Verification of a new method for PI block design of MRAS-based sensorless speed estimators

Babaei, Ebrahim; Sharifian, Mohammad Bagher Bannae; Ajabi-Farshbaf, Rahim; Hosseini, Seyed Hossein

doi:10.1109/icems.2011.6073606

Cited by 6 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Speed observation on induction machines is used in applications from induction motor fault detection to traction . Speed estimation methods are based on the measurement of electrical variables such as current and voltage from the power supply.…”

Section: Frame Of Referencementioning

confidence: 99%

Real‐time condition monitoring on VSD‐fed induction motors through statistical analysis and synchronous speed observation

Cabal-Yépez

Fernandez‐Jaramillo

García-Pérez

et al. 2014

Int. Trans. Electr. Energ. Syst.

View full text Add to dashboard Cite

SUMMARYInduction motors are key elements of every industrial process. A faulty motor produces interruptions on production lines, with consequences in cost, product quality, and safety. The relevance in induction motor monitoring is the ability to detect faults in incipient state. Many proposed methods consider direct connection of motors to the power supply; however, the common practice in industry is to connect them through variable speed drives (VSD), which introduce harmonics into the current supply signal that make the fault identification extremely difficult. This work proposes a statistical analysis through mean, variance, and information entropy computation, combined with sensorless rotating speed estimation for classifying different faults in induction motors using an artificial neural network. The proposed methodology examines the voltage and current signals provided by an industrial VSD that ensures a high certainty on identifying the treated faults at different rotational speed. A field programmable gate array-based implementation is developed to offer an online, system-on-chip solution for real-time condition monitoring.

show abstract

Section: Frame Of Referencementioning

confidence: 99%