Detecting method and monitoring system for coupling fault of rotating machine

Abstract

The invention relates to a detecting method and a monitoring system for a coupling fault of a rotating machine. The detecting method comprises the steps of using vibrating signal data which are acquired by the rotating machine in a normal working condition and a fault working condition as a training data set, establishing a deep convolutional neural network model, directly using the vibration dataas an input, introducing a self-normalizing strategy for standardizing a neuron activating value, training the parameter of the deep convolutional neural network model, storing the trained parameterdata, acquiring the data in a real-time working condition as testing data, and realizing fault detection through the deep convolutional neural network model. The detecting method and the monitoring system have advantages of realizing no requirement for an accurate mathematical model of an industrial process, facilitating actual application, realizing fault detection and fault working condition classification, effectively monitoring a specific component with mechanical damage, and realizing high detection accuracy.

Description

technical field [0001] The invention belongs to the technical field of industrial machinery monitoring and fault diagnosis, and relates to a detection method and a monitoring system for coupling faults of rotating machinery. Background technique [0002] In modern industrial production, fault diagnosis of complex electromechanical systems by collecting vibration signals of mechanical components is one of the most widely used diagnostic methods in fault diagnosis of rotating machinery. Traditional vibration diagnosis methods and theories have matured, and its core is usually divided into two parts: one is the feature extraction of vibration signals, and the other is pattern classification. [0003] Rotating machinery refers to the machinery that mainly relies on rotating actions to complete specific functions. It is widely used in mechanical equipment. Among them, gears and shafts are key components with high application rates. Their state and geometric characteristics play a...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem of poor fault detection accuracy when existing fault detection methods detect coupling faults, the present invention provides a detection method and monitoring system for rotating machinery coupling faults with high fault detection accuracy

Images