Abstract
A Fault Diagnosis Methodology Based on Non-Stationary Monitoring Signals by Extracting Features With Unknown Probability Distribution
Highlights
In practical complex industrial system, the system operation state is usually not static
The results in table 4 are the average of initial feature set using multi-domain spatial transform, and the average value of SPARSE DEEP BELIEF NEURAL NETWORK (SDBN)
WORKS In practical industrial system, a large amount of monitoring signals are generated in normal operation state, and uncertainty factors occupy main components
Summary
In practical complex industrial system, the system operation state is usually not static. The changes in market demand, external disturbances, equipment aging, sensor defects, and so on, are often regarded as the main reason where monitoring signals in industrial systems exhibit extremely obvious non-stationary characteristics [1], [2]. Fault symptoms are often overwhelmed by the non-stationary nature of monitoring signals, resulting in a large number of false alarms and missed alarms when using the traditional monitoring signal analysis and processing methods. In order to avoid the occurrence of safety accidents, more and more measurement nodes are installed for each equipment to monitor the states of modern industrial systems. With the advancement of sensor technology, the sampling frequency is getting higher and higher. From the beginning of its service to the end of its life, data collection time has been getting longer and longer, and this makes the volume of monitoring data has become larger and larger [3]–[11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
