents an adaptive PSO (Particle Swarm Optimization) algorithm as the foundation for a swarm intelligence approach in multi-UAV operations. The traditional PSO formula for particle velocity and position updates was modified to incorporate a variation strategy from Differential Evolution (DE), enabling UAVs to dynamically adjust their trajectories. The integration of deep reinforcement learning (DRL) further enhances the model's ability to optimize task offloading and computational distribution, ensuring that UAVs function as efficient edge nodes. An experimental evaluation was conducted to assess the proposed PSO-Edge method compared to other machine learning techniques, specifically Random Forest and Support Vector Machine (SVM). The experimental setup involved a simulation environment where UAVs were tasked to monitor data and execute missions over a defined area. The hardware included an Intel Xeon Gold 6248R CPU, 128 GB RAM, and an NVIDIA Tesla V100 GPU, with the simulation executed using Python 3.8. The proposed PSO-Edge algorithm demonstrated superior performance across multiple metrics: reducing task completion time by 42.1 minutes compared to Random Forest and SVM; achieving the lowest energy consumption per task at 28.9 Wh; demonstrating efficient communication with the least latency at 0.15 seconds; and achieving the highest task accuracy at 96%. The results confirm that the PSO-Edge method outperforms traditional machine learning approaches in task efficiency, energy consumption, communication latency, and accuracy. This highlights the benefits of integrating edge computing with the PSO algorithm, establishing it as a robust solution for multi-UAV operations. The findings have significant implications for optimizing UAV-based applications, particularly in environments requiring dynamic adaptation and efficient resource management.

This article focuses on classifying a variety of faults in power transformers with high precision, using an Improved Particle Swarm Optimization-Support Vector Machine (IPSO-SVM) system designed for fault diagnosis. The process begins with the identification of five distinct gases dissolved in oil, serving as diagnostic features. Minimal output encoding is then used to construct multiple binary support vector machine (SVM), facilitating a multi-class classification of transformer faults. While other studies often combine traditional Particle Swarm Optimization (PSO) algorithms with SVMs, our approach employs an enhanced PSO algorithm. This improved algorithm allows for the optimization of inertia and learning factors, values of which adapt based on iteration counts. The PSO is then leveraged on the optimization of the penalty factor and the radial basis function of SVM, thereby improving its classification performance. Simulation results indicate that our IPSO-SVM methodology achieves 90% and 92% accuracy in training and testing sets, respectively. This method significantly enhances the accurateness oftransformer malfunction diagnosis, exhibiting superior diagnostic precision compared to traditional power transformer malfunction diagnosis methods.

Celiac disease (CD) is a complex disorder whose development is affected by genetics (HLA alleles) and gluten ingestion. Its diagnosis is very difficult due to clinical manifestations complexity, latent period, and similarity to other diseases. Studies show that a high percentage of CD patients remain undiagnosed. The celiac patients who are not treated are at a high risk of cancer, malignant lymphoma, and small-bowel neoplasia. Therefore, CD diagnosis and grading is of paramount importance. This paper presents a new method for grading CD based on the combination of fuzzy cognitive map (FCM) and support vector machine. To improve the efficiency and increase classification ability of FCM, particle swarm optimization (PSO) algorithm is applied to adjust FCM weights. In this study, the newest method of grading A, B1, and B2 is used. The empirical results show that the main advantage of PSO algorithm is its speed of convergence and the ability to obtain faster possible schedules. The proposed method is tested on 89 patients. The simulation results prove the superiority of the proposed method compared with Bayesian networks based on the rules and other procedures set forth in the literature. These results show the percentages of 87%, 86%, and 84% for three grades of A, B1 and B2.

Stock trend forecasting is one of the important issues in stock market research. However, forecasting stock trend remains a challenge because of its irregular characteristic in the stock indices distribution, which changes over time. Support Vector Machine (SVM) produces a fairly good result in stock trend forecasting, but the performance of SVM can be affected by the high dimensional input features and noisy data. This paper hybridizes the Particle Swarm Optimization (PSO) algorithm to generate the optimum features set prior to facilitate SVM learning. The SVM algorithm uses the Radial Basis Function (RBF) kernel function and optimization of the gamma and large margin parameters are done using the PSO algorithm. The proposed algorithm was tested on a pre-sampled 17 years record of daily Kuala Lumpur Composite Index (KLCI) data. The PSOSVM approach is applied to eliminate unnecessary or insignificant features, and effectively determine the parameter values, in turn improving the overall prediction results. The optimized feature space of technical indicators of the algorithm is proven by the experimental results showing that PSOSVM has outperformed SVM technique significantly.