Automatic test method and system based on test tool, terminal and storage medium

Abstract

The invention relates to an automatic test method and system based on a test tool, a terminal and a storage medium, and relates to the technical field of circuit board test.The method comprises the steps that a test circuit board is loaded into the test tool through a mechanical arm in response to the situation that the test circuit board meets test conditions; using the test model to perform conventional performance test on the test circuit board, and collecting test data of the test circuit board in real time; calculating a first data difference value between the test data and the standard data; according to a comparison result of the first data difference value and a first preset threshold value, a test result of the test circuit board is generated, and the test result represents the performance standard reaching condition of the test circuit board; and placing the test circuit board in the storage area corresponding to the test result by using the manipulator. The circuit board testing method has the effects of improving the testing efficiency and reducing errors in circuit board testing.

Description

Technical Field

[0001] The present application relates to the technical field of circuit board testing, and in particular, to an automatic testing method, system, terminal, and storage medium based on a testing fixture. Background Art

[0002] In the field of automated testing, with the development of urban rail transit, in order to ensure the normal operation of urban rail transit, rapid and accurate testing of a large number of circuit boards has become necessary.

[0003] In the related art when testing a circuit board, a worker needs to first place the circuit board on a testing fixture, and then the testing fixture will automatically test the circuit board, and the circuit board will automatically output the test result of the circuit board. After that, the worker needs to place the circuit board in the corresponding area according to the test result.

[0004] Regarding the above related art, the inventor believes that the testing efficiency is relatively low, and when comprehensively testing the functions of a complex circuit board, errors are likely to be introduced due to manual participation. Summary of the Invention

[0005] In order to improve the testing efficiency and reduce errors in circuit board testing, the present application provides an automatic testing method, system, terminal, and storage medium based on a testing fixture.

[0006] In a first aspect, the present application provides an automatic testing method based on a testing fixture, adopting the following technical solution: An automatic testing method based on a testing fixture includes: In response to the test circuit board meeting the test conditions, using a manipulator to load the test circuit board into the testing fixture; Using a test model to perform a conventional performance test on the test circuit board and collect the test data of the test circuit board in real time; Calculating a first data difference between the test data and the standard data; Generating a test result of the test circuit board according to a comparison result between the first data difference and a first preset threshold, where the test result indicates the performance compliance situation of the test circuit board; Using the manipulator to place the test circuit board into a storage area corresponding to the test result.

[0007] By adopting the above technical solution, full-automatic testing of the test circuit board can be realized, the testing efficiency is improved, and the batch testing requirements of high speed and high precision are met. The whole process does not require manual participation, and errors in circuit board testing caused by humans can be reduced.

[0008] Optionally, a prediction model is used to predict the risk of the test data, and the predicted test risk of the test circuit board is obtained; According to the predicted test risk, the test process of the conventional performance test is adjusted; Collect the adjusted data generated by the adjusted test process; Using a loss function, perform a loss operation on the adjusted data and the test data to obtain a loss value; Using the gradient descent method, update the weight matrix of the prediction model to minimize the loss value.

[0009] By adopting the above technical solution, a prediction model is used to predict potential anomalies in the test process, the test process is adjusted in advance, and the robustness of the system is enhanced. And it can continuously evaluate the model performance, regularly update the training set, and optimize the model performance.

[0010] Optionally, calculate the second data difference between the adjusted data and the standard data; In the case where the second data difference is greater than a second preset threshold, according to the adjusted data, determine the problem points on the test circuit board; Perform a special performance test on the problem points to obtain special data, where the special performance test is related to the problem position of the problem points on the test circuit board and the problem components corresponding to the problem points; Generate a special test result according to the special data and the standard data.

[0011] By adopting the above technical solution, the problem points on the test circuit board can be determined, a special performance test is performed on the problem points, and a special test result is generated, thereby clarifying the problems existing in the test circuit board.

[0012] Optionally, in the case where the special test result is a test failure, extract the problem position of the problem points on the test circuit board and the problem components corresponding to the problem points; Determine the candidate performance tests related to the problem position and the problem components; Perform a feature extraction operation on the candidate performance tests to obtain candidate performance test features; Perform a feature extraction operation on the special performance test to obtain special performance test features; Calculate the vector length from the candidate performance test features to the special performance test features; Select the n candidate performance tests corresponding to the smallest vector lengths as the target candidate performance tests; Add the target candidate performance tests to the conventional performance test.

[0013] By adopting the above technical solution, the target candidate performance test can be added to the regular performance test, making the regular performance test more comprehensive, enabling a more complete test of the circuit board, and ensuring that potential anomalies during the test process can be discovered.

[0014] Optionally, detect whether there are test items in the regular performance test that are mutually exclusive with the target candidate performance test; If so, cancel adding the target candidate performance test to the regular performance test; If not, calculate the influence factor of the target candidate performance test on each test item in the regular performance test; Determine the target sorting of the target candidate performance test in the regular performance test according to the sorting of the influence factors; Add the target candidate performance test to the regular performance test according to the target sorting.

[0015] By adopting the above technical solution, adding the target candidate performance test to the regular performance test ensures that the regular performance test itself will not be affected, makes the regular performance test more reasonable, effectively improves the test speed, and shortens the test cycle.

[0016] Optionally, monitor the current state data of the test fixture; Obtain other state data of other test fixtures of the same type as the test fixture, and obtain the historical state data of the test fixture; Calculate the mean value of the other state data to obtain the mean value of the other state data; Calculate the difference between the current state data and the mean value of the other state data to obtain the first difference; Calculate the mean value of the historical state data to obtain the mean value of the historical state data; Calculate the difference between the current state data and the mean value of the historical state data to obtain the second difference; In the case where the first difference is greater than the first difference threshold and the second difference is greater than the second difference threshold, perform a maintenance operation on the test fixture.

[0017] By adopting the above technical solution, the state of the test fixture is monitored in real time, preventive maintenance of the test fixture is realized, the failure rate is reduced, the service life of the test fixture is extended, and the maintenance cost is saved.

[0018] Optionally, determine the potential failure location of the test fixture according to the current state data; Read the part data corresponding to the potential failure location; Identify the actual working state of the potential failure location through the part data; In the case where the actual working state is abnormal, retrieve the target maintenance operation using the part data in the historical maintenance record; Perform the target maintenance operation on the potential fault part.

[0019] By adopting the above technical solution, the maintenance operation is performed on the target fault part of the test tooling, enabling the target fault part to maintain a healthy state and extending the service life of the entire test tooling.

[0020] In a second aspect, the present application provides an automatic test system based on a test tooling, adopting the following technical solution: An automatic test system based on a test tooling, comprising: An acquisition module, configured to acquire test conditions, a test model, standard data, a first preset threshold, a prediction model, a loss function, a second preset threshold, special data, current state data, other state data, and historical state data; A memory, configured to store the program of the automatic test method based on the test tooling as described in any one of the above; A processor, the program in the memory can be loaded and executed by the processor and implement the automatic test method based on the test tooling as described in any one of the above.

[0021] By adopting the above technical solution, full-automatic testing of the test circuit board can be realized, the testing efficiency is improved, and the batch testing requirements of high speed and high precision are met. The whole process does not require manual participation, and errors in the circuit board testing caused by humans can be reduced.

[0022] In a third aspect, the present application provides an intelligent terminal, adopting the following technical solution: An intelligent terminal, comprising a memory and a processor, and a computer program capable of being loaded and executed by the processor is stored on the memory.

[0023] In a fourth aspect, the present application provides a computer storage medium, which can store the corresponding program and has the characteristics of facilitating the improvement of the testing efficiency and reducing errors in the circuit board testing, adopting the following technical solution: A computer-readable storage medium stores a computer program capable of being loaded and executed by the processor for any one of the above automatic test methods based on the test tooling.

[0024] In summary, the present application includes at least one of the following beneficial technical effects: Full-automatic testing of the test circuit board can be realized, the testing efficiency is improved, and the batch testing requirements of high speed and high precision are met. The whole process does not require manual participation, and errors in the circuit board testing caused by humans can be reduced; Predict potential anomalies during the test process using a prediction model, adjust the test process in advance, and enhance the robustness of the system. Continuously evaluate the model performance, update the training set regularly, and optimize the model performance; Monitor the status of the test tooling in real time, implement preventive maintenance of the test tooling, reduce the failure rate, extend the service life of the test tooling, and save maintenance costs. Description of the Drawings

[0025] Figure 1 It is a schematic flowchart of an automatic test method based on test tooling provided by an embodiment of the present application.

[0026] Figure 2 It is a schematic flowchart of a prediction method for circuit board testing provided by an embodiment of the present application.

[0027] Figure 3 It is a schematic flowchart of a special test method for testing a circuit board provided by an embodiment of the present application.

[0028] Figure 4 It is a schematic flowchart of a method for adding a performance test provided by an embodiment of the present application.

[0029] Figure 5 It is a schematic flowchart of a method for adding a performance test provided by an embodiment of the present application.

[0030] Figure 6 It is a schematic flowchart of a method for maintaining a test tooling provided by an embodiment of the present application.

[0031] Figure 7 It is a schematic flowchart of a method for maintaining a test tooling provided by an embodiment of the present application.

[0032] Figure 8 It is a schematic structural diagram of an automatic test system based on test tooling provided by an embodiment of the present application. Detailed Embodiments

[0033] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the following further details the present application in conjunction with the attached Figures 1 to 8 drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0034] An embodiment of the present application discloses an automatic test method based on test tooling. Referring to Figure 1 , the method includes: Step S101: In response to the test circuit board meeting the test conditions, use a manipulator to load the test circuit board into the test tooling.

[0035] In the embodiments of the present application, the test circuit board is any type of circuit board applied to urban rail transit.

[0036] The test fixture is a device or tool for testing the test circuit board. The test fixture can provide stable physical support and accurate electrical connection for the test circuit board. Further, several types of sensors are provided on the test fixture. For example, a camera, a voltmeter, an ammeter, a pressure sensor, a distance sensor, etc. are provided on the test fixture.

[0037] Optionally, a physical image of the test circuit board is captured by a camera. The unique identifier on the test circuit board in the physical image is identified. When the unique identifier matches the standard test configuration stored in the database, it is confirmed that the test circuit board meets the test conditions. Further, after obtaining the physical image, the spatial position of the test circuit board is determined through the physical image. Combining the spatial position, the test circuit board is loaded into the test fixture by a manipulator.

[0038] Optionally, the unique identifier on the circuit board is identified by optical recognition technology. When the unique identifier matches the standard test configuration stored in the database, it is confirmed that the test circuit board meets the test conditions.

[0039] Exemplarily, the position information of the test circuit board is monitored in real time through an encoder to ensure the position accuracy of the test circuit board, so that the test circuit board can be accurately placed into the test fixture.

[0040] Step S102: Use the test model to perform a conventional performance test on the test circuit board and collect the test data of the test circuit board in real time.

[0041] The conventional performance test includes at least one of, but is not limited to, electrical performance test, automatic optical inspection, oscilloscope inspection, function test, impedance test, reliability test.

[0042] The test data includes at least one of, but is not limited to, electrical indicators, pressure values, signal waveforms, solder joint quality, component positions.

[0043] Exemplarily, the central controller calls the pre-stored test model to automatically adjust the power supply module to simulate various electrical characteristics of the circuit board in the real working environment, such as key indicators such as power supply ripple rejection ratio and transient response time. The test model can be a test script.

[0044] Step S103: Calculate the first data difference between the test data and the standard data.

[0045] The standard data is the data of the circuit board when it meets the design standards.

[0046] Step S104: Generate the test result of the test circuit board according to the comparison result between the first data difference and the first preset threshold, where the test result indicates the performance compliance of the test circuit board.

[0047] Optionally, when the comparison result between the first data difference and the first preset threshold is greater than the preset data threshold, generate a first test result, where the first test result indicates that the performance of the test circuit board does not meet the standard.

[0048] Optionally, when the comparison result between the first data difference and the first preset threshold is less than the preset data threshold, generate a second test result, where the second test result indicates that the performance of the test circuit board meets the standard.

[0049] Further, after obtaining the test result, store the basic information and the test result of the test circuit board in a database. The database can be a cloud server or a local database. The basic information includes at least one of the type, model, batch number, and production line number of the test circuit board.

[0050] Step S105: Use a manipulator to place the test circuit board in the storage area corresponding to the test result.

[0051] Optionally, when the test result is the first test result, use a manipulator to place the test circuit board in the scrap storage area.

[0052] Optionally, when the test result is the second test result, use a manipulator to place the test circuit board in the qualified product storage area.

[0053] Further, use a manipulator to prepare the next circuit board to form a continuous cyclic operation to ensure the testing rate of the circuit board.

[0054] By adopting the above technical solution, the full-automatic testing of the test circuit board can be realized, the testing efficiency is improved, and the batch testing requirements of high speed and high precision are met. The whole process does not require manual participation, and the errors in the circuit board testing caused by humans can be reduced.

[0055] In the following embodiments, to solve the problem of ensuring the consistency and reliability of the circuit board testing, a prediction model is used to intelligently predict potential anomalies during the testing process, adjust the testing process in advance, and enhance the robustness of the system. Therefore, an embodiment of the present application discloses a prediction method for circuit board testing. Refer to Figure 2 , the method includes: Step S201: Use a prediction model to perform risk prediction on the test data to obtain the predicted test risk of the test circuit board.

[0056] The prediction model uses any one of a convolutional neural network, a recurrent neural network, a long short-term memory network, a Transformer network, and an autoencoder.

[0057] Optionally, the prediction model includes a convolutional layer to capture local features, a fully connected layer to integrate global information, and a Softmax (normalization function) layer to classify abnormal categories. Further, the prediction model is trained based on a past test data set, and the past test data set includes samples of several failure modes.

[0058] The predicted test risk is used to represent the test risk with a probability greater than a preset probability during the testing of a test circuit board. Exemplarily, the predicted test risk includes poor contact in the test circuit board, short circuit in the test circuit board line, printing error in the test circuit board, etc.

[0059] Step S202: Adjust the test process of the conventional performance test according to the predicted test risk.

[0060] Optionally, adjusting the test process of the conventional performance test includes adding or subtracting additional test items, adjusting test parameters, changing test components, etc.

[0061] Step S203: Collect the adjusted data generated by the adjusted test process.

[0062] The types of the adjusted data are the same as those of the test data, but the adjusted data is generated after the test process is adjusted, and the test data is generated before the test process is adjusted.

[0063] Step S204: Use a loss function to perform a loss operation on the adjusted data and the test data to obtain a loss value.

[0064] The loss function uses any one of cross-entropy loss, KL divergence, and Hinge loss.

[0065] Step S205: Use the gradient descent method to update the weight matrix of the prediction model to minimize the loss value.

[0066] The gradient descent method is an optimization algorithm used to train machine learning models, especially for parameter optimization in neural networks. Its goal is to continuously adjust the parameters to minimize the loss function, thereby improving the performance of the model. The core idea of the gradient descent method is to descend along the gradient of the loss function (i.e., the derivative of the loss function with respect to the parameters) to find the minimum value or optimal solution of the loss function.

[0067] By adopting the above technical solution, potential anomalies during the test are predicted using the prediction model, the test process is adjusted in advance, and the robustness of the system is enhanced. And it can continuously evaluate the model effectiveness, regularly update the training set, and optimize the model performance.

[0068] In the following embodiments, when it is determined that there is a problem with the test circuit board, a special test can be performed on the test circuit board to clarify the problem existing in the test circuit board. An embodiment of the present application discloses a special test method for a test circuit board. Refer to Figure 3 , the method includes: Step S301: Calculate the second data difference between the adjustment data and the standard data.

[0069] The second data difference is used to represent the gap between the adjustment data and the standard data.

[0070] Step S302: When the second data difference is greater than the second preset threshold, determine the problem point on the test circuit board according to the adjustment data.

[0071] The second preset threshold is a preset empirical value. Relevant personnel can adjust the value of the second preset threshold according to actual needs.

[0072] The problem point is the position on the test circuit board where the failure probability is greater than the preset probability. The problem point is a component, a connection structure or a printed board on the test circuit board. Exemplarily, determine the target adjustment data corresponding to the second data difference greater than the second preset threshold. Identify the source of the target adjustment data to obtain the source component. Determine the source component as the problem point.

[0073] Step S303: Perform a special performance test on the problem point to obtain special data. The special performance test is related to the problem position of the problem point on the test circuit board and the problem component corresponding to the problem point.

[0074] Optionally, according to the process of the special performance test, use a manipulator to test the problem point and collect data during the test through a test fixture to obtain special data.

[0075] Optionally, retrieve the special performance test in the preset database according to the problem position and the problem component, where the preset database is used to store the correspondence between the problem position, the problem component and the special performance test.

[0076] In some other embodiments, a test matching model can also be used to perform performance test matching on the problem position and the problem component to obtain the special performance test.

[0077] Step S304: Generate a special test result according to the special data and the standard data.

[0078] Optionally, when the difference between the special data and the standard data is greater than the preset data threshold, generate a first special test result, and the first special test result indicates that the problem point fails the special performance test.

[0079] Optionally, when the difference between the special data and the standard data is less than a preset data threshold, a second special test result is generated, and the second special test result indicates that the problem point passes the special performance test.

[0080] Further, after obtaining the special test result, the problem point, the basic information of the test circuit board, and the special test result are stored in a database. The database can be a cloud server or a local database. The basic information includes at least one of the type, model, batch number, and pipeline number of the test circuit board.

[0081] By adopting the above technical solution, the problem point on the test circuit board can be determined, and a special performance test is carried out for the problem point to generate a special test result, thereby clarifying the problems existing in the test circuit board.

[0082] In the following embodiments, during the process of testing the test circuit board, according to the difference of the test circuit board, additional test items can be added to the conventional test items according to the actual situation, so that the coverage of the conventional test items is more extensive. Therefore, an embodiment of the present application discloses a method for adding performance tests. Refer to Figure 4 , the method includes: Step S401: When the special test result is that the test fails, extract the problem location of the problem point on the test circuit board and the problem component corresponding to the problem point.

[0083] Optionally, when the difference between the special data and the standard data is greater than a preset data threshold, it is considered that the special test result is that the test fails.

[0084] Step S402: Determine the candidate performance tests related to the problem location and the problem component.

[0085] In some embodiments, the candidate performance test is Figure 3 exactly the same as the special performance test in the

[0086] In some embodiments, the candidate performance test has the same type as the special performance test, but the test parameters used in the candidate performance test are different from those in the special performance test. For example, some extreme performance tests may damage the circuit board itself. Therefore, when performing the candidate performance test, the test parameters need to be appropriately reduced. Taking the candidate performance test and the special performance test both being high-voltage tests as an example, if the test parameter is the pressure value, the pressure value used in the candidate performance test is less than that in the special performance test.

[0087] Step S403: Perform a feature extraction operation on the candidate performance test to obtain candidate performance test features.

[0088] Exemplarily, using the feature extraction model, perform a feature extraction operation on the candidate performance test to obtain candidate performance test features. The candidate performance test features are represented in vector form.

[0089] Step S404: Perform a feature extraction operation on the special performance test to obtain special performance test features.

[0090] Exemplarily, using the feature extraction model, perform a feature extraction operation on the special performance test to obtain special performance test features. The special performance test features are represented in vector form.

[0091] Step S405: Calculate the vector length from the candidate performance test features to the special performance test features.

[0092] The vector length can represent the similarity between the candidate performance test features and the special performance test features. The vector length is negatively correlated with the similarity, that is, the smaller the vector length, the greater the similarity between the candidate performance test and the special performance test.

[0093] Optionally, calculate the Euclidean distance from the candidate performance test features to the special performance test features to obtain the vector length.

[0094] Step S406: Select the candidate performance tests corresponding to the n candidate performance test features with the smallest vector lengths as the target candidate performance tests.

[0095] Optionally, n is a preset positive integer. Relevant personnel can adjust the value of n according to actual needs.

[0096] Furthermore, the target candidate performance tests are the candidate performance tests corresponding to the n candidate performance test features with the smallest vector lengths. Therefore, the target candidate performance tests are closest to the special performance tests, and the target candidate performance tests will not affect the test circuit board.

[0097] Step S407: Add the target candidate performance tests to the regular performance tests.

[0098] Adding the target candidate performance tests to the regular performance tests is beneficial to making the regular performance tests cover various situations.

[0099] By adopting the above technical solution, the target candidate performance tests can be added to the regular performance tests, making the regular performance tests more comprehensive, capable of performing a more complete test on the circuit board, and ensuring that potential anomalies in the test process can be detected.

[0100] In the following embodiments, when adding a target candidate performance test to a regular performance test, it is necessary to ensure that the target candidate performance test does not affect the original test items in the regular performance test. Therefore, the embodiments of the present application disclose a second method for adding a performance test. Refer to Figure 5 , the method includes: Step S501: Detect whether there are test items in the regular performance test that are mutually exclusive with the target candidate performance test.

[0101] If there are test items in the regular performance test that are mutually exclusive with the target candidate performance test, then execute Step S502; If there are no test items in the regular performance test that are mutually exclusive with the target candidate performance test, then execute Steps S503 to S505.

[0102] Step S502: If so, cancel adding the target candidate performance test to the regular performance test.

[0103] In the case where there are test items in the regular performance test that are mutually exclusive with the target candidate performance test, it means that the target candidate performance test is not conducive to being added to the regular performance test, so the behavior of curve addition.

[0104] Step S503: If not, calculate the influence factor of the target candidate performance test on each test item in the regular performance test.

[0105] In the case where there are no test items in the regular performance test that are mutually exclusive with the target candidate performance test, it means that the target candidate performance test can be added to the regular performance test.

[0106] The influence factor is used to represent the correlation degree between the target candidate performance test and each test item in the regular performance test. In some embodiments, after the test circuit board undergoes the target candidate performance test, the first state parameter of the test circuit board is obtained. After the test circuit board undergoes the target regular performance in the regular performance test, the second state parameter of the test circuit board is obtained. Calculate the difference between the first state parameter and the second state parameter to obtain the influence factor. Among them, the state parameter is used to represent the health state of the test circuit board.

[0107] Step S504: According to the sorting of the influence factors, determine the target sorting of the target candidate performance test in the regular performance test.

[0108] Optionally, obtain the arrangement order of each sub-conventional performance test in the conventional performance test. Determine two influencing factors corresponding to adjacent sub-conventional performance tests. Calculate the sum of the two influencing factors to obtain the total influencing factor. Calculate the maximum value in the total influencing factor to obtain the maximum total influencing factor and the adjacent target sub-conventional performance test corresponding to the maximum total influencing factor. Insert the target candidate performance test between the adjacent target sub-conventional performance tests.

[0109] Optionally, obtain the arrangement order of each sub-conventional performance test in the conventional performance test. When the target candidate performance test is between the i-th sub-conventional performance test and the (i + 1)-th sub-conventional performance test, determine the difference in the arrangement serial numbers from the target candidate performance test to each sub-conventional performance test according to the foregoing arrangement order, and set the weight value according to the arrangement serial number, where the weight value is negatively correlated with the difference in the arrangement serial numbers. Use the weight value and the influencing factor corresponding to the sub-conventional performance test for weighted calculation to obtain the weighted influencing factor. Traverse all possible combinations of inserting the target candidate performance test into the conventional performance test to obtain several weighted influencing factors. Determine the a-th sub-conventional performance test and the (a + 1)-th sub-conventional performance test corresponding to the maximum value among the several weighted influencing factors. Insert the target candidate performance test between the a-th sub-conventional performance test and the (a + 1)-th sub-conventional performance test.

[0110] Step S505: Add the target candidate performance test to the conventional performance test according to the target sorting.

[0111] Optionally, add the target candidate performance test to the target sorting of the conventional performance test.

[0112] By adopting the above technical solution, adding the target candidate performance test to the conventional performance test ensures that the conventional performance test itself will not be affected, makes the conventional performance test more reasonable, effectively improves the test speed, and shortens the test cycle.

[0113] In the following embodiments, the test tooling can be detected in real time, and preventive maintenance can be performed according to the state of the test tooling to reduce the failure rate of the test tooling. An embodiment of the present application discloses a maintenance method for a test tooling. Refer to Figure 6 , the method includes: Step S601: Monitor the current state data of the test tooling.

[0114] Optionally, the current state data includes electrical parameters, environmental parameters, test progress, equipment health status, physical status, user interaction status, and custom data, etc.

[0115] Further, after obtaining the current state data, the current state data can be uploaded to the server for relevant personnel to review.

[0116] Step S602: Obtain other status data of other test toolings of the same type as the test tooling, and obtain the historical status data of the test tooling.

[0117] Optionally, the other test toolings execute the same test script as the test tooling.

[0118] Optionally, download the historical status data of the test tooling from the server.

[0119] Step S603: Calculate the mean of the other status data to obtain the mean of the other status data.

[0120] Optionally, if the other status data comes from multiple different other test toolings, calculate the mean of the other status data corresponding to each other test tooling respectively to obtain a subset of means. Calculate the mean of the foregoing subset of means to obtain the mean of the other status data.

[0121] Step S604: Calculate the difference between the current status data and the mean of the other status data to obtain a first difference.

[0122] Exemplarily, calculate the difference between the current status data and the mean of the other status data to obtain a first difference.

[0123] Step S605: Calculate the mean of the historical status data to obtain the mean of the historical status data.

[0124] Optionally, classify the historical status data according to a time interval to obtain grouped historical status data. Calculate the mean for different grouped historical status data to obtain a grouped mean. Calculate the mean of the grouped means to obtain the mean of the historical status data. Among them, the time interval is Step S606: Calculate the difference between the current status data and the mean of the historical status data to obtain a second difference.

[0125] Exemplarily, calculate the difference between the current status data and the mean of the historical status data to obtain a second difference.

[0126] Step S607: In the case where the first difference is greater than the first difference threshold and the second difference is greater than the second difference threshold, perform a maintenance operation on the test tooling.

[0127] Both the first difference threshold and the second difference threshold are preset empirical values. The first difference threshold and the second difference threshold can be the same or different. Relevant personnel can adjust the first difference threshold and the second difference threshold according to actual needs.

[0128] In the case where the first difference is less than the first difference threshold or the second difference is less than the second difference threshold, it is considered that the test tooling is currently working properly and no maintenance operation is required.

[0129] By adopting the above technical solutions, the state of the test tooling is monitored in real time, preventive maintenance of the test tooling is achieved, the failure rate is reduced, the service life of the test tooling is extended, and the maintenance cost is saved.

[0130] In the following embodiments, when performing maintenance operations on the test tooling, targeted maintenance can be carried out on a certain determined part of the test tooling to improve the maintenance efficiency. Therefore, Embodiment 2 of the present application discloses a maintenance method for a test tooling. Refer to Figure 7 , the method includes: Step S701: Determine the potential failure part of the test tooling according to the current state data.

[0131] Optionally, in the case where the first difference is greater than the first difference threshold and the second difference is greater than the second difference threshold, determine the source of the current state data to obtain the source sensor. Determine the potential failure part of the test tooling according to the installation position or monitoring object of the source sensor. For example, if the source sensor is used to detect the torque of the robotic arm on the test tooling, the potential failure part is the robotic arm. Another example is that if the source sensor is installed on the fixture, the potential failure part is the fixture.

[0132] Step S702: Read the part data corresponding to the potential failure part.

[0133] Exemplarily, determine the monitoring sensor for detecting the potential failure part. Obtain the sensor data of the monitoring sensor to obtain the part data.

[0134] Step S703: Identify the actual working state of the potential failure part through the part data.

[0135] Optionally, determine the normal working data corresponding to the potential failure part. Calculate the difference between the part data and the normal working data to obtain the working data difference. If the working data difference is greater than the preset working data difference threshold, it is considered that the actual working state of the potential failure part is abnormal working. If the working data difference is less than the preset working data difference threshold, it is considered that the actual working state of the potential failure part is normal working.

[0136] Step S704: In the case where the actual working state is abnormal working, retrieve the target maintenance operation in the historical maintenance record using the part data.

[0137] In some other embodiments, in the case where the actual working state is normal working, it indicates that the potential failure part is currently working normally and no maintenance operation is required.

[0138] The historical maintenance record is used to record the maintenance operations of different parts of the test tooling under different failure scenarios.

[0139] Step S705: Perform the target maintenance operation on the potential failure part.

[0140] Exemplarily, a manipulator is used to perform target maintenance operations on potential fault locations.

[0141] By adopting the above technical solution, maintenance operations are performed on the target fault location of the test tooling, enabling the target fault location to maintain a healthy state and extending the service life of the entire test tooling.

[0142] Based on the same inventive concept, an embodiment of the present application provides an automatic test system based on a test tooling. Please refer to Figure 8 , which includes: An acquisition module 801, configured to acquire test conditions, test models, standard data, a first preset threshold, a prediction model, a loss function, a second preset threshold, special data, current state data, other state data, and historical state data; A memory 802, configured to store a program of the automatic test method based on the test tooling as described in any one of the above; A processor 803, and the program in the memory can be loaded and executed by the processor to implement the automatic test method based on the test tooling as described in any one of the above.

[0143] By adopting the above technical solution, full-automatic testing of the test circuit board can be achieved, improving the test efficiency and meeting the batch testing requirements of high speed and high precision. The entire process does not require manual participation, and errors in circuit board testing caused by humans can be reduced.

[0144] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working processes of the system, device, and unit described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0145] An embodiment of the present application provides a computer-readable storage medium storing a computer program that can be loaded and executed by a processor to perform the automatic test method based on the test tooling.

[0146] Computer storage media include, for example: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0147] Based on the same inventive concept, an embodiment of the present application provides an intelligent terminal, including a memory and a processor. A computer program capable of being loaded and executed by the processor for an automatic test method based on a test fixture is stored on the memory.

[0148] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working processes of the above-described system, device, and unit can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0149] The above are all preferred embodiments of the present application. Without limiting the protection scope of the present application accordingly, any feature disclosed in this specification (including the abstract and drawings), unless specifically described, can be replaced by other equivalent or similar-purpose alternative features. That is, unless specifically described, each feature is only an example in a series of equivalent or similar features.

Claims

1. An automatic testing method based on a testing tooling, characterized in that, The method includes: In response to the test circuit board meeting the test conditions, using a manipulator to load the test circuit board into a test fixture; Using a test model to perform a conventional performance test on the test circuit board and collect the test data of the test circuit board in real time; Calculating a first data difference between the test data and standard data; Generating a test result of the test circuit board according to a comparison result between the first data difference and a first preset threshold, where the test result indicates the performance compliance situation of the test circuit board; Using the manipulator to place the test circuit board into a storage area corresponding to the test result.

2. The automatic testing method based on a test fixture according to claim 1, wherein The method further includes: Using a prediction model to perform a risk prediction on the test data to obtain a predicted test risk of the test circuit board; Adjusting the test process of the conventional performance test according to the predicted test risk; Collecting adjustment data generated by the adjusted test process; Using a loss function to perform a loss operation on the adjustment data and the test data to obtain a loss value; Using the gradient descent method to update the weight matrix of the prediction model to minimize the loss value.

3. The automatic test method based on a test fixture according to claim 2, wherein The method further includes: Calculating a second data difference between the adjustment data and the standard data; In the case where the second data difference is greater than a second preset threshold, determining a problem point on the test circuit board according to the adjustment data; Performing a special performance test on the problem point to obtain special data, where the special performance test is related to the problem position of the problem point on the test circuit board and the problem component corresponding to the problem point; Generating a special test result according to the special data and the standard data.

4. The automatic testing method based on a testing tooling according to claim 3, wherein, The method further includes: In the case where the special test result is a test failure, extracting the problem position of the problem point on the test circuit board and the problem component corresponding to the problem point; Determining candidate performance tests related to the problem position and the problem component; Performing a feature extraction operation on the candidate performance tests to obtain candidate performance test features; Performing a feature extraction operation on the special performance test to obtain special performance test features; Calculating the vector length from the candidate performance test features to the special performance test features; Selecting the n candidate performance tests corresponding to the smallest vector lengths as target candidate performance tests; Adding the target candidate performance tests to the conventional performance test.

5. The automatic testing method based on a testing tooling according to claim 4, wherein The method further includes: Detecting whether there are test items mutually exclusive with the target candidate performance tests in the conventional performance test; If so, canceling adding the target candidate performance tests to the conventional performance test; If not, calculating the influence factors of the target candidate performance tests on each test item in the conventional performance test; Determining the target order of the target candidate performance tests in the conventional performance test according to the sorting of the influence factors; Adding the target candidate performance tests to the conventional performance test according to the target order.

6. The automatic testing method based on a testing tooling according to claim 1, wherein The method further includes: Monitoring the current state data of the test fixture; Obtain other status data of other test toolings of the same type as the test tooling, and obtain the historical status data of the test tooling; Calculate the mean value of the other status data to obtain the mean value of the other status data; Calculate the difference between the current status data and the mean value of the other status data to obtain the first difference; Calculate the mean value of the historical status data to obtain the mean value of the historical status data; Calculate the difference between the current status data and the mean value of the historical status data to obtain the second difference; In the case where the first difference is greater than the first difference threshold and the second difference is greater than the second difference threshold, perform a maintenance operation on the test tooling.

7. The automatic test method based on a test tooling according to claim 6, wherein The method further includes: Determine the potential fault location of the test tooling according to the current status data; Read the part data corresponding to the potential fault location; Identify the actual working status of the potential fault location through the part data; In the case where the actual working status is abnormal, retrieve the target maintenance operation from the historical maintenance records using the part data; Perform the target maintenance operation on the potential fault location.

8. An automatic test system based on a test tooling, characterized in that, The system includes: An acquisition module, configured to acquire test conditions, a test model, standard data, a first preset threshold, a prediction model, a loss function, a second preset threshold, special data, current status data, other status data, and historical status data; A memory, configured to store the program of the automatic test method based on a test tooling according to any one of claims 1 to 7; A processor, the program in the memory can be loaded and executed by the processor and implement the automatic test method based on a test tooling according to any one of claims 1 to 7.

9. An intelligent terminal, characterized in that, It includes a memory and a processor, and a computer program capable of being loaded and executed by the processor as described in any one of the methods of claims 1 to 7 is stored on the memory.

10. A computer-readable storage medium, characterized in that, A computer program capable of being loaded and executed by the processor as described in any one of the methods of claims 1 to 7 is stored.

Images