Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Fault Diagnosis Method for Analog Circuits Based on Echo State Network Dynamic Classification

An echo state network, a technology for simulating circuit faults, applied in the direction of measuring electricity, measuring electrical variables, measuring devices, etc., can solve problems such as low diagnostic accuracy, and achieve the effect of high diagnostic accuracy

Inactive Publication Date: 2011-11-30
HARBIN INST OF TECH
View PDF3 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention aims to solve the problem of low diagnostic accuracy of analog circuit fault diagnosis using traditional neural network, thereby providing an analog circuit fault diagnosis method based on echo state network dynamic classification

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 Method for Analog Circuits Based on Echo State Network Dynamic Classification
  • Fault Diagnosis Method for Analog Circuits Based on Echo State Network Dynamic Classification
  • Fault Diagnosis Method for Analog Circuits Based on Echo State Network Dynamic Classification

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0034] Specific implementation mode 1. Combination figure 1 Illustrating this specific embodiment, the method for diagnosing faults in an analog circuit based on dynamic classification of echo state networks described in this embodiment is implemented by the following steps:

[0035] Step 1, using the unit pulse signal to stimulate the analog circuit to work, to obtain the circuit response signal to be diagnosed; collecting the unit pulse response output signal of the analog circuit, and using the unit pulse response output signal as a fault data sample;

[0036] Step 2: Input the fault data samples obtained in step 1 into the echo state network for training, and establish an analog circuit fault diagnosis model according to the training results;

[0037] Step 3: Use the circuit to-be-diagnosed response signal obtained in Step 1 as fault data, and input it into the analog circuit fault diagnosis model established in Step 2 to obtain and output the fault diagnosis result.

[0...

specific Embodiment approach 2

[0039] Embodiment 2. This embodiment further defines step 2 in the method for diagnosing faults in analog circuits based on echo state network dynamic classification described in Embodiment 1. In step 2, the fault obtained in step 1 is The specific method for inputting data samples into the echo state network for training is as follows:

[0040] Step A, setting parameters, the parameters include the number of processing units in the reserve pool, the radius of the internal connection weight spectrum, the input scaling scale and the sparsity degree of the reserve pool;

[0041] Step B. Initialize the echo state network and input the connection weight matrix W in and the internal connection weight matrix W;

[0042] Step C, input the fault data sample into the initialized echo state network, and collect the state variable and output variable; wherein, for the state variable, only the last state variable of each fault data sample is collected;

[0043] Step D. Solve the output ...

specific Embodiment approach 3

[0045] Embodiment 3. This embodiment further defines step C in the method for diagnosing faults in analog circuits based on dynamic classification of echo state networks described in Embodiment 2. In step C, the state of the echo state network is collected. The specific methods of variables and output variables are: input the collected state variables and output variables into the activation function of the echo state network reserve pool processing unit and the output unit activation function respectively for processing: the activation function adopted by the echo state network reserve pool processing unit is: Hyperbolic tangent function, the activation function used by the output unit is an identity function, and the connection between the input layer and the output layer, the connection between the output layer and the reserve pool, and the connection between the output layer and the output layer are not 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
Login to View More

PUM

No PUM Login to View More

Abstract

An analog circuit fault diagnosis method based on echo state network dynamic classification relates to an analog circuit fault diagnosis method. It solves the problem of low diagnostic accuracy of analog circuit fault diagnosis using traditional neural network. Its method: use the unit pulse signal to excite the analog circuit to work, obtain the response signal of the circuit to be diagnosed; collect the unit pulse response output signal of the analog circuit, and use the unit pulse response output signal as the fault data sample; input the fault data sample to the echo Training is carried out in the state network, and an analog circuit fault diagnosis model is established according to the training results; the obtained circuit to-be-diagnosed response signal is used as fault data, and input into the analog circuit fault diagnosis model, and the fault diagnosis result is obtained and output. The invention is suitable for analog circuit fault diagnosis.

Description

technical field [0001] The invention relates to a fault diagnosis method for an analog circuit. Background technique [0002] In electronic equipment, the analog circuit is the weakest link that is most prone to failure. The fault diagnosis of the analog circuit can improve the maintainability of the electronic equipment. Due to the lack of a good fault model for analog circuits, the complex nonlinear relationship between circuit response and component parameters, and the limitation of the number of measurement points, the research on analog circuit fault diagnosis is not yet mature. In this context, artificial intelligence-based methods are introduced into analog circuit fault diagnosis, which treat analog circuit fault diagnosis as a pattern recognition problem. Because of its good nonlinear mapping ability, self-learning and adaptability, etc., neural network is most commonly used in the intelligent diagnosis method of analog circuit. However, the traditional neural net...

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): G01R31/3163
Inventor 彭宇杨智明郭嘉王建民王少军
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com