Fault Diagnosis of Voltage Source Inverter for Induction Motor Drives Using Decision Tree
In this paper, open-circuit faults of voltage source inverter for induction motor drives are investigated. Decision tree, which is an expert system that based on knowledge history with simple model, is applied to detect and classify the faults. Input data
- PDF / 724,918 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 66 Downloads / 239 Views
Abstract In this paper, open-circuit faults of voltage source inverter for induction motor drives are investigated. Decision tree, which is an expert system that based on knowledge history with simple model, is applied to detect and classify the faults. Input data for fault diagnosis are collected and extracted from time-domain current signals. Knowledge data are set up from simulation and experiment for building and testing decision tree, and its evaluation results illustrate the potentiality of this method.
⋅
Keywords Fault diagnosis Induction motor drives inverter Decision tree Feature extraction
⋅
⋅
⋅
Voltage source
1 Introduction Induction motors and drives have become more and more important parts in the industrial manufacturing processes. Most of drive systems use a voltage source inverter with power semiconductor switches and a control technique such as field-oriented control (FOC). The control method determines a reference voltage, and then a voltage source inverter will synthesize this voltage by modulating the power switches. Therefore, if a switch is malfunctioned, the synthesis process will be impaired, leading to failure in getting the proper voltage at motor terminals, and causes the instability of the system. To avoid that, it is important to develope predictive maintenance and fault diagnosis capabilities for these systems. It is estimated that about 38 % of the faults in AC drives are related to the failures of power switches [1]. Thus, the aim of this paper is to study on the fault detection and N.-T. Nguyen (✉) ⋅ H.-P. Nguyen (✉) Faculty of Electrical and Electronics Engineering, University of Technology, Ho Chi Minh City, Vietnam e-mail: [email protected]; [email protected] H.-P. Nguyen e-mail: [email protected] © Springer Science+Business Media Singapore 2017 H. Ibrahim et al. (eds.), 9th International Conference on Robotic, Vision, Signal Processing and Power Applications, Lecture Notes in Electrical Engineering 398, DOI 10.1007/978-981-10-1721-6_88
819
820
N.-T. Nguyen and H.-P. Nguyen
classification of a voltage source inverter feeding an induction motor, with main focuses put on open-circuit faults. There are many works carried out on fault monitoring and detection of electrical machines and drives and they have tended to move from traditional techniques to artificial intelligence (AI) techniques recently [1, 2]. An AI system, which is based on knowledge of system behaviors, can perform a fault diagnostic analysis and evaluation which can be accomplished by experts instead. Many authors use fuzzy logic [3], artificial neural network [4] to diagnose the faults. These methods are mostly based on analyzing the effects produced by typical faults in order to construct the rules of decision system. Some works used model-based techniques [5, 6] to classify patterns by comparing with reference patterns. Some works analyzed the feedback signatures with the help of AI techniques to detect and classify the faults [7–9], etc. Decision tree, which builds the diagnostic rules depending on hu
Data Loading...
