Model-Based Fault Diagnosis in Electric Drive Inverters Using Artificial Neural Network

Masrur, M. Abul; Chen, ZhiHang; Zhang, Baifang; Murphey, L.

doi:10.1109/pes.2007.385655

Cited by 37 publications

(20 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…10. For fault detection and isolation, a new index is proposed: (9) This new index is introduced to increase the robustness of the FDI scheme. An interesting property of is that the surface defined by versus time is a constant over one electrical period for both faults, as shown in Fig.…”

Section: Criterion For Fdimentioning

confidence: 99%

“…Others methods are proposed in [9]- [12]. In [9], a simulation model of the faulty drive is used to train an artificial neural network.…”

Section: Introductionmentioning

confidence: 99%

“…In [9], a simulation model of the faulty drive is used to train an artificial neural network. In [10], the authors focus the analysis on the normalized dc components of the phase currents and the number of samples close to zero current.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A signal-based technique for fault detection and isolation of inverter faults in multi-phase drives

Meinguet

Sandulescu

Aslan

et al. 2012

2012 IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES)

View full text Add to dashboard Cite

Abstract-A method for fault detection and isolation is proposed and applied to inverter faults in multi-phase drives. An analysis of simulations in faulty conditions leads to the derivation of suitable fault indices. These are based on the unbalance of the phase currents and their instantaneous frequency. The method is applied to a five-phase permanent-magnet synchronous machine drive. Simulations and experiments validate the proposed method.

show abstract

Section: Criterion For Fdimentioning

confidence: 99%

“…Others methods are proposed in [9]- [12]. In [9], a simulation model of the faulty drive is used to train an artificial neural network.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A signal-based technique for fault detection and isolation of inverter faults in multi-phase drives

Meinguet

Sandulescu

Aslan

et al. 2012

2012 IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES)

View full text Add to dashboard Cite

show abstract

“…Masrur et al (2007) proponen una estrategia basada en redes neuronales que permite detectar y aislar fallas producidas en una sola llave. En (De Araujo Ribeiro et al, 2003) se mide la tensión en determinados puntos de la carga para reconstruir las tensiones aplicadas por el inversor.…”

Section: Introductionunclassified

Diagnóstico de Fallas en el Convertidor del Rotor en Generadores de Inducción con Rotor Bobinado

González

Angelo

Forchetti

et al. 2018

Rev. iberoam. autom. inform. ind.

View full text Add to dashboard Cite

ResumenEn este artículo se presenta una estrategia de diagnóstico de fallas por llave abierta en el convertidor del rotor de generadores de inducción con rotor bobinado. La estrategia utiliza observadores de estado basados en modelos dinámicos obtenidos a partir de un enfoque que utiliza geometría diferencial. Estos observadores permiten generar señales de residuo sensibles a cada falla considerada e insensibles a perturbaciones. A partir de estas señales se construye un vector de residuos con información que puede ser utilizada para la detección y el aislamiento de la falla. La estrategia de detección consiste en la evaluación de la envolvente de la norma del vector de residuos. El aislamiento requiere evaluar la región del plano recorrida por el vector de residuos y comparar dicha región con la asociada a cada escenario de falla. Finalmente se presentan resultados donde se demuestra que la propuesta permite detectar, aislar e identificar fallas de llave abierta, simples o múltiples, aún frente a errores paramétricos y otras perturbaciones.Palabras Clave: Generador de inducción con rotor bobinado, Convertidor del rotor, Falla por llave abierta, Observadores, Escenarios de falla, Diagnóstico de fallas Detection and Isolation of Faults on the Rotor Side Converter of Doubly Fed Induction Generators AbstractThis article presents a fault diagnosis strategy for open switch faults in the rotor side converter of doubly fed induction generators. The proposed strategy uses state observers based on dynamic models obtained from an approach based on differential geometry. These observers allow the generation of residue signals that are sensitive to each fault and insensitive to disturbances. From these signals a residual vector is constructed with information that can be used for the detection and isolation of the fault. The detection strategy consists of evaluating the envelope of the module of the residual vector. Isolation requires evaluating the region of the plane described by the residual vector and comparing that region with the one associated with each fault scenario. Finally, results are presented where it is demonstrated that the proposal allows to detect, isolate and identify open switch faults, both single and multiple, even against parametric errors and other disturbances.

show abstract

“…New detection methods are often being investigated and implemented [28][29][30][31][32][33][34][35][36]. These detection methods are based on measuring currents [28][29][30][31], sensing the voltage output [32], utilizing the applied reference voltage vector [33,34] and using artificial neural networks [35,36]. These methods identify and localize one or multiple The open-switch circuit faults in a three-phase inverter lead to a DC current offset in both the healthy and the faulty phases of the inverter which is connected to the motor windings.…”

Section: Switch Faultsmentioning

confidence: 99%

Real-Time Fault Detection and Reconfiguration of a Three-Phase Electric Motor Drive

Mohammadi¹

View full text Add to dashboard Cite

Variable-frequency drives (VFDs) are widely used for control of electrical machines such as induction motors (IMs) or permanent-magnet synchronous motors (PMSMs). Similar to other electrical devices, these drives are subject to failure. Several types of faults are associated with VFDs. For instance, faults such as open-switch faults, and short-circuit switch faults are the two common faults in VFDs. These faults can yield catastrophic consequences if proper remedial action is not taken. A unique remedial topology for the post-fault period and a new pulse width modulation (PWM) strategy are proposed so that not only the motor drive can continue the operation, but also this new method reduces the common-mode voltage of the three-phase inverter. In this work, switch failures are investigated, and a new solution to detect openswitch faults are proposed using an algorithm based on current analyses. Extensive simulations are done to simulate a real system response during the post-fault phase on different operating points and at various torque loads. These simulation results were used to develop a detection algorithm. These results were also verified using an experimental setup. The proposed method offers an efficient solution for fault detection in VFDs. First, a fast response detection of a failure in an inverter is possible. Second, it provides an alternative back-up method for fault detection in electric drive systems. v

show abstract

Model-Based Fault Diagnosis in Electric Drive Inverters Using Artificial Neural Network

Cited by 37 publications

References 11 publications

A signal-based technique for fault detection and isolation of inverter faults in multi-phase drives

A signal-based technique for fault detection and isolation of inverter faults in multi-phase drives

Diagnóstico de Fallas en el Convertidor del Rotor en Generadores de Inducción con Rotor Bobinado

Real-Time Fault Detection and Reconfiguration of a Three-Phase Electric Motor Drive

Contact Info

Product

Resources

About