Automated gear case fault diagnosis method based on neural network and characteristic frequency band

Abstract

The invention relates to an automated gear case fault diagnosis method based on a neural network and a characteristic frequency band. Gear fault diagnosis comprises: S1, collecting a vibration acceleration signal of a sample gear case, and constructing a signal library; S2, performing Fourier transformation on the vibration acceleration signal to obtain a corresponding amplitude spectrum; S3, extracting several groups of frequency band samples from the amplitude spectrum, marking a gear status label of each sample, and building a gear fault frequency band library; S4, training a gear fault diagnosis neural network by using the gear fault frequency band library; S5, extracting an amplitude spectrum characteristic frequency band of a gear in a to-be-diagnosed gear case; and S6, inputting theamplitude spectrum characteristic frequency band to the gear fault diagnosis neural network to diagnose a fault probability of the gear. According to the method, an amplitude spectrum of a gear casevibration signal on a characteristic frequency band is used as a characteristic signal to separately estimate a fault status of a gear, so that a current fault can be efficiently and automatically diagnosed, persons of ordinary skill are guided to repair and maintain the gear case in time, and working efficiency is obviously improved.

Description

technical field [0001] The invention belongs to the technical field of mechanical fault diagnosis, and in particular relates to an automatic fault diagnosis method of a gearbox based on a neural network and a characteristic frequency band. Background technique [0002] Gearboxes are used to transmit forces and loads and are widely used in the field of mechanical engineering. However, the working environment of the gearbox is usually harsh and complex, resulting in it being subjected to complex alternating loads for a long time, and the operating conditions change frequently, coupled with the influence of temperature, lubrication, physical chemistry, etc., there are many factors that are easy to cause Gearbox gears (such as sun gears, planetary gears, ring gears, etc.) have faults such as pitting, peeling, and cracks. In addition, bearings are prone to failure conditions such as wear, corrosion, and indentation. [0003] When a gearbox fails, it is necessary to diagnose the...

AI Technical Summary

Problems solved by technology

[0004] However, gearbox fault diagnosis is currently mainly done by professional technicians

The manual diagnosis process often requires a large amount of manual analysis process, which is time-consuming and labor-intensive.

Moreover, the accuracy of manual diagnosis and its dependence on the professional knowledge and diagnosis experience of the diagnostic personnel, the threshold for ordinary technical personnel is relatively high

Images