The Potentiality of Integrating Model-Based Residuals and Machine-Learning Classifiers: An Induction Motor Fault Diagnosis Case

Purbowaskito, Widagdo; Lan, Chen-Yang; Fuh, Kenny

doi:10.1109/tii.2023.3299111

Cited by 5 publications

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, the challenges related to scrutinizing loss components and devising effective control strategies for synchronous reactive machines remain unaddressed. W. Purbowaskito et al [4] suggest that integrating physical measurements with mathematical models within rotor diagnostics can enhance the precision and dependability of fault detection procedures. However, the research did not address the challenge of finding a specific blend of various hybrid techniques to enhance the monitoring of induction machine conditions and their integration into the measurement system.…”

Section: Related Workmentioning

confidence: 99%

Diagnostics of Damage to the Rotor of an Induction Machine Using the Method of Instantaneous Power Analysis

Puka,

Bardhi,

Pjetri

et al. 2024

QAJ

View full text Add to dashboard Cite

The relevance of the problem under study is that in the modern world, where induction machines are widely used in industry, early detection of rotor damage using the method of instantaneous power analysis is critical to maintaining uninterrupted equipment operation and reducing operating costs. The purpose of this study is to investigate and determine the effectiveness of the method of instantaneous power analysis for diagnosing rotor damage in induction machines to improve their reliability and reduce the risk of production failures. The methods used include the analytical method, classification method, functional method, statistical method, synthesis method. In today’s industry, asynchronous machines play a central role, creating a high demand for their reliable and safe operation. Malfunctions and failures in the operation of these machines can lead to serious downtime and considerable losses that affect the financial position of the company. One method of accurate fault diagnosis is to analyses the theoretical components added to the motor current. The analysis revealed that the locus of the space vector, instantaneous power, and sideband frequencies around the fundamental frequency indicate possible broken rotor cores of an induction machine. These components become key indicators for fault detection in induction machines. Furthermore, the experiments on a three-phase squirrel-cage induction confirmed the theoretical analysis, reinforcing the practical significance of the developed methodology for early diagnosis and improving equipment reliability, as well as reducing maintenance and repair costs. The practical significance lies in the development of a methodology that can be used in the industry for early diagnosis of rotor damage in induction machines, ensuring increased equipment reliability and reduced maintenance and repair costs.

show abstract

Section: Related Workmentioning

confidence: 99%