A high-precision management method and system for managing the life of a probe card

Abstract

The present invention discloses a high-precision management method and system for managing the life of a probe card, and relates to the field of probe cards. The high-precision management method for managing the life of a probe card comprises the following steps: S1, constructing a probe card database; S2, performing electrical detection on the probe card; S3, calculating the wear value of the probe card, and predicting the service life value of the probe card; S4, optimizing and adjusting the predicted service life value of the probe card; S5, outputting the optimized and adjusted predicted service life value and wear value of the probe card, and optimizing and adjusting the use process of the probe card based on the wear value of the probe card. The present invention constructs a probe card database, collects environmental influencing factors, electrical characteristics and wear data, and then uses these data to predict the wear and service life value, thereby improving the accuracy of the prediction, and the meticulous data analysis and model matching ensure that the prediction results are closely related to the actual situation.

Description

Technical Field

[0001] The present invention relates to the field of probe cards, and in particular to a high-precision management method and system for managing the life of a probe card. Background Art

[0002] The probe card is a key interface device used to connect the test equipment and the chip to be tested during the semiconductor manufacturing process, especially in the wafer testing stage before chip packaging. Its basic function is to ensure that the test signal is accurately transmitted from the tester to the specific test point on the chip, and also transmits the chip's response signal back to the tester, allowing manufacturers to comprehensively test the performance of the chip before it is put on the market, identify and eliminate unqualified chips.

[0003] As the number of uses increases, the probes of the probe card may wear, deform or be damaged, affecting the quality of contact between the probe card and the test points on the wafer and reducing the accuracy and reliability of the test results. High-precision management is crucial in the field of probe cards, which not only involves precise control of the production process, equipment performance and data analysis, but also product quality, safety, efficiency and cost control. At the same time, high-precision management helps companies stay ahead in the fierce market competition and gain competitive advantages by providing high-quality products and services that exceed industry standards.

[0004] However, the existing methods for managing the life of probe cards are unable to accurately obtain the wear conditions and predict the life of the probe cards during use, resulting in unsatisfactory accuracy in data management of the probe cards during use. In addition, the use process of the probe cards cannot be optimized based on the results of the analysis of the wear values ​​of the probe cards and environmental factors, which greatly reduces the effectiveness of the methods for managing the life of probe cards during use. As a result, the accuracy and effectiveness of the existing methods for managing the life of probe cards during use are unsatisfactory.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a high-precision management method and system for managing the life of a probe card to overcome the above-mentioned technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] According to one aspect of the present invention, there is provided a high-precision management method for managing the life of a probe card, comprising the following steps:

[0009] S1. Obtaining the regional environmental influencing factors and the use process of the probe card when the probe card is used, and building a probe card database based on the pre-collected probe card parameters and the probe card historical parameters;

[0010] S2, presetting a probe card detection process, and performing electrical detection on the probe card based on the probe card detection process;

[0011] S3, establishing a wear analysis model, bringing the electrical detection results into the wear analysis model to calculate the wear value of the probe card, and predicting the service life of the probe card based on the wear value of the probe card to obtain the predicted service life of the probe card;

[0012] S4, comparing the predicted service life value of the probe card with the actual service life value of the probe card in the probe card database, and optimizing and adjusting the predicted service life value of the probe card based on the comparison result;

[0013] S5. Output the predicted service life value and wear value of the probe card after optimization and adjustment, and optimize and adjust the use process of the probe card based on the wear value of the probe card.

[0014] As a preferred solution, a probe card detection process is preset, and electrical detection of the probe card based on the probe card detection process includes the following steps:

[0015] S21, defining a probe card electrical characteristic detection target, and configuring a probe card detection instrument based on the probe card detection target;

[0016] S22, setting a probe card detection process according to an electrical characteristic detection target, and detecting electrical characteristics of the probe card based on the probe card detection process and a probe card detection instrument;

[0017] S23 , verifying and adjusting the electrical characteristic test result of the probe card, recording and storing the electrical characteristic test result of the probe card after verification and adjustment, and outputting the stored electrical characteristic test result of the probe card.

[0018] As a preferred solution, a wear analysis model is established, the electrical detection results are brought into the wear analysis model to calculate the wear value of the probe card, and the service life value of the probe card is predicted based on the wear value of the probe card. Obtaining the predicted service life value of the probe card includes the following steps:

[0019] S31, data cleaning the electrical characteristic test results of the probe card, and sorting the electrical characteristic test results of the probe card after data cleaning based on a time series;

[0020] S32, based on the electrical characteristic test result of the probe card after data cleaning, obtaining the electrical characteristic factor of the probe card of the electrical characteristic test result of the probe card after data cleaning by feature extraction;

[0021] S33, presetting a wear analysis model library and feature matching rules, and matching the electrical feature factors of the probe card with the wear analysis model in the wear analysis model library based on the feature matching rules;

[0022] S34, bringing the electrical characteristic factors of the probe card into a matching wear analysis model to calculate the wear value of the probe card;

[0023] S35. Analyze the wear value of the probe card using a statistical analysis model according to the wear value of the probe card, and predict the service life value of the probe card based on the analysis result.

[0024] As a preferred solution, presetting a wear analysis model library and feature matching rules, and matching the electrical feature factors of the probe card with the wear analysis model in the wear analysis model library based on the feature matching rules includes the following steps:

[0025] S331, presetting a wear analysis model library and multi-dimensional matching rules, defining a probe card wear mechanism, and classifying wear analysis models within the wear analysis model library based on the probe card wear mechanism;

[0026] S332, quantifying the electrical characteristic factors of the probe card through characteristic conversion, and analyzing the change trend of the electrical characteristics of the probe card based on the quantification results;

[0027] S333, matching the wear analysis model in the wear analysis model library through a multi-dimensional matching rule based on the change trend of the electrical characteristics of the probe card and the time series sorting result of the electrical characteristics detection result of the probe card;

[0028] S334, verifying the matching value between the electrical characteristic factors of the probe card and the wear analysis model, and optimizing and adjusting the wear analysis model based on the matching value.

[0029] As a preferred solution, quantifying the electrical characteristic factors of the probe card by characteristic conversion, and analyzing the change trend of the electrical characteristics of the probe card based on the quantification results includes the following steps:

[0030] S3321, normalizing the electrical characteristic factors of the probe card, and performing quantitative feature extraction on the normalized electrical characteristic factors of the probe card;

[0031] S3322, analyzing the mutual influence parameters of the electrical characteristics of the probe card among the electrical characteristic factors of the probe card;

[0032] S3323. Based on the quantitative feature extraction results and the mutual influence parameters, the electrical feature change trend of the probe card is obtained through a regression analysis algorithm.

[0033] As a preferred solution, the electrical characteristic factors of the probe card are brought into the matching wear analysis model to calculate the wear value of the probe card:

[0034] ;

[0035] in, W is the probe card wear value;

[0036] is the intercept coefficient of the wear analysis model;

[0037] It is the influence coefficient of the first electrical characteristic factor of the wear analysis model;

[0038] It is the influence coefficient of the second electrical characteristic factor of the wear analysis model;

[0039] is the influence coefficient of the nth electrical characteristic factor of the wear analysis model;

[0040] It is the first electrical characteristic factor value among the electrical characteristic factors of the probe card;

[0041] It is the second electrical characteristic factor value among the electrical characteristic factors of the probe card;

[0042] is the nth electrical characteristic factor value among the electrical characteristic factors of the probe card;

[0043] is the error term of the wear analysis model.

[0044] As a preferred solution, according to the wear value of the probe card, the wear value of the probe card is analyzed by a statistical analysis model, and the service life value of the probe card is predicted based on the analysis result, including the following steps:

[0045] S351, constructing a statistical analysis model library and matching screening rules, and matching the probe card wear value with the statistical analysis model in the statistical analysis model library based on the matching screening rules;

[0046] S352, analyzing the influence of regional environmental factors on the probe card according to the wear value of the probe card, and obtaining the wear environmental factors of the probe card;

[0047] S353, inputting the probe card wear value and the probe card wear environment factor into a statistical analysis model to predict the probe card service life value;

[0048] S354, verifying the optimized predicted service life value of the probe card, and outputting the verified and optimized predicted service life value of the probe card.

[0049] As a preferred solution, according to the wear value of the probe card, analyzing the influence of the regional environmental factors on the probe card, and obtaining the wear environmental factors of the probe card includes the following steps:

[0050] S3521. Define regional environmental impact rules and obtain regional environmental impact factors based on the regional environmental impact rules;

[0051] S3522. Obtaining an environmental wear impact correlation value through statistical testing according to regional environmental impact factors and probe card wear values;

[0052] S3523, establishing an environmental impact model, and inputting the regional environmental impact factors and environmental wear impact association values ​​into the environmental impact model to obtain the impact of the probe card;

[0053] S3524. Perform quantitative analysis based on the impact of the probe card to obtain the wear environment factors of the probe card.

[0054] As a preferred solution, quantitative analysis is performed based on the influence of the probe card to obtain the wear environment factors of the probe card, including the following steps:

[0055] S35241. quantify the impact of the probe card, and analyze the impact value of the environmental impact factor in the impact of the probe card based on the quantified result of the impact of the probe card;

[0056] S35242, sorting the impact values ​​of the environmental impact factors in the probe card impact situation in order of magnitude;

[0057] S35243. According to the ranking results of the environmental impact factors, set the environmental sensitivity threshold of the environmental impact factors;

[0058] S35244. Set the influencing factor classification rules, and compare the influencing factor classification rules with the environmental sensitivity threshold, and adjust the probe card wear environment factors according to the comparison results.

[0059] According to another aspect of the present invention, there is provided a high-precision management system for managing the life of a probe card, the system comprising:

[0060] A data acquisition module, used to acquire the regional environmental influencing factors and the use process of the probe card when the probe card is used, and to build a probe card database based on the pre-collected probe card parameters and the probe card historical parameters;

[0061] An electrical detection module, used for presetting a probe card detection process and performing electrical detection on the probe card based on the probe card detection process;

[0062] A wear analysis module is used to establish a wear analysis model, bring the electrical detection results into the wear analysis model to calculate the wear value of the probe card, and predict the service life value of the probe card based on the wear value of the probe card to obtain the predicted service life value of the probe card;

[0063] A life prediction module, used to compare the predicted service life value of the probe card with the actual service life value of the probe card in the probe card database, and optimize and adjust the predicted service life value of the probe card based on the comparison result;

[0064] An output optimization module, used for outputting the predicted service life value and the wear value of the probe card after optimization and adjustment, and optimizing and adjusting the use process of the probe card based on the wear value of the probe card;

[0065] The data acquisition module, the electrical detection module, the wear analysis module, the life prediction module and the output optimization module are connected.

[0066] The beneficial effects of the present invention are:

[0067] 1. The present invention constructs a probe card database to collect environmental factors, electrical characteristics and wear data, and then uses these data to predict wear and service life values, thereby improving the accuracy of the prediction. The meticulous data analysis and model matching ensure that the prediction results are closely related to the actual situation. According to the results of the probe card wear value and environmental factor analysis, the use process of the probe card is optimized to reduce wear and extend the service life value. At the same time, by extending the service life value of the probe card and improving its use efficiency, the replacement frequency caused by excessive wear is reduced, thereby reducing the overall cost and reducing excessive maintenance and inspection due to inaccurate predictions.

[0068] 2. The present invention maintains the excellent performance of the probe card through precise wear analysis and life prediction, ensures the quality and reliability of the test data, and makes the management of the probe card more responsive to market and technological changes by regularly updating and optimizing the prediction model. It also improves the flexibility of the entire testing process by adapting to environmental changes and new testing requirements. At the same time, by taking regional environmental factors into consideration, the external factors that affect the performance and life of the probe card are fully understood, so that adaptive measures can be taken to reduce adverse effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0070] Figure 1 is a method flow chart of a high-precision management method for managing the life of a probe card according to an embodiment of the present invention;

[0071] Figure 2 4 is a system block diagram of a high-precision management system for managing the life of a probe card according to an embodiment of the present invention.

[0072] In the figure:

[0073] 1. Data acquisition module; 2. Electrical detection module; 3. Wear analysis module; 4. Life prediction module; 5. Output optimization module. DETAILED DESCRIPTION

[0074] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0075] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] According to an embodiment of the present invention, a high-precision management method and system for managing the life of a probe card are provided.

[0077] The present invention is further described with reference to the accompanying drawings and specific embodiments. According to one embodiment of the present invention, Figure 1 As shown, the high-precision management method for managing the life of a probe card according to an embodiment of the present invention includes the following steps:

[0078] S1. Obtaining the regional environmental influencing factors and the use process of the probe card when the probe card is used, and building a probe card database based on the pre-collected probe card parameters and the probe card historical parameters;

[0079] Specifically, collect environmental data of the place where the probe card is used, including temperature, humidity, dust level, chemical exposure, etc., and record the use process of the probe card, including test frequency, test duration, voltage and current, cleaning and maintenance procedures, etc., and integrate the collected environmental data and usage process data into a unified data management system to ensure the accessibility and analyzability of the data. The pre-collected probe card parameters include the physical properties of the probe card, such as material, size, design specifications, and electrical properties, such as resistance, capacitance, inductance, etc., and then extract key parameters from the historical usage records of the probe card, such as wear degree, failure history, maintenance records, etc.

[0080] Use a database management system, such as a database, to build a probe card database, design the database structure, including tables, fields, and relationships, to store environmental data, usage process data, probe card parameters, and historical parameters, ensure that the database has good data integrity, consistency, and security, and then clean and preprocess the collected data to ensure data accuracy and completeness, apply data analysis techniques, such as statistical analysis and data mining, to identify the relationship between probe card performance and environmental and process factors, and based on the data analysis results, establish a relationship model between probe card performance and environmental and process factors, and regularly update the database to reflect the latest probe card usage and environmental changes, maintain database performance and security, and ensure long-term data availability.

[0081] S2, presetting a probe card detection process, and performing electrical detection on the probe card based on the probe card detection process;

[0082] Specifically, presetting a probe card detection process, and performing electrical detection on the probe card based on the probe card detection process includes the following steps:

[0083] S21, defining a probe card electrical characteristic detection target, and configuring a probe card detection instrument based on the probe card detection target;

[0084] Specifically, determine the electrical characteristics detection targets of the probe card, such as resistance, capacitance, inductance, impedance, signal integrity, etc., and refer to industry standards or specifications to ensure that the detection targets meet industry requirements. Consider the specific application scenarios of the probe card, such as high-speed signal testing, high-density integrated circuit testing, etc., determine appropriate electrical characteristics indicators, and select instruments that can meet the detection targets, such as high-precision multimeters, impedance analyzers, signal generators, etc. Ensure that the instrument's measurement range, resolution, accuracy and other specifications can meet the requirements of the detection targets, consider the instrument's interface and software compatibility, and ensure that it can be seamlessly integrated with the existing test system.

[0085] Calibrate the instrument according to the instructions to ensure the accuracy of the measurement results, then set the instrument's measurement parameters according to the test objectives, such as frequency, voltage range, measurement points, etc., develop or select appropriate test procedures to automate the measurement process, improve efficiency and consistency, and perform pre-tests before actual application to verify whether the instrument configuration is correct and ensure the accuracy of the measurement results, evaluate the consistency of the measurement results with the expected goals, and make adjustments if necessary.

[0086] S22, setting a probe card detection process according to an electrical characteristic detection target, and detecting electrical characteristics of the probe card based on the probe card detection process and a probe card detection instrument;

[0087] Specifically, clarify the specific goals of electrical characteristics testing, such as measuring the resistance, capacitance, inductance, etc. of the probe card, design the testing process according to the testing goals, including the test sequence, test environment, test conditions, etc., select instruments that can meet the testing goals, such as impedance analyzers, signal generators, oscilloscopes, etc., configure the instrument's measurement range, resolution, accuracy and other parameters to ensure that the required electrical characteristics can be accurately measured, and then ensure that the test environment meets the testing requirements, such as temperature, humidity, anti-static, etc., prepare the tools and materials required for the test, such as probe card fixtures, test boards, connecting wires, etc.

[0088] Connect the probe card to the test instrument correctly to ensure that the connection is stable and reliable. Set the measurement parameters of the instrument according to the test target, such as frequency, voltage range, measurement points, etc., start the test program, execute the measurement process, obtain the actual measurement values ​​of the electrical characteristics, record the measurement results, including values, measurement time, environmental conditions, etc., and analyze the measurement results to evaluate whether the electrical characteristics of the probe card meet the design requirements, verify the credibility and accuracy of the measurement results, repeat the measurement or calibration when necessary, and adjust the test process or instrument configuration according to the measurement results.

[0089] S23 , verifying and adjusting the electrical characteristic test result of the probe card, recording and storing the electrical characteristic test result of the probe card after verification and adjustment, and outputting the stored electrical characteristic test result of the probe card.

[0090] Specifically, first review the original test data to check whether there are outliers, missing data or obvious errors, compare the test results with known standards or expected values, and verify the rationality of the results. If the test results deviate significantly from expectations, consider calibrating the testing equipment to ensure that the equipment is accurate, repeat the tests on questionable results and verify consistency, mark or eliminate outliers to prevent them from affecting the overall analysis, smooth the data points with large fluctuations to improve the credibility of the data, and correct the data with large deviations based on experience and expertise.

[0091] Record the verified and adjusted data in detail, including metadata such as test conditions, test time, and testers. Store the adjusted data in a database to ensure long-term preservation and traceability of the data. Create a detailed test report, including test results, verification and adjustment process, conclusions and suggestions. Display the test results through charts or graphs to make the data more intuitive and easy to understand. Share the test report and results with relevant stakeholders, such as engineers, quality management personnel, etc.

[0092] S3, establishing a wear analysis model, bringing the electrical detection results into the wear analysis model to calculate the wear value of the probe card, and predicting the service life of the probe card based on the wear value of the probe card to obtain the predicted service life of the probe card;

[0093] Specifically, a wear analysis model is established, the electrical detection results are brought into the wear analysis model to calculate the wear value of the probe card, and the service life value of the probe card is predicted based on the wear value of the probe card. Obtaining the predicted service life value of the probe card includes the following steps:

[0094] S31, data cleaning the electrical characteristic test results of the probe card, and sorting the electrical characteristic test results of the probe card after data cleaning based on a time series;

[0095] Specifically, the electrical characteristics test data of the probe card is imported into a data analysis tool or database, and a preliminary review is conducted to understand the overall situation of the data, including data format, range, missing values, etc., identify and process missing data points, choose to fill missing values, delete records with missing values, or use interpolation methods, then identify and evaluate outliers, and delete, replace or retain outliers. The measures taken depend on the cause of the outliers and the context of the data. The data is standardized, such as unifying units, scaling data to the same range, etc., non-numeric data is converted to numerical data, such as converting categorical data into numerical codes, and duplicate records are deleted to ensure the uniqueness of the data.

[0096] Make sure that the data set contains a timestamp or date field that records the time of each detection. Sort the data according to the timestamp to ensure that the order of the data corresponds to the actual detection time. After cleaning and sorting the data, perform data verification to ensure the accuracy and completeness of the data. Use visualization tools such as line charts and scatter plots to visually check the time series pattern of the data, and then output the cleaned and sorted data to a new data set or database.

[0097] S32, based on the electrical characteristic test result of the probe card after data cleaning, obtaining the electrical characteristic factor of the probe card of the electrical characteristic test result of the probe card after data cleaning by feature extraction;

[0098] Specifically, ensure that the data cleaning process has been completed, the data quality meets the analysis requirements, and ensure that the data set contains complete time series information. Perform preliminary exploration of the data, including descriptive statistics, drawing charts such as scatter plots, box plots, etc., identify potential electrical features, and then use expert knowledge to identify key features related to the electrical characteristics of the probe card. Select appropriate feature extraction methods, such as statistical methods, signal processing methods, or machine learning methods. When calculating electrical features, use the frequency components, time domain statistics, power spectral density, etc. of the signal. Analyze the correlation between features, eliminate highly correlated features, avoid multicollinearity problems, and then use statistical tests or machine learning models to evaluate the importance of features. According to business needs and model performance, set feature selection thresholds, such as information gain, Gini coefficient, etc. Use cross-validation methods to evaluate the stability and reliability of the feature extraction and screening process. By establishing a prediction model and evaluating model performance, the extracted and screened features are stored in a database or data warehouse to ensure the reusability and manageability of the features, and maintain feature metadata, including feature definitions, calculation methods, data types, sources, etc.

[0099] S33, presetting a wear analysis model library and feature matching rules, and matching the electrical feature factors of the probe card with the wear analysis model in the wear analysis model library based on the feature matching rules;

[0100] Specifically, presetting a wear analysis model library and feature matching rules, and matching the electrical feature factors of the probe card with the wear analysis model in the wear analysis model library based on the feature matching rules includes the following steps:

[0101] S331, presetting a wear analysis model library and multi-dimensional matching rules, defining a probe card wear mechanism, and classifying wear analysis models within the wear analysis model library based on the probe card wear mechanism;

[0102] Specifically, determine the required type of wear analysis model, such as statistical model, machine learning model, physical model, etc., consider the characteristics that the model should have, such as the ability to handle nonlinear relationships and time series analysis capabilities, investigate relevant academic literature, industry reports and technical standards, collect existing wear analysis models, and consult field experts to obtain effective models used in practice, organize the collected models, ensure the integrity and availability of the models, and then determine the feature dimensions used for matching, such as wear rate, material properties, usage environment, etc., select or develop matching algorithms, such as nearest neighbor algorithms, clustering algorithms, decision trees, etc., set thresholds and weights for matching rules, and guide model selection and classification.

[0103] Study the physical and chemical changes that may occur in the probe card during operation, such as material fatigue, corrosion, wear, etc., identify the main factors that affect the wear of the probe card, such as temperature, humidity, frequency of use, etc., map the wear mechanism with the models in the model library, determine which models can simulate specific wear mechanisms, and formulate classification standards such as simple models, medium models, and complex models according to the complexity of the wear mechanism and the capabilities of the model. Then verify the accuracy of the model through experimental data or historical data, and adjust the model classification according to the verification results, optimize the matching and use of the model, and create detailed documents for each model and classification, including model description, applicable conditions, usage guidelines, etc., regularly update the model library, introduce new models and research results, and maintain the timeliness and practicality of the model library.

[0104] S332, quantifying the electrical characteristic factors of the probe card through characteristic conversion, and analyzing the change trend of the electrical characteristics of the probe card based on the quantification results;

[0105] Specifically, quantifying the electrical characteristic factors of the probe card by characteristic conversion, and analyzing the change trend of the electrical characteristics of the probe card based on the quantification results include the following steps:

[0106] S3321, normalizing the electrical characteristic factors of the probe card, and performing quantitative feature extraction on the normalized electrical characteristic factors of the probe card;

[0107] Specifically, select an appropriate normalization method, such as minimum-maximum normalization, Z-score normalization, maximum absolute value normalization, etc. According to the selected method, calculate the parameters required for normalization, such as maximum value, minimum value, mean and standard deviation, normalize each feature, scale the feature value to a predefined range, such as [0, 1] or a mean of 0 and a standard deviation of 1, select features that have a significant impact on the electrical characteristics of the probe card from the normalized features, convert non-numerical features into numerical features, such as using one-hot encoding to process categorical features, create new features, such as combining existing features to capture more complex patterns and relationships, scale the quantized features, such as using standardization methods to eliminate the impact of different dimensions, use statistical tests to verify the importance of the features, and then verify the validity of the features by building a predictive model and evaluating the model performance, and check the stability of the features in different data sets and time points.

[0108] Store the quantitative features in a database or feature storage system, record the metadata of the features, including feature name, description, data type, source, etc., use the quantitative features to train machine learning models, such as random forests and support vector machines, evaluate the performance and generalization ability of the models through methods such as cross-validation, and deploy the trained models to the production environment for actual electrical characteristics analysis of probe cards.

[0109] S3322, analyzing the mutual influence parameters of the electrical characteristics of the probe card among the electrical characteristic factors of the probe card;

[0110] Specifically, ensure that the electrical characteristic data of the probe card has been cleaned and normalized, and the data quality meets the analysis requirements. Ensure that the data set contains a complete feature set, including all relevant electrical characteristics. Calculate the Pearson correlation coefficient, Spearman rank correlation coefficient or Kendall rank correlation coefficient between the features, evaluate the linear or nonlinear relationship between the features, visualize the correlation coefficient through a heat map, intuitively display the mutual influence between the features, and build a multivariate linear regression model to analyze the relationship between a dependent variable and multiple independent variables. Analyze whether different features have a significant impact on the electrical characteristics of the probe card. Use machine learning models to evaluate the importance of features, and use model interpretability tools to analyze the contribution of features to model predictions.

[0111] Based on previous knowledge and experience, propose hypotheses about the mutual influence of features, use statistical tests to verify these hypotheses, display the relationship between features through visualization tools such as scatter plots, box plots, parallel coordinates plots, etc., write detailed reports, summarize the analysis results, including the parameters and conclusions of the mutual influence of features, verify the accuracy of the analysis model through cross-validation or leaving out validation sets, and continuously iterate and improve the analysis model based on new data or feedback.

[0112] S3323. Based on the quantitative feature extraction results and the mutual influence parameters, the electrical feature change trend of the probe card is obtained through a regression analysis algorithm.

[0113] Specifically, ensure that the quantitative feature data is cleaned, normalized and standardized, ensure that the data set contains complete time series information, use the correlation coefficient to analyze the correlation between features, eliminate highly correlated features, and then use the machine learning model to evaluate the importance of features, select features that have a significant impact on the prediction target, and select appropriate regression models based on the characteristics of the probe card electrical characteristics, such as linear regression, polynomial regression, ridge regression or lasso regression, etc. Use time series regression models such as autoregressive models, moving average models or autoregressive moving average models. If the features change over time, use historical data to train the regression model, use the quantitative features as independent variables, and the probe card electrical characteristics as dependent variables, and then adjust the model parameters, such as regularization strength, polynomial order, etc., to optimize model performance.

[0114] Evaluate the predictive power of the model through cross-validation or leaving out a validation set, use statistical metrics to measure the performance of the model, use the trained regression model to make predictions on new or future data, and analyze the prediction results to identify the changing trends of the electrical characteristics of the probe card.

[0115] S333, matching the wear analysis model in the wear analysis model library through a multi-dimensional matching rule based on the change trend of the electrical characteristics of the probe card and the time series sorting result of the electrical characteristics detection result of the probe card;

[0116] Specifically, through regression analysis or other time series analysis methods, the changing trend of the electrical characteristics of the probe card is identified, and the key features of the changing trend, such as slope, inflection point, periodicity, etc., are described. The time series analysis of the electrical characteristics test results of the probe card is performed to identify characteristics such as periodicity, trend and seasonality. According to the time series analysis results, the sorting rules of the electrical characteristics test results of the probe card are determined, and then the dimensions used for matching, such as wear rate, material properties, usage environment, etc., are determined. Weights are assigned to each dimension to reflect its importance to the selection of the wear analysis model. According to the changing trend of the electrical characteristics of the probe card and the time series sorting results, potential matching wear analysis models are screened out, and then multi-dimensional matching rules are applied to match the screened models to select the most appropriate model.

[0117] Prepare a validation data set, including historical wear data, electrical characteristics test results, etc. Use the validation data set to evaluate the performance of the matching wear analysis model, adjust model parameters or select other models based on the validation results, and deploy the validated wear analysis model to actual applications for wear monitoring and prediction of probe cards. Monitor the status of probe cards in real time and update the model regularly to adapt to new data.

[0118] S334, verifying the matching value between the electrical characteristic factors of the probe card and the wear analysis model, and optimizing and adjusting the wear analysis model based on the matching value.

[0119] Specifically, ensure that the electrical characteristic data of the probe card is cleaned, normalized and standardized, ensure that the data set contains a complete feature set, including all relevant electrical characteristics, extract key features from the electrical characteristic data of the probe card, use the wear analysis model to predict the characteristic data, calculate the difference between the predicted value and the actual value, obtain the matching value, use statistical tests to verify the significance of the matching value, and use performance indicators to evaluate the matching degree of the model.

[0120] According to the matching value and verification results, adjust the model parameters, such as regularization strength, learning rate, etc., re-evaluate the importance of features, select features that have a significant impact on the prediction target, consider using different model structures or algorithms, such as switching from linear models to nonlinear models, evaluate the performance and generalization ability of the optimized model through cross-validation, use an independent verification data set to evaluate the matching value and performance of the model, deploy the optimized wear analysis model to actual applications, monitor the status of the probe card in real time, and update the model regularly to adapt to new data.

[0121] S34, bringing the electrical characteristic factors of the probe card into a matching wear analysis model to calculate the wear value of the probe card;

[0122] Specifically, the electrical characteristic factors of the probe card are brought into the matching wear analysis model to calculate the wear value of the probe card. The calculation formula is:

[0123] ;

[0124] in, W is the probe card wear value;

[0125] is the intercept coefficient of the wear analysis model;

[0126] It is the influence coefficient of the first electrical characteristic factor of the wear analysis model;

[0127] It is the influence coefficient of the second electrical characteristic factor of the wear analysis model;

[0128] is the influence coefficient of the nth electrical characteristic factor of the wear analysis model;

[0129] It is the first electrical characteristic factor value among the electrical characteristic factors of the probe card;

[0130] It is the second electrical characteristic factor value among the electrical characteristic factors of the probe card;

[0131] is the nth electrical characteristic factor value among the electrical characteristic factors of the probe card;

[0132] is the error term of the wear analysis model.

[0133] S35. Analyze the wear value of the probe card using a statistical analysis model according to the wear value of the probe card, and predict the service life value of the probe card based on the analysis result.

[0134] Specifically, according to the wear value of the probe card, the wear value of the probe card is analyzed by a statistical analysis model, and based on the analysis result, the service life value of the probe card is predicted, including the following steps:

[0135] S351, constructing a statistical analysis model library and matching screening rules, and matching the probe card wear value with the statistical analysis model in the statistical analysis model library based on the matching screening rules;

[0136] Specifically, various statistical analysis models are collected, including linear regression, logistic regression, time series analysis, survival analysis, etc. The collected models are organized, including model description, applicable conditions, parameter settings, etc. The models are verified using historical data to ensure the accuracy and stability of the models, and the features used for matching are determined, such as wear rate, material properties, usage environment, etc. Matching rules are set according to the features, such as thresholds and weights, and historical data are used to verify the effectiveness of the matching rules. Wear value data of the probe card is collected, including wear degree, wear rate, etc. The collected data is cleaned to remove outliers and noise.

[0137] Extract key features from the probe card wear value data, apply matching screening rules, screen and classify the probe card wear values, select the most appropriate statistical analysis model based on the screening results, verify the matched statistical analysis model using the validation data set, use performance indicators to evaluate the performance of the model, adjust the model based on the validation results, such as adjusting parameters, changing the model structure, etc., deploy the matched and verified statistical analysis model to actual applications, monitor the status of the probe card in real time, and regularly update the model to adapt to new data.

[0138] S352, analyzing the influence of regional environmental factors on the probe card according to the wear value of the probe card, and obtaining the wear environmental factors of the probe card;

[0139] Specifically, according to the wear value of the probe card, the influence of the regional environmental factors on the probe card is analyzed, and obtaining the wear environmental factors of the probe card includes the following steps:

[0140] S3521. Define regional environmental impact rules and obtain regional environmental impact factors based on the regional environmental impact rules;

[0141] Specifically, study existing environmental impact assessment methods and cases, understand the main factors affecting the regional environment, consult experts in environmental science, urban planning, ecology and other fields to obtain their views on the factors affecting the regional environment, identify the main factors that may affect the regional environment, such as industrial emissions, traffic flow, land use, population density, etc., and formulate specific rules and standards based on the identified environmental factors, such as emission restrictions, traffic management measures, land use planning, etc., verify the effectiveness of the formulated environmental impact rules through case studies or simulation experiments, collect data related to regional environmental impact factors, including statistical data, monitoring data, satellite images, etc., use data analysis methods, such as statistical analysis, geographic information system analysis, etc., to analyze the collected data, and identify the environmental factors that have a significant impact on the regional environment based on the data analysis results.

[0142] Incorporate the formulated environmental impact rules into regional development plans, policies and regulations, establish an environmental monitoring system, regularly monitor changes in environmental factors, and provide feedback and adjustments to the environmental impact rules based on the monitoring results to cope with environmental changes. Record the entire process of formulating environmental impact rules and obtaining environmental impact factors, including the methods, data and results used, and write a detailed report summarizing the identification results of environmental impact rules and environmental impact factors.

[0143] S3522. Obtaining an environmental wear impact correlation value through statistical testing according to regional environmental impact factors and probe card wear values;

[0144] Specifically, collect data on regional environmental influencing factors, such as temperature, humidity, pollutant concentration, etc., as well as probe card wear value data for the corresponding time period, ensure data integrity, remove missing values ​​and outliers, perform data normalization or standardization processing to eliminate dimensional effects, conduct descriptive statistical analysis on the collected data, understand the basic distribution of the data, and then visualize the relationship between environmental factors and wear values ​​through scatter plots, box plots, etc., preliminarily determine whether there is potential correlation, calculate the correlation coefficient between environmental factors and wear values, and evaluate the linear or nonlinear relationship between the two. If the correlation analysis results show that there is a significant correlation, the impact of environmental factors on wear values ​​can be further quantitatively analyzed through a regression analysis model.

[0145] According to the distribution characteristics of the data and the purpose of the research, select appropriate statistical test methods to verify whether the correlation between environmental factors and wear values ​​is statistically significant. According to the hypothesis test results, define environmental wear impact association values, such as coefficients and determination coefficients in regression analysis, to reflect the degree and direction of the impact of environmental factors on wear values. Explain the obtained environmental wear impact association values ​​in detail to clarify how environmental factors affect probe card wear and the possible mechanisms of the impact. Verify the analysis results through cross-validation or use of independent data sets to ensure the stability and reliability of the results. Perform sensitivity analysis to evaluate the impact of different environmental factors on model results and verify the robustness of the model.

[0146] S3523, establishing an environmental impact model, and inputting the regional environmental impact factors and environmental wear impact association values ​​into the environmental impact model to obtain the impact of the probe card;

[0147] Specifically, the purpose of the model should be clarified, that is, to evaluate the impact of environmental factors on the performance of the probe card. According to the goal and available data, the appropriate model type should be selected, such as linear model, nonlinear model, machine learning model, etc. Based on previous research and data analysis, the environmental factors that have a significant impact on the probe card should be identified, and environmental factor data in relevant areas, such as temperature and humidity, air quality, etc., should be collected. Then, the wear data of the probe card under the corresponding environmental conditions should be collected, including cleaning, standardization or normalization, to ensure that the data quality is suitable for the model input, and the input variables (environmental factors) and output variables (probe card wear or performance indicators) of the model should be defined. According to the framework design, statistical or machine learning methods should be used to construct the environmental impact model. According to the data characteristics and model requirements, appropriate model parameters should be set.

[0148] The environmental wear impact correlation values ​​are used as the input or weight of the model, and the model is adjusted to reflect the impact of these correlation values ​​on the performance of the probe card. The model is optimized based on the correlation values ​​to improve the accuracy and sensitivity of the model's response to environmental changes. The model is trained using historical data sets to determine the model parameters and structure. The predictive ability and accuracy of the model are then verified through cross-validation or the use of independent test sets. The established environmental impact model is used to input current or predicted environmental factors to obtain the impact of the probe card, analyze the model output results, understand the specific impact of different environmental factors on the performance of the probe card, and based on the model results, formulate or adjust the use and maintenance strategies of the probe card to cope with environmental impacts.

[0149] S3524. Perform quantitative analysis based on the impact of the probe card to obtain the wear environment factors of the probe card.

[0150] Specifically, the quantitative analysis based on the influence of the probe card to obtain the wear environment factors of the probe card includes the following steps:

[0151] S35241. quantify the impact of the probe card, and analyze the impact value of the environmental impact factor in the impact of the probe card based on the quantified result of the impact of the probe card;

[0152] Specifically, performance indicators that reflect the impact of the probe card are determined, such as wear rate, electrical performance parameters such as resistance and capacitance, and failure rate, and quantification methods are defined for each performance indicator, such as using the rate of change or absolute change of the performance parameter as the quantification result, and collecting the performance data of the probe card under different environmental conditions and the environmental factor data within the corresponding time period. According to the defined quantitative indicators, the collected performance data are quantified to obtain the quantitative results of the impact of the probe card under various environmental conditions, and the quantitative results are classified and summarized according to environmental factors for subsequent analysis.

[0153] Statistical methods such as the Pearson correlation coefficient and the Spearman rank correlation coefficient are used to analyze the correlation between environmental factors and the quantitative results of the probe card performance indicators. For environmental factors and performance indicators that show significant correlation, regression analysis is performed to quantitatively evaluate the impact intensity of environmental factors. If there is interaction between environmental factors, multivariate analysis methods such as multiple regression analysis and principal component analysis are used to comprehensively evaluate the comprehensive impact of various environmental factors on the performance of the probe card. The analysis results are interpreted to clarify which environmental factors have a significant impact on the performance of the probe card and the extent of the impact. Based on the impact value analysis results, improvement measures or optimization suggestions are proposed, such as adjusting the probe card use environment, optimizing the maintenance plan, etc., to reduce the impact of adverse environmental factors.

[0154] S35242, sorting the impact values ​​of the environmental impact factors in the probe card impact situation in order of magnitude;

[0155] Specifically, ensure that the impact values ​​of various environmental factors on the performance of the probe card have been analyzed and calculated according to the previous steps, including the coefficients in the regression analysis, the correlation coefficients in the correlation analysis, or other quantitative evaluation indicators, and organize the various environmental influencing factors and their corresponding impact values ​​into a list or table, ensure that the correspondence between each environmental factor and its impact value is clear, and sort them according to the absolute value of the impact value. Use the sorting function of the spreadsheet software, or the sorting function in the programming language to implement it. When sorting, pay attention to the positive and negative of the impact value, because positive and negative values ​​indicate different directions of impact, positive values ​​indicate promotion, and negative values ​​indicate inhibition. Then clearly point out which environmental factors have the greatest impact on the performance of the probe card and which have less impact. Pay special attention to the factors with the largest impact values, which are the factors that should be paid most attention to when optimizing the performance of the probe card. According to the sorting results, give priority to controlling and improving the environmental factors that have the greatest impact on performance.

[0156] S35243. According to the ranking results of the environmental impact factors, set the environmental sensitivity threshold of the environmental impact factors;

[0157] Specifically, we need to have an in-depth understanding of the sorting results, identify which environmental factors have a greater impact on the performance of the probe card and which have a smaller impact, evaluate the degree of impact of each environmental factor on the performance of the probe card based on the size of the impact value, and divide the environmental factors into several levels based on the size of the impact value, such as high sensitivity, medium sensitivity, and low sensitivity. Based on the distribution and specific values ​​of the impact values, determine the thresholds for dividing each level. Based on the statistical analysis of the impact values, set specific values ​​as thresholds for distinguishing different sensitivity levels. When setting the thresholds, we should take into account the positive and negative directions of the impact values, because different impact directions may mean different sensitivity responses.

[0158] Use historical data to test the rationality of the thresholds and check whether the classification can accurately reflect the actual impact of environmental factors. Experts in fields such as environmental science and materials science are required to review the set thresholds to ensure their scientificity and practicality. Deploy a monitoring system in the probe card use and storage environment to monitor key environmental factors in real time. When the monitored environmental factors exceed the set sensitivity threshold, take corresponding early warning or adjustment measures, such as adjusting the use environment of the probe card, increasing the maintenance frequency, etc.

[0159] S35244. Set the influencing factor classification rules, and compare the influencing factor classification rules with the environmental sensitivity threshold, and adjust the probe card wear environment factors according to the comparison results.

[0160] Specifically, according to the impact of environmental factors on the performance of the probe card, the environmental factors are divided into high-impact, medium-impact and low-impact categories, and specific rules for dividing each category are defined, such as based on the size of the impact value, the statistical significance of the impact or the actual degree of impact on the performance of the probe card. According to the previous steps, the thresholds of different environmental sensitivities are determined, such as high sensitivity, medium sensitivity and low sensitivity. Based on the impact value analysis, specific numerical thresholds are set to distinguish different sensitivity levels.

[0161] Compare the impact of environmental factors with the corresponding environmental sensitivity thresholds, identify which environmental factors have an impact that exceeds their corresponding sensitivity thresholds, determine the environmental factors that have a greater impact on the performance of the probe card, and for environmental factors with impact values ​​higher than the sensitivity thresholds, consider taking measures to optimize them, such as improving the storage environment or use conditions of the probe card, formulating and implementing specific measures, such as adjusting temperature and humidity control, improving air quality or reducing mechanical vibration, etc., monitor the implementation effect of environmental optimization measures, and evaluate the improvement of probe card performance.

[0162] S353, inputting the probe card wear value and the probe card wear environment factor into a statistical analysis model to predict the probe card service life value;

[0163] Specifically, the wear values ​​of the probe card under different environmental conditions are collected, including the wear rate and degree of wear of the probe, and the environmental factor data corresponding to the wear data, such as temperature, humidity, vibration intensity, etc., are collected. The correlation between the environmental factors and the wear values ​​of the probe card is analyzed to determine which environmental factors have a high correlation with the wear values. Based on the correlation analysis results, the environmental factors that have a significant impact on the wear of the probe card are selected as the input features of the model. According to the data characteristics and analysis objectives, appropriate statistical analysis models are selected, such as linear regression, survival analysis, and machine learning models. The selected features and historical wear data are used to build a model to predict the wear trend and service life value of the probe card.

[0164] The data is divided into a training set and a test set. The training set data is used to train the model, and the test set data is used to verify the model accuracy. The cross-validation method is used to evaluate the stability and generalization ability of the model to avoid overfitting. The probe card wear environmental factors are used as model input to predict the wear rate and final service life of the probe card under specific environmental conditions. The prediction results are analyzed to determine the main environmental factors that affect the service life of the probe card, providing guidance for the use and maintenance of the probe card.

[0165] S354, verifying the optimized predicted service life value of the probe card, and outputting the verified and optimized predicted service life value of the probe card.

[0166] Specifically, the model is used to predict historical cases and compared with the actual service life values. By comparing the prediction results with the actual situation, the accuracy and reliability of the model are evaluated. Cross-validation methods, such as K-fold cross-validation, are used to ensure the robustness of the model. The data set is divided into several parts, the model is trained on one part, and the model is tested on another part. Then the average accuracy of the model is calculated, the proportion of correct predictions by the model is calculated, and the average sum of squares of the differences between the model prediction values ​​and the actual values ​​is evaluated. It is particularly suitable for continuous variables to measure the degree of correlation between the variables predicted by the model and the actual observed variables.

[0167] Re-evaluate the input features, including adding new features, removing irrelevant features, or transforming existing features. Use grid search, random search, and other methods to optimize the model parameters and find the optimal parameter combination. If the current model cannot achieve satisfactory prediction performance, consider using other statistical analysis or machine learning algorithms to verify whether model optimization has improved the accuracy of the prediction. Apply the final optimized model to current or future environmental data, output the predicted service life value of the probe card, and prepare a detailed report, including model description, verification process, performance indicators, and predicted service life value. In actual applications, monitor the prediction results of the model in real time to detect prediction deviations in a timely manner, review model performance regularly, and update the model based on newly collected data and feedback to ensure that the prediction accuracy is maintained at a high level in the long term.

[0168] S4, comparing the predicted service life value of the probe card with the actual service life value of the probe card in the probe card database, and optimizing and adjusting the predicted service life value of the probe card based on the comparison result;

[0169] Specifically, ensure that the probe card database contains the actual service life value data of the probe card and the corresponding environmental factor data, synchronize the predicted service life value data, ensure that the predicted data and the actual service life value data are consistent in timeline and environmental factors, directly compare the predicted service life value of each probe card with its actual service life value, calculate the difference, and use statistical methods to analyze the difference between the predicted life and the actual life, including indicators such as mean error and mean square error, to identify the systematic errors and random errors between the predicted life and the actual life.

[0170] Conduct in-depth analysis on cases with large errors to identify possible causes of inaccurate predictions, such as changes in environmental factors, data quality issues, or model inadaptability. Based on the results of the error analysis, evaluate whether the feature set used by the current model is reasonable and whether it is necessary to introduce new features or exclude certain existing features. Based on the results of error identification and cause analysis, adjust model parameters or try different model algorithms. Use cross-validation methods to re-evaluate the accuracy of the optimized model to ensure that the model optimization is effective. Apply the optimized model to the probe card database, update the predicted service life value, establish a continuous monitoring mechanism, regularly compare the predicted life with the actual life, and promptly discover changes in the predicted performance. Feedback the new comparison results into the model adjustment process to form a continuous optimization cycle. As time goes by, continuously update the actual service life value data in the probe card database for future analysis and model optimization.

[0171] S5. Output the predicted service life value and wear value of the probe card after optimization and adjustment, and optimize and adjust the use process of the probe card based on the wear value of the probe card.

[0172] Specifically, the optimized and adjusted prediction model is applied to the current environmental data of the probe card use to obtain the latest predicted service life value and wear value, and the predicted service life value and wear value are organized into a report or database form to ensure that the data is easy to understand and access. The trend of the predicted service life value and wear value of the probe card is analyzed to identify possible abnormal values ​​or significant changes, and the wear rates of different probe cards are compared to evaluate their performance and identify probe cards with faster wear. The relationship between the wear value and the probe card usage process is analyzed to identify the key process parameters that affect wear.

[0173] Optimize the process parameters that have a greater impact on wear, such as adjusting the test pressure, reducing the number of tests, or optimizing the test procedure. Propose specific process improvement measures based on the wear trend, such as changing the contact method between the probe card and the chip, using more wear-resistant materials, etc. Test the new usage process on some probe cards, monitor the changes in wear values ​​and the impact on probe card performance, compare the wear values ​​before and after adjustment, and evaluate the effect of the process adjustment. If the test results show that the process adjustment can significantly reduce wear and increase the service life value, implement these adjustments on all relevant probe cards, continue to monitor the service life and wear values ​​of the probe cards after the process adjustment, and regularly review and further optimize the probe card usage process based on the monitoring results and new forecast data. Record the process adjustment process, parameter settings, and effect evaluation results in detail to provide guidance for subsequent operation and maintenance, and train relevant personnel on the new process to ensure that the process adjustment is implemented correctly.

[0174] According to another embodiment of the present invention, Figure 2 A high-precision management system for managing the life of a probe card is provided, the system comprising:

[0175] The data acquisition module 1 is used to acquire the regional environmental influencing factors and the use process of the probe card when the probe card is used, and to construct a probe card database based on the pre-collected probe card parameters and the probe card historical parameters;

[0176] An electrical detection module 2, used for presetting a probe card detection process and performing electrical detection on the probe card based on the probe card detection process;

[0177] Wear analysis module 3, used to establish a wear analysis model, bring the electrical detection results into the wear analysis model to calculate the wear value of the probe card, and predict the service life value of the probe card based on the wear value of the probe card to obtain the predicted service life value of the probe card;

[0178] The life prediction module 4 is used to compare the predicted service life value of the probe card with the actual service life value of the probe card in the probe card database, and optimize and adjust the predicted service life value of the probe card based on the comparison result;

[0179] An output optimization module 5 is used to output the predicted service life value and the wear value of the probe card after optimization and adjustment, and optimize and adjust the use process of the probe card based on the wear value of the probe card;

[0180] The data acquisition module 1, the electrical detection module 2, the wear analysis module 3, the life prediction module 4 and the output optimization module 5 are connected.

[0181] To summarize, with the aid of the above-mentioned technical scheme of the present invention, the present invention constructs a probe card database, collects environmental influencing factors, electrical characteristics and wear data, and then uses these data to predict wear and service life values, thereby improving the accuracy of the prediction. The meticulous data analysis and model matching ensure that the prediction results are closely related to the actual situation. According to the results of the analysis of the probe card wear values ​​and environmental factors, the use process of the probe card is optimized to reduce wear and extend the service life value. At the same time, by extending the service life value of the probe card and improving its use efficiency, the replacement frequency caused by excessive wear is reduced, thereby reducing the overall cost and reducing excessive maintenance and inspection due to inaccurate predictions.

[0182] In addition, the present invention maintains the excellent performance of the probe card through precise wear analysis and life prediction, ensures the quality and reliability of the test data, and makes the management of the probe card more responsive to market and technological changes by regularly updating and optimizing the prediction model. It also improves the flexibility of the entire testing process by adapting to environmental changes and new testing requirements. At the same time, by taking regional environmental factors into consideration, the external factors that affect the performance and life of the probe card are fully understood, so that adaptive measures can be taken to reduce adverse effects.

[0183] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A high-precision management method for managing the life of a probe card, characterized in that: The following steps are involved: S1. Obtaining the regional environmental influencing factors and the use process of the probe card when the probe card is used, and building a probe card database based on the pre-collected probe card parameters and the probe card historical parameters; S2, presetting a probe card detection process, and performing electrical detection on the probe card based on the probe card detection process; S3, establishing a wear analysis model, bringing the electrical detection results into the wear analysis model to calculate the wear value of the probe card, and predicting the service life of the probe card based on the wear value of the probe card to obtain the predicted service life of the probe card; The wear analysis model is established, the electrical detection results are brought into the wear analysis model to calculate the wear value of the probe card, and the service life value of the probe card is predicted based on the wear value of the probe card. The acquisition of the predicted service life value of the probe card includes the following steps: S31, data cleaning the electrical characteristic test results of the probe card, and sorting the electrical characteristic test results of the probe card after data cleaning based on a time series; S32, based on the electrical characteristic test result of the probe card after data cleaning, obtaining the electrical characteristic factor of the probe card of the electrical characteristic test result of the probe card after data cleaning by feature extraction; S33, presetting a wear analysis model library and feature matching rules, and matching the electrical feature factors of the probe card with the wear analysis model in the wear analysis model library based on the feature matching rules; S34, bringing the electrical characteristic factors of the probe card into a matching wear analysis model to calculate the wear value of the probe card; S35, analyzing the probe card wear value through a statistical analysis model according to the probe card wear value, and predicting the probe card service life value based on the analysis result; S4, comparing the predicted service life value of the probe card with the actual service life value of the probe card in the probe card database, and optimizing and adjusting the predicted service life value of the probe card based on the comparison result; S5. Output the predicted service life value and wear value of the probe card after optimization and adjustment, and optimize and adjust the use process of the probe card based on the wear value of the probe card.

2. A high-precision management method for managing the life of a probe card according to claim 1, characterized in that: The preset probe card detection process and the electrical detection of the probe card based on the probe card detection process include the following steps: S21, defining a probe card electrical characteristic detection target, and configuring a probe card detection instrument based on the probe card detection target; S22, setting a probe card detection process according to an electrical characteristic detection target, and detecting electrical characteristics of the probe card based on the probe card detection process and a probe card detection instrument; S23 , verifying and adjusting the electrical characteristic test result of the probe card, recording and storing the electrical characteristic test result of the probe card after verification and adjustment, and outputting the stored electrical characteristic test result of the probe card.

3. A high-precision management method for managing the life of a probe card according to claim 1, characterized in that: The preset wear analysis model library and feature matching rules, and matching the electrical feature factors of the probe card with the wear analysis model in the wear analysis model library based on the feature matching rules, include the following steps: S331, presetting a wear analysis model library and multi-dimensional matching rules, defining a probe card wear mechanism, and classifying wear analysis models within the wear analysis model library based on the probe card wear mechanism; S332, quantifying the electrical characteristic factors of the probe card through characteristic conversion, and analyzing the change trend of the electrical characteristics of the probe card based on the quantification results; S333, matching the wear analysis model in the wear analysis model library through a multi-dimensional matching rule based on the change trend of the electrical characteristics of the probe card and the time series sorting result of the electrical characteristics detection result of the probe card; S334, verifying the matching value between the electrical characteristic factors of the probe card and the wear analysis model, and optimizing and adjusting the wear analysis model based on the matching value.

4. A high-precision management method for managing the life of a probe card according to claim 3, characterized in that: The step of quantifying the electrical characteristic factors of the probe card by characteristic conversion and analyzing the change trend of the electrical characteristics of the probe card based on the quantification results comprises the following steps: S3321, normalizing the electrical characteristic factors of the probe card, and performing quantitative feature extraction on the normalized electrical characteristic factors of the probe card; S3322, analyzing the mutual influence parameters of the electrical characteristics of the probe card among the electrical characteristic factors of the probe card; S3323. Based on the quantitative feature extraction results and the mutual influence parameters, the electrical feature change trend of the probe card is obtained through a regression analysis algorithm.

5. A high-precision management method for managing the life of a probe card according to claim 1, characterized in that: The calculation formula for calculating the probe card wear value by bringing the probe card electrical characteristic factors into the matching wear analysis model is: ； in, W is the probe card wear value; is the intercept coefficient of the wear analysis model; It is the influence coefficient of the first electrical characteristic factor of the wear analysis model; It is the influence coefficient of the second electrical characteristic factor of the wear analysis model; is the influence coefficient of the nth electrical characteristic factor of the wear analysis model; It is the first electrical characteristic factor value among the electrical characteristic factors of the probe card; It is the second electrical characteristic factor value among the electrical characteristic factors of the probe card; is the nth electrical characteristic factor value among the electrical characteristic factors of the probe card; is the error term of the wear analysis model.

6. A high-precision management method for managing the life of a probe card according to claim 1, characterized in that: The method of analyzing the probe card wear value by using a statistical analysis model according to the probe card wear value and predicting the probe card service life value based on the analysis result includes the following steps: S351, constructing a statistical analysis model library and matching screening rules, and matching the probe card wear value with the statistical analysis model in the statistical analysis model library based on the matching screening rules; S352, analyzing the influence of regional environmental factors on the probe card according to the wear value of the probe card, and obtaining the wear environmental factors of the probe card; S353, inputting the probe card wear value and the probe card wear environment factor into the statistical analysis model to predict the probe card service life value; S354, verifying the optimized predicted service life value of the probe card, and outputting the verified and optimized predicted service life value of the probe card.

7. A high-precision management method for managing the life of a probe card according to claim 6, characterized in that: Analyzing the influence of regional environmental factors on the probe card according to the wear value of the probe card and obtaining the wear environmental factors of the probe card comprises the following steps: S3521. Define regional environmental impact rules and obtain regional environmental impact factors based on the regional environmental impact rules; S3522. Obtaining an environmental wear impact correlation value through statistical testing according to regional environmental impact factors and probe card wear values; S3523, establishing an environmental impact model, and inputting the regional environmental impact factors and environmental wear impact association values ​​into the environmental impact model to obtain the impact of the probe card; S3524. Perform quantitative analysis based on the impact of the probe card to obtain the wear environment factors of the probe card.

8. A high-precision management method for managing the life of a probe card according to claim 7, characterized in that: The quantitative analysis based on the influence of the probe card to obtain the wear environment factors of the probe card includes the following steps: S35241. quantify the impact of the probe card, and analyze the impact value of the environmental impact factor in the impact of the probe card based on the quantified result of the impact of the probe card; S35242, sorting the impact values ​​of the environmental impact factors in the probe card impact situation in order of magnitude; S35243. According to the ranking results of the environmental impact factors, set the environmental sensitivity threshold of the environmental impact factors; S35244. Set the influencing factor classification rules, and compare the influencing factor classification rules with the environmental sensitivity threshold, and adjust the probe card wear environment factors according to the comparison results.

9. A high-precision management system for managing the life of a probe card, used to implement the high-precision management method for managing the life of a probe card as claimed in any one of claims 1 to 8, characterized in that: The system includes: A data acquisition module, used to acquire the regional environmental influencing factors and the use process of the probe card when the probe card is used, and to build a probe card database based on the pre-collected probe card parameters and the probe card historical parameters; An electrical detection module, used for presetting a probe card detection process and performing electrical detection on the probe card based on the probe card detection process; A wear analysis module is used to establish a wear analysis model, bring the electrical detection results into the wear analysis model to calculate the wear value of the probe card, and predict the service life value of the probe card based on the wear value of the probe card to obtain the predicted service life value of the probe card; A life prediction module, used to compare the predicted service life value of the probe card with the actual service life value of the probe card in the probe card database, and optimize and adjust the predicted service life value of the probe card based on the comparison result; An output optimization module, used for outputting the predicted service life value and the wear value of the probe card after optimization and adjustment, and optimizing and adjusting the use process of the probe card based on the wear value of the probe card; The data acquisition module, the electrical detection module, the wear analysis module, the life prediction module and the output optimization module are connected.

Images