Applications of a Strong Track Filter and LDA for On-Line Identification of a Switched Reluctance Machine Stator Inter-Turn Shorted-Circuit Fault

Abstract

Reliability is pivotal significance for switched reluctance machine drives (SRD) applied to safety essential transportation and industrial fields. An inter-turn shorted-circuit fault (ISCF) could incite the machine to operate in unbalanced status, resulting in the noise increases. In the event such a fault remains untreated, the fault will further destroy the rest of the normal phases, even leading to a tragic incident for the entire drive application. To improve the reliability of SRD, an efficient on-line fault diagnosis method for ISCF should be proposed. This paper is focused on employing the strong track filter (STF) to achieve real-time phase resistance differences between before and after ISCF, which are used as features to diagnose the fault occurrence and the fault phase. Furthermore, a classification namely as linear discriminant analysis (LDA) is selected to estimate fault severity. Finally, simulation and experiments correspond to various running statuses are executed and their results can verify that the diagnosis method has accuracy and robustness.

Predictive Maintenance Model Based on Multisensor Data Fusion of Hybrid Fuzzy Rough Set Theory Feature Selection and Stacked Ensemble for Fault Classification

“…LDA is among the predominantly used classifiers for predictive maintenance purposes [29,31,101]. Its concept exploits the statistical features such as the mean and covariance matrix of each class and then utilises mathematical processes and functions for classifying multiple classes.…”

“…In the context of developing intelligent frameworks for PdM purposes, the utilisation of supervised machine learning algorithms such as the Artificial Neural Networks (ANNs) [21][22][23][24], Support Vector Machine (SVM) [25][26][27][28], Linear Discriminant Analysis (LDA) [29][30][31], and the Bayes classifiers [32][33][34][35] have been well studied. Although these algorithms have yielded to some extent satisfactory results, they are prone to local optima entrapment when their required hyperparameters are not appropriately fine-tuned.…”

Empowering Predictive Maintenance: A Hybrid Method to Diagnose Abnormal Situations

“…This approach looks for the component axes that maximize data variance; likewise, for axes that maximize multiple class separation. However, there are many cases where LDA is the classification algorithm due to all those characteristics [23]. The analyst may choose several other different techniques to reduce the dimensionality of input data based on feature extraction techniques that do not depend on the specialized knowledge about the system.…”

A novel hybrid predictive maintenance model based on clustering, smote and multi-layer perceptron neural network optimised with grey wolf algorithm

“…However, these two major forms of learning possess their strength and limitations. For instance, among the widely used supervised algorithms for fault classification like Artificial Neural Networks (ANNs) [6][7][8], Support Vector Machine (SVM) [2,9,10], Linear Discriminant Analysis (LDA) [11][12][13] and Bayes classifiers [3,14,15] are considered superior in producing labels, but assumes that the objects classified are drawn from an independent and identical distribution, and as such does not consider their interdependencies [16].…”