Hybrid Rotating Machinery Fault Diagnosis and Prognosis

Description:

Reference #: 01570

The University of South Carolina is offering licensing opportunities for Hybrid Rotating Machinery Fault Diagnosis and Prognosis

Background:

Bearing faults are the top contributor to the failure of rotating machinery systems. In wind energy systems, about 80% of gearbox failures are caused by bearing faults. According to verified market report, the global rotating machinery market was valued at USD 1,20,725.2 million in 2018 and is projected to reach USD 1,75,184.2 Million by 2026, growing at a compound annual growth rate of 4.87% from 2019 to 2026. The FDP techniques in this patent can be used for the FDP tasks for many rotating machinery systems.

Invention Description:

This patent is a hybrid system FDP approach, which include failure identification, automatic fault model selection, and Bayesian estimation-based FDP framework. First, the continuous wavelet transform (CWT) is adopted to analyze the vibration data in the time frequency domain. To improve the fault detection accuracy and model selection accuracy, the system domain knowledge and operating conditions are involved in this process.

Potential Applications:

This proposed method can be used for the FDP tasks for different types of rotating machinery systems or components, such as bearings, gearboxes, power transformers, etc. These systems are spread throughout all aspects of production and life.

Advantages and Benefits:

This patent presents a hybrid bearing FDP framework that integrates CWCM-CNN based system failure identification and fault model selection, Bayesian estimation-based FDP, and prognosis result fusion. This approach takes advantage of the strong learning and pattern identification ability of CNN, uncertainty representation ability of the Monte Carlo method.

Patent Information:
For Information, Contact:
Omar Iyile
Technology Associate
University of South Carolina
oiyile@email.sc.edu
Inventors:
Guangxing Niu
Bin Zhang
Keywords:
Continuous wavelet transform
convolutional neural network
Fault model selection
Particle filter
Rotating machinery systems
STP estimation
© 2024. All Rights Reserved. Powered by Inteum