A method for analyzing consistency of test data quality of a resistance detector
The automated process of GRUBBS verification and consistency calculation solves the problems of time-consuming and unreliable resistance test data quality analysis, and achieves efficient and accurate quality consistency management, which is suitable for both single-variable and multi-variable testing scenarios.
Patent Information
- Application Number
- CN202611097501.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-23
- Publication Date
- 2026-11-06
AI Technical Summary
In existing technologies, resistance measurement data quality analysis is time-consuming and the results are unreliable. There is a lack of systematic quality consistency analysis methods, resulting in low utilization of test data and difficulty in supporting the quality control of electronic products.
An automated process combining GRUBBS testing and consistency calculation is adopted to achieve fully automated processing of resistance test data through data filtering, statistical testing, and result display, and to identify and manage abnormal data in a hierarchical manner.
It improves analysis efficiency and result accuracy, optimizes data utilization and quality control, enhances user experience and decision-making efficiency, and is suitable for both univariate and multivariate detection scenarios.
Smart Images
Figure CN123046779A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of data analysis technology, specifically relating to a method for analyzing the consistency of test data quality from a resistance meter. Background Technology
[0002] With the rapid development of the electronics industry, the quality control requirements for various electronic products are becoming increasingly stringent. As a key performance parameter of electronic products, the quality analysis of resistance data is particularly important. Currently, the industry mainly uses resistance meters or resistance testing tools to measure resistance, followed by manual comparison and analysis of the data. This traditional method is prone to losing crucial data information during manual processing, and also suffers from time-consuming analysis and poor reliability. Furthermore, it lacks systematic methods for analyzing the consistency of resistance measurement data, resulting in low data utilization and difficulty in effectively supporting the quality control of electronic products. Summary of the Invention
[0003] To address the aforementioned technical problems, the purpose of this invention is to provide a method for analyzing the consistency of resistance test data. This method, combined with GRUBBS testing, enables automated consistency analysis of test data, resulting in accurate and efficient analysis that effectively improves the product's resistance data consistency management capabilities.
[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for analyzing the consistency of test data quality from a resistance meter includes the following steps: Step 1: Data Acquisition and Screening: The resistance tester completes the acquisition of resistance data of the product being tested. The acquired data is then screened using a pre-determined pass / fail range. The screened data is then collected into either the pass / fail resistance test dataset or the abnormal resistance test dataset. Step 2, Statistical Testing and Analysis: Perform a GRUBBS test on the qualified resistance test dataset to identify outlier data exceeding the critical value, and perform consistency calculations on the outlier data. Step 3: Result Display: Convert the consistency calculation results into quality consistency analysis results and display them on the quality analysis interface of the resistance tester's host computer; Step two, the GRUBBS test and the consistency calculation for outlier sample data, includes the following steps: (a) Calculate the mean and standard deviation of the qualified resistance test dataset, which includes qualified data from the current test and previous tests; (b) Calculate the absolute value of the difference between each data point in the qualified resistance test dataset and the mean, and then divide it by the standard deviation to obtain the corresponding statistic. (c) Set the significance level to 0.05, set the test type to one-sided, and determine the critical value by consulting the GRUBBS test critical value table based on the total number of samples in the qualified resistance test dataset. (d) If the statistical measure of the sample data to be tested is less than the critical value, it is determined that there is no abnormality. If the statistical measure is greater than the critical value, it is calculated whether the data exceeds 25% of the mean of the sample set. If it is greater than 25%, it is determined that the product corresponding to the data exceeds the quality consistency requirements and is listed as a key product for quality control. If it is not greater than 25%, it is determined that although the data is a statistical outlier, it does not exceed the quality consistency requirements.
[0005] Furthermore, the initial sample size of the qualified resistance test dataset is no less than 6 pieces.
[0006] Furthermore, if the sample data is multivariate, the GRUBBS test and consistency calculation process in step two are repeated for each variable.
[0007] Compared with the prior art, the beneficial effects of the present invention are: (1) Improved analysis efficiency and automation: The present invention replaces the traditional manual comparison and analysis method with the automated process of GRUBBS test and consistency calculation, realizing the full-process automated processing of resistance test data from collection and screening to quality consistency judgment, which greatly reduces the cost of manual intervention, significantly improves the efficiency of data analysis, and shortens the quality control cycle. (2) Enhanced accuracy and reliability of analysis results: The present invention uses GRUBBS test to screen outliers in the test data, and combines the consistency judgment standard of 25% deviation of the sample mean to realize the dual verification of data quality, effectively avoids the misjudgment that may be caused by a single statistical method, significantly improves the accuracy and reliability of analysis results, and provides solid data support for the quality control of electronic products. (3) Optimized data utilization and control targeting: The present invention further divides the abnormal data of GRUBBS test into two categories: "abnormal test but qualified consistency" and "exceeding the quality consistency requirements", realizing the refined hierarchical management of data quality. This grading mechanism not only improves the utilization rate of test data, but also accurately locates key products for quality control, making quality management more targeted and effectively reducing production costs and quality risks. (4) Improve user experience and decision-making efficiency: This invention displays the quality consistency analysis results intuitively on the upper computer interface of the resistance tester. Through clear text labels and color reminders, operators can quickly identify abnormal data and key products, greatly improving the response speed and decision-making efficiency of quality control and reducing the possibility of human error. (5) Good scalability and applicability: The analysis method of this invention is not only applicable to single-variable resistance data, but can also be extended to multi-variable testing scenarios. GRUBBS test and consistency calculation are performed on each variable separately, which has good scalability and applicability and can meet the quality testing needs of different types of electronic products. Attached Figure Description
[0008] Figure 1 is a schematic diagram of the overall process for analyzing the consistency of the test results of the resistance detector of the present invention; Figure 2 is a schematic diagram of the preliminary analysis process of the data collected by the resistance detector of the present invention; Figure 3 is a schematic diagram of the univariate outlier consistency test process of the present invention; Figure 4 is a schematic diagram of the quality consistency analysis interface of the present invention. Detailed Implementation
[0009] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.
[0010] See Figure 1-3 A method for analyzing the consistency of test data quality of a resistance meter includes the following steps: Step 1: Data Acquisition and Screening: The resistance tester completes the acquisition of resistance data of the product being tested. The acquired data is then screened using a pre-determined pass / fail range. The screened data is then collected into either the pass / fail resistance test dataset or the abnormal resistance test dataset. Step 2, Statistical Testing and Analysis: Perform a GRUBBS test on the qualified resistance test dataset to identify outlier data exceeding the critical value, and perform consistency calculations on the outlier data. Step 3: Result Display: Convert the consistency calculation results into quality consistency analysis results and display them on the quality analysis interface of the resistance tester's host computer; Step two, the GRUBBS test and the consistency calculation for outlier sample data, includes the following steps: (a) Calculate the mean and standard deviation of the qualified resistance test dataset, which includes qualified data from the current test and previous tests; (b) Calculate the absolute value of the difference between each data point in the qualified resistance test dataset and the mean, and then divide it by the standard deviation to obtain the corresponding statistic. (c) Set the significance level to 0.05, set the test type to one-sided, and determine the critical value by consulting the GRUBBS test critical value table based on the total number of samples in the qualified resistance test dataset. (d) If the statistical measure of the sample data to be tested is less than the critical value, it is determined that there is no abnormality. If the statistical measure is greater than the critical value, it is calculated whether the data exceeds 25% of the mean of the sample set. If it is greater than 25%, it is determined that the product corresponding to the data exceeds the quality consistency requirements and is listed as a key product for quality control. If it is not greater than 25%, it is determined that although the data is a statistical outlier, it does not exceed the quality consistency requirements.
[0011] The initial sample size of the qualified resistance test dataset is no less than 6 pieces.
[0012] If the sample data is multivariate, repeat the GRUBBS test and consistency calculation process in step two for each variable.
[0013] Example 1: Figure 1 This document outlines the overall process for analyzing the consistency of resistance test results. The method described in this implementation is based on a sample of qualified data from the same product model accumulated over historical testing cycles. The sample size should be no less than six samples. For ease of understanding, this process is presented using univariate data; for multivariate samples, the process can be repeated.
[0014] This embodiment includes the following steps: Step 1, such as Figure 2As shown, the host computer sends a test command to the resistance tester to collect the raw resistance data of the product under test. A pre-set range of ±10% of the nominal resistance value is used to initially screen the collected data: data falling within the acceptable range are collected into the acceptable resistance test dataset, while data falling outside the range are directly collected into the abnormal resistance test dataset. If the number of acceptable data sets is less than 6, the system automatically retrieves acceptable test data for the same model of product from the local quality traceability database for the next 3 months to supplement the data, ensuring that the total number of acceptable samples (n≥6) participating in subsequent inspections. Step 2: The system performs a GRUBBS test on the constructed qualified dataset, which meets the requirements of the national standard GB / T 4883-2008 "Statistical Processing and Interpretation of Data - Judgment of Outliers in Normal Samples". (a) Calculate the arithmetic mean and unbiased sample standard deviation s of the qualified dataset; (b) Iterate through each resistance value sample x in the dataset. i Calculate the one-sided GRUBBS test statistic: (c) Set the significance level α = 0.05 (corresponding to 95% confidence level), the test type is one-sided, and find the GRUBBS critical value G as specified in GB / T 4883-2008 based on the total sample size n. 0.05,n ; (d) Decision logic: If G i ≤G 0.05,n If the sample is determined to have no statistical outliers (no anomalies) in the test, it will be included in the valid data pool. If G i >G 0.05,n If the sample is identified as a statistical outlier, consistency calculations are performed only on such samples: calculate the relative deviation × 100%. If δ > 25%, the sample is considered an anomaly, the corresponding product exceeds quality consistency requirements, and the associated product's unique number and production batch are added to the key control list; if δ ≤ 25%, the sample is considered to have a statistical outlier, but does not exceed the engineering tolerance threshold, and is included in the valid data pool.
[0015] Step 3: Convert the consistency calculation results into quality consistency analysis results and display them on the quality analysis interface of the resistance tester's host computer.
Claims
1. A method for analyzing the consistency of test data quality from a resistance meter, characterized in that, Includes the following steps: Step 1: Data Acquisition and Screening: The resistance tester completes the acquisition of resistance data of the product being tested. The acquired data is then screened using a pre-determined pass / fail range. The screened data is then collected into either the pass / fail resistance test dataset or the abnormal resistance test dataset. Step 2, Statistical Testing and Analysis: Perform a GRUBBS test on the qualified resistance test dataset to identify outlier data exceeding the critical value, and perform consistency calculations on the outlier data. Step 3: Result Display: Convert the consistency calculation results into quality consistency analysis results and display them on the quality analysis interface of the resistance tester's host computer; Step two, the GRUBBS test and the consistency calculation for outlier sample data, includes the following steps: (a) Calculate the mean and standard deviation of the qualified resistance test dataset, which includes qualified data from the current test and previous tests; (b) Calculate the absolute value of the difference between each data point in the qualified resistance test dataset and the mean, and then divide it by the standard deviation to obtain the corresponding statistic. (c) Set the significance level to 0.05, set the test type to one-sided, and determine the critical value by consulting the GRUBBS test critical value table based on the total number of samples in the qualified resistance test dataset. (d) If the statistical value of the sample data to be tested is less than the critical value, it is determined that there is no abnormality. If the statistical value is greater than the critical value, it is calculated whether the data exceeds 25% of the mean of the sample set. If it is greater than 25%, it is determined that the product corresponding to the data exceeds the quality consistency requirements and is listed as a key product for quality control. If the percentage is no greater than 25%, the data is considered a statistical outlier, but it does not exceed the quality consistency requirements.
2. The method for analyzing the consistency of test data quality of a resistance meter according to claim 1, characterized in that, The initial sample size of the qualified resistance test dataset is no less than 6 pieces.
3. The method for analyzing the consistency of test data quality of a resistance meter according to claim 1, characterized in that, If the sample data is multivariate, repeat the GRUBBS test and consistency calculation process in step two for each variable.