Unlock instant, AI-driven research and patent intelligence for your innovation.

Fault diagnosis model self-learning method based on asynchronous parallel reinforcement learning

A technology of fault diagnosis model and self-learning method, applied in neural learning method, biological neural network model, neural architecture, etc., can solve the problems of high resource consumption and low self-learning efficiency of fault diagnosis model, and achieve the goal of improving self-learning efficiency. Effect

Pending Publication Date: 2021-05-14
BEIHANG UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a fault diagnosis model self-learning method based on asynchronous parallel reinforcement learning, which is used to solve the technical problems of low efficiency and excessive resource consumption of fault diagnosis model self-learning in multi-parameter and high-dimensional space

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Fault diagnosis model self-learning method based on asynchronous parallel reinforcement learning
  • Fault diagnosis model self-learning method based on asynchronous parallel reinforcement learning
  • Fault diagnosis model self-learning method based on asynchronous parallel reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0142] Example 1: Self-learning of gearbox fault diagnosis model based on asynchronous parallel reinforcement learning

[0143] The gearbox data used in the test comes from the power transmission failure prediction test bench manufactured by Spectra Quest in the United States. The test bench consists of a drive motor, a planetary gearbox for testing, a parallel gearbox for testing, a load parallel gearbox (2), a load motor, a driver and a number of supporting sensors.

[0144] In this case study, the fault injection site is located in the planetary gearbox, and the injected fault modes include gear cracks, wear, missing teeth, and broken teeth.

[0145] The working condition of data acquisition is gear speed 20Hz, load 0Nm (no load). The acceleration sensor is installed on the outer end cover of the input shaft of the planetary gearbox through a threaded connection, and the vibration signal of the gearbox is sampled for 40 seconds at a frequency of 12.8kHz. Therefore, there ...

Embodiment 2

[0156] Example 2: Self-learning of hydraulic pump fault diagnosis model based on asynchronous parallel reinforcement learning

[0157] Piston hydraulic pump test equipment The acceleration sensor used to collect vibration signals is installed on the end face of the pump, and the sampling frequency is 1024Hz. During the test, the motor speed was set at 5280rpm, and a total of seven pumps were used. The collected data included normal conditions, wear failures of the sliding shoe and swash plate, and wear failure of the valve plate. Two sets of samples were collected for each pump, each set had 1024 points, and a total of 14 sets of samples were obtained. When used to evaluate the effect of fault diagnosis, each group of samples is reorganized through a sliding window with a length of 512, that is, the length of a single sample is 512 points, and the data of each pump forms a fault sample containing 1536 through the sliding window. Data set, as shown in Table 10.

[0158] Table...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a fault diagnosis model self-learning method based on asynchronous parallel reinforcement learning. The method comprises the following steps: configuring N fault diagnosis model self-learning agents for simultaneously operating respective actor-judge reinforcement learning algorithms on multiple threads of a CPU; configuring N fault diagnosis model self-learning interaction environments, wherein each fault diagnosis model self-learning interaction environment interacts with a corresponding fault diagnosis model self-learning agent; configuring a global network used for synchronizing the global network parameters to the N fault diagnosis model self-learning agents; and performing self-learning on network parameters of the fault diagnosis models through multiple operations among each fault diagnosis model self-learning agent, each fault diagnosis model self-learning interaction environment and the global network.

Description

technical field [0001] The invention relates to the technical field of fault diagnosis, in particular to a fault diagnosis model self-learning method based on asynchronous parallel reinforcement learning. Background technique [0002] Fault diagnosis is a technical means to detect, isolate and identify faults. With the rapid development of science and technology, the structure of various mechanical equipment is becoming more and more complex, and the monitoring parameters tend to be massive and multi-dimensional. Higher requirements are put forward for fault diagnosis methods. Require. Because deep learning technology has a high ability to approximate high-dimensional nonlinear parameters, a series of fault diagnosis methods combined with deep learning methods are widely used in industrial engineering. The fault diagnosis method based on deep learning can integrate fault feature extraction and fault mode classification, realize the algorithm to independently extract useful ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06N3/04G06N3/08
CPCG06N3/04G06N3/08
Inventor 丁宇王超马剑杨帆吕琛
Owner BEIHANG UNIV