Method and system for detecting and analyzing performance quality of touch screen

Abstract

The invention relates to the technical field of touch screen performance quality detection, in particular to a touch screen performance quality detection analysis method and system, and the detection method comprises the steps: obtaining a preset unloading operation simulation algorithm and test data of a touch screen; processing the test data of the touch screen by adopting an unloading operation simulation algorithm to obtain a preliminary data cleaning result; extracting data residual features according to the data cleaning capability evaluation result of the touch screen; inputting the data residue features into a preset fault positioning model to obtain position and type information of data residues; generating a fault positioning report according to the position and type information of the data residue; acquiring environmental adaptability test data of the touch screen under extreme conditions of high temperature, humidity, vibration and the like; constructing an environmental adaptability evaluation model according to the environmental adaptability test data; and predicting data residue conditions in different environments by adopting the environment adaptability evaluation model to obtain a change trend of the data residue.

Description

Technical Field

[0001] The present invention relates to the technical field of touch screen performance quality detection, and particularly to a touch screen performance quality detection and analysis method and system thereof. Background Art

[0002] In a touch screen performance quality detection and analysis system, the uninstallation technology of the user interface is a key link, which directly affects the security of user data and privacy protection. However, there are still some technical problems to be solved in the current uninstallation technology. First, the existing uninstallation operation simulation method is inefficient and cannot quickly and accurately evaluate the data cleaning ability of the touch screen, resulting in a long detection process. Second, during the uninstallation process, if there is a risk of data residue or privacy leakage on the touch screen, the existing detection and analysis system lacks accurate fault location ability and cannot quickly locate the root cause of the problem, bringing difficulties to subsequent problem repair. Moreover, the usage environment of the touch screen is variable, and the existing detection and analysis system is not comprehensive enough in environmental adaptability testing and fails to fully consider data security risks under various complex environments, such as data residue problems under extreme conditions like high temperature, humidity, and vibration. The existence of these technical problems leads to the inability to effectively guarantee the data security and privacy protection level of the touch screen, and the user's private data is at risk of leakage. Therefore, it is urgent to develop new technical solutions to improve the detection efficiency of uninstallation operations, enhance the accuracy of fault location, and expand the coverage of environmental adaptability testing, so as to comprehensively improve the data security guarantee ability of the touch screen performance quality detection and analysis system and maximize the protection of user private data security. Summary of the Invention

[0003] To solve the technical problems mentioned in the background art, the present invention provides a touch screen performance quality detection and analysis method, which mainly includes the following steps:

[0004] Obtain a preset uninstallation operation simulation algorithm and test data of the touch screen;

[0005] Process the test data of the touch screen using the uninstallation operation simulation algorithm to obtain a preliminary data cleaning result;

[0006] According to the preliminary data cleaning result, optimize the parameters of the uninstallation operation simulation algorithm through multiple rounds of iteration to obtain an optimized uninstallation operation simulation algorithm;

[0007] Process the test data of the touch screen using the optimized uninstallation operation simulation algorithm to obtain an evaluation result of the data cleaning ability of the touch screen;

[0008] Extract data residue characteristics according to the evaluation result of the data cleaning ability of the touch screen;

[0009] Input the data residue features into a preset fault location model to obtain the location and type information of the data residue;

[0010] Generate a fault location report based on the location and type information of the data residue;

[0011] Obtain the environmental adaptability test data of the touch screen under extreme conditions such as high temperature, humidity, and vibration;

[0012] Construct an environmental adaptability evaluation model based on the environmental adaptability test data;

[0013] Use the environmental adaptability evaluation model to predict the data residue situation under different environments and obtain the change trend of the data residue.

[0014] Optionally, the obtaining of the preset uninstall operation simulation algorithm and the test data of the touch screen includes:

[0015] Obtain the preset uninstall operation simulation algorithm and the touch screen test data and use them as input data;

[0016] For the uninstall operation and touch screen test requirements in the business content, use the simulation algorithm to analyze and calculate the input data to obtain the uninstall operation simulation results and touch screen test result data that meet the business requirements;

[0017] Based on the uninstall operation simulation results, determine whether the uninstall operation meets the preset operation specifications and process requirements. If not, perform optimization and adjustment to obtain the optimized uninstall operation simulation results;

[0018] Based on the touch screen test result data, use a data mining algorithm to perform statistical analysis on it to obtain the evaluation results of the touch screen performance and stability;

[0019] Summarize the optimized uninstall operation simulation results and the touch screen test evaluation results to generate a comprehensive test analysis report that meets the business requirements;

[0020] Based on the comprehensive test analysis report, determine whether the uninstall operation simulation algorithm and the touch screen test process need to be further optimized. If so, return to the analysis and calculation processing steps for iterative optimization.

[0021] Optionally, the using of the uninstall operation simulation algorithm to process the test data of the touch screen to obtain preliminary data cleaning results includes:

[0022] Obtain the touch screen test data and input the test data into a preset uninstall operation simulation algorithm for processing to obtain preliminary processing results;

[0023] For the preliminary processing result, data cleaning is performed using a preset data cleaning technique to obtain a processed result after cleaning;

[0024] According to a preset data quality assessment rule, the quality of the processed result after cleaning is evaluated. If the quality assessment result meets a preset threshold, the processed result after cleaning is determined as the final test data processing result;

[0025] If the quality assessment result does not reach the preset threshold, the processed result after cleaning is re - input into the unloading operation simulation algorithm for secondary processing to obtain a secondary processing result;

[0026] Repeat data cleaning and quality assessment until the quality assessment result meets the preset threshold, and determine the final processed result after cleaning as the processing result of the touch - screen test data;

[0027] Store the processing result of the touch - screen test data in a preset result database;

[0028] Through a preset data visualization technique, visually display the processing result of the touch - screen test data.

[0029] Optionally, according to the preliminary data cleaning result, optimizing the parameters of the unloading operation simulation algorithm through multiple rounds of iteration, the steps include:

[0030] Obtain the initial unloading operation simulation algorithm parameters as the input for the first - round iteration optimization;

[0031] For the algorithm parameters of the current round, simulate the execution of the unloading operation to obtain operation result data;

[0032] Perform cleaning processing on the operation result data to remove outliers and noise data, obtaining cleaned result data;

[0033] Compare the cleaned result data with a preset optimization target, and use the support vector machine algorithm to calculate the optimization degree of the current parameters;

[0034] If the optimization degree does not reach the preset threshold, adjust the parameters of the unloading operation simulation algorithm according to the cleaning result of the current round, and use the adjusted algorithm parameters as the input for the next - round iteration optimization;

[0035] If the optimization degree reaches the preset threshold, determine the current algorithm parameters as the optimized unloading operation simulation algorithm parameters;

[0036] According to the optimized unloading operation simulation algorithm parameters, re - execute the unloading operation simulation to obtain optimized operation result data.

[0037] Optionally, process the test data of the touch screen using the optimized offloading operation simulation algorithm to obtain an evaluation result of the data cleaning ability of the touch screen, including:

[0038] Obtain the test data of the touch screen, perform preprocessing on the test data to remove noise and outliers, and obtain the cleaned test data;

[0039] According to a preset offloading operation model, extract features from the cleaned test data to obtain a feature vector of the test data;

[0040] Input the feature vector of the test data into the optimized offloading operation simulation algorithm, simulate the offloading operation process of the touch screen through the algorithm, and obtain the simulated offloading operation result;

[0041] Perform statistical analysis on the simulated offloading operation result, calculate indicators such as the success rate, failure rate, and average response time of the offloading operation, and obtain the performance evaluation result of the offloading operation;

[0042] According to a preset data cleaning ability evaluation rule, compare the performance evaluation result of the offloading operation with the rule to determine whether the data cleaning ability of the touch screen meets the standard, and obtain a preliminary evaluation result;

[0043] Adopt the method of cross-validation, divide the test data into multiple subsets, and use different subsets to train and test the offloading operation simulation algorithm respectively to obtain multiple evaluation results;

[0044] Perform comprehensive analysis on multiple evaluation results, calculate the mean and variance of the evaluation results, and determine the final evaluation result of the data cleaning ability of the touch screen according to the magnitudes of the mean and variance.

[0045] Optionally, extract data residue features according to the evaluation result of the data cleaning ability of the touch screen, including:

[0046] Obtain multiple data cleaning logs of the touch screen, and determine the evaluation result of the data cleaning ability of the touch screen according to a preset data cleaning ability evaluation rule;

[0047] For the evaluation result of the data cleaning ability, adopt a data residue feature extraction algorithm to obtain the data residue features of the touch screen;

[0048] According to the data residue features, combine a pre-constructed data residue risk assessment model to judge the data residue risk level of the touch screen;

[0049] If the data residue risk level exceeds a preset threshold, trigger a data cleaning reminder, and determine a targeted data cleaning strategy according to the data residue features;

[0050] Send the data cleaning strategy to the touch screen, and execute the data cleaning strategy through the touch screen to improve the data cleaning effect of the touch screen;

[0051] After completing the data cleaning, obtain the updated data cleaning log of the touch screen, and re-evaluate the data cleaning ability and data residue risk level of the touch screen;

[0052] According to the change of the evaluation result, dynamically adjust the data residue feature extraction algorithm and the data residue risk assessment model, and continuously optimize the accuracy of data cleaning ability evaluation and data residue risk warning.

[0053] Optionally, inputting the data residue feature into a preset fault location model to obtain the location and type information of the data residue, including:

[0054] Obtain the feature information of the data residue;

[0055] Input the feature information into a preset fault location model to obtain the location information and type information of the data residue;

[0056] If the location information meets the first preset condition, add the location information to the candidate set;

[0057] If the type information meets the second preset condition, add the type information to the candidate set;

[0058] Classify according to the location information and type information in the candidate set by using the support vector machine algorithm to obtain the target data residue location and the target data residue type;

[0059] Use the target data residue location and the target data residue type as the output result for subsequent data recovery or security protection.

[0060] Optionally, generating a fault location report according to the location and type information of the data residue, including:

[0061] Obtain the location information and type information of the data residue;

[0062] According to the location information and type information, use data analysis technology to extract the features of the data residue to obtain a feature vector;

[0063] Input the feature vector into a pre-established fault location model, and through model inference, judge the location and cause of the fault;

[0064] If the confidence levels of the fault location and cause output by the fault location model are greater than the preset threshold, write the fault location and cause information into the fault location report;

[0065] If the confidence levels of the fault location and cause output by the fault location model are less than or equal to the preset threshold, then according to the location information and type information, use the rule reasoning method to determine the fault location and cause, and write the reasoning result into the fault location report;

[0066] According to the fault location information in the fault location report, obtain the device model and configuration parameters at this location;

[0067] By querying the pre-established device knowledge base, obtain the fault diagnosis method and repair plan corresponding to the device model;

[0068] Associate and store the fault location report, fault diagnosis method and repair plan to form a complete fault handling plan, and output it to relevant technical personnel;

[0069] Obtain the feedback information on fault diagnosis and repair for optimizing the fault location model and device knowledge base.

[0070] Optionally, the obtaining of the environmental adaptability test data of the touch screen under extreme conditions such as high temperature, humidity, and vibration includes:

[0071] According to the preset temperature, humidity, and vibration frequency thresholds, design multiple groups of extreme environmental condition parameter combinations to construct a touch screen environmental adaptability test plan;

[0072] For each group of extreme environmental condition parameter combinations, obtain a touch screen to be tested, place it in the corresponding high temperature, humidity, and vibration environment, and connect signal acquisition equipment at the same time;

[0073] When the touch screen reaches the preset duration under the current environmental conditions, use the signal acquisition equipment to record the performance parameters of the touch screen in real time to obtain the time series data of the touch screen performance under this environmental condition;

[0074] Clean and extract features from the touch screen performance time series data collected under each group of extreme environmental conditions to obtain a feature vector set of touch screen performance under different environmental conditions;

[0075] Use the support vector machine algorithm to train the above feature vector set to establish an association model between touch screen performance and extreme environmental conditions for quantitatively evaluating the environmental adaptability of the touch screen;

[0076] Use the established evaluation model to predict the performance of a given touch screen under various extreme environmental conditions and judge whether it meets the environmental adaptability requirements;

[0077] If the prediction result shows that the environmental adaptability of the touch screen does not meet the requirements, optimization suggestions are put forward for the weak links, and the optimized touch screen is further tested until the requirements are met.

[0078] Optionally, constructing an environmental adaptability evaluation model according to the environmental adaptability test data includes:

[0079] Obtain the test data of the product under various environmental conditions, where the environmental conditions include temperature, humidity, vibration and shock factors, and the test data includes the influence data of environmental factors on the product performance;

[0080] Preprocess the test data, remove outliers and noise, and normalize the data format;

[0081] According to the preprocessed test data, use at least one machine learning algorithm to train the environmental adaptability evaluation model, and the machine learning algorithms include support vector machine and random forest;

[0082] Use the cross-validation method to evaluate the performance of the evaluation model, calculate the accuracy, precision and recall rate of the evaluation model, and judge the generalization ability and prediction effect of the evaluation model;

[0083] If the performance of the evaluation model does not reach the preset threshold, use feature engineering and parameter tuning methods to optimize the evaluation model;

[0084] Deploy the optimized environmental adaptability evaluation model to the production environment, and predict the performance and adaptability of the new product under different environmental conditions according to the test data of the new product;

[0085] Continuously collect the environmental data and performance feedback of the new product in actual application for evaluating the online performance of the evaluation model;

[0086] Incrementally train and update the evaluation model according to the environmental data and performance feedback.

[0087] On the other hand, the present application also provides a touch screen performance quality detection and analysis system, which detects and analyzes the touch screen performance quality according to the touch screen performance quality detection and analysis method described above.

[0088] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0089] The present invention discloses a method for detecting and analyzing the performance and quality of a touch screen. The method obtains a preset uninstall operation simulation algorithm and touch screen test data, preliminarily cleans the data, and optimizes the algorithm parameters through multiple rounds of iteration. The optimized algorithm is used to process the test data to obtain an evaluation result of the data cleaning ability of the touch screen. Further, the data residue characteristics are extracted and input into a preset fault location model to obtain the residue position and type information, and a fault location report is generated. The present invention also considers extreme environmental factors. By constructing an environmental adaptability evaluation model, it predicts the change trend of data residue under different environments and reduces the risk of user privacy leakage. This method can not only accurately evaluate the data cleaning ability of the touch screen, but also effectively locate faults and predict the performance under different environmental conditions, protect user privacy, and provide an important basis for the quality improvement and reliability enhancement of the touch screen. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 It is a flowchart of a method for detecting and analyzing the performance and quality of a touch screen according to the present invention.

[0091] Figure 2 It is a schematic diagram of a method for detecting and analyzing the performance and quality of a touch screen according to the present invention.

[0092] Figure 3 It is another schematic diagram of a method for detecting and analyzing the performance and quality of a touch screen according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0093] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0094] Such as Figures 1-3 , a method for detecting and analyzing the performance and quality of a touch screen in this embodiment may specifically include:

[0095] Step S101, obtain a preset uninstall operation simulation algorithm and test data of the touch screen.

[0096] Obtain the preset uninstall operation simulation algorithm and touch screen test data, and use them as input data; for the uninstall operation and touch screen test requirements in the business content, use the simulation algorithm to analyze and calculate the input data to obtain the uninstall operation simulation result and touch screen test result data that meet the business requirements; according to the uninstall operation simulation result, judge whether the uninstall operation meets the preset operation specifications and process requirements, and if not, perform optimization and adjustment to obtain the optimized uninstall operation simulation result; according to the touch screen test result data, use the data mining algorithm to perform statistical analysis on it to obtain the evaluation result of the touch screen performance and stability; summarize the optimized uninstall operation simulation result and the touch screen test evaluation result to generate a comprehensive test analysis report that meets the business requirements; according to the comprehensive test analysis report, determine whether the uninstall operation simulation algorithm and the touch screen test process need to be further optimized, and if so, return to the analysis and calculation processing step for iterative optimization.

[0097] Exemplarily, first, extract 1000 pieces of data from the preset uninstall operation simulation algorithm and touch screen test database as input. Among them, the uninstall operation data includes operation time, operation steps, operation results, etc., and the touch screen test data includes touch point coordinates, touch pressure, response time, etc. Then, use the rule-based simulation algorithm to analyze the uninstall operation data. By comparing with the preset operation specifications, identify the operations that do not meet the specifications, such as the operation time exceeding 10 seconds, the operation steps being missing, etc., and perform optimization and adjustment to make the accuracy of the uninstall operation simulation result reach 99%. At the same time,

[0098]

[0099] $D_i$ represents the distance from the $i$-th data point to the cluster center, $x_{ij}$ represents the $j$-th feature value of the $i$-th data point, $c_j$ represents the $j$-th feature value of the cluster center, and $n$ represents the number of features. This formula is used to calculate the Euclidean distance from the data point to the cluster center. Use the K-means clustering algorithm to analyze the touch screen test data. By calculating the mean and variance of different parameters, evaluate the performance indicators such as the sensitivity and stability of the touch screen, and obtain a comprehensive touch screen performance score of 95 points. Finally,

[0100]

[0101] ROI represents the return on investment, $B_0$ represents the system benefit before optimization, $B_1$ represents the system benefit after optimization, and $C$ represents the optimization input cost. This formula is used to evaluate the economic benefits of system optimization.

[0102]

[0103] Let U represent the user experience satisfaction score, n represent the number of evaluation metrics, w_i represent the weight of the i-th metric, and s_i represent the score of the i-th metric. This formula is used to calculate the overall satisfaction of users with the optimized system.

[0104] P = αE + βR. P represents the comprehensive performance score, α and β are the weight coefficients of the offloading operation efficiency and the touch screen response speed respectively, and E and R are the percentage improvements in efficiency and response speed calculated from the first two formulas. This formula is used to evaluate the degree of improvement in the overall system performance.

[0105]

[0106] R represents the percentage improvement in response speed, S_0 represents the average response time before optimization, and S_1 represents the average response time after optimization. This formula calculates the improvement in response speed after optimizing the touch screen performance.

[0107]

[0108] E represents the percentage improvement in efficiency, T_0 represents the average operation time before optimization, and T_1 represents the average operation time after optimization. This formula calculates the improvement in efficiency after optimizing the offloading operation process. The simulation optimization results of the offloading operation and the touch screen test evaluation results are summarized to generate a test analysis report. The report shows that after optimizing the offloading operation process, the user operation efficiency has increased by 20%, and after optimizing the touch screen performance, the response speed has increased by 15%. According to the report results, the offloading operation simulation algorithm is further optimized by introducing an intelligent algorithm based on deep learning, which improves the generalization ability and robustness of the algorithm, enabling it to adapt to more operation scenarios and data types.

[0109] Step S102: Process the test data of the touch screen using the offloading operation simulation algorithm to obtain a preliminary data cleaning result.

[0110] Obtain the test data of the touch screen, input the test data into a preset unloading operation simulation algorithm for processing to obtain a preliminary processing result; for the preliminary processing result, use a preset data cleaning technique to clean the data to obtain a cleaned processing result; according to a preset data quality evaluation rule, evaluate the quality of the cleaned processing result. If the quality evaluation result meets the preset threshold, determine the cleaned processing result as the final test data processing result; if the quality evaluation result does not reach the preset threshold, re-input the cleaned processing result into the unloading operation simulation algorithm for secondary processing to obtain a secondary processing result; repeat the data cleaning and quality evaluation until the quality evaluation result meets the preset threshold, and determine the final cleaned processing result as the processing result of the touch screen test data; store the processing result of the touch screen test data in a preset result database; through a preset data visualization technique, visually display the processing result of the touch screen test data.

[0111] Exemplarily, during the touch screen test, the original test data of the touch screen is collected by a high-precision sensor, including parameters such as touch point coordinates and pressure values. The collected test data is input into a neural network-based unloading operation simulation algorithm for processing. This algorithm can accurately simulate the user's unloading operation and preprocess the test data through learning and training on a large amount of historical test data. For the result data processed by the unloading operation simulation algorithm, a data cleaning technique, such as removing outliers and noise, is used to initially clean the data and improve the data quality. According to a preset data quality evaluation rule, such as data integrity and consistency, the quality of the cleaned data is evaluated. If the data quality reaches more than 95%, it is determined as the final test data processing result. If the quality evaluation result is lower than 95%, the cleaned data is re-input into the unloading operation simulation algorithm for secondary processing, and continuously iteratively optimized until the data quality meets the requirements. Finally, the processed high-quality test data is stored in the result database to provide data support for subsequent touch screen performance analysis, algorithm optimization, etc. At the same time, through the data visualization technique, the test data processing result is visually displayed in the form of charts, heat maps, etc., facilitating the R & D personnel to quickly evaluate and analyze the effect of the test data processing and optimize the touch screen test process and algorithm.

[0112] Step S103, according to the preliminary data cleaning result, iteratively optimize the parameters of the unloading operation simulation algorithm through multiple rounds to obtain an optimized unloading operation simulation algorithm.

[0113] Acquire initial unloading operation simulation algorithm parameters as input for the first round of iterative optimization; simulate and execute the unloading operation for the algorithm parameters of the current round to obtain operation result data; clean the operation result data, remove outliers and noise data, and obtain cleaned result data; compare the cleaned result data with the preset optimization target, and use the support vector machine algorithm to calculate the current parameter optimization degree; if the optimization degree does not reach the preset threshold, adjust the unloading operation simulation algorithm parameters through the gradient descent algorithm according to the current round of cleaning results, and use the adjusted algorithm parameters as input for the next round of iterative optimization; if the optimization degree reaches the preset threshold, determine the current algorithm parameters as the optimized unloading operation simulation algorithm parameters; re-execute the unloading operation simulation according to the optimized unloading operation simulation algorithm parameters to obtain the optimized operation result data.

[0114] Exemplarily, first, according to the preliminary data cleaning results, the initial parameters of the unloading operation simulation algorithm, including unloading rate, unloading amount, etc., are obtained by least squares fitting as the input of the first round of iterative optimization. Then, for the algorithm parameters of the current round, the Monte Carlo method is used to simulate and execute 1000 unloading operations to obtain the operation result data, including unloading time, unloading success rate, etc. Next, the simulation operation result data is cleaned, the outliers are eliminated using the 3σ principle, and the noise data is removed by wavelet transform to obtain the cleaned result data. The cleaned result data is compared with the preset optimization target, and the optimization degree of the current parameters is calculated by the support vector machine algorithm. If the optimization degree does not reach the preset 90% threshold, then according to the cleaning results of the current round, the unloading rate and other parameters are adjusted by the gradient descent algorithm with a learning rate of 0.1 as the input of the next round of iterative optimization; if the optimization degree reaches the preset threshold, the current parameters are determined as the optimized parameters of the unloading operation simulation algorithm. Finally, based on the optimized algorithm parameters, 500 unloading operation simulations were re-executed, and the average unloading time was shortened by 20%, and the unloading success rate was increased by 5%. The final optimization result data was obtained, completing the algorithm optimization process.

[0115] X clean ={x i |outlier i =0}

[0116] X_clean represents the cleaned data set, x_i represents the i-th data sample, and outlier_i represents whether the i-th sample is an outlier.

[0117]

[0118] outlier_i indicates whether the i-th sample is an outlier, d_i represents the Mahalanobis distance of the i-th sample, and θ represents the outlier detection threshold.

[0119]

[0120] θ represents the outlier detection threshold, χ 2 represents the chi-square distribution, p represents the data dimension, and α represents the significance level.

[0121]

[0122] d_i represents the Mahalanobis distance of the i-th sample, x_i represents the i-th data sample, μ represents the data mean vector, and Σ represents the covariance matrix.

[0123]

[0124] Σ represents the covariance matrix, μ represents the data mean vector, and x_i represents the i-th data sample.

[0125] X = {x 1 , x 2 , ##x n}

[0126] X represents the input data set, x_i represents the i-th data sample, and n represents the total number of samples. The outlier detection algorithm EllipticEnvelope is used to automatically identify and remove outliers to ensure data quality; the main frequency range of the noise is determined through spectrum analysis, and an appropriate wavelet basis is designed to remove high-frequency noise and improve the signal-to-noise ratio of the data. During the parameter optimization process, a regularization term is introduced to avoid overfitting and improve the generalization performance of the algorithm; at the same time, the optimal hyperparameters are determined through grid search, such as the kernel function type and penalty coefficient of the support vector machine, to further improve the optimization effect.

[0127] Step S104, use the optimized offloading operation simulation algorithm to process the test data of the touch screen to obtain the evaluation result of the data cleaning ability of the touch screen.

[0128] Obtain the test data of the touch screen, preprocess the test data, remove noise and outliers, and obtain the cleaned test data. According to the preset uninstall operation model, extract the features of the cleaned test data to obtain the feature vector of the test data. Input the feature vector of the test data into the optimized uninstall operation simulation algorithm, and simulate the uninstall operation process of the touch screen through the algorithm to obtain the simulated uninstall operation result. Conduct statistical analysis on the simulated uninstall operation result, calculate indicators such as the success rate, failure rate, and average response time of the uninstall operation, and obtain the performance evaluation result of the uninstall operation. According to the preset data cleaning ability evaluation rules, compare the performance evaluation result of the uninstall operation with the rules to determine whether the data cleaning ability of the touch screen meets the standard, and obtain the preliminary evaluation result. Adopt the cross-validation method, divide the test data into multiple subsets, and use different subsets to train and test the uninstall operation simulation algorithm respectively to obtain multiple evaluation results. Conduct comprehensive analysis on the multiple evaluation results, calculate the mean and variance of the evaluation results, and determine the final evaluation result of the data cleaning ability of the touch screen according to the magnitudes of the mean and variance.

[0129] Exemplarily, first, collect 1000 groups of touch data through a touch screen test tool. Each group of data contains the coordinates and timestamps of 10 touch points. Then, use the median filtering algorithm to preprocess the data, remove the coordinates outside the screen range and the data with abnormal time intervals, and obtain 800 groups of valid data. Next, according to the characteristics of the uninstall operation, extract 10 features such as the number of touch points, sliding speed, and sliding direction of each group of data to form a feature vector. Input the feature vector into the uninstall operation simulation algorithm based on SVM, simulate 1000 uninstall operations through the algorithm, and statistically obtain that the uninstall success rate is 95% and the average response time is 5 seconds. According to the preset rules, when the uninstall success rate is greater than 90% and the average response time is less than 1 second, it is determined that the data cleaning ability of the touch screen meets the standard. To further verify the reliability of the evaluation result, adopt the 5-fold cross-validation method, randomly divide the 800 groups of data into 5 subsets, select 1 subset as the test set each time, and the remaining 4 subsets as the training set, repeat the experiment 5 times to obtain 5 evaluation results. Calculate that the mean of the 5 evaluation results is 95% and the variance is 01. The mean is close to the overall evaluation result and the variance is small, indicating that the evaluation result is stable and reliable. Finally, it is determined that the data cleaning ability of this touch screen reaches an excellent level.

[0130] Step S105, extract the data residue features according to the evaluation result of the data cleaning ability of the touch screen.

[0131] Obtain multiple data cleaning logs of the touch screen, determine the evaluation result of the data cleaning ability of the touch screen according to the preset data cleaning ability evaluation rules; for the evaluation result of the data cleaning ability, use a data residue feature extraction algorithm to obtain the data residue features of the touch screen; according to the data residue features, combined with the pre-constructed data residue risk assessment model, judge the data residue risk level of the touch screen; if the data residue risk level exceeds the preset threshold, trigger a data cleaning reminder, and determine a targeted data cleaning strategy according to the data residue features; send the data cleaning strategy to the touch screen, and execute the data cleaning strategy through the touch screen to improve the data cleaning effect of the touch screen; after completing the data cleaning, obtain the updated data cleaning logs of the touch screen, and re-evaluate the data cleaning ability and data residue risk level of the touch screen; according to the change of the evaluation result, dynamically adjust the data residue feature extraction algorithm and the data residue risk assessment model to continuously optimize the accuracy of data cleaning ability evaluation and data residue risk warning.

[0132] Exemplarily, first, the system will automatically obtain 50 data cleaning logs of the touch screen. According to the preset data cleaning ability evaluation rules (such as cleaning frequency, amount of data cleaned, etc.), using a weighted average algorithm, calculate that the data cleaning ability score of the touch screen is 85 points. Then, for this evaluation result, the system uses a data residue feature extraction algorithm based on association rule mining. By analyzing the changes before and after data cleaning, extract the data residue feature vector of the touch screen. Then,

[0133]

[0134] P(Risk|F) represents the risk probability when the feature vector F is given, w represents the weight vector, and b represents the bias term.

[0135] Risk(F)=sign(w T F + b)

[0136] Risk(F) represents the risk level judgment function, F represents the input feature vector, w represents the trained weight vector, and b represents the bias term.

[0137]

[0138] w represents the weight vector, b represents the bias term, C represents the penalty parameter, x_i represents the i-th training sample, y_i represents the corresponding label, and m represents the number of training samples.

[0139] F=[f 1 , f 2 ,##f nLet \(F\) denote the extracted feature vector, \(f_i\) denote the \(i\)-th feature component, and \(n\) denote the feature dimension. The extracted feature vector is input into a pre-trained data residual risk assessment SVM model to determine that the data residual risk level of the touch screen is medium risk. Since this risk level exceeds the preset warning threshold, the system automatically triggers a data cleaning reminder and, based on the data residual characteristics, matches the optimal data cleaning strategy from the policy library, such as using a secure deletion algorithm and increasing the number of random overwrites, and issues the policy to the touch screen for execution. After the cleaning is completed, the system obtains the data cleaning log of the touch screen again, re-evaluates its data cleaning ability as 95 points, and the residual risk level drops to low risk. By comparing the changes in the evaluation results, the system uses the reinforcement learning algorithm to dynamically adjust the association rule threshold in the data residual feature extraction algorithm and the SVM kernel function parameters in the risk assessment model, and incorporates the feedback data of this cleaning effect to continuously optimize the evaluation and warning models, improving their accuracy from the original 90% to 95%.

[0140] Step S106: Input the data residual characteristics into a preset fault location model to obtain the location and type information of the data residual.

[0141] Obtain the feature information of the data residual; input the feature information into a preset fault location model to obtain the location information and type information of the data residual; if the location information meets the first preset condition, add the location information to the candidate set; if the type information meets the second preset condition, add the type information to the candidate set; classify using the support vector machine algorithm based on the location information and type information in the candidate set to obtain the target data residual location and the target data residual type; use the target data residual location and the target data residual type as the output results for subsequent data recovery or security protection.

[0142] Exemplarily, according to the data residue characteristics, the characteristic information of the residual data can be obtained. For example, the size of the residual data is 10 MB, the creation time is March 1, 2022, and the modification time is March 5, 2022, etc. These characteristic information are input into a preset fault location model for processing. The model uses a decision tree algorithm. By analyzing the characteristic information, it is determined that the location information of the residual data is under the Windows directory of drive C, and the type information is a system log file. Then, the residual location information is compared with preset conditions. For example, the preset condition is that the residual data is located on a non-system disk and the data volume is less than 1 GB. Then the residual location information meets the preset condition and is added to the candidate set. At the same time, the residual type information is compared with the preset condition. For example, the preset condition is that the residual data is a non-critical file. Then the residual type information meets the preset condition and is also added to the candidate set. Then, according to the location information and type information in the candidate set, a support vector machine algorithm is used for classification. Through the training of the candidate set, the final data residue location is the system32 sub-directory under the Windows directory of drive C, and the data residue type is a system log file. Finally, the data residue location and type are used as outputs for subsequent data recovery or security protection processing, etc. For example, the residual file can be encrypted or deleted to prevent data leakage.

[0143] Step S107, generate a fault location report according to the location and type information of the data residue.

[0144] Obtain the location information and type information of the data residue; according to the location information and type information, use data analysis techniques to extract the characteristics of the data residue to obtain a feature vector; input the feature vector into a pre-established fault location model, and through model inference, judge the location and cause of the fault; if the confidence levels of the fault location and cause output by the fault location model are greater than a preset threshold, then write the fault location and cause information into the fault location report; if the confidence levels of the fault location and cause output by the fault location model are less than or equal to the preset threshold, then according to the location information and type information, use a rule inference method to determine the fault location and cause, and write the inference result into the fault location report; according to the fault location information in the fault location report, obtain the device model and configuration parameters of this location; by querying a pre-established device knowledge base, obtain the fault diagnosis method and repair plan corresponding to the device model; associate and store the fault location report, fault diagnosis method, and repair plan to form a complete fault handling plan and output it to relevant technical personnel; obtain the feedback information of fault diagnosis and repair for optimizing the fault location model and device knowledge base.

[0145] Exemplarily, according to the location information and type information of data remnants, the K-means clustering algorithm is used to classify the data remnants, extract key features such as the size, format, creation time, etc. of the data remnants, and form a 128-dimensional feature vector. The feature vector is input into a pre-trained fault location model based on a convolutional neural network. The model outputs the confidence levels of the fault location and cause through operations such as convolution, pooling, and fully connected of the feature vector. If the confidence level is greater than 8, the fault location and cause are written into the fault location report; if the confidence level is less than or equal to 8, a rule reasoning method based on an expert knowledge base is adopted. According to the location and type of the data remnants, combined with the rules in the expert knowledge base, such as "if the data remnants are located in the system log and the type is an error log, it may be a system software fault", the possible fault location and cause are inferred and the inference result is written into the fault location report. According to the fault location in the report, such as "main board of server A", information such as the server model is Dell R740, CPU model is Intel Xeon Gold 6148, and memory is 32GB DDR4 is queried from the device knowledge base, and common fault diagnosis methods and repair solutions for this model of server are obtained, such as "use Dell diagnostic tools to detect the main board. If the detected fault code is 0142, the main board needs to be replaced", etc. The fault location report, diagnosis method, and repair solution are associated and stored to form a complete fault handling solution, which is output to technicians through an automated operation and maintenance platform. Technicians detect the main board using Dell diagnostic tools according to the solution guidance, confirm the fault location, and replace the new main board according to the repair solution to complete the fault repair. Finally, the fault handling process and results are fed back to the system for optimizing the fault location model and device knowledge base. For example, the feature vector and diagnosis results of this fault are added to the training set to retrain the fault location model to improve the generalization ability and accuracy of the model. At the same time, information such as the device model, fault phenomenon, diagnosis method, and repair solution of this fault is added to the device knowledge base to enrich the content of the knowledge base and improve the efficiency of fault location and handling.

[0146] Step S108, obtain the environmental adaptability test data of the touch screen under extreme conditions such as high temperature, humidity, and vibration.

[0147] According to the preset temperature, humidity, and vibration frequency thresholds, design multiple sets of extreme environmental condition parameter combinations to construct a touch screen environmental adaptability test plan. For each set of extreme environmental condition parameter combinations, obtain a touch screen to be tested, place it in the corresponding high-temperature, humid, and vibrating environment, and connect a signal acquisition device at the same time. When the touch screen reaches the preset duration under the current environmental conditions, use the signal acquisition device to record the performance parameters of the touch screen in real time to obtain the time-series data of the touch screen performance under these environmental conditions. Clean and extract features from the time-series data of the touch screen performance collected under each set of extreme environmental conditions to obtain a feature vector set of the touch screen performance under different environmental conditions. Use the support vector machine algorithm to train the above feature vector set to establish an association model between the touch screen performance and the extreme environmental conditions for quantitatively evaluating the environmental adaptability of the touch screen. Use the established evaluation model to predict the performance of a given touch screen under various extreme environmental conditions and determine whether it meets the environmental adaptability requirements. If the prediction result shows that the environmental adaptability of the touch screen does not meet the requirements, put forward optimization suggestions for the weak links and further test the optimized touch screen until the requirements are met.

[0148] Exemplarily, first, according to the preset temperature, humidity, and vibration frequency thresholds, multiple sets of extreme environmental condition parameter combinations are designed. For example, the temperature range is -20°C to 60°C, the humidity range is 30% to 90%, and the vibration frequency range is 50 Hz to 500 Hz. Multiple sets of extreme environmental conditions such as (60°C, 90%, 500 Hz) and (-20°C, 30%, 50 Hz) can be combined. Then, for each set of extreme environmental conditions, a touch screen to be tested is obtained, and it is placed in a comprehensive environmental test chamber composed of equipment such as a high and low temperature test chamber, a constant temperature and humidity test chamber, and a vibration table, and signal acquisition equipment such as an oscilloscope is connected. When the touch screen continuously operates in the current environmental conditions for 8 hours, performance parameters such as the capacitance value and response time of the touch screen are recorded through the oscilloscope, and the time-series data of the touch screen performance changing with time under this environmental condition is obtained. The collected time-series data is preprocessed such as denoising and normalization, and characteristic parameters such as the mean value, variance, and peak value are extracted to obtain a set of feature vectors characterizing the touch screen performance under different environmental conditions. The support vector machine algorithm is used to train the set of feature vectors, a non-linear regression model of the touch screen performance and environmental parameters such as temperature, humidity, and vibration frequency is established, and the model parameters are optimized through 5-fold cross-validation to improve the generalization ability of the model for quantitatively evaluating the environmental adaptability of the touch screen. Using the trained evaluation model, for a new touch screen, key performance indicators such as its capacitance value and response time under specific environmental conditions such as (50°C, 80%, 300 Hz) are predicted to determine whether it meets the preset environmental adaptability requirements. If the prediction result shows that the environmental adaptability of the touch screen does not meet the standard, the key environmental factors affecting the touch screen performance are determined through sensitivity analysis, and optimization suggestions are put forward from aspects such as the touch screen structure design and material selection, such as using a high-temperature-resistant ITO conductive film and reducing the electrode spacing to improve the signal-to-noise ratio. The optimized touch screen is re-tested for environmental adaptability to evaluate the optimization effect until the environmental adaptability requirements are met.

[0149] Step S109, construct an environmental adaptability evaluation model according to the environmental adaptability test data.

[0150] Obtain the test data of the product under various environmental conditions, where the environmental conditions include temperature, humidity, vibration, and shock factors, and the test data includes the data on the impact of environmental factors on the product performance; preprocess the test data to remove outliers and noise and normalize the data format; according to the preprocessed test data, use at least one machine learning algorithm to train an environmental adaptability evaluation model, and the machine learning algorithms include support vector machine and random forest; use the cross-validation method to evaluate the performance of the evaluation model, calculate the accuracy, precision, and recall rate of the evaluation model, and judge the generalization ability and prediction effect of the evaluation model; if the performance of the evaluation model does not reach the preset threshold, use feature engineering and parameter tuning methods to optimize the evaluation model; deploy the optimized environmental adaptability evaluation model to the production environment, and predict the performance and adaptability of the new product under different environmental conditions for the test data of the new product; continuously collect the environmental data and performance feedback of the new product in actual applications to evaluate the online performance of the evaluation model; perform incremental training and update on the evaluation model according to the environmental data and performance feedback.

[0151] Exemplarily, first, collect the performance data of the product under environmental conditions such as a temperature range of -40°C to 85°C, a humidity range of 20% to 95%, vibration in the frequency range of 0 to 50 Hz, and an impact of up to 50G through sensors. Clean the collected data to remove outliers beyond the physical thresholds, and normalize the data from different sources and with different units, uniformly mapping them to the interval of 0 to 1. Select the support vector machine algorithm and use the Gaussian kernel function to train the processed data to obtain a non-linear mapping model between environmental factors and product performance. Adopt the method of 5-fold cross-validation to evaluate the performance of the model on the test set. The accuracy reaches 92%, the precision is 97%, and the recall rate is 83%, indicating that the model has good generalization ability. For the underfitting problem of the model under extreme environmental conditions, optimize the feature space of the model through feature combination and selection, and use grid search to tune the penalty coefficient C of SVM, the kernel function parameter γ, etc. Finally, increase the accuracy of the model to 95%. Package the trained model as a RESTful interface and deploy it to the cloud server to achieve real-time online evaluation of the product environmental adaptability. At the same time, continuously collect the environmental parameters and performance data of the product in actual application scenarios, continuously train and optimize the model, and continuously improve the adaptability and robustness of the model to ensure that the product works stably and reliably in a complex and changeable environment.

[0152] Step S1010, use the environmental adaptability evaluation model to predict the data residue situation under different environments to obtain the change trend of the data residue.

[0153] Obtain the data residue situation in different environments as the input data for the environmental adaptability evaluation model. Use machine learning algorithms such as support vector machines and random forests to establish an environmental adaptability evaluation model for predicting the data residue situation. Train and optimize the environmental adaptability evaluation model based on environmental factors and historical data residue situations to improve the prediction accuracy of the model. Use the trained environmental adaptability evaluation model to predict and analyze the data residue situation in a new environment. Compare the predicted data residue situation with the actual situation to evaluate the prediction performance of the environmental adaptability evaluation model. If the prediction performance of the model does not reach the preset threshold, return to step 3 to further optimize the model; otherwise, apply the model to the actual environment. Based on the prediction results of the environmental adaptability evaluation model, obtain the change trend of data residue in different environments, providing a decision-making basis for data security management.

[0154] Exemplarily, in order to obtain the data residue situation in different environments, web crawler technology can be used to automatically collect data residue information from devices and systems in various environments. For example, a Python script can be written to use the Scrapy framework to collect data from devices under different operating systems, different storage media, and different network environments, obtaining information such as the type, quantity, and distribution of residual data to form an input data set. When establishing the environmental adaptability evaluation model, the support vector machine (SVM) algorithm can be selected. Utilize the non-linear mapping ability of the SVM algorithm to map high-dimensional environmental factors to a low-dimensional space, find the optimal classification hyperplane, and achieve the prediction of the data residue situation. In the model training stage, methods such as grid search and cross-validation can be used to optimize hyperparameters such as the kernel function type and penalty coefficient of the SVM to improve the prediction accuracy of the model. For example, 5-fold cross-validation can be used, randomly dividing the data set into 5 subsets, taking turns using 4 of the subsets as the training set and the remaining 1 subset as the test set for model training and evaluation, and finally selecting the combination of hyperparameters with the highest average prediction accuracy. Use the trained SVM model to predict the environmental factor data in the new environment and obtain the prediction result of the data residue situation. Compare the prediction result with the actually collected data residue situation and calculate evaluation metrics such as the prediction accuracy, recall rate, and F1 value of the model. If the evaluation metrics do not reach the preset threshold (such as the accuracy being lower than 90%), it is necessary to return to the model training stage and further optimize the model by increasing training data, adjusting feature selection, optimizing algorithm parameters, etc. until the performance requirements are met. Finally, deploy the optimized SVM model to the production environment to predict and analyze the environmental factor data in the actual scenario. Display the change trend of the data residue situation in different environments through visualization tools, such as the change curve of the residual data volume over time and the pie chart of the residual distribution of different types of data, providing intuitive decision-making support for data security managers.

[0155] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A touch screen performance quality detection and analysis method, characterized in that: The method comprises the following steps: Obtaining a preset uninstallation operation simulation algorithm and touch screen test data; Using the uninstall operation simulation algorithm to process the test data of the touch screen to obtain a preliminary data cleaning result; According to the preliminary data cleaning result, optimizing the parameters of the unloading operation simulation algorithm through multiple rounds of iterations to obtain an optimized unloading operation simulation algorithm; Using the optimized uninstallation operation simulation algorithm to process the test data of the touch screen to obtain a data cleaning capability evaluation result of the touch screen; Extracting data residual features according to the data cleaning capability evaluation result of the touch screen; Inputting the data residue feature into a preset fault location model to obtain location and type information of the data residue; Generate a fault location report based on the location and type information of the data residue; Obtain environmental adaptability test data of touch screens under extreme conditions such as high temperature, humidity, and vibration; Constructing an environmental adaptability evaluation model according to the environmental adaptability test data; The environmental adaptability evaluation model is used to predict the data residual conditions in different environments to obtain the changing trend of the data residual.

2. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The obtaining of the preset uninstallation operation simulation algorithm and the test data of the touch screen includes: Obtaining a preset uninstallation operation simulation algorithm and touch screen test data as input data; According to the uninstallation operation and touch screen test requirements in the business content, the simulation algorithm is used to analyze and calculate the input data to obtain the uninstallation operation simulation results and touch screen test result data that meet the business requirements; According to the unloading operation simulation result, judging whether the unloading operation meets the preset operation specifications and process requirements, and if not, performing optimization adjustment to obtain an optimized unloading operation simulation result; According to the touch screen test result data, a data mining algorithm is used to perform statistical analysis on it to obtain an evaluation result of the touch screen performance and stability; Summarize the optimized uninstall operation simulation results and the touch screen test evaluation results to generate a comprehensive test analysis report that meets business needs; According to the comprehensive test analysis report, determine whether the uninstall operation simulation algorithm and the touch screen test process need to be further optimized. If necessary, return to the analysis and calculation processing step for iterative optimization.

3. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The method of using the uninstall operation simulation algorithm to process the test data of the touch screen to obtain a preliminary data cleaning result includes: Acquire test data of the touch screen, input the test data into a preset uninstallation operation simulation algorithm for processing, and obtain preliminary processing results; Based on the preliminary processing results, a preset data cleaning technology is used to clean the data to obtain a cleaned processing result; According to the preset data quality assessment rules, the quality of the cleaned processing result is assessed, and if the quality assessment result meets the preset threshold, the cleaned processing result is determined as the final test data processing result; If the quality assessment result does not reach the preset threshold, the cleaned processing result is re-input into the unloading operation simulation algorithm for secondary processing to obtain a secondary processing result; Repeating data cleaning and quality assessment until the quality assessment result meets the preset threshold, and determining the final cleaned processing result as the processing result of the touch screen test data; Storing the processing results of the touch screen test data in a preset result database; The processing results of the touch screen test data are visualized through a preset data visualization technology.

4. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The method of optimizing the parameters of the uninstallation operation simulation algorithm through multiple rounds of iterations according to the preliminary data cleaning result to obtain an optimized uninstallation operation simulation algorithm includes: Obtaining initial unloading operation simulation algorithm parameters as input for the first round of iterative optimization; For the algorithm parameters of the current round, simulate the uninstallation operation and obtain the operation result data; Cleaning the operation result data, removing abnormal values ​​and noise data, and obtaining cleaned result data; Compare the cleaned result data with the preset optimization target, and use the support vector machine algorithm to calculate the current parameter optimization degree; If the optimization degree does not reach the preset threshold, the parameters of the unloading operation simulation algorithm are adjusted by a gradient descent algorithm according to the current round of cleaning results, and the adjusted algorithm parameters are used as input for the next round of iterative optimization; If the optimization degree reaches a preset threshold, the current algorithm parameters are determined as optimized uninstallation operation simulation algorithm parameters; According to the optimized uninstallation operation simulation algorithm parameters, the uninstallation operation simulation is re-executed to obtain optimized operation result data.

5. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The method of using the optimized uninstall operation simulation algorithm to process the test data of the touch screen to obtain a data cleaning capability evaluation result of the touch screen includes: Acquire test data of the touch screen, perform preprocessing on the test data, remove noise and abnormal values, and obtain cleaned test data; According to the preset unloading operation model, feature extraction is performed on the cleaned test data to obtain a feature vector of the test data; The feature vector of the test data is input into the optimized uninstallation operation simulation algorithm, and the uninstallation operation process of the touch screen is simulated by the algorithm to obtain the simulated uninstallation operation result; Perform statistical analysis on the simulated offloading operation results, calculate indicators such as the success rate, failure rate, and average response time of the offloading operation, and obtain the performance evaluation results of the offloading operation; According to the preset data cleaning capability evaluation rules, the performance evaluation results of the uninstallation operation are compared with the rules to determine whether the data cleaning capability of the touch screen meets the standards and obtain a preliminary evaluation result; The test data is divided into multiple subsets by using the cross-validation method. Different subsets are used to train and test the unloading operation simulation algorithm, and multiple evaluation results are obtained. A comprehensive analysis is performed on multiple evaluation results, the mean and variance of the evaluation results are calculated, and the final evaluation result of the data cleaning capability of the touch screen is determined according to the size of the mean and variance.

6. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The extracting data residual features according to the data cleaning capability evaluation result of the touch screen comprises: Acquire multiple data cleaning logs of the touch screen, and determine a data cleaning capability evaluation result of the touch screen according to a preset data cleaning capability evaluation rule; According to the data cleaning capability evaluation result, a data residual feature extraction algorithm is used to obtain the data residual feature of the touch screen; According to the data residue characteristics, combined with a pre-built data residue risk assessment model, the data residue risk level of the touch screen is determined; If the data residual risk level exceeds a preset threshold, a data cleaning reminder is triggered, and a targeted data cleaning strategy is determined based on the data residual characteristics; Sending the data cleaning strategy to the touch screen, executing the data cleaning strategy through the touch screen, and improving the data cleaning effect of the touch screen; After completing the data cleaning, obtaining the updated data cleaning log of the touch screen, and re-evaluating the data cleaning capability and data residual risk level of the touch screen; According to the changes in the evaluation results, the data residual feature extraction algorithm and the data residual risk evaluation model are dynamically adjusted to continuously optimize the accuracy of the data cleaning capability evaluation and the data residual risk warning.

7. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The step of inputting the data residue feature into a preset fault location model to obtain the location and type information of the data residue includes: Obtaining residual characteristic information of data; Inputting the characteristic information into a preset fault location model to obtain location information and type information of the data residue; If the location information satisfies a first preset condition, adding the location information to a candidate set; If the type information satisfies a second preset condition, adding the type information to the candidate set; According to the position information and type information in the candidate set, a support vector machine algorithm is used to perform classification to obtain the target data residual position and the target data residual type; The target data residual position and the target data residual type are used as output results for subsequent data recovery or security protection.

8. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The generating of a fault location report according to the location and type information of the data residue comprises: Obtain the location and type information of data residues; According to the location information and type information, a data analysis technique is used to extract features of data residues to obtain a feature vector; Input the feature vector into a pre-established fault location model, and determine the location and cause of the fault through model reasoning; If the confidence of the fault location and cause output by the fault location model is greater than a preset threshold, the fault location and cause information is written into the fault location report; If the confidence of the fault location and cause output by the fault location model is less than or equal to the preset threshold, the fault location and cause are determined by using a rule-based reasoning method based on the location information and type information, and the reasoning result is written into the fault location report; According to the fault location information in the fault location report, obtain the device model and configuration parameters at the location; By querying a pre-established equipment knowledge base, the fault diagnosis method and maintenance plan corresponding to the equipment model are obtained; The fault location report, fault diagnosis method and maintenance plan are stored in association to form a complete fault handling plan, which is output to relevant technicians; Obtaining feedback information on fault diagnosis and maintenance for optimizing the fault location model and equipment knowledge base.

9. The touch screen performance quality detection and analysis method according to claim 1, characterized in that: The acquisition of environmental adaptability test data of the touch screen under extreme conditions such as high temperature, humidity, vibration, etc. includes: According to the preset temperature, humidity and vibration frequency thresholds, multiple sets of extreme environmental condition parameter combinations are designed to build a touch screen environmental adaptability test plan; For each set of extreme environmental condition parameter combinations, obtain a touch screen to be tested, place it in the corresponding high temperature, humidity and vibration environment, and connect it to the signal acquisition equipment; When the touch screen is in the current environmental conditions for a preset period of time, the various performance parameters of the touch screen are recorded in real time by the signal acquisition device to obtain the time series data of the touch screen performance under the environmental conditions; Data cleaning and feature extraction are performed on the touch screen performance time series data collected under each group of extreme environmental conditions to obtain a feature vector set of touch screen performance under different environmental conditions; The support vector machine algorithm is used to train the above feature vector set, and a correlation model between touch screen performance and extreme environmental conditions is established to quantitatively evaluate the environmental adaptability of the touch screen; Using the established evaluation model, for a given touch screen, predict its performance under various extreme environmental conditions and determine whether it meets the environmental adaptability requirements; If the prediction results show that the environmental adaptability of the touch screen does not meet the requirements, optimization suggestions are made for the weak links, and the optimized touch screen is further tested until it meets the requirements.

10. A touch screen performance quality detection and analysis system, characterized in that: The touch screen performance quality detection and analysis method according to any one of claims 1 to 9 is used to detect and analyze the touch screen performance quality.

Images