Electronic product health status assessment method based on component degradation data

Abstract

The invention discloses an electronic product health state evaluation method based on component degradation data, which comprises the following steps of: referring to an electronic product function principle and circuit composition, and establishing a corresponding simulation model in Simulink; obtaining the performance degradation of the electronic product model by injecting the degradation parameters of the component; health degree indexes are constructed according to electronic product performance degradation, a time sequence prediction model is developed based on a deep learning method, health degree prediction is achieved, and health state evaluation is completed. According to the electronic product health state evaluation method based on the component degradation data, an electronic product digital simulation model is established, component-level degradation injection is carried out, and single machine performance degradation data is obtained; the product health degree is predicted in combination with a deep learning mode, and the problem of residual life prediction of existing spaceflight electronic products is solved.

Description

Technical Field

[0001] This invention relates to the field of lifespan prediction and health management, and in particular to a method for assessing the health status of electronic products based on component degradation data. Background Technology

[0002] Whether in service or in storage, aerospace electronic products are subject to continuous environmental stresses such as temperature, humidity, and mechanical stress, which can lead to performance changes or even failure, ultimately affecting the safety and reliability of the entire system. Therefore, conducting relevant reliability or life tests on aerospace electronic products is necessary. However, due to limitations in prototype cost, the number of prototypes available for testing is limited, resulting in insufficient data sample size and making effective reliability or life assessments difficult.

[0003] With the development of digital prototyping technology, establishing digital models of electronic products to obtain simulation data under corresponding requirements has become an effective way to solve the problem of small print. However, how to evaluate the consistency between the digital model of electronic products and the actual system has become the primary problem to be solved. Summary of the Invention

[0004] The purpose of this invention is to provide a method for assessing the health status of electronic products based on component degradation data. This method uses a digital model to simulate component degradation injection, obtains single-unit performance degradation data of electronic products, and combines deep learning to predict product health, thus solving the problem of predicting the remaining lifespan of existing aerospace electronic products.

[0005] To achieve the above objectives, the present invention provides a method for assessing the health status of electronic products based on component degradation data, comprising the following steps: S1. Establish an electronic product simulation model based on the Simulink platform; S2. Write MATLAB scripts to perform component-level degradation injection; S3. Predict health status based on deep learning methods.

[0006] Preferably, in step S1, the electronic product simulation model is built in Simulink with reference to the functional principles and circuit composition of the electronic product.

[0007] Preferably, in step S1, the electronic components included in the simulation model should include, but are not limited to, resistors, capacitors, inductors, diodes, transistors, MOSFETs, IGBTs, relays, optocouplers, integrated circuits, etc.

[0008] Preferably, in step S2, the component-level degradation injection process is as follows: S21. Modeling of raw degradation data of components; S22. Construct a component degradation database; S23, Batch injection simulation of component degradation data.

[0009] Preferably, in step S21, the degradation data modeling should be performed by constructing linear functions, logarithmic functions, exponential functions, and power functions, respectively, and then using the least squares method to fit them, and the optimal fitting function should be selected based on the goodness of fit.

[0010] Preferably, in step S22, the component degradation database should include component name, ID, time, degradation parameters, reference value, and degradation injection path.

[0011] Preferably, in step S23, the batch injection simulation process is as follows: S231. Write MATLAB scripts for batch injection and reading MATLAB scripts for electronic product performance degradation data. S232. Construct a database of electronic product performance degradation.

[0012] Preferably, in step S231, the batch injection MATLAB script uses the component ID row vector as input to inject specific component degradation data according to its degradation injection path, and the electronic product performance degradation data reading MATLAB script is used to obtain various performance degradation data of electronic products.

[0013] Preferably, in step S232, the electronic product performance degradation database should include the type of injected component, the name of the injected component, the time, the single-unit performance degradation parameters, and the reference value.

[0014] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0015] Preferably, in step S3, the health status is calculated by weighting and normalizing the various performance degradation parameters of a single machine at each sampling time point.

[0016] Preferably, in step S3, the deep learning method constructs a time-series prediction model based on LSTM, defines the input window size and network structure, inputs health data, and outputs future health. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for assessing the health status of electronic products based on component degradation data, according to the present invention. Figure 2 This is a schematic diagram of a power distribution unit performance model, representing an embodiment of a method for assessing the health status of electronic products based on component degradation data according to the present invention. Figure 3This is an embodiment of the present invention, which is a method for injecting component degradation data based on MATLAB scripts, into an electronic product health status assessment method based on component degradation data. Figure 4 This is a schematic diagram of the health prediction result of an embodiment of the electronic product health status assessment method based on component degradation data according to the present invention; Detailed Implementation

[0018] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0020] A method for assessing the health status of electronic products based on component degradation data is proposed. This method uses MATLAB functions to define model parameters and integrates Simulink functional modules to build a performance model. MATLAB scripts are written to implement the component degradation database and degradation injection simulation. Deep learning methods are combined to predict product health, solving the problem of predicting the remaining lifespan of existing aerospace electronic products. The method includes the following steps: S1, such as Figure 2 As shown, for reference Figure 1 The power distribution unit's structural components are described, and its corresponding performance model is built in Simulink.

[0021] S2, such as Figure 3 As shown, a MATLAB script was written to perform component-level degradation injection.

[0022] S3, such as Figure 4 As shown, a MATLAB script is written to construct a single-machine health index, train a time-series prediction model, and complete the single-machine health prediction.

[0023] Therefore, this invention employs the aforementioned method for digitizing electronic products to establish a power distribution performance simulation model in Simulink. Simultaneously, this invention also performs component degradation injection for individual power distribution units, combining deep learning to predict product health, thus solving the problem of predicting the remaining lifespan of existing aerospace electronic products.

[0024] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for assessing the health status of electronic products based on component degradation data, characterized in that, Includes the following steps: S1. Establish an electronic product simulation model based on the Simulink platform; S2. Write MATLAB scripts to perform component-level degradation injection; S3. Predict health status based on deep learning methods.

2. The method for assessing the health status of electronic products based on component degradation data according to claim 1, characterized in that, In step S1, the electronic product simulation model is built in Simulink with reference to the functional principles and circuit composition of the electronic product.

3. The method for assessing the health status of electronic products based on component degradation data according to claim 1, characterized in that, In step S1, the electronic components included in the simulation model should include, but are not limited to, resistors, capacitors, inductors, diodes, transistors, MOSFETs, IGBTs, relays, optocouplers, integrated circuits, etc.

4. The method for assessing the health status of electronic products based on component degradation data according to claim 1, characterized in that, In step S2, the component-level degradation injection is performed, and the specific process is as follows: S21. Modeling of raw degradation data of components; S22. Construct a component degradation database; S23, Batch injection simulation of component degradation data.

5. The method for assessing the health status of electronic products based on component degradation data according to claim 4, characterized in that, In step S21, the degradation data modeling should be performed by constructing linear functions, logarithmic functions, exponential functions, and power functions, respectively, and then using the least squares method to fit them, and the optimal fitting function should be selected based on the goodness of fit.

6. The method for assessing the health status of electronic products based on component degradation data according to claim 4, characterized in that, In step S22, the component degradation database should include component name, ID, time, degradation parameters, reference value, and degradation injection path.

7. The method for assessing the health status of electronic products based on component degradation data according to claim 4, characterized in that, In step S23, the batch injection simulation process is as follows: S231. Write MATLAB scripts for batch injection and reading MATLAB scripts for electronic product performance degradation data. S232. Construct a database of electronic product performance degradation.

8. The method for assessing the health status of electronic products based on component degradation data according to claim 7, characterized in that, In step S231, the batch injection MATLAB script takes the component ID row vector as input and injects the degradation data of specific components according to their degradation injection path. The electronic product performance degradation data reading MATLAB script is used to obtain the various performance degradation data of electronic products.

9. The method for assessing the health status of electronic products based on component degradation data according to claim 7, characterized in that, In step S232, the electronic product performance degradation database should include the type of injected component, the name of the injected component, the time, the single-unit performance degradation parameters, and the reference value.

10. The method for assessing the health status of electronic products based on component degradation data according to claim 1, characterized in that, In step S3, the deep learning method is described in the following specific process: A health index is constructed, and a weighted average and normalized is performed on each performance degradation parameter of a single machine at each sampling time point. A time-series prediction model is built based on deep learning, defining the input window size and network structure, taking health data as input, and outputting future health status.

Images