Sensorless, online motor diagnostics

Kliman, G.B.; Premerlani, W.; Yazici, B.; Koegl, R.A.; Mazereeuw, J.

doi:10.1109/67.582451

Cited by 76 publications

(17 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such intelligent systems typically consist of three main parts: data acquisition, feature extraction and selection, and data classification. Time-domain [4], frequency-domain [5][6][7][8], enhanced frequency [9][10][11][12], and time-scale analysis [13][14][15][16] are four main areas where signal processing techniques are used in the feature extraction [17]. The extracted features are used to both train and operate Artificial Neural Networks (ANNs) or other decision support systems (DSS) [18][19][20][21][22][23][24][25][26][27][28][29][30][31].…”

Section: Introductionmentioning

confidence: 99%

Bearing Fault Detection by One‐Dimensional Convolutional Neural Networks

Eren

2017

Mathematical Problems in Engineering

198

106

View full text Add to dashboard Cite

Bearing faults are the biggest single source of motor failures. Artificial Neural Networks (ANNs) and other decision support systems are widely used for early detection of bearing faults. The typical decision support systems require feature extraction and classification as two distinct phases. Extracting fixed features each time may require a significant computational cost preventing their use in realtime applications. Furthermore, the selected features for the classification phase may not represent the most optimal choice. In this paper, the use of 1D Convolutional Neural Networks (CNNs) is proposed for a fast and accurate bearing fault detection system. The feature extraction and classification phases of the bearing fault detection are combined into a single learning body with the implementation of 1D CNN. The raw vibration data (signal) is fed into the proposed system as input eliminating the need for running a separate feature extraction algorithm each time vibration data is analyzed for classification. Implementation of 1D CNNs results in more efficient systems in terms of computational complexity. The classification performance of the proposed system with real bearing data demonstrates that the reduced computational complexity is achieved without a compromise in fault detection accuracy.

show abstract

Section: Introductionmentioning

confidence: 99%

Bearing Fault Detection by One‐Dimensional Convolutional Neural Networks

Eren

2017

Mathematical Problems in Engineering

198

106

View full text Add to dashboard Cite

show abstract

“…Thus it has similar manifestation as the lamination material anisotropy and can hardly be separated. Since the separation could be very useful for example in the online condition-monitoring (about 50% of all machine breakdowns are based on bearing faults [2]) it is of great importance to know the mechanisms of the creation of the eccentricity dependent 2nd harmonic in the control signal. Fig.…”

Section: Rotor Eccentricitymentioning

confidence: 99%

Separating anisotropic material effects in the control signals of inverter-fed AC machines

Wolbank

Machl

Hauser

2004

Journal of Magnetism and Magnetic Materials

View full text Add to dashboard Cite

“…Using the stator current of electric rotating machinery in order to perform Condition Monitoring (CM) has gained a lot of interest the last decades [1]- [10]. Reasonably, because the CM-technology is cheap, robust, both electrical and mechanical faults can be detected, easy to implement.…”

Section: Introductionmentioning

confidence: 99%

Misalignment and unbalance fault severity estimation using stator current measurements

Corne

Knockaert

Desmet

2017

2017 IEEE 11th International Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives (SDEMPED)

View full text Add to dashboard Cite

Abstract-The feasibility of using stator current measurements in order to detect mechanical problems in electric rotating machinery has been proven by many authors several decades ago. However, this way of detecting problems has never been exploited as a full-covering and reliable prognostic monitoring technology. Although the imposed current-signatures of different evolving mechanical faults in electric machines are well characterized, the relation between the severity of those faults and the corresponding faulty signatures is still not elucidated thoroughly. As the authors built an innovative test-rig on which all kinds of mechanical faults can be emulated with a wide variety of controllable environmental conditions, those specific relations can finally be investigated. The test-rig contains an 11kW induction machine with an integrated active magnetic bearing at drive-end side. The magnetic bearing can easily manipulate the rotor's position corresponding to a real manifesting and evolving mechanical fault with a high accuracy and reproducibility. In this paper, the relation of the current fault-signatures and the severity of misalignment and mechanical unbalance is investigated. After theoretically describing the different eccentricity-related rotorfaults, the faults are induced and emulated on the induction machine under test. The fault-components in the stator currents' spectra are monitored while the severity of the faults is being increased. By fitting corresponding trending-functions on the obtained results, the severity of misalignment and unbalance can be estimated out of stator current measurements.

show abstract

Sensorless, online motor diagnostics

Cited by 76 publications

References 3 publications

Bearing Fault Detection by One‐Dimensional Convolutional Neural Networks

Bearing Fault Detection by One‐Dimensional Convolutional Neural Networks

Separating anisotropic material effects in the control signals of inverter-fed AC machines

Misalignment and unbalance fault severity estimation using stator current measurements

Contact Info

Product

Resources

About