Analog parameter test system based on digital test equipment

Abstract

The application discloses an analog parameter test system based on digital test equipment and relates to the field of analog test, which comprises a verification control module, a rule screening module and a complete calculation module.The verification control module is used for obtaining the acquisition authority of the running and setting parameters of the specified test equipment and sending control instructions to each functional module.The rule screening module is used for establishing a database, storing the measurable item data of the test equipment, predefining the screening rules of the measurable items of the test equipment, screening the corresponding data based on the rules and processing the data in categories.The complete calculation module is used for setting various parameter indexes for each test item, calculating the completeness of various categories of parameters under the current test item, counting the missing data of the indexes, identifying and preferentially supplementing the high-sensitive parameters with higher values for the test items, avoiding the interference of low-value parameters on the model while ensuring the prediction efficiency, identifying the factors that have the greatest impact on the test results, predicting potential risks in advance, and automatically adjusting the direction and amplitude of the missing parameters.

Description

Technical Field

[0001] This invention relates to the field of simulation testing technology, specifically to a simulation parameter testing system based on digital testing equipment. Background Technology

[0002] In the process of electronic product research and development, testing and quality control, testing equipment often involves a large number of complex parameters and data. The lack or inaccuracy of these parameters directly affects the reliability and validity of the test results. With the improvement of semiconductor process precision and the popularization of testing scenarios, traditional testing methods face challenges such as a large parameter system, dynamic changes in data gaps, and low testing efficiency. In existing technologies, actual testing costs are high, and simulation testing is often affected by missing data. When data is missing, it is difficult to obtain reliable supplementary predictions. Existing technologies cannot accurately locate the missing correlations between multi-dimensional parameters, and cannot effectively predict the thresholds of key parameters when data is incomplete. It is also difficult to effectively identify the direction and magnitude of data gaps, and cannot correct preset indicators based on the data performance trends in simulation results. It is also impossible to clearly determine which missing parameters have a greater impact on system testing, resulting in insufficient credibility of the test model and failure to adjust the test strategy in a timely manner. Intelligent testing scenarios require the system to have self-correcting capabilities to adapt to factors such as equipment aging and environmental drift, and dynamic optimization needs to be achieved by combining historical data modeling and real-time trend analysis. Summary of the Invention

[0003] (a) Technical problems to be solved In view of the above-mentioned shortcomings of the prior art, the present invention provides a simulation parameter testing system based on digital testing equipment, which can effectively solve the problems of the prior art.

[0004] (II) Technical Solution To achieve the above objectives, the present invention is implemented through the following technical solutions: This invention discloses a simulation parameter testing system based on digital testing equipment, comprising: The verification control module is used to obtain the permission to access the operation and setting parameters of the specified test equipment and to send control commands to each functional module. The rule filtering module is used to establish a database to store data on each testable item of the test equipment, predefine the filtering rules for the testable items of the test equipment, filter the corresponding data based on the rules, and classify and process them. The complete calculation module is used to set various parameters for each test item, calculate the completeness of each category of parameters under the current test item, and count the missing data of the indicators. The missing data analysis module receives data submitted by the complete calculation module, uses matrix analysis to locate data gaps, and identifies the direction and magnitude of missing parameters. The correction model module is used to build a correction model through a time series network algorithm. The correction model predicts the range of values ​​for missing parameters based on the currently available parameter data, and forms a supplementary parameter set based on the predicted range of values. The simulation analysis unit is used to construct a Verilog-A model of the current test equipment behavior based on the supplemented parameter set, obtain the changing trend of the predefined key indicators of the current test project within a specific time period, identify key influencing factors, and output the weights of several key indicators. The simulation output module is used to input the trend analysis results and preset indicator weights into the Verilog-A model to simulate the performance of each test item indicator under a specific period, calculate the peak, mean and fluctuation range, and output the simulation results data and corresponding evaluation reports. The optimization unit is used to evaluate the contribution of supplementary prediction data to the final simulation test results, adjust the direction and magnitude of missing indicators in the current state data, and correct them according to a preset scale.

[0005] Furthermore, the simulation analysis unit is further equipped with sub-modules, including: a test model module, a trend analysis module, and a weight allocation module. The test model module and the trend analysis module are interconnected via a wireless network, and the trend analysis module and the weight allocation module are interconnected via a wireless network. The test model module is used to build a basic Verilog-A model for simulation testing based on the attribute information of the testable items of the test equipment and historical test case data. The supplemented parameter set after the supplemented and corrected model is input into the basic Verilog-A model, key project indicators are defined within a specific test period, and corresponding test scenarios are built for simulation. The trend analysis module is used to perform time-series trend analysis on historical data and current test data, and output the changing trend of each indicator within a specific period. The weighting module is used to identify key factors affecting the performance of measurement indicators in the trend analysis results, classify them into different factor categories according to the nature of the influencing factors, and calculate the weights of several key indicators under the current test item based on the classification results.

[0006] Furthermore, the process by which the weight allocation module calculates the weights of several key indicators under the current test item includes: Step a: Based on the trend analysis results, extract the change characteristics of each test indicator within a specific time period. According to the mapping relationship between the characteristics and the predefined factor categories, identify the key factors that have a significant impact on the performance of the indicators, and classify them into the corresponding factor categories according to physical attributes, time-series correlation and equipment parameter type. Step b: For each factor category, establish a weight allocation matrix by combining the contribution distribution of similar factors in historical data; Step c: Process the weight allocation matrix using a normalization algorithm to generate an initial weight set, and dynamically adjust it by introducing the constraints of the current test environment; Step d: Input the calculated weights into the Verilog-A model for simulation verification. If the deviation between the performance of the indicators and the preset target in the simulation results exceeds the threshold, repeat steps a to c until the weight set converges to a stable state.

[0007] Furthermore, the predefined factor categories in step a include inherent device parameters, environmental noise parameters, and test timing parameters. When classifying, the Euclidean distance algorithm is used to match feature vectors with category templates.

[0008] Furthermore, the weight allocation matrix in step b includes the sensitivity coefficients of each factor category to the test index, the intensity of time-series influence, and physical constraints. The sensitivity coefficients of the weight allocation matrix are determined by covariance analysis, the intensity of time-series influence is calculated by an autoregressive model, and the physical constraints are generated by mapping through a database of equipment parameter thresholds.

[0009] Furthermore, during the process of locating data gaps, the missing data analysis module maps the parameter completeness data of each test item into a two-dimensional matrix, where the row vectors represent parameter categories, the column vectors represent test items, and the matrix element values ​​represent the existence status and quantitative completeness of the corresponding parameter in a specific test item. Principal components of the matrix are extracted by singular value decomposition, the projection distribution of parameter missing directions in the feature vector space is identified, the column missing density and row missing density of each parameter category are calculated, and the missing direction vector is established based on the joint analysis of projection distribution and missing density-intensity. The overall missing proportion is calculated based on the matrix sparsity index, and a missing magnitude assessment value with directional weights is generated by combining the magnitude of the missing direction vector.

[0010] Furthermore, the construction process of the correction model in the correction model module is as follows: A parameter prediction architecture is constructed based on a temporal network algorithm. A multi-layer neural network structure with temporal correlation is established. The network structure includes an input layer, a hidden layer, and an output layer. The input layer receives standardized data of currently available parameters, and the output layer generates predicted values ​​of missing parameters. The complete parameter set is extracted from the historical database as training samples. The samples are divided into training set and validation set according to time series. The network weights are optimized by backpropagation algorithm to minimize the root mean square error between the predicted value output by the corrected model and the actual historical data. Define a parameter sensitivity evaluation function, calculate the sensitivity coefficient based on the response gradient of the test item to each parameter, and generate a sensitivity ranking list. The function satisfies the following: ; in Let i be the sensitivity of the i-th parameter. To test project metrics, The current parameter value. For the standard deviation of the parameter, The average of the indicators; Set threshold constraints to truncate the range of predicted parameter values; Missing parameters are assigned priority values ​​based on the sensitivity ranking list, and parameters with sensitivity higher than a preset threshold are added to the parameter set first, generating a supplemented parameter set containing predicted values.

[0011] Furthermore, the optimization unit has sub-modules deployed at its lower levels. These sub-modules include a contribution evaluation module, a dynamic adjustment module, and a configuration application module. The contribution evaluation module interacts with the dynamic adjustment module and the configuration application module via a wireless network. The contribution evaluation module is used to evaluate the contribution of supplementary prediction data to the final simulation test results. When the contribution is lower than the preset threshold, a data correction reminder is triggered, and the deviation between the supplemented parameter set and the preset target data is compared. The dynamic adjustment module is used to obtain the evaluation results of the contribution of the predicted data. When the contribution evaluation module determines that the contribution of the predicted data is lower than the preset threshold, it adjusts the direction and magnitude of the missing indicators in the current state data and corrects them according to the preset scale. The configuration application module is used to re-input the state data adjusted by the dynamic adjustment module as input to the correction model module, and then use the correction model for the next round of prediction.

[0012] Furthermore, the correction logic for the direction and magnitude of missing indicators in the dynamic adjustment module is as follows: Obtain the projection components of the missing direction vector in each feature vector space, extract the amplitude and polarity of the projection components, obtain the main direction weight of the missing parameters, and combine the column missing density and row missing density to output the missing correlation strength of each parameter category in the continuous test items. Based on the product of the matrix sparsity index and the magnitude of the gap direction vector, a direction-weighted missing magnitude coefficient is generated, and the preset scale is defined as a combination function of dynamically adjusting the step size and the direction weight factor. Based on the aforementioned direction weight factor, the direction vector of the missing parameter is normalized, and the adjustment step size is allocated to the missing direction of the corresponding parameter category according to the weight, thereby generating a correction amount with direction bias. The magnitude range of the correction amount is constrained by a preset threshold. If the correction amount exceeds the threshold, it is compressed proportionally to the maximum allowable value. The corrected direction vector and amplitude value are superimposed on the missing position of the current state data, the magnitude and sparsity index of the gap direction vector are updated, and the process is iterated until the contribution reaches the preset threshold.

[0013] Furthermore, the verification control module is interconnected with the rule filtering module and the complete calculation module via a wireless network; the missing data analysis module is interconnected with the complete calculation module, the correction model module, and the simulation analysis unit via a wireless network; and the simulation output module is interconnected with the simulation analysis unit and the optimization unit via a wireless network.

[0014] (III) Beneficial Effects Compared with known prior art, the technical solution provided by this invention has the following beneficial effects: By automatically filtering and classifying data based on test item attributes, the system quickly locates parameter incompleteness gaps, accurately identifies the direction and magnitude of missing indicators, and establishes a correction model with temporal correlation. Through sensitivity ranking and physical threshold constraints, it identifies and prioritizes the addition of highly sensitive parameters that are of high value to the test items. This ensures prediction efficiency while avoiding interference from low-value parameters, improving the effectiveness of the supplemented parameter set. It achieves efficient correction of missing parameter priority assignment and supplemented prediction data, ensuring continuous optimization of prediction values ​​under physical constraints.

[0015] By using simulation results output by the Verilog-A model through the simulation analysis unit, time-specific indicators are defined and trends are analyzed. The weights of key influencing factors are quantified, and the factors with the greatest impact on test results are identified. The system can further simulate future test scenarios, predict potential risks in advance, and dynamically adjust the indicator weights based on the classification of influencing factors, making the test model more in line with actual working conditions, avoiding deviations caused by static weights, and improving the test adaptability in complex scenarios.

[0016] The effectiveness of the predicted data is judged by introducing a contribution threshold through optimization unit. Based on the projection component of the gap direction vector and the correlation strength of the missing data, a dynamic adjustment step size and direction weight factor are defined. The missing data is updated by threshold compression and iterative superposition until the contribution reaches the standard. If the supplementary data does not improve the trend analysis sufficiently, the direction and magnitude of the missing parameters are automatically adjusted and the predicted value is regenerated to improve the accuracy of data correction and reduce manual intervention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 This is a schematic diagram of the overall framework of the present invention; Figure 2 This is a schematic diagram of the simulation analysis unit in this invention; Figure 3 This is a schematic diagram of the framework of the optimization unit in this invention.

[0019] The labels in the diagram represent: 1. Verification and control module; 2. Rule filtering module; 3. Complete calculation module; 4. Missing data analysis module; 5. Correction model module; 6. Simulation analysis unit; 61. Test model module; 62. Trend analysis module; 63. Weight allocation module; 7. Simulation output module; 8. Optimization unit; 81. Contribution evaluation module; 82. Dynamic adjustment module; 83. Configuration and application module. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] The present invention will be further described below with reference to embodiments.

[0022] ① Example 1 This embodiment presents a simulated parameter testing system based on digital testing equipment, such as... Figure 1 - Figure 3 As shown, it includes: Verification control module 1 is used to obtain the permission to access the operation and setting parameters of the specified test equipment and to send control commands to each functional module; Rule filtering module 2 is used to establish a database to store data on each testable item of the test equipment, predefine filtering rules for the testable items of the test equipment, filter the corresponding data based on the rules and classify them; during the classification process, rule filtering module 2 classifies the data according to function category, test mode category and historical test record category; Complete calculation module 3 is used to set various parameters for each test item, calculate the completeness of each category of parameters under the current test item, and count the missing data of the indicators; The missing data analysis module 4 is used to receive data submitted by the complete calculation module 3, and to locate data gaps using matrix analysis to identify the direction and magnitude of missing parameters. During the process of locating data gaps, the missing data analysis module 4 maps the parameter index completeness data under each test item into a two-dimensional matrix, where the row vectors represent parameter categories, the column vectors represent test items, and the matrix element values ​​represent the existence status and quantitative completeness of the corresponding parameter in a specific test item. Principal components of the matrix are extracted by singular value decomposition, the projection distribution of parameter missing directions in the feature vector space is identified, and the column missing density (number of missing items / total number of test items) and row missing density (number of consecutive missing test items / total number of test items) of each parameter category are calculated. Based on the joint analysis of projection distribution and missing density-intensity, a gap direction vector is established. Based on the matrix sparsity index (1 - number of non-empty elements / total number of matrix elements), the overall missing ratio is calculated, and a missing magnitude assessment value with directional weights is generated by combining the magnitude of the missing direction vector. By constructing a parameter and test item matrix and integrating singular value decomposition and multi-dimensional missing data quantification algorithms, this approach overcomes the limitations of traditional single statistical or local imputation methods. Existing technologies typically rely solely on linear analysis based on discrete parameter missing rates, making it difficult to capture the correlation gap patterns between parameters and test items. In contrast, this solution combines directional vector projection analysis with joint calculation of dynamic density and intensity. This not only accurately locates the direction of systematic parameter missing across test items but also quantifies the spatial distribution characteristics of missing magnitude through the magnitude of the missing direction vector. This significantly improves the ability to identify latent data gaps and provides structured missing data assessment data with spatial weight attributes for subsequent model correction. It also solves the problem of error accumulation caused by neglecting the topological relationship between parameters in existing technologies. The correction model module 5 is used to build a correction model through a time series network algorithm. The correction model is trained using historical data as training samples. Based on the currently available parameter data, the correction model predicts the numerical range of missing parameters and forms a supplemented parameter set based on the predicted numerical range. Priority is given to supplementing parameters that are highly sensitive to the test items, and the supplemented values ​​are limited to a physically reasonable threshold. The process of constructing the corrected model is as follows: A parameter prediction architecture is constructed based on the temporal network algorithm. A multi-layer neural network structure with temporal correlation is established. The network structure includes an input layer, a hidden layer, and an output layer. The input layer receives the standardized data of the currently available parameters, and the output layer generates the predicted values ​​of the missing parameters. The complete parameter set is extracted from the historical database as training samples. The samples are divided into training set and validation set according to time series. The network weights are optimized by backpropagation algorithm to minimize the root mean square error between the predicted value output by the corrected model and the actual historical data. Define a parameter sensitivity evaluation function that calculates sensitivity coefficients based on the response gradients of the test items to each parameter, and generates a sensitivity ranking list. The function satisfies the following: ; in Let i be the sensitivity of the i-th parameter. To test project metrics, The current parameter value. For the standard deviation of the parameter, The average of the indicators; Set threshold constraints to truncate the range of predicted parameter values; Missing parameters are assigned priority values ​​based on the sensitivity ranking list, and parameters with sensitivity higher than the preset threshold are added to the parameter set first, generating a supplemented parameter set containing predicted values; Simulation analysis unit 6 is used to construct a Verilog-A model of the current test equipment behavior based on the supplemented parameter set, obtain the changing trend of the predefined key indicators of the current test project within a specific time period, identify key influencing factors, output the weights of several key indicators, embed test environment constraints, support user-defined test indicators, and generate adapted test vectors. The simulation analysis unit 6 has sub-modules deployed below it, including: a test model module 61, a trend analysis module 62, and a weight allocation module 63. The test model module 61 and the trend analysis module 62 are interconnected via a wireless network, and the trend analysis module 62 and the weight allocation module 63 are interconnected via a wireless network. The test model module 61 is used to build a basic Verilog-A model for simulation testing based on the attribute information of the testable items of the test equipment and historical test case data. The supplemented parameter set after the supplemented and corrected model is input into the basic Verilog-A model, key project indicators are defined within a specific test period, and corresponding test scenarios are constructed for simulation operation. The trend analysis module 62 is used to perform time-series trend analysis on historical data and current test data, and output the changing trend of each indicator within a specific time period. The weight allocation module 63 is used to identify key factors affecting the performance of measurement indicators in the trend analysis results, classify them into different factor categories according to the nature of the influencing factors, and calculate the weights of several key indicators under the current test item based on the classification results. The simulation output module 7 is used to input the trend analysis results and preset indicator weights into the Verilog-A model, simulate the performance of each test item indicator under a specific period, calculate the peak, mean and fluctuation range, and output the simulation results data and corresponding evaluation report. Optimization unit 8 is used to evaluate the contribution of supplementary prediction data to the final simulation test results, adjust the direction and magnitude of missing indicators in the current state data, and correct them according to a preset scale. The optimization unit 8 has sub-modules deployed below it, including a contribution evaluation module 81, a dynamic adjustment module 82, and a configuration application module 83. The contribution evaluation module 81 interacts with the dynamic adjustment module 82 and the configuration application module 83 via a wireless network. The contribution evaluation module 81 is used to evaluate the contribution of the supplementary prediction data to the final simulation test results. When the contribution is lower than the preset threshold, a data correction reminder is triggered to compare the deviation between the supplemented parameter set and the preset target data. The dynamic adjustment module 82 is used to obtain the evaluation results of the contribution of the predicted data. When the contribution evaluation module 81 determines that the contribution of the predicted data is lower than the preset threshold, it adjusts the direction and magnitude of the missing indicators in the current state data and corrects them according to the preset scale. The configuration application module 83 is used to re-input the state data adjusted by the dynamic adjustment module 82 as the input of the correction model module 5, and then perform the next round of prediction through the correction model. Verification control module 1 is interconnected with rule filtering module 2 and complete calculation module 3 via a wireless network. Missing data analysis module 4 is interconnected with complete calculation module 3, correction model module 5 and simulation analysis unit 6 via a wireless network. Simulation output module 7 is interconnected with simulation analysis unit 6 and optimization unit 8 via a wireless network.

[0023] Compared with the prior art, this embodiment classifies test data according to functional categories, test modes and historical test records, and uses the matrix singular value decomposition method to accurately locate parameter gaps. This not only achieves a quantitative assessment of data integrity, but also generates a missing magnitude assessment value with directional weights by combining the gap direction vector and matrix sparsity index, thereby accurately predicting and correcting missing parameters while ensuring the rationality of the data. By constructing a modified model of a multi-layer neural network using temporal network algorithms, the parameter value range can be effectively truncated based on backpropagation weight optimization. Parameter sensitivity can be ranked according to the response gradient of the test items, thus prioritizing the addition of parameters with high sensitivity to the test items. Through a multi-level strategy of Verilog-A model construction and trend analysis, key factor identification, and weight allocation, the performance of key indicators of each test item within a specific time period can be simulated and evaluated in real time and accurately. This significantly improves the integrity and reliability of the data and the prediction accuracy of the test indicators, and has higher adaptability, prediction accuracy, and operating efficiency compared to existing technologies.

[0024] ② Example 2 At other levels, this embodiment also provides another optimization mechanism based on embodiment 1, specifically a scheme for calculating the weights of several key indicators under the current test item, the specific process of which includes: Step a: Based on the trend analysis results, extract the change characteristics of each test indicator within a specific time period. According to the mapping relationship between the characteristics and the predefined factor categories, identify the key factors that have a significant impact on the performance of the indicators, and classify them into the corresponding factor categories according to physical attributes, time series correlation, and equipment parameter types. The predefined factor categories include inherent equipment parameters, environmental noise parameters, and test time series parameters. When classifying, the Euclidean distance algorithm is used to match the feature vector with the category template. Step b: For each factor category, establish a weight allocation matrix by combining the contribution distribution of similar factors in historical data; the weight allocation matrix includes the sensitivity coefficient, time-series influence strength and physical constraints of each factor category to the test index. The sensitivity coefficient of the weight allocation matrix is ​​determined by covariance analysis, the time-series influence strength is calculated by the autoregressive model, and the physical constraints are generated by mapping through the equipment parameter threshold library. Step c: Process the weight allocation matrix using a normalization algorithm to generate an initial weight set, and dynamically adjust it by incorporating the constraints of the current test environment; ensure that the weights of each key indicator meet the requirements. =1, and ∈[0,1], where This represents the final weight value of the i-th indicator; dynamic adjustment uses the Lagrange multiplier method, with environmental constraints as boundary conditions to optimize the initial weights; Step d: Input the calculated weights into the Verilog-A model for simulation verification. If the deviation between the performance of the indicators and the preset target in the simulation results exceeds the threshold, repeat steps a to c until the weight set converges to a stable state.

[0025] In this embodiment, feature extraction is performed on the trend analysis results, and precise factor matching is achieved using the Euclidean distance algorithm. The significant influencing factors of each test index are scientifically classified according to the inherent parameters of the equipment, environmental noise parameters, and test time series parameters, thereby optimizing the identification process of the influencing factors of key indicators. The contribution information of each factor in historical data is fully utilized, and physical constraints are generated by mapping the physical parameter threshold library, making the weight calculation more consistent with the actual test environment and equipment attributes. The calculated weights are fed back to the Verilog-A model for simulation verification, and automatic iterative optimization is performed when the deviation exceeds the standard, which significantly improves the accuracy of the simulation results and the overall robustness of the system. Compared with the prior art, this embodiment has obvious advantages in improving the prediction accuracy of key indicators, dynamic adaptive capability, and data processing efficiency.

[0026] ③ Example 3 This embodiment provides a correction logic for the direction and magnitude of missing indicators, specifically as follows: Obtain the projection components of the gap direction vector in each feature vector space, extract the magnitude and polarity of the projection components, obtain the main direction weight of the missing parameters, and combine the column missing density and row missing density to output the missing correlation strength of each parameter category in the continuous test items. Based on the product of the matrix sparsity index and the magnitude of the missing direction vector, a direction-weighted missing amplitude coefficient is generated. The preset scale is defined as a combination function of dynamic adjustment step size and direction weight factor. Adjustment step size = missing amplitude coefficient × log(1 + missing correlation strength); direction weight factor = projection component amplitude / ∑ projection component amplitude × polarity coefficient. Based on the direction weight factor, the direction vector of the missing parameter is normalized, and the adjustment step size is allocated to the missing direction of the corresponding parameter category according to the weight, generating a correction amount with direction bias. The magnitude range of the correction amount is constrained by a preset threshold. If the correction amount exceeds the threshold, it is compressed proportionally to the maximum allowable value to ensure the physical feasibility of supplementing the predicted value. The corrected direction vector and amplitude value are superimposed on the missing position of the current state data, the magnitude and sparsity index of the gap direction vector are updated, and the process is iterated until the contribution reaches the preset threshold.

[0027] This embodiment organically combines the directionality and amplitude dynamic adjustment of missing indicators, which can not only accurately capture multidimensional missing information in test data, but also take into account physical constraints during the correction process, achieving efficient, precise and adaptive data completion. Compared with the existing technology, it has significant advantages in terms of missing information accuracy, model robustness and iterative optimization efficiency.

[0028] Working Principle: When the system of this invention is installed, the verification control module 1 obtains control authority over the test equipment. The rule filtering module 2 filters the corresponding available data in the database based on the testable items of the test equipment and classifies them. The completeness calculation module 3 calculates the completeness of each category of parameters on a certain test item. The missing data analysis module 4 analyzes the state data of the direction and magnitude of the missing indicators of the current completeness. The correction model module 5 constructs a supplementary correction model based on the original missing parameter state data. Using historical data as a reference, it predicts and supplements the missing parts of the currently available parameters. The test model module 61 constructs a simulation test model based on the testable item attributes of the test equipment and its historical test case data. The supplemented parameters are input. The trend analysis module 62 defines the project indicators required for a specific period. Based on the defined indicators, trend analysis is performed. Based on the trend analysis results, the weight allocation module 63 classifies the influencing factors in the trend analysis and calculates the weights of several indicators under the current test item based on the classification results. Based on the trend analysis results and several preset indicator weights, the simulation output module 7 extracts the trained test model and runs it in a simulation to output the data performance of a certain project under a specific period. The contribution assessment module 81 assesses the contribution of the predicted supplementary data based on the trend analysis results. When the contribution is insufficient, the dynamic adjustment module 82 adjusts the direction and magnitude of the missing indicators of the status data according to the preset scale. The configuration application module 83 resubmits the adjusted status data to the missing analysis module 4 and the correction model module 5 for the next round of prediction. This system addresses the core pain points of low data reliability and poor model adaptability in complex testing environments by constructing a closed loop for parameter integrity assessment, matrix gap location, and prediction correction. It meets the urgent need for automated and intelligent testing systems in the R&D verification and mass production testing of high-precision digital equipment.

[0029] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A simulation parameter testing system based on digital testing equipment, characterized in that, include: The verification control module (1) is used to obtain the permission to obtain the operation and setting parameters of the specified test equipment and to send control commands to each functional module. The rule filtering module (2) is used to establish a database, store the data of each testable item of the test equipment, predefine the filtering rules of the testable items of the test equipment, filter the corresponding data based on the rules and classify and process them. The complete calculation module (3) is used to set various parameter indicators for each test item, calculate the completeness of each type of parameter under the current test item, and count the missing data of the indicators. The missing data analysis module (4) is used to receive data submitted by the complete calculation module (3), and to locate data gaps and identify the direction and magnitude of missing parameters using matrix analysis. The correction model module (5) is used to establish a correction model through a time series network algorithm. The correction model predicts the range of values ​​of missing parameters based on the currently available parameter data and forms a supplementary parameter set based on the predicted range of values. The simulation analysis unit (6) is used to construct a Verilog-A model of the behavior of the current test equipment based on the supplemented parameter set, obtain the changing trend of the key indicators predefined in the current test project in a specific period, identify key influencing factors, and output the weights of several key indicators. The simulation output module (7) is used to input the trend analysis results and preset indicator weights into the Verilog-A model, simulate the performance of each test item indicator under a specific period, calculate the peak, mean and fluctuation range, and output the simulation results data and corresponding evaluation report. The optimization unit (8) is used to evaluate the contribution of supplementary prediction data to the final simulation test results, adjust the direction and magnitude of missing indicators in the current state data, and correct them according to the preset scale.

2. The analog parameter testing system based on digital testing equipment according to claim 1, characterized in that, The simulation analysis unit (6) has sub-modules deployed below it, including: a test model module (61), a trend analysis module (62), and a weight allocation module (63). The test model module (61) and the trend analysis module (62) are interconnected via a wireless network, and the trend analysis module (62) and the weight allocation module (63) are interconnected via a wireless network. The test model module (61) is used to construct a basic Verilog-A model for simulation testing based on the attribute information of the testable items of the test equipment and historical test case data. The supplemented parameter set after the supplemented and corrected model is input into the basic Verilog-A model, key project indicators within a specific test period are defined, and corresponding test scenarios are constructed for simulation operation. The trend analysis module (62) is used to perform time-series trend analysis on historical data and current test data, and output the changing trend of each indicator within a specific period. The weight allocation module (63) is used to identify key factors affecting the performance of measurement indicators in the trend analysis results, classify them into different factor categories according to the nature of the influencing factors, and calculate the weights of several key indicators under the current test item based on the classification results.

3. The analog parameter testing system based on digital testing equipment according to claim 2, characterized in that, The process by which the weight allocation module (63) calculates the weights of several key indicators under the current test item includes: Step a: Based on the trend analysis results, extract the change characteristics of each test indicator within a specific time period. According to the mapping relationship between the characteristics and the predefined factor categories, identify the key factors that have a significant impact on the performance of the indicators, and classify them into the corresponding factor categories according to physical attributes, time-series correlation and equipment parameter type. Step b: For each factor category, establish a weight allocation matrix by combining the contribution distribution of similar factors in historical data; Step c: Process the weight allocation matrix using a normalization algorithm to generate an initial weight set, and dynamically adjust it by introducing the constraints of the current test environment; Step d: Input the calculated weights into the Verilog-A model for simulation verification. If the deviation between the performance of the indicators and the preset target in the simulation results exceeds the threshold, repeat steps a to c until the weight set converges to a stable state.

4. The analog parameter testing system based on digital testing equipment according to claim 3, characterized in that, In step a, the predefined factor categories include inherent equipment parameters, environmental noise parameters, and test timing parameters. When classifying, the Euclidean distance algorithm is used to match the feature vector with the category template.

5. The analog parameter testing system based on digital testing equipment according to claim 3, characterized in that, The weight allocation matrix in step b includes the sensitivity coefficients of each factor category to the test index, the intensity of time-series influence, and physical constraints. The sensitivity coefficients of the weight allocation matrix are determined by covariance analysis, the intensity of time-series influence is calculated by an autoregressive model, and the physical constraints are generated by mapping through a database of equipment parameter thresholds.

6. The analog parameter testing system based on digital testing equipment according to claim 1, characterized in that, In the process of locating data gaps, the missing analysis module (4) maps the parameter index completeness data under each test item into a two-dimensional matrix, where the row vector represents the parameter category, the column vector represents the test item, and the matrix element value represents the existence status and quantitative completeness of the corresponding parameter in a specific test item. Principal components of the matrix are extracted by singular value decomposition, the projection distribution of parameter missing directions in the feature vector space is identified, the column missing density and row missing density of each parameter category are calculated, and the missing intensity represented by the projection distribution, row missing density, column missing density and vector magnitude is jointly analyzed to establish the missing direction vector. The column missing density is calculated by dividing the number of missing items by the total number of test items, and the row missing density is calculated by dividing the number of consecutive missing test items by the total number of test items. The overall missing proportion is calculated based on the matrix sparsity index, and a missing magnitude assessment value with directional weights is generated by combining the magnitude of the missing direction vector.

7. The analog parameter testing system based on digital testing equipment according to claim 1, characterized in that, The construction process of the correction model in the correction model module (5) is as follows: A parameter prediction architecture is constructed based on a temporal network algorithm. A multi-layer neural network structure with temporal correlation is established. The network structure includes an input layer, a hidden layer, and an output layer. The input layer receives standardized data of currently available parameters, and the output layer generates predicted values ​​of missing parameters. The complete parameter set is extracted from the historical database as training samples. The samples are divided into training set and validation set according to time series. The network weights are optimized by backpropagation algorithm to minimize the root mean square error between the predicted value output by the corrected model and the actual historical data. Define a parameter sensitivity evaluation function, calculate the sensitivity coefficient based on the response gradient of the test item to each parameter, and generate a sensitivity ranking list. The function satisfies the following: ； in Let i be the sensitivity of the i-th parameter. To test project metrics, The current parameter value. For the standard deviation of the parameter, The average of the indicators; Set threshold constraints to truncate the range of predicted parameter values; Missing parameters are assigned priority values ​​based on the sensitivity ranking list, and parameters with sensitivity higher than a preset threshold are added to the parameter set first, generating a supplemented parameter set containing predicted values.

8. The analog parameter testing system based on digital testing equipment according to claim 6, characterized in that, The optimization unit (8) has sub-modules deployed below it, including a contribution evaluation module (81), a dynamic adjustment module (82), and a configuration application module (83). The contribution evaluation module (81), the dynamic adjustment module (82), and the configuration application module (83) are interconnected via a wireless network. The contribution evaluation module (81) is used to evaluate the contribution of the supplementary prediction data to the final simulation test results. When the contribution is lower than the preset threshold, a data correction reminder is triggered to compare the deviation between the supplemented parameter set and the preset target data. The dynamic adjustment module (82) is used to obtain the evaluation result of the contribution of the prediction data. When the contribution evaluation module (81) determines that the contribution of the prediction data is lower than the preset threshold, it adjusts the direction and magnitude of the missing indicators in the current state data and corrects them according to the preset scale. The configuration application module (83) is used to re-input the state data adjusted by the dynamic adjustment module (82) as the input of the correction model module (5) and perform the next round of prediction through the correction model.

9. The analog parameter testing system based on digital testing equipment according to claim 8, characterized in that, The correction logic for the direction and magnitude of missing indicators in the dynamic adjustment module (82) is as follows: Obtain the projection components of the missing direction vector in each feature vector space, extract the amplitude and polarity of the projection components, obtain the main direction weight of the missing parameters, and combine the column missing density and row missing density to output the missing correlation strength of each parameter category in the continuous test items. Based on the product of the matrix sparsity index and the magnitude of the gap direction vector, a direction-weighted missing magnitude coefficient is generated, and the preset scale is defined as a combination function of dynamically adjusting the step size and the direction weight factor. Based on the aforementioned direction weight factor, the direction vector of the missing parameter is normalized, and the adjustment step size is allocated to the missing direction of the corresponding parameter category according to the weight, thereby generating a correction amount with direction bias. The magnitude range of the correction amount is constrained by a preset threshold. If the correction amount exceeds the threshold, it is compressed proportionally to the maximum allowable value. The corrected direction vector and amplitude value are superimposed on the missing position of the current state data, the magnitude and sparsity index of the gap direction vector are updated, and the process is iterated until the contribution reaches the preset threshold.

10. The analog parameter testing system based on digital testing equipment according to claim 1, characterized in that, The verification control module (1) is interconnected with the rule filtering module (2) and the complete calculation module (3) via a wireless network. The missing analysis module (4) is interconnected with the complete calculation module (3), the correction model module (5), and the simulation analysis unit (6) via a wireless network. The simulation output module (7) is interconnected with the simulation analysis unit (6) and the optimization unit (8) via a wireless network.

Images