Exploring evolutionary-tuned autoencoder-based architectures for fault diagnosis in a wind turbine gearbox DOI Creative Commons
Samuel M. Gbashi, Obafemi O. Olatunji, Paul A. Adedeji

и другие.

Smart Science, Год журнала: 2024, Номер unknown, С. 1 - 21

Опубликована: Июнь 11, 2024

Vibration-based fault diagnosis from rotary machinery requires prior feature extraction, selection, or dimensionality reduction. Feature extraction is tedious, and computationally expensive. selection presents unique challenges intrinsic to the method adopted. Nonlinear reduction may be achieved through kernel transformations, however there often a trade-off in information achieve this. Given above, this study proposes novel autoencoder (AE) pre-processing framework for vibration-based wind turbine (WT) gearboxes. In study, AEs are used learn features of WT gearbox vibration data while simultaneously compressing data, obviating need costly engineering The effectiveness proposed was evaluated by training genetically optimized linear discriminant analysis (LDA), multilayer perceptron (MLP), random forest (RF) models, with AE's latent space features. models were using known classification metrics. results showed that performance depends on size space. As increased, quality extracted improved until plateau observed at dimension 10. AE pre-processed RF, MLP, LDA designated AE-Pre-GO-RF, AE-Pre-GO-MLP, AE-Pre-GO-LDA, accuracy, sensitivity, specificity seven (7) conditions. AE-Pre-GO-RF model outperformed its counterparts, scoring 100% all metrics, though longest time (239.50 sec). Comparable found comparing similar investigations involving traditional processing techniques. More so, it established effective can manifold learning without expensive engineering.

Язык: Английский

Computer-aided diagnosis system for predicting liver cancer disease using modified Genghis Khan Shark Optimizer algorithm DOI

Marwa M. Emam,

Reham R. Mostafa, Essam H. Houssein

и другие.

Expert Systems with Applications, Год журнала: 2025, Номер 285, С. 128017 - 128017

Опубликована: Май 11, 2025

Язык: Английский

Процитировано

0
Feature Niching Based Differential Evolution for Feature Selection on High-Dimensional Data DOI
Biyu Yin, Mingwei Wang, Maolin Chen

и другие.

Communications in computer and information science, Год журнала: 2025, Номер unknown, С. 44 - 54

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
A Hybrid Adaptive Particle Swarm Optimization Algorithm for Enhanced Performance DOI Creative Commons
Zhengfeng Jiang, Daoheng Zhu,

Xiao-Yu Li

и другие.

Applied Sciences, Год журнала: 2025, Номер 15(11), С. 6030 - 6030

Опубликована: Май 27, 2025

The traditional particle swarm optimization (PSO) algorithm often exhibits defects such as of slow convergence and easily falling into a local optimum. To overcome these problems, this paper proposes an enhanced variant featuring adaptive selection. Initially, composite chaotic mapping model integrating Logistic Sine mappings is employed to initialize the population for diversity exploration capability. Subsequently, global search capabilities are balanced through introduction inertia weights. then divided three subpopulations—elite, ordinary, inferior particles—based on their fitness values, with each group employing distinct position update strategy. Finally, mutation strategy incorporated avoid optima. Experimental results demonstrate that our outperforms existing algorithms standard benchmark functions. In practical engineering applications, also has demonstrated better performance than other meta heuristic algorithms.

Язык: Английский

Процитировано

0
FTDZOA: An Efficient and Robust FS Method with Multi-Strategy Assistance DOI Creative Commons

Fuqiang Chen,

Shitong Ye, Lijuan Xu

и другие.

Biomimetics, Год журнала: 2024, Номер 9(10), С. 632 - 632

Опубликована: Окт. 17, 2024

Feature selection (FS) is a pivotal technique in big data analytics, aimed at mitigating redundant information within datasets and optimizing computational resource utilization. This study introduces an enhanced zebra optimization algorithm (ZOA), termed FTDZOA, for superior feature dimensionality reduction. To address the challenges of ZOA, such as susceptibility to local optimal subsets, limited global search capabilities, sluggish convergence when tackling FS problems, three strategies are integrated into original ZOA bolster its performance. Firstly, fractional order strategy incorporated preserve from preceding generations, thereby enhancing ZOA's exploitation capabilities. Secondly, triple mean point guidance introduced, amalgamating point, random current effectively augment exploration prowess. Lastly, capacity further elevated through introduction differential strategy, which integrates disparities among different individuals. Subsequently, FTDZOA-based method was applied solve 23 problems spanning low, medium, high dimensions. A comparative analysis with nine advanced methods revealed that FTDZOA achieved higher classification accuracy on over 90% secured winning rate exceeding 83% terms execution time. These findings confirm reliable, high-performance, practical, robust method.

Язык: Английский

Процитировано

3
Exploring evolutionary-tuned autoencoder-based architectures for fault diagnosis in a wind turbine gearbox DOI Creative Commons
Samuel M. Gbashi, Obafemi O. Olatunji, Paul A. Adedeji

и другие.

Smart Science, Год журнала: 2024, Номер unknown, С. 1 - 21

Опубликована: Июнь 11, 2024

Vibration-based fault diagnosis from rotary machinery requires prior feature extraction, selection, or dimensionality reduction. Feature extraction is tedious, and computationally expensive. selection presents unique challenges intrinsic to the method adopted. Nonlinear reduction may be achieved through kernel transformations, however there often a trade-off in information achieve this. Given above, this study proposes novel autoencoder (AE) pre-processing framework for vibration-based wind turbine (WT) gearboxes. In study, AEs are used learn features of WT gearbox vibration data while simultaneously compressing data, obviating need costly engineering The effectiveness proposed was evaluated by training genetically optimized linear discriminant analysis (LDA), multilayer perceptron (MLP), random forest (RF) models, with AE's latent space features. models were using known classification metrics. results showed that performance depends on size space. As increased, quality extracted improved until plateau observed at dimension 10. AE pre-processed RF, MLP, LDA designated AE-Pre-GO-RF, AE-Pre-GO-MLP, AE-Pre-GO-LDA, accuracy, sensitivity, specificity seven (7) conditions. AE-Pre-GO-RF model outperformed its counterparts, scoring 100% all metrics, though longest time (239.50 sec). Comparable found comparing similar investigations involving traditional processing techniques. More so, it established effective can manifold learning without expensive engineering.

Язык: Английский

Процитировано

2