Sound equipment production defect intelligent test method and system

Abstract

The invention relates to the technical field of equipment testing, in particular to an intelligent testing method and system for sound equipment production defects, and the method comprises the steps: obtaining hardware parameter data corresponding to a to-be-tested sound equipment, inputting the hardware parameter data into a preset automatic testing platform, the automatic test platform performs preliminary hardware function detection on the to-be-tested sound based on the hardware parameter data to obtain a corresponding hardware detection result; when the production line tests the sound equipment to be tested in different working states, key parameter sets of the sound equipment in different working states are collected in real time based on an embedded sensor, and a hardware detection result is associated with each key parameter set; and calling a pre-trained detection model, inputting the associated key parameter set and the hardware detection result into the detection model to identify a potential defect mode, obtaining a corresponding detection analysis result, and generating a detailed test report based on the detection analysis result. According to the invention, the accuracy of sound comprehensive detection can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of equipment testing, and more particularly to an intelligent testing method and system for audio production defects. Background Art

[0002] As consumers' requirements for audio equipment performance continue to increase, audio manufacturers need to ensure that their products meet high standards in all aspects. Traditional manual inspection methods are inefficient and prone to missing problems, especially when it comes to complex functions such as Bluetooth connectivity, audio processing, and waterproof performance.

[0003] Therefore, it is particularly important to develop an intelligent testing method that can automatically, quickly and accurately identify various defects that may occur in the production process. From the above, it can be seen that how to improve the accuracy of comprehensive audio testing still needs to be solved. Summary of the invention

[0004] In order to improve the accuracy of comprehensive audio testing, the present application provides an intelligent testing method and system for audio production defects.

[0005] In the first aspect, the present application provides an intelligent testing method and system for audio production defects, which adopts the following technical solutions: A method for intelligently testing production defects of audio equipment, comprising: obtaining hardware parameter data corresponding to an audio equipment to be tested, inputting the hardware parameter data into a pre-set automated testing platform, wherein the automated testing platform performs preliminary hardware function detection on the audio equipment to be tested based on the hardware parameter data, and obtains corresponding hardware detection results, wherein the hardware function detection includes audio quality, Bluetooth connectivity, battery performance, and waterproof level; when the audio equipment to be tested is tested in different working states on the production line, a set of key parameters of the audio equipment under different working states is collected in real time based on embedded sensors, and the hardware detection results are associated with each key parameter set, wherein one working state corresponds to one key data set, and the key parameters include audio output level, Bluetooth signal strength, battery voltage, and temperature; a pre-trained detection model is retrieved, the associated key parameter set and the hardware detection result are input into the detection model to identify potential defect patterns, corresponding detection analysis results are obtained, and a detailed test report is generated based on the detection analysis results, wherein the detailed test report includes specific improvement measures or suggestions.

[0006] By adopting the above technical solutions, integrating hardware function detection and real-time working status monitoring, combined with pre-trained detection models to identify potential defect patterns, a more comprehensive and accurate evaluation of the quality of audio products is achieved. It not only ensures the basic performance indicators in multiple dimensions, from audio quality to waterproof performance, but also enhances the ability to discover hidden problems or abnormal situations through detailed analysis of changes in key parameters in actual usage scenarios, effectively improving the accuracy and reliability of overall detection.

[0007] Optionally, the method also includes: obtaining a corresponding comprehensive environmental parameter set to simulate a variety of actual usage scenarios, wherein the comprehensive environmental parameter set includes humidity level, temperature range, vibration intensity, impact force, electromagnetic field strength, dust concentration and light intensity; adjusting a comprehensive environmental parameter in the comprehensive environmental parameter set for simulation, and obtaining detection and analysis results corresponding to different comprehensive environmental parameter conditions; based on all detection and analysis results, obtaining a corresponding audio performance score, and if the audio performance score is higher than a preset performance score threshold, it is determined to be qualified.

[0008] By adopting the above technical solutions, a comprehensive set of environmental parameters is obtained and adjusted to simulate a variety of actual usage scenarios, and the performance of the speakers under different environmental conditions is comprehensively tested and scored. This method not only ensures that the performance of the speakers meets the standards under standard test conditions, but also improves the comprehensiveness and accuracy of the test by simulating various complex conditions in the real usage environment, thereby more reliably evaluating the overall performance of the speakers and ensuring that they perform well under various actual usage conditions.

[0009] Optionally, the method also includes: continuously collecting test data of each audio test to be tested and determining it as historical test data, wherein the test data includes test results, environmental conditions, and equipment status; analyzing the historical test data based on statistical analysis to identify key factors that affect the test results; adjusting test parameters based on the key factors, and applying the adjusted test parameters to new tests, wherein the test parameters include test frequency and increasing the weight of specific test items.

[0010] By adopting the above technical solutions, by continuously collecting and analyzing historical test data, the key factors that affect the test results are identified, and the test parameters (such as test frequency and the weight of specific test items) are dynamically adjusted accordingly to optimize the future test process. This method improves the pertinence and effectiveness of the test, ensures that each test can more accurately reflect the actual performance of the audio system, and enhances the accuracy and reliability of quality control.

[0011] Optionally, the method also includes: retrieving corresponding historical test data to obtain the corresponding current production conditions on the current production line, wherein the production conditions include temperature, humidity, and material batches; analyzing the historical test data and the current production conditions based on statistical analysis to identify the performance characteristics and change trends of different batches of products; and adjusting the thresholds of the test parameters based on the performance characteristics and change trends.

[0012] By adopting the above technical solution, by retrieving historical test data and analyzing it in combination with current production conditions, the performance characteristics and change trends of different batches of products can be identified, and the thresholds of test parameters can be dynamically adjusted accordingly. This method ensures that the test parameters can adapt to the characteristics of the current production batch, improves the consistency and accuracy of the test results, and thus more effectively guarantees product quality.

[0013] Optionally, the method also includes: retrieving corresponding historical test data, and determining corresponding key areas and high-risk areas based on the historical test data; adjusting sampling points and sampling frequencies based on the key areas and the high-risk areas; and applying the adjusted sampling frequencies and sampling points to new tests.

[0014] By adopting the above technical solutions, key areas and high-risk areas are determined by analyzing historical test data, and sampling points and sampling frequencies are adjusted intelligently accordingly, so that resources can be more effectively concentrated on these important areas in new tests. This method improves the pertinence and efficiency of data collection, ensures the accuracy and reliability of test results, and helps to discover and solve potential quality problems earlier.

[0015] Optionally, the method also includes: retrieving corresponding historical test data, classifying the historical test data by batches, retrieving historical key performance indicators in the historical test data, wherein the key performance indicators include audio quality, Bluetooth connection stability, and battery life; obtaining current key performance indicators of the current batch, comparing the current key performance indicators with the historical key performance indicators, and obtaining corresponding indicator comparison results; identifying abnormal data points in the current batch through an anomaly detection algorithm based on the indicator comparison results, and issuing an early warning based on the abnormal data points.

[0016] By adopting the above technical solution, by comparing the key performance indicators of the current batch with the historical batches and using the anomaly detection algorithm to identify abnormal data points in the current batch, potential quality fluctuations can be warned in advance. This method ensures continuous monitoring of product quality, improves the timeliness and accuracy of problem discovery, and helps to take quick measures to maintain product consistency and reliability.

[0017] Optionally, the method also includes: identifying possible key factors affecting product quality based on the historical test data, drawing the causal relationship between possible key factors, and constructing a corresponding causal relationship diagram, wherein the possible key factors include temperature, humidity, material batch, and production process parameters; using a causal inference method to analyze the causal path from input factors to output results, and obtaining corresponding causal analysis results, wherein the causal analysis results include critical paths and the greatest influencing factors; based on the causal relationship diagram and the causal analysis results, identifying the root causes of defects, and generating improvement suggestions based on the root causes.

[0018] By adopting the above technical solution, by constructing and analyzing the cause-and-effect relationship diagram, the key factors affecting product quality and the causal paths between them are identified, so as to determine the root cause of the defect and generate improvement suggestions accordingly. This method provides a systematic root cause analysis, which helps to accurately locate the source of the problem and take targeted improvement measures, thereby improving product quality and the reliability of the production process.

[0019] In a second aspect, the present application provides an intelligent testing system for audio production defects, which adopts the following technical solutions: An intelligent testing system for audio production defects, comprising: The hardware detection result acquisition module obtains the hardware parameter data corresponding to the audio to be tested, and inputs the hardware parameter data into a pre-set automated testing platform. The automated testing platform performs a preliminary hardware function test on the audio to be tested based on the hardware parameter data, and obtains the corresponding hardware detection result, wherein the hardware function test includes audio quality, Bluetooth connectivity, battery performance, and waterproof level. An association module, when the tested speaker is tested in different working states on the production line, collects key parameter sets of the speaker in different working states in real time based on the embedded sensor, and is used to associate the hardware detection result with each key parameter set, wherein one working state corresponds to one key data set, and the key parameters include audio output level, Bluetooth signal strength, battery voltage and temperature; The detection and analysis result acquisition module calls up a pre-trained detection model, inputs the associated key parameter set and the hardware detection result into the detection model to identify potential defect patterns, and is used to obtain corresponding detection and analysis results, and generate a detailed test report based on the detection and analysis results, wherein the detailed test report includes specific improvement measures or suggestions.

[0020] In a third aspect, the present application provides an intelligent testing method for audio production defects, which adopts the following technical solution: A method for intelligently testing audio production defects comprises a processor, wherein a program of any one of the above-mentioned methods for intelligently testing audio production defects is running on the processor.

[0021] In a fourth aspect, the present application provides a storage medium, which adopts the following technical solution: A storage medium stores a program of any one of the above-mentioned intelligent testing methods for acoustic production defects.

[0022] In summary, the present application includes at least one of the following beneficial technical effects: 1. By integrating hardware function detection with real-time working status monitoring, and combining pre-trained detection models to identify potential defect patterns, it not only ensures that the performance of the audio system meets the standards under standard test conditions, but also enhances the ability to discover hidden problems or abnormal situations by simulating various complex conditions in the real use environment. In addition, by conducting statistical analysis of historical data and constructing causal relationship diagrams, the impact assessment of key factors and the location of root causes are achieved, so that targeted improvement measures can be taken.

[0023] 2. By continuously collecting and analyzing historical test data to dynamically adjust test parameters, sampling strategies, and performance indicator comparisons, the solution improves the pertinence and efficiency of the test process, ensuring that each test can more accurately reflect the actual performance of the audio system. This approach not only improves the consistency and accuracy of test results, but also strengthens the early warning capabilities for product quality fluctuations, thereby comprehensively improving the accuracy and reliability of audio product testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The present invention is a flowchart of a method for intelligently testing audio production defects according to an exemplary embodiment.

[0025] Figure 2 The figure is a structural block diagram of an intelligent testing system for audio production defects according to an exemplary embodiment. DETAILED DESCRIPTION

[0026] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.

[0027] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0028] The present application embodiment discloses an intelligent testing method for acoustic production defects, referring to Figure 1 ,include: S100, obtaining hardware parameter data corresponding to the audio to be tested, inputting the hardware parameter data into a pre-set automated testing platform, and the automated testing platform performs a preliminary hardware function test on the audio to be tested based on the hardware parameter data to obtain a corresponding hardware test result.

[0029] Among them, special test equipment or built-in self-test program is used to collect various hardware parameter data of the audio to be tested, which usually include audio quality, Bluetooth connectivity, battery performance and waterproof level, etc.; the collected hardware parameter data is input into a pre-set automated test platform. It should be pointed out here that the automated test platform is usually an integrated software system that can automatically execute a series of predefined test processes; then, the automated test platform performs preliminary hardware function detection based on the input data, for example, testing whether the audio quality meets the standards, whether the Bluetooth connection is stable, whether the battery performance is up to standard, whether the waterproof level meets the requirements, etc.

[0030] Finally, the preliminary hardware detection results are generated and recorded in the database to provide a basis for subsequent analysis. Automated testing improves efficiency and consistency, reduces human intervention, and ensures that the basic functions of each speaker can be evaluated in a standardized manner.

[0031] S110, when the production line tests the speakers under test in different working states, key parameter sets of the speakers under different working states are collected in real time based on embedded sensors, and the hardware detection results are associated with each key parameter set.

[0032] Among them, embedded sensors are deployed on the production line to monitor the key parameters of the audio in different working states in real time. It should be pointed out here that different working states are defined, such as playing music, talking, standby, etc. Each working state corresponds to a specific test scenario, and one working state corresponds to a key data set.

[0033] Based on embedded sensors, a set of key parameters of the audio system under different working conditions are collected in real time. These key parameters include audio output level, Bluetooth signal strength, battery voltage and temperature, etc. By real-time monitoring of key parameters under different working conditions, the performance of the audio system in actual use can be more comprehensively evaluated and potential problems can be discovered.

[0034] In addition, the preliminary hardware test results are associated with the key parameter sets under each working state. For example, the audio quality test results are associated with parameters such as the audio output level when playing music; it is necessary to ensure that each key parameter set can correspond to the corresponding hardware test results, and by associating the data, a complete data set is established, including all relevant hardware test results and key parameter sets.

[0035] S120, retrieve the pre-trained detection model, input the associated key parameter set and hardware detection results into the detection model to identify potential defect patterns, obtain the corresponding detection analysis results, and generate a detailed test report based on the detection analysis results.

[0036] Among them, a pre-trained detection model is retrieved. It should be pointed out here that the detection model is usually trained based on a large amount of historical data and machine learning algorithms (such as neural networks, support vector machines, etc.); the associated key parameter set and hardware detection results are input into the detection model, and the detection model will identify potential defect patterns and generate corresponding detection analysis results. The output detection analysis results include potential defect patterns and related indicators. For example, the model may identify certain abnormal audio output levels or unstable Bluetooth signal strength problems.

[0037] Generate a detailed test report based on the test and analysis results. The report content includes the test results, potential defect modes, specific improvement measures or suggestions, etc. The report can be presented in the form of charts, text descriptions, etc., which is convenient for technical personnel to understand and take action. For example, the report may include a trend chart of audio quality changes over time, as well as specific improvement suggestions; the generated test report is distributed to relevant personnel, such as production line managers, quality control personnel and technical support teams.

[0038] Based on the above execution steps, by obtaining accurate hardware parameter data, automated preliminary detection, real-time working status testing, data association, calling pre-trained detection models to identify potential defect patterns, and generating detailed test reports, it is ensured that the performance of the audio system meets the standards under standard test conditions. By simulating the complex conditions in the real use environment, the ability to discover hidden problems and abnormal situations is enhanced, thereby comprehensively improving the accuracy and reliability of detection and providing a solid foundation for product quality control.

[0039] In the solution in the embodiment of the present application, it is necessary to consider the situation of the audio in different actual usage scenarios, so the method also includes: S1211, obtaining a corresponding comprehensive environmental parameter set to simulate a variety of actual usage scenarios.

[0040] Among them, determine the actual usage scenarios that need to be simulated, and define the corresponding comprehensive environmental parameter sets, which include humidity level, temperature range, vibration intensity, impact force, electromagnetic field intensity, dust concentration and light intensity, etc.; then collect the specific numerical ranges of these environmental parameters from historical data, standard test specifications or actual usage feedback, and set these comprehensive environmental parameters in the test platform to ensure that a variety of actual usage scenarios can be simulated. For example, different humidity levels (such as 30% to 90%) and temperature ranges (such as -20°C to 50°C) can be set.

[0041] S1212, simulating the adjustment of a comprehensive environmental parameter in the comprehensive environmental parameter set to obtain detection and analysis results that do not correspond to different comprehensive environmental parameter conditions.

[0042] Among them, a comprehensive environmental parameter is selected for adjustment, such as adjusting the humidity level or temperature range, and a simulation test is performed under the adjusted environmental parameters to collect a set of key parameters of the audio in real time, including audio output level, Bluetooth signal strength, battery voltage and temperature, etc., and record the test data under different environmental parameters. It should be pointed out here that similar adjustments and tests need to be made to each comprehensive environmental parameter to obtain the detection and analysis results under different environmental parameters.

[0043] S1213, obtaining a corresponding sound performance score based on all the detection and analysis results, and if the sound performance score is higher than a preset performance score threshold, determining that it is qualified.

[0044] Among them, all the detection and analysis results under different environmental parameters are integrated to form a complete data set, and a pre-set scoring model is applied. The scoring model is usually based on multiple key performance indicators (such as audio quality, Bluetooth connection stability, battery life, etc.) and the influence of environmental parameters; the audio performance score under each environmental parameter is calculated according to the scoring model. The score can be based on a weighted average, a comprehensive index or other statistical methods, and the performance score under each environmental parameter is recorded to form a detailed scoring report.

[0045] Then, according to product standards and quality requirements, a performance score threshold is pre-set. This threshold can be based on industry standards, customer requirements or internal company regulations, and the performance score under each environmental parameter is compared with the preset threshold. If the performance score under all environmental parameters is higher than the preset threshold, the batch of speakers is determined to be qualified; otherwise, it is marked as unqualified, and specific reasons for unqualified and improvement suggestions are generated. The qualified or unqualified results are fed back to the production line management and technical team so that appropriate measures can be taken.

[0046] Based on the above steps S1211 to S1213, by obtaining a comprehensive set of environmental parameters and conducting simulation tests, the system evaluates the performance of the audio system under various actual use conditions, calculates the performance score based on the test analysis results, and compares it with the preset threshold to determine whether the product is qualified. This method not only ensures that the performance of the audio system meets the standards under standard test conditions, but also enhances the ability to discover hidden problems and abnormal situations by simulating complex conditions in the real use environment, thereby comprehensively improving the accuracy and reliability of the detection.

[0047] In addition, in the embodiment of the present application, the method further includes: S1221, continuously collect the test data of each test of the audio system to be tested and determine it as historical test data.

[0048] Each time the audio system to be tested is tested, all relevant test data is recorded, including but not limited to test results (such as audio quality, Bluetooth connection stability, battery life, etc.), environmental conditions (such as temperature, humidity, electromagnetic field strength, etc.) and device status (such as sensor readings, device operation logs, etc.); the collected data is stored in a central database to ensure data security and accessibility, and structured databases (such as SQL databases) or unstructured databases (such as NoSQL databases) can be used to store these data. In addition, the stored data is verified regularly to ensure data accuracy and completeness.

[0049] S1222, analyzing historical test data based on statistical analysis to identify key factors that affect test results.

[0050] Among them, the historical test data is preprocessed, including data cleaning, missing value processing, outlier detection, etc., to ensure the quality of the data; then the features related to the test results, such as environmental conditions, equipment status, etc., are selected as input variables for analysis; statistical analysis methods (such as correlation analysis, regression analysis, variance analysis, etc.) are applied to identify the key factors affecting the test results. For example, the relationship between temperature and audio quality can be discovered through correlation analysis.

[0051] S1223, adjusting the test parameters based on the key factors, and applying the adjusted test parameters to a new test.

[0052] Among them, the test parameters are adjusted according to the identified key factors. For example, if temperature is found to be a key factor affecting audio quality, the test frequency at different temperatures can be increased; if certain specific test items (such as waterproof performance) are found to have frequent problems, the weights of these test items can be increased.

[0053] Optimize test parameters through experimental design (such as orthogonal experiments, response surface methodology, etc.) to find the best test plan; apply the adjusted test parameters to new tests to ensure that the new tests can more effectively reflect the performance under actual use conditions, and incorporate the new test results into historical data again to form a closed-loop feedback mechanism to continuously optimize test parameters.

[0054] Based on the above steps S1221 to S1223, by continuously collecting test data, identifying key factors through statistical analysis, and dynamically adjusting test parameters (such as test frequency and weights of specific test items) based on these key factors, the scientific nature and accuracy of the test are ensured, and the targetedness and efficiency of the test are improved.

[0055] The method also includes: S1231, retrieve the corresponding historical test data to obtain the corresponding current production conditions on the current production line.

[0056] Among them, the production condition data on the current production line is collected in real time, including temperature, humidity, material batches, etc. These data can be obtained through sensors, monitoring systems and production records, and historical test data and current production condition data are integrated.

[0057] S1232, analyze historical test data and current production conditions based on statistical analysis to identify performance characteristics and change trends of different batches of products.

[0058] Among them, historical test data and current production condition data are preprocessed, including data cleaning, missing value processing, outlier detection, etc., to ensure the quality of the data; characteristics related to product performance, such as temperature, humidity, material batch, etc., are selected as input variables for analysis.

[0059] Calculate the mean, variance, standard deviation and other statistics in historical data to understand the basic distribution of the data; use correlation coefficients (such as Pearson correlation coefficient) to identify the correlation between different production conditions and product performance; apply time series analysis methods (such as moving average, exponential smoothing, etc.) to identify the changing trends of performance indicators over time; establish regression models (such as linear regression, polynomial regression, etc.) to predict performance indicators under different production conditions; use clustering algorithms (such as K-means, hierarchical clustering, etc.) to classify different batches of products according to performance characteristics.

[0060] Through statistical analysis and machine learning methods, the performance characteristics and change trends of different batches of products can be systematically identified, providing a scientific basis for subsequent parameter adjustments. This helps to improve the pertinence and effectiveness of the test.

[0061] S1233, adjusting the threshold of the test parameter based on the performance characteristics and the change trend.

[0062] Among them, the threshold of the test parameters is reset according to the identified performance characteristics and change trends. For example, if it is found that some batches have poor performance within a specific temperature range, the qualified threshold of the temperature range can be lowered; the test parameters are optimized through experimental design (such as orthogonal experiments, response surface method, etc.) to find the best test plan. The adjusted test parameter threshold is applied to the new test to ensure that the new test can more effectively reflect the performance under actual use conditions; the new test results are incorporated into the historical data again to form a closed-loop feedback mechanism to continuously optimize the test parameter threshold.

[0063] By dynamically adjusting the thresholds of test parameters, the pertinence and efficiency of the test can be improved, ensuring that each test can more accurately reflect the actual performance of the audio system. This helps to promptly discover and resolve potential quality issues and improve product quality.

[0064] In addition, the method includes: S1241, retrieve corresponding historical test data, and determine corresponding key areas and high-risk areas based on the historical test data.

[0065] Among them, historical test data related to the current production batch is retrieved from the database, and the historical test data is preprocessed, including data cleaning, missing value processing, outlier detection, etc., to ensure the quality of the data.

[0066] Use time series analysis methods (such as moving average, exponential smoothing, etc.) to identify the changing trends of performance indicators over time; apply clustering algorithms (such as K-means, hierarchical clustering, etc.) to classify different batches of products according to performance characteristics and identify areas with consistent performance; use anomaly detection algorithms (such as isolated forests, local anomaly factors, etc.) to identify anomalies in historical data, which may indicate potential high-risk areas; analyze key performance indicators (such as audio quality, Bluetooth connection stability, battery life, etc.) to determine in which areas or time periods these indicators frequently show abnormalities or fluctuate greatly.

[0067] Based on the above analysis results, key areas (areas where performance is particularly important) and high-risk areas (areas where performance is unstable or prone to problems) are identified. Through detailed analysis of historical data, key areas and high-risk areas can be systematically identified, which helps to improve the pertinence and effectiveness of testing.

[0068] S1242, adjust sampling points and sampling frequency based on key areas and high-risk areas.

[0069] Among them, for sampling points, more sampling points are added in key areas to ensure that data collection in these areas is more intensive and comprehensive; sampling points are added in high-risk areas, while ensuring that the sampling points are evenly distributed to cover all potential problem points.

[0070] For sampling frequency, increase the sampling frequency in key areas, for example, from once an hour to once every half hour, to obtain more detailed performance data; dynamically adjust the sampling frequency in high-risk areas based on the frequency of problems found in historical data, for example, increase the sampling frequency in certain time periods and reduce the sampling frequency in other time periods.

[0071] Combined with the characteristics of key areas and high-risk areas, the optimal sampling strategy is formulated to ensure the efficiency and accuracy of data collection. By adjusting the sampling points and sampling frequency, resources can be more effectively concentrated on key areas and high-risk areas, improving the pertinence and efficiency of data collection, thereby more accurately reflecting the actual performance of the audio system.

[0072] S1243, apply the adjusted sampling frequency and sampling points to a new test.

[0073] Among them, the test plan is updated according to the adjusted sampling points and sampling frequency to ensure that the new test plan can cover all key areas and high-risk areas, and the settings of the test equipment are adjusted to ensure that it can collect data according to the new sampling points and sampling frequency.

[0074] According to the updated test plan and configuration, new tests are conducted to collect more detailed and targeted data, record new test data, and conduct preliminary verification to ensure the accuracy and completeness of the data. The new test results are incorporated into the historical data again to form a closed-loop feedback mechanism to continuously optimize the sampling points and sampling frequency.

[0075] By applying the adjusted sampling frequency and sampling points to new tests, it can be ensured that each test can more accurately reflect the actual performance of the audio system, improve the effectiveness and efficiency of data collection, and thus better discover and solve potential quality problems.

[0076] In the scheme of the embodiment of the present application, corresponding abnormality detection is also performed, and the method further includes: S1251, retrieve corresponding historical test data, classify the historical test data by batches, and retrieve historical key performance indicators in the historical test data.

[0077] Among them, the historical test data is classified by batches to ensure that the data of each batch can be analyzed independently. Identifiers such as timestamps and batch numbers can be used for classification to extract key performance indicators from historical test data, such as audio quality, Bluetooth connection stability, battery life, etc. These indicators are usually important factors affecting product quality. The extracted key performance indicators are preprocessed, including data cleaning, missing value processing, outlier detection, etc., to ensure data quality.

[0078] S1252, obtaining the current key performance indicators of the current batch, comparing the current key performance indicators with the historical key performance indicators, and obtaining corresponding indicator comparison results.

[0079] Among them, the key performance indicator data of the current production batch is collected in real time, which can be obtained through automated testing platforms, sensors and other monitoring systems. The key performance indicator data of the current batch needs to be preprocessed to ensure the consistency and comparability of the data.

[0080] Directly compare the key performance indicators of the current batch with the corresponding indicators of the historical batches, and calculate the differences and change rates; apply statistical methods (such as mean, variance, standard deviation, etc.) to analyze the performance differences between the current batch and the historical batches; use time series analysis methods (such as moving average, exponential smoothing, etc.) to identify the change trend of performance indicators over time. Record the results of the comparative analysis, including the specific values, differences and change trends of each indicator.

[0081] S1253, based on the indicator comparison results, an abnormal data point in the current batch is identified through an anomaly detection algorithm, and an early warning is issued based on the abnormal data point.

[0082] Among them, choose a suitable anomaly detection algorithm, such as Isolation Forest, Local Outlier Factor (LOF), Boxplot, etc.; if you are using an anomaly detection algorithm based on machine learning, you need to train the model with historical data first.

[0083] Input the key performance indicator data of the current batch into the anomaly detection algorithm to identify abnormal data points. For example, if the value of some indicators deviates significantly from the normal range of historical data, they will be marked as abnormal. According to the anomaly detection results, set warning rules. For example, when the abnormal value of a key performance indicator exceeds a certain threshold, trigger the warning and generate specific warning information, including a detailed description of the abnormal data point, possible cause analysis and recommended improvement measures. Send the warning information to relevant personnel, such as production line managers, quality control personnel and technical support teams, so that timely action can be taken.

[0084] Based on the above execution steps S1251 to S1253, by retrieving and classifying historical test data, comparing key performance indicators of current and historical batches, and using anomaly detection algorithms to identify and warn abnormal data points in the current batch, the scientificity and accuracy of the test are improved, and the ability to discover potential quality problems is enhanced, thereby more effectively ensuring the consistency and reliability of the quality of audio products.

[0085] Finally, there may be multiple key factors for product quality, and there are correlations between the key factors, so the method also includes: S1261, identify possible key factors affecting product quality based on historical test data, draw the cause-effect relationship between possible key factors, and construct the corresponding cause-effect relationship diagram.

[0086] Among them, historical test data are retrieved from the database and preprocessed, including data cleaning, missing value processing, outlier detection, etc., to ensure the quality of the data, and features related to product quality, such as temperature, humidity, material batch, production process parameters, etc., are selected as input variables for analysis.

[0087] Use statistical methods (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) to identify the correlation between different factors, which helps to preliminarily screen out possible key factors, and use causal inference methods (such as causal diagrams, structural equation models, Bayesian networks, etc.) to identify the causal relationship between different factors. For example, a causal diagram can intuitively show how temperature changes affect audio quality; based on the identified causal relationship, draw a causal diagram. Causal diagrams are usually composed of nodes (representing factors) and directed edges (representing causal relationships) to help understand the interaction between factors.

[0088] S1262, using causal inference methods to analyze the causal path from input factors to output results, and obtain corresponding causal analysis results.

[0089] Among them, causal inference methods (such as causal diagrams, structural equation models, potential outcome models, etc.) are used to analyze the causal path from input factors to output results. These methods can help identify which factors have a direct or indirect impact on the quality of the final product. Through causal path analysis, the critical path with the greatest impact on product quality is identified. For example, if it is found that temperature changes directly affect audio quality, and humidity changes indirectly affect audio quality through temperature changes, then temperature changes are the critical path.

[0090] Based on the results of causal path analysis, determine the factors that have the greatest impact on product quality, which may be the main cause of defects. By changing the values ​​of certain factors and observing their impact on other factors and the final result, determine which factors are most sensitive to product quality.

[0091] S1263, based on the cause-effect diagram and cause-effect analysis results, identifies the root causes of defects and generates improvement suggestions based on the root causes.

[0092] Among them, the root causes of product quality defects are systematically identified by combining the cause-effect relationship diagram and the cause-effect analysis results. For example, if temperature change is found to be the main factor causing the audio quality to deteriorate, then the temperature control problem is the root cause. The specific manifestations and formation mechanisms of the root causes are further analyzed. For example, the temperature control problem may be caused by the failure of the temperature control system of the production equipment or the large temperature fluctuations in the production environment.

[0093] Based on the results of the root cause analysis, specific improvement suggestions are generated, for example, suggestions to improve the performance of the temperature control system, strengthen temperature monitoring of the production environment, or optimize production process parameters; a detailed implementation plan is formulated, including specific steps, timetables, and responsibility allocations for improvement measures to ensure that the improvement suggestions can be effectively implemented; the effects of the improvement measures after implementation are incorporated into historical data again to form a closed-loop feedback mechanism to continuously optimize the improvement measures.

[0094] Based on the execution steps from S1261 to S1263 above, through in-depth mining of historical data and construction of causal relationship diagrams, this method can more scientifically identify the key factors affecting product quality, thereby improving the accuracy and reliability of quality testing; using advanced causal inference technology, it not only helps to determine which factors directly affect product quality, but also reveals the complex interactions between these factors, deepening the understanding of the causes behind quality problems; based on causal relationship diagrams and detailed causal path analysis, it can more accurately identify the root causes of product defects; by generating specific improvement suggestions and combining feedback after implementation to continuously adjust strategies, a closed-loop quality management system is formed, which helps companies continuously improve production processes and improve product quality.

[0095] The present application embodiment discloses an intelligent testing system for acoustic production defects, referring to Figure 2 , the system includes but is not limited to: The hardware detection result acquisition module 200 acquires the hardware parameter data corresponding to the audio to be tested, and inputs the hardware parameter data into a pre-set automated testing platform. The automated testing platform performs a preliminary hardware function test on the audio to be tested based on the hardware parameter data, and acquires the corresponding hardware detection result, wherein the hardware function test includes audio quality, Bluetooth connectivity, battery performance, and waterproof level. The association module 210 collects key parameter sets of the audio system in different working states in real time based on the embedded sensor when the audio system to be tested is tested in the production line, and is used to associate the hardware detection result with each key parameter set, wherein one working state corresponds to one key data set, and the key parameters include audio output level, Bluetooth signal strength, battery voltage and temperature; The detection and analysis result acquisition module 220 calls up a pre-trained detection model, inputs the associated key parameter set and hardware detection results into the detection model to identify potential defect patterns, so as to obtain corresponding detection and analysis results, and generates a detailed test report based on the detection and analysis results, wherein the detailed test report includes specific improvement measures or suggestions.

[0096] Furthermore, the system includes but is not limited to: A simulation module, used to obtain a corresponding comprehensive environmental parameter set to simulate a variety of actual usage scenarios, wherein the comprehensive environmental parameter set includes humidity level, temperature range, vibration intensity, impact force, electromagnetic field intensity, dust concentration and light intensity; The detection and analysis result acquisition module simulates the adjustment of a comprehensive environmental parameter in the comprehensive environmental parameter set to obtain detection and analysis results that do not correspond to different comprehensive environmental parameter conditions; The audio performance score acquisition module is used to obtain the corresponding audio performance score based on all the detection and analysis results. If the audio performance score is higher than the preset performance score threshold, it is determined to be qualified.

[0097] Furthermore, the system includes but is not limited to: A historical test data determination module, which continuously collects test data of each test of the audio system to be tested for determining historical test data, wherein the test data includes test results, environmental conditions, and device status; The key factor identification module analyzes historical test data based on statistical analysis to identify key factors that affect test results; The test parameter adjustment module is used to adjust the test parameters based on key factors and apply the adjusted test parameters to new tests, wherein the test parameters include test frequency and increasing the weight of specific test items.

[0098] Furthermore, the system includes but is not limited to: The current production condition acquisition module retrieves the corresponding historical test data to obtain the current production conditions corresponding to the current production line, where the production conditions include temperature, humidity, and material batches; The change trend identification module analyzes historical test data and current production conditions based on statistical analysis to identify the performance characteristics and change trends of different batches of products; The threshold adjustment module is used to adjust the threshold of the test parameter based on the performance characteristics and change trends.

[0099] Furthermore, the system includes but is not limited to: The area determination module retrieves the corresponding historical test data and determines the corresponding key areas and high-risk areas based on the historical test data; Sampling adjustment module, used to adjust sampling points and sampling frequency based on key areas and high-risk areas; The application module is used to apply the adjusted sampling frequency and sampling points to the new test.

[0100] Furthermore, the system includes but is not limited to: A historical key performance indicator retrieval module retrieves the corresponding historical test data and classifies the historical test data by batches, and is used to retrieve historical key performance indicators in the historical test data, where the key performance indicators include audio quality, Bluetooth connection stability, and battery life; The indicator comparison result acquisition module obtains the current key performance indicators of the current batch and compares the current key performance indicators with the historical key performance indicators to obtain the corresponding indicator comparison results; The abnormal data point identification module is used to identify abnormal data points in the current batch through anomaly detection algorithm based on the indicator comparison results, and issue an early warning based on the abnormal data points.

[0101] Furthermore, the system includes but is not limited to: The possible key factor identification module is used to identify possible key factors affecting product quality based on historical test data, draw the causal relationship between possible key factors, and construct the corresponding causal relationship diagram, where the possible key factors include temperature, humidity, material batch, and production process parameters; A causal analysis result acquisition module uses a causal inference method to analyze the causal path from input factors to output results to obtain corresponding causal analysis results, wherein the causal analysis results include critical paths and the most influential factors; The root cause identification module is used to identify the root causes of defects based on the cause-effect diagram and cause-effect analysis results, and to generate improvement suggestions based on the root causes.

[0102] An embodiment of the present application further discloses an intelligent testing method for acoustic production defects, comprising a processor, wherein a program of any one of the above-mentioned intelligent testing methods for acoustic production defects is running on the processor.

[0103] The embodiment of the present application further discloses a storage medium storing a program of any one of the above-mentioned intelligent testing methods for acoustic production defects.

[0104] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An intelligent testing method for acoustic production defects, characterized in that: include: Obtaining hardware parameter data corresponding to the audio to be tested, inputting the hardware parameter data into a pre-set automated testing platform, wherein the automated testing platform performs a preliminary hardware function test on the audio to be tested based on the hardware parameter data, and obtains a corresponding hardware test result, wherein the hardware function test includes audio quality, Bluetooth connectivity, battery performance, and waterproof level; When the speaker to be tested is tested in different working states on the production line, the key parameter sets of the speaker in different working states are collected in real time based on the embedded sensor, and the hardware detection result is associated with each key parameter set, wherein one working state corresponds to one key data set, and the key parameters include audio output level, Bluetooth signal strength, battery voltage and temperature; Retrieve a pre-trained detection model, input the associated key parameter set and the hardware detection results into the detection model to identify potential defect patterns, obtain corresponding detection analysis results, and generate a detailed test report based on the detection analysis results, wherein the detailed test report includes specific improvement measures or suggestions.

2. The intelligent testing method for acoustic production defects according to claim 1 is characterized in that: The method also includes: Obtaining a corresponding comprehensive environmental parameter set to simulate a variety of actual usage scenarios, wherein the comprehensive environmental parameter set includes humidity level, temperature range, vibration intensity, impact force, electromagnetic field intensity, dust concentration, and light intensity; A comprehensive environmental parameter in the comprehensive environmental parameter set is adjusted for simulation, and detection and analysis results corresponding to different comprehensive environmental parameter conditions are obtained; The corresponding sound performance score is obtained based on all the detection and analysis results. If the sound performance score is higher than a preset performance score threshold, it is determined to be qualified.

3. The intelligent testing method for acoustic production defects according to claim 1 is characterized in that: The method also includes: Continuously collect test data of each test of the audio system to be tested and determine it as historical test data, wherein the test data includes test results, environmental conditions, and device status; Analyze the historical test data based on statistical analysis to identify key factors affecting test results; The test parameters are adjusted based on the key factors, and the adjusted test parameters are applied to new tests, wherein the test parameters include test frequency and increasing the weight of specific test items.

4. The intelligent testing method for acoustic production defects according to claim 3 is characterized in that: The method also includes: Retrieve the corresponding historical test data and obtain the current production conditions corresponding to the current production line, where the production conditions include temperature, humidity, and material batches; Analyze the historical test data and the current production conditions based on statistical analysis to identify performance characteristics and change trends of different batches of products; The thresholds of the test parameters are adjusted based on the performance characteristics and the changing trends.

5. The intelligent testing method for acoustic production defects according to claim 3 is characterized in that: The method also includes: Retrieving corresponding historical test data, and determining corresponding key areas and high-risk areas based on the historical test data; Adjusting the sampling points and sampling frequencies based on the key areas and the high-risk areas; Apply the adjusted sampling frequency and sampling points to the new test.

6. The intelligent testing method for acoustic production defects according to claim 3 is characterized in that: The method also includes: Retrieving corresponding historical test data, classifying the historical test data by batches, and retrieving historical key performance indicators in the historical test data, wherein the key performance indicators include audio quality, Bluetooth connection stability, and battery life; Obtaining a current key performance indicator of a current batch, comparing the current key performance indicator with the historical key performance indicator, and obtaining a corresponding indicator comparison result; Based on the indicator comparison results, abnormal data points in the current batch are identified through an anomaly detection algorithm, and an early warning is issued based on the abnormal data points.

7. The intelligent testing method for acoustic production defects according to claim 3 is characterized in that: The method also includes: Based on the historical test data, possible key factors affecting product quality are identified, causal relationships between possible key factors are plotted, and corresponding causal relationship diagrams are constructed, wherein possible key factors include temperature, humidity, material batches, and production process parameters; Use causal inference methods to analyze the causal path from input factors to output results and obtain corresponding causal analysis results, where the causal analysis results include critical paths and the most influential factors; Based on the cause-effect relationship diagram and the cause-effect analysis results, the root cause of the defect is identified, and improvement suggestions are generated based on the root cause.

8. An intelligent testing system for acoustic production defects, characterized in that: include: The hardware detection result acquisition module obtains the hardware parameter data corresponding to the audio to be tested, and inputs the hardware parameter data into a pre-set automated testing platform. The automated testing platform performs a preliminary hardware function test on the audio to be tested based on the hardware parameter data, and obtains the corresponding hardware detection result, wherein the hardware function test includes audio quality, Bluetooth connectivity, battery performance, and waterproof level. An association module, when the tested speaker is tested in different working states on the production line, collects key parameter sets of the speaker in different working states in real time based on the embedded sensor, and is used to associate the hardware detection result with each key parameter set, wherein one working state corresponds to one key data set, and the key parameters include audio output level, Bluetooth signal strength, battery voltage and temperature; The detection and analysis result acquisition module calls up a pre-trained detection model, inputs the associated key parameter set and the hardware detection result into the detection model to identify potential defect patterns, and is used to obtain corresponding detection and analysis results, and generate a detailed test report based on the detection and analysis results, wherein the detailed test report includes specific improvement measures or suggestions.

9. An intelligent testing method for acoustic production defects, characterized in that: It comprises a processor, in which a program of the intelligent testing method for acoustic production defects as described in any one of claims 1 to 7 is run.

10. A storage medium, characterized in that: A program for an intelligent testing method for acoustic production defects as described in any one of claims 1 to 7 is stored.

Images