Automated Testing Method and System for Dynamically Compensated Simulated Switch Turn-Off Isolation

Abstract

This invention discloses an automated testing method and system for the dynamic compensation of analog switch turn-off isolation, relating to the field of isolation testing technology. It includes information acquisition: acquiring test information on previous switch turn-off isolation, including test switch information, test environment information, interference signal information, and test results. This invention systematically utilizes environmental and interference factors from previous test information to classify and store test data, constructing multiple dedicated data sub-libraries for each category. This enables the isolation testing model to accurately test the target switch information under different test scenarios and interference information. Furthermore, the output not only provides multi-level test results including fuzzy and precise outputs, improving the comprehensiveness and accuracy of the test, but also effectively combines target switch information with fuzzy output technology, enhancing the flexibility and robustness of the test, and improving the automation level and reliability of the test process.

Description

Technical Field

[0001] This invention relates to the field of isolation testing technology, specifically to an automated testing method and system for the isolation of simulated switch turn-off with dynamic compensation. Background Technology

[0002] In switch isolation, a switch refers to a device used to control the connection and disconnection of a circuit. In its off state, it should effectively block the passage of signals or current to ensure system safety and stability. Isolation testing evaluates the switch's ability to prevent signal leakage in the off state, ensuring it meets system requirements for safety and interference suppression. The purpose of isolation testing is to verify whether the switch's turn-off performance meets standards, preventing signal penetration or interference propagation, thereby ensuring normal equipment operation, improving system reliability, and complying with relevant safety regulations.

[0003] A test circuit and platform for accurately testing the isolation of a simulated switch cutoff state, with patent publication number CN119125866A, relates to the field of electronic circuit technology. The test platform specifically includes a sine wave generator circuit, a signal amplification circuit, and a two-stage filter circuit. The platform comprises a sine wave generator circuit, a DUT (Distributed Under Test), a signal amplification circuit, a two-stage filter circuit, an integrated circuit test bench, and an oscilloscope. The addition of the amplification and filtering circuit not only improves test accuracy but also allows for direct testing using the test bench. The addition of a two-stage bandpass filter design processes most of the noise signal at the detection end. The integrated circuit test bench automatically generates the test waveform and automatically controls the switching of different test channels. Simultaneously, through amplification and filtering, the signal can be automatically tested using the integrated circuit test bench, increasing both test accuracy and efficiency.

[0004] Currently, automated testing processes for analog switch isolation based on existing technologies generally rely on fixed test parameters (such as preset interference frequency, amplitude, and duration) and a limited range of interference signal types. This method lacks flexibility and is difficult to cope with the changing environment and interference sources in complex real-world application scenarios. Consequently, the test results cannot truly reflect the isolation performance of the system under different operating conditions. As a result, the testing efficiency is low, the testing time is long, and it is difficult to quickly determine whether the switch isolation performance meets the safety and performance requirements of the engineering design, affecting the reliability assessment and optimization design of the equipment. Therefore, this invention is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide an automated testing method and system for the isolation degree of simulated switch turn-off with dynamic compensation, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an automated testing method for the turn-off isolation of a dynamically compensated analog switch, the method comprising:

[0007] Information Acquisition: Acquire test information on the isolation of previous switches, including test switch information, test environment information, interference signal information, and test results;

[0008] Its features include:

[0009] Information processing: The test information is classified using a classification method to obtain classification results. The classification results include fixed classification items and classification information. The classification information includes several sub-classification information, and the fixed classification items include fixed environment and fixed signal.

[0010] Database establishment: Establish a sub-database for storing sub-classification information, integrate all sub-databases to obtain the database, and use information exploration methods to explore the correlation between different test switch information and interference signal information based on test information to obtain auxiliary information;

[0011] Directory creation: Based on the data sub-database, the isolation test models of different test switches under different test environments are established through the directory creation method and integrated to obtain the directory table. The directory table includes reference information, isolation test models and association relationships.

[0012] Output method determination: Preset output method, including fuzzy output and precise output, obtain user requirements, and obtain the result method based on the output method and user requirements;

[0013] Simulated isolation test: Obtain target switch information, and obtain the target result through test methods based on the target switch information and the result method;

[0014] The testing method includes: when the result is fuzzy output, obtaining a fuzzy output result based on the target switch information using the fuzzy output method, and the fuzzy output result is the test result; when the result is precise output, obtaining target information by selecting the corresponding reference information based on the catalog table and user requirements, obtaining the target model by extracting the isolation degree test model corresponding to the target information based on the association relationship, importing the target switch information into the target model to obtain the simulation result, and integrating the simulation result and the target information to obtain the test result.

[0015] Furthermore, the classification method includes: pre-setting fixed classification items, which include fixed environment and fixed signal; extracting test environment information and interference signal information from the test information based on the fixed classification items to obtain selected test environment and selected interference signal; integrating the selected test environment and selected interference signal to obtain basic items; extracting test switch information and test results corresponding to the basic items from the test information to obtain selected items; establishing a correspondence between basic items and selected items; integrating the correspondence, basic items, and selected items to obtain sub-classification information; and integrating all sub-classification information to obtain classification information.

[0016] Furthermore, the information exploration method includes: ranking the test switch information based on the performance of the test switch information to obtain a first ranking result; setting invariants based on the same test environment information and the same interference signal information; extracting the test switch information from the first ranking result based on the invariants to obtain adjacent switch groups; extracting the test results corresponding to the adjacent switch groups from the test information to obtain the result range; establishing the correlation between the invariants and the adjacent switch groups; and integrating the correlation, adjacent switch groups, result range, and invariants to obtain auxiliary information.

[0017] Furthermore, the directory creation method includes:

[0018] Training data acquisition: Extract information from the data sub-database to obtain basic items, selected items and corresponding relationships. Based on the corresponding relationships, extract the corresponding basic items and selected items to obtain preliminary data. Preprocess the preliminary data to obtain training data.

[0019] Model selection and training: Select a deep learning model as the model base, and import the training data into the model base for the model base to learn and obtain the initial model;

[0020] Model Adjustment and Output: Preset verification information and verification results, import the verification information into the initial model to obtain the result information, compare the result information and the verification results to obtain the comparison results, and adjust and optimize the initial model based on the comparison results to obtain the isolation test model;

[0021] Directory creation: The relationship between the isolation test model and different switches and different test environments is established through the association creation method to obtain the association relationship and reference information. The association relationship, reference information and isolation test model are integrated to obtain the directory table.

[0022] Furthermore, the association establishment method includes: extracting data sub-databases from the database and numbering them to obtain numbering information; acquiring training data based on the numbering information; recording the training order of the isolation test model to obtain order information; establishing a first sub-association between the isolation test model and the data sub-databases based on the order information and the numbering information; extracting basic items from the data sub-databases to obtain reference information; establishing a second sub-association between the reference information and the isolation model based on the first sub-association; and integrating the first and second sub-associations to obtain the association relationship.

[0023] Furthermore, the fuzzy output method includes: splitting auxiliary information to obtain correlation, adjacent switch groups, result range and invariants; extracting adjacent switch groups containing target switch information to obtain target switch groups; extracting invariants corresponding to target switch groups based on correlation to obtain target quantity; extracting result ranges corresponding to target switch groups from auxiliary information to obtain target range; and integrating target quantity and target range to obtain fuzzy output results.

[0024] The automated test system for the dynamic compensation analog switch turn-off isolation degree uses the aforementioned automated test method for the dynamic compensation analog switch turn-off isolation degree.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] This dynamically compensated automated testing method and system for analog switch isolation systematically utilizes environmental and interference factors from previous test information to classify and store test data, constructing multiple dedicated data sub-libraries for each category. This enables the isolation testing model to accurately test the target switch information under different test scenarios and interference information. Furthermore, the system can provide multi-level test results, including fuzzy and precise outputs, improving the comprehensiveness and accuracy of the test. It can also effectively combine target switch information with fuzzy output technology to enhance the flexibility and robustness of the test, thereby improving the automation level of the testing process and the reliability of the results.

[0027] Meanwhile, during the testing process, by importing target switch information and user requirements, test results corresponding to user needs can be simulated. The test results can include isolation results under different environments and different interference signals. An isolation test model can be built using a set directory method to facilitate automatic testing of switch-off isolation. A single isolation test model is built for each data sub-database. The set directory table allows managers or users to select the corresponding isolation test model according to test requirements. By building a single isolation test model for each data sub-database, the efficiency of switch-off isolation testing can be improved, while reducing the efficiency of model training and improving the accuracy of test results.

[0028] Meanwhile, by setting the test method, different test results are output according to different result methods. When the result method is fuzzy output, the isolation result of the range value is output. When the result method is precise output, the precise isolation result is output, so as to improve the efficiency of isolation test and facilitate the rapid determination of the isolation information of target switch information. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the overall process structure of the present invention;

[0030] Figure 2 This is a schematic diagram of the classification method of the present invention;

[0031] Figure 3 This is a schematic diagram of the association establishment method of the present invention;

[0032] Figure 4 This is a schematic diagram of the test method structure of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] The main purpose of automated testing of the isolation of simulated switch turn-off with dynamic compensation is to ensure that the switch has good isolation performance in the off state, so as to ensure the safety and reliability of the system. Through automated testing, the isolation effect of the equipment under various operating conditions can be evaluated efficiently and accurately, potential faults can be detected in time, equipment performance can be optimized, and compliance with industry standards and safety specifications can be ensured, thereby ensuring the stable operation and safety of the power system.

[0035] like Figures 1-4 As shown, the present invention provides a technical solution: an automated testing method for the turn-off isolation of a dynamically compensated analog switch, the method comprising:

[0036] Information Acquisition: Acquire test information on the isolation of previous switches, including test switch information, test environment information, interference signal information, and test results;

[0037] It is important to note that there are multiple test information entries. Specific methods for obtaining this information include extracting it from previous switch-off isolation test logs. This involves extracting the performance information of the test switches recorded in the logs, extracting the test environment information, extracting interference source information, and extracting the isolation results. Further methods can be used depending on the actual usage. Test switch information refers to the specific parameters and characteristics of the switchgear used in historical tests, including: Model and specifications: switch model, manufacturer, rated current and voltage, etc.; Performance parameters: on-resistance, off-voltage, isolation index, response time, etc.; Control status: switch on / off state, control signal type (voltage control, pulse control, etc.); Manufacturing batch and service life: production time, batch information, cumulative usage time, etc.; Test environment information refers to the specific environmental conditions during the test, including temperature and humidity. : Ambient temperature and humidity levels during testing; Power supply conditions: Power supply voltage and its stability or fluctuation; Electromagnetic environment: Surrounding electromagnetic interference levels, shielding measures, etc.; Other environmental factors: Conditions affecting performance, such as vibration, noise, and air pressure; Interference signal information refers to the specific parameters of the interference source or interference signal applied during testing, including: Interference signal type: Electromagnetic interference, radio frequency interference, power supply noise, etc.; Signal frequency: The spectral range of the interference signal (e.g., hundreds of kHz to hundreds of MHz); Amplitude and intensity: The amplitude or power level of the interference signal; Duration and modulation method: The duration and modulation method (amplitude modulation, frequency modulation, etc.) of the interference signal; Refers to the shutdown isolation performance indicators measured during testing, including: Isolation value: Signal isolation effect in decibels (dB); Leakage current: Leakage current or leakage signal strength in the shutdown state; Performance indicators combined with environmental conditions: Specific performance under specific environmental and interference conditions; Other related indicators: Such as switch opening / closing response time, insulation level, etc.

[0038] Its features include:

[0039] Information processing: The test information is classified using a classification method to obtain classification results. The classification results include fixed classification items and classification information. The classification information includes several sub-classification information, and the fixed classification items include fixed environment and fixed signal.

[0040] It is important to note that the classification method is set to define fixed classification items, and the test information is classified according to these fixed classification items to obtain the classification results. The fixed classification items are fixed environment and fixed signal. That is, the test information is classified based on the fixed environment and fixed signal as the reference information. The fixed environment and fixed signal are selected from the test environment information and interference signal information in the test information.

[0041] Database establishment: Establish a sub-database for storing sub-classification information, integrate all sub-databases to obtain the database, and use information exploration methods to explore the correlation between different test switch information and interference signal information based on test information to obtain auxiliary information;

[0042] It is important to note that a data sub-library is established to store sub-classification information generated under a single category, which facilitates subsequent model building. The data sub-library can be categorized and named with fixed items to facilitate the selection of the corresponding data sub-library. The data sub-library is stored by establishing a database, and the set information exploration method is used to explore the correlation between different test switch information and interference signal information, so as to facilitate the subsequent fuzzy output isolation test results.

[0043] Directory creation: Based on the data sub-database, the isolation test models of different test switches under different test environments are established through the directory creation method and integrated to obtain the directory table. The directory table includes reference information, isolation test models and association relationships.

[0044] It is important to note that the directory creation method is used to establish isolation test models to facilitate automatic switching isolation testing. Each data sub-database corresponds to a single isolation test model. The directory table allows administrators or users to select the appropriate isolation test model according to their testing needs. By establishing a single isolation test model for each data sub-database, the efficiency of switching isolation testing can be improved, while reducing the efficiency of model training and increasing the accuracy of test results.

[0045] Output method determination: Preset output method, including fuzzy output and precise output, obtain user requirements, and obtain the result method based on the output method and user requirements;

[0046] It should be noted that by setting the output method, which includes fuzzy output and precise output, isolation test information at different levels can be obtained. In actual use, user requirements are obtained by establishing an information receiving interface, where users send information to the information receiving interface, and the result method is obtained by matching the user requirements with the output method. The result method includes either fuzzy output or precise output.

[0047] Simulated isolation test: Obtain target switch information, and obtain the target result through test methods based on the target switch information and the result method.

[0048] It should be noted that the target switch information can also be used to establish an information receiving interface in actual use. It can be obtained by receiving information sent by users, and by using the set information exploration method to obtain the target result corresponding to the user's needs based on the target switch information and the result method. The target result is the isolation test result of the target switch information.

[0049] like Figure 2 As shown, the classification method includes: pre-setting fixed classification items, which include fixed environment and fixed signal; extracting test environment information and interference signal information from the test information based on the fixed classification items to obtain selected test environment and selected interference signal; integrating the selected test environment and selected interference signal to obtain basic items; extracting test switch information and test results corresponding to the basic items from the test information to obtain selected items; establishing a correspondence between basic items and selected items; integrating the correspondence, basic items, and selected items to obtain sub-classification information; and integrating all sub-classification information to obtain classification information.

[0050] It is important to note that the classification fixed items are determined based on the actual usage. Specifically, the test environment information and interference signal information in the test information are selected as the classification fixed items in turn. Through the set classification method, the corresponding test switch information and test results are extracted from the test information according to the set classification fixed items and integrated to obtain sub-classification information, so as to facilitate the subsequent training of the model.

[0051] The information exploration method includes: ranking the test switch information based on its performance to obtain a first ranking result; setting invariants based on the same test environment information and the same interference signal information; extracting the test switch information from the first ranking result based on the invariants to obtain adjacent switch groups; extracting the test results corresponding to the adjacent switch groups from the test information to obtain the result range; establishing the correlation between the invariants and the adjacent switch groups; and integrating the correlation, adjacent switch groups, result range, and invariants to obtain auxiliary information.

[0052] It is important to note that the performance ranking based on test switch information involves a comprehensive evaluation of different switches in terms of response speed, stability, durability, energy consumption, anti-interference capability, and safety, resulting in a ranking from best to worst. The top-ranked test switch exhibits superior performance across all indicators, possessing characteristics such as fast response, stability and reliability, long lifespan, low energy consumption, strong anti-interference capability, and good safety assurance, making it most suitable for practical applications. The other switches, in turn, perform slightly worse in these indicators, thus achieving a scientific ranking of various switch performances. The specific performance ranking is determined based on actual usage conditions. By setting invariants using the same test environment information and the same interference signal information, several adjacent switch groups with the same test environment information and the same interference signal information can be obtained. Each adjacent switch group consists of two test switch information sets, and the result range is composed of the test results corresponding to the two test switch information sets, i.e., the range value.

[0053] Directory creation methods include:

[0054] Training data acquisition: Extract information from the data sub-database to obtain basic items, selected items and corresponding relationships. Based on the corresponding relationships, extract the corresponding basic items and selected items to obtain preliminary data. Preprocess the preliminary data to obtain training data.

[0055] It is important to note that by using the correspondence, the corresponding basic items and selected items can be extracted to facilitate the acquisition of training data. The preprocessing process includes normalization, missing value imputation, outlier removal, and feature engineering (such as PCA dimensionality reduction).

[0056] Model selection and training: Select a deep learning model as the model base, and import the training data into the model base for the model base to learn and obtain the initial model;

[0057] It is important to note that when selecting a deep learning framework, such as a convolutional neural network, based on the task type, the training data can be imported into the model matrix in batches according to actual usage needs to obtain the initial model.

[0058] Model Adjustment and Output: Preset verification information and verification results, import the verification information into the initial model to obtain the result information, compare the result information and the verification results to obtain the comparison results, and adjust and optimize the initial model based on the comparison results to obtain the isolation test model;

[0059] It is important to note that the verification information is the specific switch information, and the verification result is the isolation test result information that is consistent with the test result. The model is adjusted according to the difference analysis (such as confusion matrix): underfitting: increase network complexity or training epochs; overfitting: introduce dropout layer or data augmentation, and finally output the isolation test model that passes the verification.

[0060] Directory creation: The relationship between the isolation test model and different switches and different test environments is established through the association creation method to obtain the association relationship and reference information. The association relationship, reference information and isolation test model are integrated to obtain the directory table.

[0061] It is important to note that the relationship between the isolation test model and different switches and test environments is established through the configured association method. This makes it easier to find the corresponding isolation test model according to the usage requirements later. By creating a directory table, it is convenient to find the isolation test model.

[0062] like Figure 3As shown, the association establishment method includes: extracting data sub-databases from the database and numbering them to obtain numbering information; acquiring training data based on the numbering information; recording the training order of the isolation test model to obtain order information; establishing the first sub-association relationship between the isolation test model and the data sub-database based on the order information and the numbering information; extracting basic items from the data sub-database to obtain reference information; establishing the second sub-association relationship between the reference information and the isolation model based on the first sub-association relationship; and integrating the first and second sub-association relationships to obtain the association relationship.

[0063] It's important to note the following processes: First, extracting and numbering sub-databases from the database to obtain identification information. Second, recording the training order of the isolation test models to obtain sequence information. Third, establishing the first sub-association between the isolation test models and the sub-databases based on the sequence information and identification information. This involves selecting the isolation test models and sub-databases corresponding to the sequence and identification information, establishing a connection between them, and then using this first sub-association with reference information to establish a second sub-association. Establishing these associations facilitates the selection of appropriate isolation test models based on actual usage requirements. Specifically, the corresponding identification information and sequence information are as follows: Figure 3 As shown.

[0064] like Figure 4 As shown, the test method includes: when the result is fuzzy output, a fuzzy output result is obtained based on the target switch information using the fuzzy output method, and the fuzzy output result is the test result; when the result is precise output, the target information is obtained by selecting the corresponding reference information based on the catalog table and user requirements, the target model is obtained by extracting the isolation test model corresponding to the target information based on the association relationship, the target switch information is imported into the target model to obtain the simulation result, and the simulation result and the target information are integrated to obtain the test result.

[0065] It is important to note that different test results are output depending on the set test method and the result method. When the result method is fuzzy output, the isolation result of the range value is output; when the result method is precise output, the precise isolation result is output. This improves the efficiency of isolation testing and facilitates the rapid determination of the isolation information of the target switch. The process of obtaining the target information by selecting the corresponding reference information based on the catalog table and user needs is the information matching process. The process of obtaining the target information by selecting the required reference information according to user needs, extracting the isolation test model corresponding to the target information based on the correlation relationship, obtaining the target model, and importing the target switch information into the target model to obtain the simulation result can obtain the isolation result under the target information. This allows for obtaining isolation test results under different environmental information and interference signals to adapt to different actual situations. In actual use, the user's feedback address information can be obtained, including but not limited to mobile phone number, email, and IP address, and the target result can be sent to the user through the feedback address information.

[0066] The fuzzy output method includes: splitting auxiliary information to obtain correlation, adjacent switch groups, result range and invariants; extracting adjacent switch groups containing target switch information to obtain target switch groups; extracting invariants corresponding to target switch groups based on correlation to obtain target quantities; extracting result ranges corresponding to target switch groups from auxiliary information to obtain target ranges; and integrating target quantities and target ranges to obtain fuzzy output results.

[0067] It is important to note that the process of extracting adjacent switch groups containing target switch information to obtain the target switch group involves information matching, selecting adjacent switch groups where the target switch information is located, and extracting the invariants corresponding to the target switch group based on correlation to obtain the target quantity. This process involves extracting environmental information and interference signal information of the target range, so as to understand the environment and interference sources under which the target range was generated when obtaining fuzzy output results, in order to further understand the test results.

[0068] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended embodiments and their equivalents.

Claims

1. An automated testing method for the turn-off isolation of a dynamically compensated analog switch, the method comprising: Information Acquisition: Acquire test information on the isolation of previous switches, including test switch information, test environment information, interference signal information, and test results; The method is characterized in that it further includes: Information processing: The test information is classified using a classification method to obtain classification results. The classification results include fixed classification items and classification information. The classification information includes several sub-classification information, and the fixed classification items include fixed environment and fixed signal. Database establishment: Establish a sub-database for storing sub-classification information, integrate all sub-databases to obtain the database, and use information exploration methods to explore the correlation between different test switch information and interference signal information based on test information to obtain auxiliary information; Directory creation: Based on the data sub-database, the isolation test models of different test switches under different test environments are established through the directory creation method and integrated to obtain the directory table. The directory table includes reference information, isolation test models and association relationships. Output method determination: Preset output method, including fuzzy output and precise output, obtain user requirements, and obtain the result method based on the output method and user requirements; Simulated isolation test: Obtain target switch information, and obtain the target result through test methods based on the target switch information and the result method; The testing method includes: when the result is fuzzy output, obtaining a fuzzy output result based on the target switch information using the fuzzy output method, and the fuzzy output result is the test result; when the result is precise output, obtaining target information by selecting the corresponding reference information based on the catalog table and user requirements, obtaining the target model by extracting the isolation degree test model corresponding to the target information based on the association relationship, obtaining the target model by importing the target switch information into the target model, obtaining the simulation result, and integrating the simulation result and the target information to obtain the test result. The fuzzy output method includes: splitting auxiliary information to obtain correlation, adjacent switch groups, result range and invariants; extracting adjacent switch groups containing target switch information to obtain target switch groups; extracting invariants corresponding to target switch groups based on correlation to obtain target quantities; extracting result ranges corresponding to target switch groups from auxiliary information to obtain target ranges; and integrating target quantities and target ranges to obtain fuzzy output results.

2. The automated testing method for the dynamic compensation analog switch turn-off isolation degree according to claim 1, characterized in that: The classification method includes: pre-setting fixed classification items, which include fixed environment and fixed signal; extracting test environment information and interference signal information from test information based on the fixed classification items to obtain selected test environment and selected interference signal; integrating the selected test environment and selected interference signal to obtain basic items; extracting test switch information and test results corresponding to the basic items from the test information to obtain selected items; establishing a correspondence between basic items and selected items; integrating the correspondence, basic items, and selected items to obtain sub-classification information; and integrating all sub-classification information to obtain classification information.

3. The automated testing method for the dynamic compensation analog switch turn-off isolation degree according to claim 1, characterized in that: The information exploration method includes: ranking the test switch information based on the performance of the test switch information to obtain a first ranking result; setting invariants based on the same test environment information and the same interference signal information; extracting the test switch information from the first ranking result based on the invariants to obtain adjacent switch groups; extracting the test results corresponding to the adjacent switch groups from the test information to obtain the result range; establishing the correlation between the invariants and the adjacent switch groups; and integrating the correlation, adjacent switch groups, result range, and invariants to obtain auxiliary information.

4. The automated testing method for the dynamic compensation analog switch turn-off isolation degree according to claim 1, characterized in that: The directory creation method includes: Training data acquisition: Extract information from the data sub-database to obtain basic items, selected items and corresponding relationships. Based on the corresponding relationships, extract the corresponding basic items and selected items to obtain preliminary data. Preprocess the preliminary data to obtain training data. Model selection and training: Select a deep learning model as the model base, and import the training data into the model base for the model base to learn and obtain the initial model; Model Adjustment and Output: Preset verification information and verification results, import the verification information into the initial model to obtain the result information, compare the result information and the verification results to obtain the comparison results, and adjust and optimize the initial model based on the comparison results to obtain the isolation test model; Directory creation: The relationship between the isolation test model and different switches and different test environments is established through the association creation method to obtain the association relationship and reference information. The association relationship, reference information and isolation test model are integrated to obtain the directory table.

5. The automated testing method for the dynamic compensation analog switch turn-off isolation degree according to claim 4, characterized in that: The association establishment method includes: extracting data sub-databases from the database and numbering them to obtain numbering information; acquiring training data based on the numbering information; recording the training order of the isolation test model to obtain order information; establishing a first sub-association between the isolation test model and the data sub-databases based on the order information and the numbering information; extracting basic items from the data sub-databases to obtain reference information; establishing a second sub-association between the reference information and the isolation model based on the first sub-association; and integrating the first and second sub-association to obtain the association relationship.

6. An automated testing system for the simulated switch turn-off isolation degree with dynamic compensation, characterized in that: An automated test method for analog switch turn-off isolation with dynamic compensation as described in any one of claims 1-5 was used.

Images