Adaptive fault diagnosis method based on one-dimensional convolutional neural network

A convolutional neural network and fault diagnosis technology, which is applied in the fault diagnosis of parts and mechanical equipment, can solve problems such as loss, achieve the effect of small model, save diagnosis time, and save labor costs

CN110991295AActive Publication Date: 2020-04-10UNIV OF ELECTRONICS SCI & TECH OF CHINA
9 Cites 13 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2020-04-10

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses an adaptive fault diagnosis method based on a one-dimensional convolutional neural network. According to the adaptive fault diagnosis method, the one-dimensional convolutionalneural network is used as a main model structure; a high-precision diagnosis result can be output immediately by directly inputting a one-dimensional time sequence signal; compared with a traditionalmode of diagnosis by manual feature extraction, the adaptive fault diagnosis method provided by the invention is more convenient and efficient and has lower requirements for professional knowledge andexperience of diagnosis personnel; meanwhile, compared with the mode of fault diagnosis based on neural network structures, the one-dimensional convolutional neural network model is simpler and faster in calculation while maintaining the diagnosis precision, and can provide good guarantee for real-time fault diagnosis. The real-time and rapid adaptive fault diagnosis method provided by the invention can deal with the diversity of fault types easily, and can be widely applied to the fields of machine manufacturing, aerospace, electricity, metallurgy and the like.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of fault diagnosis and computer artificial intelligence, in particular to a method for fault diagnosis of mechanical equipment and parts based on artificial intelligence convolutional neural network (CNN). Background technique

[0002] Existing mechanical equipment and parts are developing in the direction of high speed, high efficiency, and precision. However, in actual operation, the working conditions are complicated, and it is often necessary to analyze and troubleshoot equipment performance and potential faults to achieve early detection, Early detection, early avoidance. Among them, key components of mechanical equipment such as gears, bearings, compressor valves, and motors are often the key objects in the field of mechanical fault diagnosis and monitoring. In this field, traditional time-frequency analysis and other methods have achieved good results in small data environments and single-fault tasks...

Examples

Embodiment Construction

[0026] The present invention will be further described in detail below with reference to the accompanying drawings and in combination with specific embodiments. It should be understood that the examples are only for illustration and not limitation of the present invention.

[0027] Such as figure 1 As shown, the present invention provides a kind of equipment, parts fault diagnosis method based on one-dimensional convolutional neural network, comprising the following steps:

[0028] Table 1 Rolling bearing fault type list

[0029]

[0030] Step 1. Collect the historical data of vibration acceleration signals under various operating states to diagnose faults through the acceleration sensor, and then perform the same adaptive filtering and normalization processing on the total sequence data of various faults to ensure the same distribution of data samples, and then Segment the total sequence data. In this embodiment, the faults are classified according to the different faul...