Method, system, computer device, readable storage medium and program product for automated acceptance of equipment

Abstract

The application relates to a method and system for automatically checking equipment, a computer device, a computer readable storage medium and a computer program product. The method comprises the following steps: obtaining checking data, matching an initial checking process, adjusting a strategy according to real-time dynamic data, obtaining an optimized checking strategy through screening and parameter optimization, and finally realizing automatic and accurate checking of the equipment. In this way, the low-voltage equipment checking process is changed from static rule driving to dynamic intelligent self-adaptation. Through data-driven strategy construction, real-time dynamic adjustment, and process screening and parameter determination based on optimization, the adaptability, accuracy and efficiency of the low-voltage equipment checking work are improved.

Description

Technical Field

[0001] This application relates to the field of electrical engineering technology, and in particular to a method, system, computer equipment, computer-readable storage medium, and computer program product for automated equipment acceptance testing. Background Technology

[0002] With the continuous development of electrical technology, low-voltage equipment is widely used in many fields such as industrial production, civil buildings and power distribution. The quality and operational safety of low-voltage equipment directly affect many aspects of production and life. The acceptance of low-voltage equipment is a key link to ensure equipment quality and operational safety. It is necessary to test the equipment parameters and performance indicators to verify whether the equipment meets the standards and specifications.

[0003] Because low-voltage equipment is widely used and comes in many varieties, the acceptance standards for different models or applications vary significantly. Traditional acceptance methods rely heavily on manual operation, requiring acceptance personnel to select standards and procedures based on personal experience. Manual processes are prone to errors in standard matching and omissions, leading to inconsistent acceptance accuracy. Furthermore, acceptance standards need continuous revision as actual application conditions change, and traditional methods struggle to adapt quickly to these dynamic changes. Consequently, low-voltage equipment acceptance suffers from low efficiency, slow response, and poor adaptability. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, system, computer equipment, computer-readable storage medium, and computer program product for automated acceptance testing of low-voltage equipment, which can improve the acceptance accuracy, efficiency, and adaptability of low-voltage equipment.

[0005] Firstly, this application provides a method for automated acceptance testing of equipment, including:

[0006] Obtain the acceptance data set; the acceptance data set includes the attribute information of the equipment to be accepted, the acceptance standards corresponding to each attribute information, and historical acceptance records;

[0007] Based on the acceptance criteria and historical acceptance records, multiple initial acceptance processes corresponding to the attribute information are matched, and an initial acceptance strategy is constructed based on these multiple initial acceptance processes.

[0008] The system acquires real-time dynamic data of the equipment to be accepted and adjusts the initial acceptance strategy accordingly to obtain the first acceptance strategy. The first acceptance strategy includes multiple first target acceptance processes and first target process parameters corresponding to each first target acceptance process.

[0009] Multiple second-target acceptance processes are selected from multiple first-target acceptance processes, and process parameters of each second-target acceptance process are optimized to determine the second acceptance strategy; wherein, the second acceptance strategy includes multiple second-target acceptance processes and the second-target process parameters corresponding to each second-target acceptance process;

[0010] The equipment to be accepted is inspected based on the second target acceptance process and the corresponding second target process parameters, and the acceptance result data is obtained.

[0011] In one embodiment, the initial acceptance strategy includes multiple initial acceptance processes and initial process parameters corresponding to each acceptance process. The steps of matching multiple initial acceptance processes corresponding to attribute information based on acceptance criteria and historical acceptance records, and constructing the initial acceptance strategy based on the multiple initial acceptance processes include:

[0012] A judgment matrix is ​​constructed based on the acceptance criteria to determine the subjective weight of attribute information, and the objective weight is calculated based on the data dispersion of data information in historical acceptance records. The weighting coefficients of subjective and objective weights are adjusted according to the priority of attribute information in the acceptance scenario, and the subjective and objective weights are weighted and summed based on the weighting coefficients to obtain the comprehensive weight.

[0013] Based on the comprehensive weight, feature vectors corresponding to the attribute information of the equipment to be accepted are constructed, and the feature vectors are matched with the attribute requirement vectors of multiple preset acceptance standards to determine multiple initial acceptance processes for the equipment to be accepted.

[0014] Based on the priority and logical relationship of multiple initial acceptance processes, the testing sequence is planned, multiple initial acceptance processes are planned, and the pass threshold of the testing indicators corresponding to each acceptance process is determined, so as to obtain the initial process parameters corresponding to each acceptance process.

[0015] In one embodiment, the steps of acquiring real-time dynamic data of the equipment to be accepted and adjusting the initial acceptance strategy according to the real-time dynamic data to obtain a first acceptance strategy include:

[0016] Acquire real-time dynamic data of the equipment to be accepted during the acceptance process; wherein, the real-time dynamic data includes equipment status data to represent the real-time operating parameters of the equipment to be accepted, acceptance progress data to represent the acceptance implementation, and deviation data to represent the difference between actual data and standard data;

[0017] If at least one of the equipment status data, acceptance progress data, and deviation data is in the target state, adjust each initial acceptance process and the corresponding initial process parameters in the initial acceptance strategy to obtain the adjusted first target acceptance process and the corresponding adjusted first target process parameters.

[0018] In one embodiment, the step of selecting multiple second-target acceptance processes from multiple first-target acceptance processes includes:

[0019] The sample dataset is constructed by extracting the attribute information, status data, progress data, and deviation data of historically accepted equipment from historical acceptance records as input features, and using the historically accepted equipment's actual triggered and marked as valid historical acceptance processes as output labels.

[0020] The sample dataset is input into the random forest algorithm to calculate the importance score of each feature and sort them.

[0021] Based on a preset importance score threshold, a subset of key attributes is selected;

[0022] A subset of key attributes is input into a support vector machine algorithm, which maps it to a high-dimensional feature space through a kernel function to find the optimal classification hyperplane, thereby determining the second-target acceptance process for the equipment to be accepted in multiple first-target acceptance processes.

[0023] In one embodiment, the steps for optimizing the process parameters of each second target acceptance process include:

[0024] The initial positions of the sparrow population are randomly generated based on piecewise linear chaotic mapping, so that the sparrows are evenly distributed in the search space. The process parameters of the second target acceptance process are used as the position coordinates of the sparrows. The fitness value of each sparrow is calculated with the classification accuracy as the objective function.

[0025] Based on a non-inertial weighting mechanism, the search step size is adjusted according to the number of iterations and fitness value to search the search space. The iteration is repeated to output the position coordinates corresponding to the target fitness value as the second target process parameter of the second target acceptance process.

[0026] In one embodiment, the step of obtaining the acceptance data set includes:

[0027] Obtain basic data for equipment acceptance; this data should include at least equipment attribute data, acceptance standards and specifications, and historical acceptance records.

[0028] Data cleaning is performed on the basic data for equipment acceptance to remove invalid and abnormal data.

[0029] And / or perform data deduplication on the equipment acceptance baseline data to remove duplicate acceptance data from the equipment acceptance baseline data;

[0030] Standardize the format of equipment attribute data, acceptance standards and specifications, and historical acceptance records, and convert them into structured data to obtain an acceptance data set.

[0031] Secondly, this application also provides an automated equipment acceptance system, comprising:

[0032] The data acquisition and preprocessing module is used to acquire the acceptance data set; the acceptance data set includes the attribute information of the equipment to be accepted, the acceptance standards corresponding to each attribute information, and historical acceptance records.

[0033] The adaptive matching module is used to match multiple initial acceptance processes corresponding to attribute information based on acceptance criteria and historical acceptance records, and to construct an initial acceptance strategy based on multiple initial acceptance processes.

[0034] The dynamic adjustment module is used to acquire real-time dynamic data of the equipment to be accepted and adjust the initial acceptance strategy according to the real-time dynamic data to obtain the first acceptance strategy; wherein, the first acceptance strategy includes multiple first target acceptance processes and first target process parameters corresponding to each first target acceptance process;

[0035] An adaptive adjustment module is used to filter multiple second-target acceptance processes from multiple first-target acceptance processes, and to optimize the process parameters of each second-target acceptance process to determine the second acceptance strategy; wherein, the second acceptance strategy includes multiple second-target acceptance processes and the second-target process parameters corresponding to each second-target acceptance process;

[0036] The acceptance execution module is used to perform acceptance testing on the equipment to be accepted based on the second target acceptance process and the corresponding second target process parameters, and to obtain acceptance result data.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the device automated acceptance method provided in the first aspect.

[0038] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for automated acceptance testing of the device provided in the first aspect.

[0039] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the method for automated acceptance testing of the device provided in the first aspect.

[0040] The aforementioned methods, systems, computer equipment, computer-readable storage media, and computer program products for automated equipment acceptance acquire acceptance data, match the initial acceptance process, adjust strategies based on real-time dynamic data, and then obtain optimized acceptance strategies through screening and parameter optimization, ultimately achieving automated and precise equipment acceptance. This transforms the low-voltage equipment acceptance process from static rule-driven to dynamic, intelligent, and adaptive, improving the adaptability, accuracy, and efficiency of low-voltage equipment acceptance through data-driven strategy construction, real-time dynamic adjustment, and optimized process screening and parameter determination. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart illustrating a method for automated equipment acceptance testing in one embodiment.

[0043] Figure 2 This is a flowchart illustrating step 12 in one embodiment;

[0044] Figure 3 This is a flowchart illustrating step 13 in one embodiment;

[0045] Figure 4 This is a flowchart illustrating step 14 in one embodiment;

[0046] Figure 5 This is a schematic diagram illustrating a scenario of an automated equipment acceptance method in one embodiment;

[0047] Figure 6 This is a schematic diagram of the structure of an automated equipment acceptance system in one embodiment;

[0048] Figure 7 This is a schematic diagram of the internal structure of a computer device in one embodiment. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0050] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0051] In the traditional low-voltage equipment acceptance process, due to the diversity of equipment types and significant differences in acceptance standards, reliance on manual methods frequently leads to errors in standard matching and omissions in procedures, resulting in reduced accuracy of acceptance results. At the same time, the dynamic updating nature of acceptance standards and the continuous emergence of new equipment make it difficult for existing methods to adapt quickly, leading to decreased acceptance efficiency and increased response lag. In addition, historical acceptance data has not been systematically utilized, making it impossible to extract inherent patterns through data mining to optimize acceptance strategies, thus preventing continuous iterative improvement of acceptance capabilities.

[0052] For example, when accepting intelligent low-voltage circuit breakers with communication functions at industrial production sites, the equipment's attribute information includes programmable protection thresholds and remote monitoring interfaces. Its acceptance criteria need to be dynamically adjusted according to the latest industry standards. However, the acceptance personnel selected a fixed process based on historical experience, which failed to cover the testing requirements of the new functions. At the same time, they monitored the parameter drift phenomenon of the equipment in real time during the test, but the existing system could not correct the acceptance strategy in time based on the dynamic data, resulting in the omission of key test items. In the end, the acceptance results failed to accurately reflect the actual performance status of the equipment.

[0053] If the above problems are not addressed, errors in standard matching during the acceptance process will increase the potential risk of equipment failure after it is put into operation, which may lead to instability in the power system. At the same time, the lag in the acceptance process will prolong the equipment deployment cycle and affect the continuous execution of production plans. In addition, the lack of a learning mechanism for historical data makes it impossible to optimize the acceptance strategy over time, which continuously restricts the overall effectiveness of the acceptance work and ultimately leads to systemic challenges to equipment operation safety and production reliability.

[0054] Based on this, this embodiment proposes a method for automated equipment acceptance testing, such as... Figure 1 As shown, steps 11 to 15 are included, wherein:

[0055] Step 11: Obtain the acceptance data set.

[0056] The acceptance data set includes the attribute information of the equipment to be accepted, the acceptance standards corresponding to each attribute information, and historical acceptance records.

[0057] Here, the acceptance data set can refer to all relevant data used in the equipment acceptance process. The acceptance data set may include the inherent attribute information of the equipment to be accepted, the acceptance standards used to measure the performance and quality of the equipment, and the historical acceptance records accumulated in the past equipment acceptance process.

[0058] Specifically, the attribute information of the equipment to be accepted can refer to data describing the characteristics of the equipment itself, such as the model, specifications, manufacturer, manufacturing date, rated parameters, functional modules, and other attribute parameters of the equipment to be accepted. The attribute information can be used to identify the equipment type of the equipment to be accepted and match the corresponding acceptance standards.

[0059] In addition, acceptance criteria can refer to specific requirements and judgment criteria for equipment performance, function, safety and other aspects. They can be national standards, industry standards, enterprise standards or customized standards. Acceptance criteria can be used to guide the execution of the acceptance process and the judgment of the acceptance results.

[0060] In addition, historical acceptance records can refer to data records generated during previous acceptance of similar or identical equipment. Historical acceptance records can include the acceptance process, test results, problems found, handling measures, and final acceptance conclusions, which are used for learning and optimizing acceptance strategies.

[0061] As one example, equipment model, production batch, rated power, and other attribute information can be manually entered into the system; simultaneously, acceptance standards in paper or electronic document form can be manually imported, and historical acceptance records can be exported from the existing database. As another example, partial attribute information can be obtained by scanning the equipment nameplate or QR code, and acceptance standards and historical records can be imported in batches using preset templates.

[0062] In one embodiment, the step of obtaining the acceptance data set includes:

[0063] (1) Obtain basic data for equipment acceptance.

[0064] The basic data for equipment acceptance includes at least equipment attribute data, acceptance standards and specifications, and historical acceptance records.

[0065] Here, equipment attribute data can cover information such as equipment model, specifications, production date, and manufacturer; acceptance standards and specifications can refer to industry standards, internal enterprise standards, or national standards; historical acceptance records can include past equipment acceptance results, problems found, and handling measures.

[0066] As an example, basic equipment acceptance data can be automatically extracted or synchronized by interfaceing with existing information systems such as equipment management systems, manufacturing execution systems, and enterprise resource planning systems. Alternatively, data can be obtained through manual entry, scanning of paper documents, or sensor data collection.

[0067] (2) Perform data cleaning on the basic data of equipment acceptance to remove invalid and abnormal data; and / or perform data deduplication on the basic data of equipment acceptance to delete duplicate acceptance data.

[0068] After obtaining the basic data for equipment acceptance, data cleaning can be performed on the basic data to remove invalid and abnormal data.

[0069] It's important to note that data cleaning is the process of identifying, correcting, or removing corrupt or inaccurate records in a dataset, aiming to improve data quality and ensure the reliability of subsequent analysis. For example, statistical methods, such as using interquartile range or Z-score to detect outliers, can be employed; machine learning algorithms, such as isolation forests or local anomaly detection, can also be used. Invalid data can be identified and removed through pre-defined business rules or data type validation.

[0070] Furthermore, data deduplication can be performed on the basic equipment acceptance data to remove duplicate acceptance data, thereby identifying and eliminating redundant data copies and avoiding deviations in statistical analysis or subsequent processing caused by duplicate records.

[0071] As an example, this can be achieved using exact matching algorithms, such as comparing hash values ​​or primary keys, to identify and delete identical records. Alternatively, fuzzy matching algorithms, such as those based on string similarity (e.g., Levenshtein distance, Jaccard similarity) or machine learning clustering methods, can be used to identify data that are semantically repetitive but slightly different in expression.

[0072] (3) Unify the format of equipment attribute data, acceptance standards and specifications and historical acceptance records, and convert the equipment attribute data, acceptance standards and specifications and historical acceptance records into structured data to obtain the acceptance data set.

[0073] The acceptance dataset can be structured data in a predefined format, such as tables, JSON, or XML, to facilitate efficient storage, retrieval, and processing.

[0074] Specifically, a unified data model and data dictionary can be defined, and data transformation tools or programming scripts can be used to map data from different sources and in different formats to a unified structured template. Furthermore, natural language processing techniques can be employed to extract information and recognize entities from unstructured text data, converting it into structured information that machines can understand and process.

[0075] In this embodiment, all kinds of original basic data required for equipment acceptance, including equipment attributes, acceptance standards, and historical records, are comprehensively acquired. Then, this original data undergoes rigorous data cleaning to effectively remove invalid and abnormal information, ensuring the authenticity and accuracy of the data. Furthermore, data deduplication is performed to eliminate redundant and duplicate records, avoiding analytical biases caused by data duplication. Finally, the cleaned and deduplicated data undergoes format standardization and structuring to conform to a preset data model, facilitating efficient machine recognition, storage, and processing. This transforms raw, heterogeneous, and potentially quality-prone equipment acceptance basic data into a high-quality, structured, and consistent acceptance data set, significantly improving the accuracy and reliability of matching the initial acceptance process, constructing the initial acceptance strategy, and dynamically adjusting the initial acceptance strategy.

[0076] For example, the data acquisition and preprocessing module for low-voltage equipment acceptance data can comprehensively collect basic attribute information, corresponding acceptance standard and specification data, and historical acceptance record data of different types of low-voltage equipment through various methods such as sensors, database interfaces, and manual input.

[0077] The basic attribute information includes equipment model, manufacturer, rated voltage, rated current, rated power, service life, and installation environment; the acceptance standard and specification data includes national low-voltage equipment acceptance standards, industry standards, and custom acceptance standards developed by enterprises according to their own needs; the historical acceptance record data includes acceptance time, acceptance personnel, acceptance items, test data of each item, acceptance results, descriptions of non-conformities, and rectification records.

[0078] The low-voltage equipment acceptance data acquisition module also incorporates a data processing algorithm to preprocess the collected basic equipment acceptance data. This includes: data cleaning, which removes invalid data and handles outliers; data deduplication, which removes duplicate acceptance records and standard data; and data standardization, which converts test data of different formats and units into a unified standard format.

[0079] Step 12: Based on the acceptance criteria and historical acceptance records, match multiple initial acceptance processes corresponding to the attribute information, and construct an initial acceptance strategy based on the multiple initial acceptance processes.

[0080] The initial acceptance strategy can refer to an overall acceptance plan consisting of multiple initial acceptance procedures and their corresponding initial procedure parameters. The first acceptance strategy is formulated before the equipment acceptance begins and can be used to guide early acceptance activities.

[0081] Here, the initial acceptance process can refer to a series of testing steps and operation sequences that are preliminarily determined based on the attribute information, acceptance standards and historical acceptance records of the equipment to be accepted, and are used to accept the equipment.

[0082] One approach is to perform keyword matching within a pre-defined acceptance process library based on the attribute information of the equipment to be accepted, identifying all acceptance processes that contain the equipment's attribute characteristics. For example, for a "three-phase asynchronous motor," the system might match "insulation resistance test process," "winding DC resistance test process," etc. Then, based on the preset priority or manually set order of these matched processes, they can be combined to form an initial acceptance strategy, and a default testing parameter, such as testing duration or pass threshold, can be set for each process.

[0083] Step 13: Obtain real-time dynamic data of the equipment to be accepted, and adjust the initial acceptance strategy according to the real-time dynamic data to obtain the first acceptance strategy.

[0084] The first acceptance strategy includes multiple first target acceptance processes and the first target process parameters corresponding to each first target acceptance process.

[0085] Here, the first acceptance strategy refers to the acceptance plan obtained after adjusting it based on the initial acceptance strategy and real-time dynamic data. The first acceptance strategy aims to better adapt to the actual situation of the equipment during the acceptance process.

[0086] It should be noted that the first target acceptance process can refer to the acceptance process included in the first acceptance strategy, which has undergone preliminary adjustments. These first acceptance processes are testing steps optimized for specific equipment and the current acceptance stage. The first target process parameters can refer to parameters such as testing index thresholds, testing sequence, or operating instructions that have been preliminarily adjusted and correspond to the first target acceptance process.

[0087] It is understandable that real-time dynamic data can refer to data collected in real time during the equipment acceptance process that reflects the current operating status of the equipment and the progress of the acceptance. Real-time dynamic data can include the operating parameters implemented by the equipment to be accepted, sensor readings, and the completion status of the acceptance tasks.

[0088] For example, during equipment acceptance testing, real-time operating parameters such as temperature and current of the equipment can be manually read using sensors. If a certain operating parameter of the equipment is found to be continuously exceeding the preset normal range, manual intervention can be initiated to determine, based on experience, whether the current acceptance process needs to be adjusted. For example, the testing frequency of a specific test item can be increased, or the threshold of a certain test parameter can be modified. Thus, the initial acceptance strategy is adjusted to the first acceptance strategy.

[0089] Step 14: Select multiple second-target acceptance processes from multiple first-target acceptance processes, and optimize the process parameters of each second-target acceptance process to determine the second acceptance strategy.

[0090] The second acceptance strategy includes multiple second target acceptance processes and the second target process parameters corresponding to each second target acceptance process.

[0091] Here, the second acceptance strategy can refer to the final acceptance plan obtained after further screening and parameter optimization based on the first acceptance strategy. The second acceptance strategy aims to achieve a balance between acceptance efficiency and accuracy.

[0092] It should be noted that the second-target acceptance process can refer to the refined acceptance process included in the second acceptance strategy. These second-target acceptance processes are the core testing steps ultimately used to accept the equipment. The second-target process parameters refer to the optimized and determined parameters such as testing index thresholds, testing sequence, or operating instructions corresponding to the second-target acceptance process.

[0093] After the first acceptance strategy is determined, several acceptance processes that play an important role in the acceptance can be selected as the second target acceptance processes according to the actual situation. The optimal acceptance parameters can be determined by simulation and iteration of several second target acceptance processes. During the simulation and iteration process, the acceptance parameters of these second target acceptance processes can be fine-tuned to obtain the corresponding second target process parameters, in order to improve the acceptance effect.

[0094] Step 15: Based on the second target acceptance process and the corresponding second target process parameters, the equipment to be accepted is accepted to obtain acceptance result data.

[0095] The acceptance result data refers to the final judgment information generated after the equipment to be accepted has been completed. The acceptance result data may include a qualified / unqualified conclusion, detailed data of various test indicators, description of unqualified items, and suggested handling opinions, etc. The acceptance result data is the basis for evaluating the quality and performance of the equipment.

[0096] As one implementation method, automated testing equipment can be driven to perform various testing tasks according to the second target acceptance process and corresponding second target process parameters determined in the second acceptance strategy. For example, if the second target acceptance process includes a "withstand voltage test," the withstand voltage tester can be controlled to apply a specified voltage to the equipment, and the test result can be judged as qualified according to the pass threshold set in the second target process parameters. After all acceptance process tests in the second acceptance process are completed, the test results can be summarized to generate the final acceptance result data. This acceptance result data can be an acceptance report after the equipment to be accepted has been accepted, which includes the test values ​​of various indicators, the judgment results, and the overall acceptance conclusion.

[0097] The aforementioned method for automated equipment acceptance involves acquiring acceptance data, matching it with the initial acceptance process, adjusting strategies based on real-time dynamic data, and then obtaining an optimized acceptance strategy through screening and parameter optimization. This ultimately achieves automated and precise equipment acceptance. In this way, the low-voltage equipment acceptance process is transformed from static rule-driven to dynamic, intelligent, and adaptive. Through data-driven strategy construction, real-time dynamic adjustment, and optimized process screening and parameter determination, the adaptability, accuracy, and efficiency of low-voltage equipment acceptance are improved, while reducing labor costs.

[0098] In an exemplary embodiment, the initial acceptance strategy includes multiple initial acceptance processes and initial process parameters corresponding to each acceptance process, such as... Figure 2 As shown, step 12 includes steps 121 to 123, wherein:

[0099] Step 121: Construct a judgment matrix based on the acceptance criteria to determine the subjective weight of the attribute information and calculate the objective weight based on the data dispersion of the data information in the historical acceptance records. Adjust the weighting coefficients of the subjective and objective weights according to the priority of the attribute information in the acceptance scenario, and sum the subjective and objective weights according to the weighting coefficients to obtain the comprehensive weight.

[0100] Subjective weights can be used to quantify the subjective judgment of experts or domain knowledge on the importance of equipment attributes.

[0101] In this embodiment, the expert experience corresponding to the attribute information can be quantified by constructing a judgment matrix. Specifically, the subjective weight of each device attribute can be calculated by comparing the relative importance of different attribute information and forming a matrix corresponding to the attribute information, so as to transform the experience of experts or fields into quantitative evaluation weights.

[0102] As an example, the Analytic Hierarchy Process (AHP) can be used to construct the judgment matrix. The expert team compares the equipment attributes pairwise and gives a relative importance score. Then, the subjective weight of each attribute is determined by calculating the largest eigenvalue and the corresponding eigenvector of the judgment matrix.

[0103] As another example, fuzzy comprehensive evaluation can be used. By constructing a fuzzy judgment matrix, the importance of the constructed judgment matrix corresponding to the attribute information can be evaluated using fuzzy mathematics methods, thereby obtaining subjective weights.

[0104] In addition, objective weights can be determined based on the amount of information in the historical data itself. The greater the dispersion of data information in historical acceptance records, the more information corresponding to attribute information is contained in the historical acceptance records, and the greater the impact on decision-making.

[0105] In this embodiment, the importance of the attribute information in the actual acceptance process can be objectively reflected by analyzing the fluctuations or differences in the attribute data in the historical acceptance records.

[0106] As one implementation method, the entropy weight method can be used. By calculating the information entropy of each attribute information in the historical acceptance records, the smaller the information entropy, the greater the data dispersion of the attribute information, and the greater its objective weight.

[0107] As another implementation method, the standard deviation method can also be used to calculate the standard deviation of each attribute information in historical acceptance records. The larger the standard deviation, the greater the data volatility, and the higher the objective weight assigned to that attribute.

[0108] It should be noted that weighting coefficients can be used to balance the influence of subjective experience and objective data in the overall weight calculation. Different acceptance scenarios may have different emphases on subjective experience or objective data. For example, for new equipment or equipment lacking historical data, subjective experience may be more important; while for mature equipment, historical data may be more valuable.

[0109] In this embodiment, by adjusting the weighting coefficients, the overall weight can be made more in line with actual acceptance requirements. This step can set different weighting coefficients for different equipment types or acceptance stages based on preset rules or expert experience. For example, for critical performance attributes, the subjective weighting coefficient can be higher; for routine testing attributes, the objective weighting coefficient can be higher. Alternatively, a machine learning model can be used to dynamically learn and optimize the weighting coefficients of subjective and objective weights based on historical acceptance results and scenario characteristics to maximize the accuracy of the acceptance strategy.

[0110] Furthermore, the comprehensive weight can be used to represent a combination of subjective and objective weights, taking into account expert experience and historical data, providing a more comprehensive and accurate importance assessment for each device attribute. Specifically, the subjective weight can be directly multiplied by its weighting coefficient, the objective weight multiplied by its weighting coefficient, and then the two added together to obtain the comprehensive weight.

[0111] Step 122: Based on the comprehensive weight, construct the feature vector corresponding to the attribute information of the equipment to be accepted, and perform similarity matching between the feature vector and the attribute requirement vector of multiple preset acceptance standards to determine multiple initial acceptance processes for the equipment to be accepted.

[0112] The feature vector can be a quantitative representation of the attribute information of the equipment to be accepted. Each component in the feature vector represents an attribute information, and the value of the feature vector is determined by the comprehensive weight of that attribute information.

[0113] As an example, the attribute information of all devices to be accepted and their corresponding comprehensive weights can be arranged into a multi-dimensional vector. For instance, if a device has N attributes, the feature vector is an N-dimensional vector, with each dimension corresponding to the comprehensive weight of an attribute.

[0114] Furthermore, the feature vector can be matched with the attribute requirement vectors of multiple preset acceptance criteria to determine multiple initial acceptance procedures for the equipment to be accepted. By comparing the similarity between the feature vector of the equipment to be accepted and the attribute requirement vectors of different acceptance criteria, the initial acceptance procedure that best matches the characteristics and acceptance requirements of the current equipment can be identified. The higher the similarity, the more suitable the acceptance procedure is for the current equipment to be accepted.

[0115] One approach is to use cosine similarity calculation, which calculates the cosine value between the feature vector of the equipment to be inspected and the vectors of each acceptance standard attribute requirement. The larger the cosine value, the higher the similarity. Alternatively, Euclidean distance or Manhattan distance can be used as distance metrics; the smaller the distance, the higher the similarity.

[0116] Step 123: Based on the priority and logical association of multiple initial acceptance processes, plan the testing sequence, plan multiple initial acceptance processes, determine the qualified threshold of the testing indicators corresponding to each acceptance process, and obtain the initial process parameters corresponding to each acceptance process.

[0117] After determining multiple initial acceptance processes, the execution order of these processes can be arranged according to their priority (indicating process importance) and logical relationships (indicating interdependencies) to ensure the rationality and efficiency of the acceptance process.

[0118] For example, some basic tests may need to be performed before advanced functional tests. A pre-defined rule engine can be used to assign priorities to each acceptance process based on its type (e.g., security testing, performance testing, functional testing) and the device's key attributes. Simultaneously, preconditions and postconditions between processes can be defined to form a logical relationship graph. Alternatively, a Directed Acyclic Graph (DAG) can be used to represent the logical relationships and dependencies between initial acceptance processes, and a topology sorting algorithm can be used to plan the testing order.

[0119] Here, the qualified threshold for the test indicators can be a specific quantitative limit for judging whether the equipment meets the acceptance standards.

[0120] After determining multiple initial acceptance procedures, accurate acceptance thresholds can be set for each initial acceptance procedure to ensure the objectivity and consistency of the acceptance results.

[0121] One approach is to directly extract or calculate the pass thresholds for each testing indicator based on the explicit provisions in the acceptance standards and specifications. Alternatively, historical acceptance data can be combined with statistical analysis (such as mean, standard deviation, percentiles) or machine learning methods (such as anomaly detection models) to dynamically determine or adjust the pass thresholds for the testing indicators, thereby obtaining the initial process parameters corresponding to each acceptance process.

[0122] For example, the adaptive matching module takes the structured dataset generated by the data acquisition and preprocessing module as input and completes the initial matching of the acceptance strategy through an adaptive learning algorithm. The adaptive matching module includes a weight allocation unit, a standard matching unit, and a strategy generation unit. These units work together to achieve full automation from equipment attribute analysis to the output of the initial acceptance strategy.

[0123] The weight allocation unit employs a combined weighting strategy of analytic hierarchy process (AHP) and entropy weighting to determine the comprehensive weight of each equipment attribute. Specifically, firstly, subjective weights are determined by constructing a judgment matrix using AHP based on the core requirements of equipment acceptance; then, objective weights are calculated using entropy weighting based on the degree of data dispersion; finally, the weighting coefficients are dynamically adjusted according to the priority of the acceptance scenario, and the subjective and objective weights are summed to obtain the comprehensive weight. This enables differentiated acceptance of equipment under different scenarios, improving acceptance efficiency.

[0124] The standard matching unit uses the comprehensive weight output by the weight allocation unit as its core basis to achieve accurate matching between equipment attributes and acceptance standards. Specifically, firstly, the basic attribute vector of the equipment is weighted to construct a feature vector; then, the attribute requirement vectors of each preset acceptance standard are extracted; the matching degree between the equipment and each standard is calculated through a weighted similarity algorithm, and the acceptance standard with the highest matching degree is selected as the initial acceptance standard. At the same time, key information such as its acceptance items and testing indicators are extracted and integrated to form a standard subset (multiple initial acceptance processes).

[0125] The strategy generation unit automatically generates the initial acceptance process and judgment rules based on the acceptance criterion subset output by the standard matching unit. Specifically, it can plan the testing sequence according to the priority and logical relationship of each item in the criterion subset, clarify the testing methods and data collection frequency of each item, and build a standardized acceptance process framework; at the same time, it extracts the pass threshold of each testing indicator, constructs the corresponding acceptance logic rules, and formulates re-inspection rules for non-conforming items, thereby ensuring the rigor and operability of the initial acceptance strategy.

[0126] In this embodiment, a refined initial acceptance strategy addresses the potential for blindness and subjectivity in determining initial acceptance procedures and parameters using traditional methods. Specifically, a judgment matrix is ​​first constructed based on acceptance criteria, incorporating expert experience and domain knowledge to determine the subjective weights of attribute information. This ensures the acceptance strategy fully considers the key characteristics of the equipment and industry standards. Furthermore, by analyzing the dispersion of data in historical acceptance records, objective weights of attributes are calculated. This ensures the acceptance strategy is based on actual operational data and reflects the importance of attributes in historical performance. Subsequently, the weighting coefficients of subjective and objective weights are flexibly adjusted according to the specific acceptance scenario, and a weighted sum is obtained to obtain a more comprehensive and accurate integrated weight that combines experience and data. Based on this integrated weight, a feature vector corresponding to the attribute information of the equipment to be accepted is constructed. This vector comprehensively and quantitatively represents the characteristics of the equipment. Next, this feature vector is matched with the attribute requirement vectors of multiple preset acceptance criteria to intelligently identify the initial acceptance process that best matches the characteristics of the equipment to be accepted, avoiding the bias of manual selection. After determining the initial acceptance process set, a reasonable testing sequence is planned based on the priority and logical relationships of these processes to ensure the scientific and efficient nature of the acceptance process. Finally, a pass threshold for each initial acceptance process is determined, thus overcoming the inaccuracies and incompleteness of traditional methods in constructing initial acceptance strategies due to a lack of quantitative assessment of the importance of equipment attributes. By combining subjective experience and objective data, and flexibly adjusting weights according to the acceptance scenario, the matching of initial acceptance processes becomes more precise, and the setting of initial process parameters becomes more reasonable. This significantly improves the scientific nature and adaptability of the initial acceptance strategy, effectively enhancing the efficiency and accuracy of automated equipment acceptance and reducing the risks and costs arising from inappropriate initial strategies.

[0127] In some of the above implementation methods, although an initial acceptance strategy is constructed based on acceptance criteria and historical acceptance records, this initial acceptance strategy is based on preset conditions and historical experience, and may not fully consider the dynamic changes and abnormal situations that may occur in the actual acceptance process of the equipment to be accepted. This may result in insufficient adaptability of the acceptance strategy, and an inability to respond in a timely manner to deviations between equipment status, acceptance progress, or actual data and standard data, thereby affecting the accuracy and efficiency of the acceptance.

[0128] Based on this, a method for automated equipment acceptance is further proposed, such as... Figure 3 As shown, step 13 includes steps 131 and 132, wherein:

[0129] Step 131: Obtain real-time dynamic data of the equipment to be accepted during the acceptance process.

[0130] The real-time dynamic data includes equipment status data to represent the real-time operating parameters of the equipment to be accepted, acceptance progress data to represent the acceptance implementation, and deviation data to represent the difference between actual data and standard data.

[0131] Here, real-time dynamic data refers to data continuously collected during the equipment acceptance process that reflects the current operating status of the equipment, the execution of acceptance tasks, and the differences between actual performance and expected standards. Real-time dynamic data can be acquired in various ways. As an example, it can be read directly from sensors, actuators, or controllers integrated into the equipment to be accepted; it can be obtained through data interface communication with the equipment control system, manufacturing execution system, etc.; or it can be collected through manual input or visual recognition systems.

[0132] It should be noted that equipment status data can be a type of real-time dynamic data, specifically used to represent the real-time operating parameters of the equipment to be accepted. Equipment status data can reflect the health status and performance of the equipment to be accepted. For example, equipment status data can include physical quantities such as temperature, pressure, current, voltage, speed, and vibration frequency, as well as logical states such as operating mode, fault codes, and alarm information.

[0133] To further clarify, acceptance progress data can be a type of real-time dynamic data, specifically used to represent the current stage and completion status of the acceptance process. Acceptance progress data can be used to monitor the execution efficiency and workflow smoothness of acceptance tasks. For example, acceptance progress data may include currently executing acceptance process steps, the percentage of completed acceptance tasks, estimated remaining time, and the execution status or test results of specific test cases.

[0134] It should also be noted that deviation data can be a type of real-time dynamic data, specifically used to represent the difference between actual data and standard data. Deviation data can be used to quantify the degree of deviation between the equipment under acceptance and the preset acceptance standards in specific performance indicators. For example, deviation data can be expressed as the absolute difference between the measured value and the standard value, the relative percentage deviation, or whether it exceeds the preset tolerance range, etc.

[0135] Step 132: If at least one of the equipment status data, acceptance progress data, and deviation data is in the target state, adjust each initial acceptance process and the corresponding initial process parameters in the initial acceptance strategy to obtain the adjusted first target acceptance process and the corresponding adjusted first target process parameters.

[0136] Here, the target state can refer to the situation where any of the equipment status data, acceptance progress data, or deviation data enters a preset specific condition or threshold.

[0137] When any one or more of the real-time dynamic data reaches or exceeds this target state, the adjustment of the initial acceptance strategy is triggered. For example, the target state can be a specific value (such as temperature exceeding 80°C), a range (such as vibration frequency exceeding the safe range), a trend (such as deviation continuously increasing), or a logical judgment (such as a key test step not being completed for a long time).

[0138] Adjusting the initial acceptance procedures and their corresponding parameters within the initial acceptance strategy can be achieved by modifying or optimizing pre-set initial acceptance procedures and their associated parameters based on real-time dynamic data reflecting the actual equipment situation. For example, the order of tests can be adjusted, certain test steps can be added or removed, or the intensity or duration of tests can be modified based on equipment status data; the priority of procedures can be adjusted based on acceptance progress data, and diagnostic procedures can be inserted to identify anomalies; or the pass thresholds for testing indicators can be adjusted based on deviation data, and measuring equipment can be recalibrated, thereby determining the adjusted first target acceptance procedure and its corresponding adjusted parameters.

[0139] For example, the dynamic adjustment module can adapt to changes in equipment attributes, updates to acceptance standards, and deviations in the acceptance process in real time by constructing a dynamic adjustment mechanism, thereby ensuring the timeliness, accuracy, and adaptability of the acceptance strategy. This module achieves dynamic optimization of the acceptance strategy by monitoring key data throughout the entire acceptance process in real time, accurately identifying adjustment trigger conditions, and executing standardized adjustment procedures.

[0140] In terms of real-time data monitoring, the dynamic adjustment module collects and analyzes three types of core data during the acceptance process to provide data support for adjustment decisions. Specifically, the first type is equipment status data, which covers real-time operating parameters such as voltage fluctuations and current stability of the equipment during the acceptance process, reflecting the current operating condition of the equipment; the second type is acceptance progress data, including the completion status of each acceptance item, testing time, and data acquisition quality, used to evaluate the efficiency of acceptance implementation; the third type is deviation data, which includes the degree of deviation between the actual test data and the pass threshold, as well as the dispersion of multiple test results for the same item, used to accurately capture deviations between the acceptance results and standard requirements.

[0141] Regarding triggering the decision, the module automatically initiates the acceptance strategy adjustment process when the monitoring data meets any of the following conditions: Changes in the basic attributes of the equipment, such as changes in the equipment model or adjustments to the installation environment leading to a decrease in the adaptability of the original strategy; updates to the acceptance standards, such as new national or industry acceptance specifications or indicator requirements; acceptance deviations exceeding a preset threshold, i.e., the deviation between the test data and the pass threshold is greater than 10%, or the dispersion of multiple test results for the same project is greater than 5%; and the acceptance progress lags behind the preset progress.

[0142] Regarding the standardized adjustment process, after triggering the adjustment, the device attribute vector and acceptance standard vector of the current acceptance scenario are matched with the corresponding vectors of historical acceptance scenarios to quantify the degree of fit between the current and historical scenarios. Secondly, a strategy adjustment method is selected: if the similarity is greater than or equal to a preset threshold, local fine-tuning is performed based on the historical acceptance strategy with the highest matching degree, such as adjusting the detection order, modifying the data collection frequency, or correcting the pass threshold; if the similarity is less than the preset threshold, the adaptive learning algorithm is triggered to re-execute the weight allocation, standard matching, and strategy generation process to generate a new acceptance strategy adapted to the current scenario. Finally, the adjusted acceptance strategy is applied to a small-scale pilot acceptance to verify its feasibility and accuracy: if the fluctuation of the pilot acceptance pass rate is within a preset range, the adjusted acceptance strategy is officially implemented; if the fluctuation exceeds the preset range, it is returned to readjustment until the acceptance requirements are met.

[0143] In this embodiment, a continuous monitoring and feedback mechanism for real-time dynamic data of the equipment effectively compensates for the shortcomings of the initial acceptance strategy in dealing with actual dynamic environments. Specifically, after constructing the initial acceptance strategy, real-time dynamic data can be acquired, including equipment status data, acceptance progress data, and deviation data of the equipment to be accepted during the acceptance process. This data comprehensively and instantly reflects the actual operating status of the equipment, the execution efficiency of the acceptance task, and the degree of conformity between the equipment performance and the standards. When any one or more of these real-time dynamic data reach a preset target state, such as abnormal equipment operating parameters, stagnant acceptance progress, or significant deviation between actual performance and standards, adjustments to the initial acceptance strategy will be immediately triggered. These adjustments include modifications to the initial acceptance process, such as reordering, adding or deleting process steps, and optimization of initial process parameters, such as adjusting detection thresholds, test intensity, or duration. In this way, the initial acceptance strategy is dynamically revised into a first acceptance strategy, enabling it to more accurately adapt to the actual situation of the equipment to be accepted in the current environment, thereby ensuring the effectiveness and accuracy of the acceptance process.

[0144] In one exemplary embodiment, such as Figure 4 As shown, step 14 includes filtering multiple second-target acceptance processes from multiple first-target acceptance processes and optimizing the process parameters of each second-target acceptance process to determine the second-target process parameters. Specifically, it includes steps 141 to 144, and steps 145 and 146, wherein:

[0145] Step 141: Extract the attribute information, status data, progress data and deviation data of the historical acceptance equipment from the historical acceptance records as input features, and use the historical acceptance processes that were actually triggered and marked as valid by the historical acceptance equipment as output labels to construct a sample dataset.

[0146] The input features describe the various performance and characteristics of the historically accepted equipment during the acceptance process. For example, attribute information can be static information such as equipment model, production batch, and configuration parameters; status data can be dynamic performance indicators such as real-time sensor data, CPU utilization, and memory usage during equipment operation; progress data can be the percentage of acceptance task completion and the number of acceptance steps executed; and deviation data can be the difference between the actual measured value and the standard value, the error rate, etc.

[0147] Output labels can indicate the necessary acceptance procedures for historically accepted equipment in the historical acceptance process.

[0148] After determining the input features and output labels, a sample dataset can be constructed to provide training data for subsequent machine learning algorithms, enabling them to learn the correlation between input features and effective acceptance processes. For example, database queries and data cleaning tools can be used to automatically extract and format the required data from historical acceptance records; or historical data can be reviewed and labeled manually or semi-automatically to ensure the accuracy of the output labels.

[0149] Step 142: Input the sample dataset into the random forest algorithm, calculate the importance score of each feature and sort them.

[0150] Random forest is an ensemble learning method that makes predictions by constructing multiple decision trees and combining their results.

[0151] Here, the importance score can be used to evaluate the importance of input features to predicting effective acceptance processes. The importance score of a feature represents the degree to which each input feature contributes to the accuracy of the model's prediction. The higher the score, the more important the feature is in distinguishing effective acceptance processes.

[0152] It's worth noting that the Random Forest algorithm can naturally calculate feature importance during training, for example, by measuring the information gain or Gini impurity reduction that a feature brings when splitting the decision tree. For instance, feature importance can be directly obtained by calling the feature importance attribute of the Random Forest classifier or Random Forest regressor from Python's scikit-learn library; alternatively, other feature selection algorithms, such as L1-regularized linear models or XGBoost, can be used to provide feature importance assessments, from which corresponding importance scores are calculated.

[0153] Step 143: Select a subset of key attributes based on the preset importance score threshold.

[0154] The importance score threshold is a pre-set value that can be used to distinguish between important and unimportant features. Features that are above this threshold are considered key features.

[0155] Filtering a subset of key attributes based on this importance score threshold aims to reduce model complexity, improve generalization ability, and remove redundant or noisy features. For example, a fixed threshold can be set through expert experience to retain only the top N features by importance score, or features with importance scores above a certain percentage; alternatively, methods such as cross-validation can be used to experimentally determine the optimal threshold to achieve a balance between model performance and the number of features.

[0156] Step 144: Input the subset of key attributes into the support vector machine algorithm, map it to the high-dimensional feature space through the kernel function, and find the optimal classification hyperplane to determine the second target acceptance process of the equipment to be accepted in multiple first target acceptance processes.

[0157] It should be noted that the Support Vector Machine (SVM) algorithm is a binary classification model that can find an optimal classification hyperplane in the feature space to maximize the margin between sample points of different classes.

[0158] Furthermore, the kernel function maps data from the original feature space to a higher-dimensional feature space, making linearly inseparable data in the original space linearly separable in the higher-dimensional space. The optimal classification hyperplane, in the higher-dimensional feature space, separates sample points of different categories to the greatest extent. The SVM model then classifies and predicts the features (key attribute subset) of the new equipment to be inspected, thereby determining the most suitable and effective inspection process selected from the first objective inspection process—the second objective inspection process. For example, the SVC (Support Vector Classifier) ​​from Python's scikit-learn library can be used, selecting an appropriate kernel function; or other classification algorithms, such as neural networks or logistic regression, can also be used to classify and predict the inspection process based on the key attribute subset.

[0159] To improve the efficiency and accuracy of automated equipment acceptance testing, this application proposes a method for selecting the primary target acceptance process. This method first extracts multi-dimensional data related to equipment acceptance from a large number of historical acceptance records, including attribute information, status data, progress data, and deviation data of historically accepted equipment. This data is then used as input features for a machine learning model. Simultaneously, acceptance processes that have actually been triggered and verified as effective in historical acceptance are used as output labels for the model, thus constructing a comprehensive sample dataset. After constructing the sample dataset, this method inputs it into a random forest algorithm. The random forest algorithm, by integrating the advantages of multiple decision trees, can effectively evaluate the importance of each input feature in predicting an effective acceptance process. By calculating and ranking the importance score of each feature, the influence of each feature on the selection of the acceptance process can be quantified. Subsequently, based on a preset importance score threshold, the system can intelligently select the subset of attributes most critical to the acceptance process decision, thereby effectively removing redundant or irrelevant features and reducing the complexity and computational burden of subsequent models. Finally, the selected subset of key attributes is input into a support vector machine algorithm. Furthermore, a support vector machine algorithm is used to map these key attributes to a high-dimensional feature space using kernel functions to find an optimal classification hyperplane. This hyperplane effectively classifies different acceptance processes, thereby accurately determining the most suitable secondary acceptance process for the current equipment among multiple primary acceptance processes. In this way, based on historical experience and data-driven analysis, the most effective and relevant acceptance process can be intelligently selected from the initial, potentially redundant, primary acceptance processes. This makes the acceptance strategy more refined and efficient, avoids unnecessary testing steps, and improves the overall automation and accuracy of the acceptance process.

[0160] Step 145: Based on piecewise linear chaotic mapping, the initial positions of the sparrow population are randomly generated so that the sparrow individuals are evenly distributed in the search space. The process parameters of the second target acceptance process are used as the position coordinates of the sparrow individuals. The fitness value of each sparrow individual is calculated with the classification accuracy as the objective function.

[0161] Piecewise linear chaotic mapping is a nonlinear dynamic system that generates sequences that are random, ergodic, and aperiodic, which can more evenly cover the search space, thereby increasing the diversity of the initial population and preventing the algorithm from getting trapped in local optima too early.

[0162] As an example, chaotic sequences can be generated iteratively using a specific formula, and these sequence values ​​can be mapped to the range of process parameters as the initial positions of the sparrow population. Alternatively, other chaotic mappings, such as the Logistic chaotic mapping or the Tent chaotic mapping, can also be used to generate sequences with good ergodicity for initializing the sparrow population, ensuring the diversity and uniformity of the initial solutions.

[0163] Using the process parameters of the second-target acceptance process as the position coordinates of individual sparrows means that in the optimization algorithm, each sparrow represents a potential solution, and its position coordinates correspond to the set of parameters to be optimized. These process parameters can be encoded as multi-dimensional vectors, where each dimension represents a specific process parameter, such as the pass threshold for detection indicators, detection order, and detection frequency. For discrete parameters, integer or binary encoding can be used, and then the encoding can be converted into actual process parameter values ​​through a decoding function; for continuous parameters, floating-point numbers can be used directly.

[0164] The fitness value can be used to measure the quality of a solution. The higher the fitness value, the higher the classification accuracy of the set of process parameters, and the better the solution.

[0165] Specifically, historical acceptance data (including known equipment status and acceptance results) can be input into the acceptance model defined by the current sparrow's individual position coordinates (i.e., process parameters). The classification accuracy of this model on historical data can then be calculated. Furthermore, techniques such as cross-validation can be used to calculate the classification accuracy, thereby improving the robustness of fitness value evaluation. After calculating the classification accuracy, the fitness value for each sparrow is calculated.

[0166] Step 146: Based on the non-inertial weight mechanism, the search step size is adjusted according to the number of iterations and fitness value to search the search space. The iteration is repeated to output the position coordinates corresponding to the target fitness value as the second target process parameter of the second target acceptance process.

[0167] Here, the non-inertial weight mechanism dynamically adjusts the search step size according to the algorithm's iteration process and the quality (fitness value) of the current solution, enabling the algorithm to perform a large-scale global exploration in the early stage and a more refined local exploration in the later stage, thereby improving the convergence speed and optimization accuracy.

[0168] As an example, non-inertial weights can be designed to have a non-linear relationship with the number of iterations, while introducing fitness values ​​as adjustment factors. When the fitness values ​​of the population vary greatly, the weights can be appropriately increased to promote exploration; when the fitness values ​​tend to be consistent, the weights can be decreased to promote development. In addition, adaptive inertial weight strategies can also be adopted, such as fuzzy logic-controlled inertial weights, which adjust the weights in real time based on indicators such as population diversity and convergence speed.

[0169] The entire optimization process is carried out through repeated iterations until a preset termination condition is met. When the preset termination condition is reached (e.g., the maximum number of iterations, the fitness value reaching a threshold, or the fitness value not significantly improving for several consecutive generations), the algorithm stops and outputs the coordinates of the sparrow's position corresponding to the current optimal fitness value, which is the second target process parameter obtained by optimizing each target acceptance process.

[0170] In this embodiment, to overcome the problems of uneven population distribution and poor convergence that may be caused by traditional random initialization, a piecewise linear chaotic mapping is used to generate the initial positions of the sparrow population. The ergodicity and randomness of the chaotic mapping ensure that sparrow individuals are evenly distributed throughout the search space, thereby significantly enhancing the algorithm's global search capability. The various process parameters of the second objective acceptance process to be optimized are mapped to the position coordinates of sparrow individuals in the search space, and the fitness value of each sparrow individual is calculated using classification accuracy as the objective function. During the iterative optimization process, this scheme introduces a non-inertial weight mechanism to dynamically adjust the search step size of sparrow individuals. This dynamic adjustment allows the algorithm to perform a large-scale global exploration in the early stages of optimization, while enabling more refined local development in the later stages, thus better balancing global exploration and local development, effectively improving optimization efficiency and result accuracy. The entire optimization process is repeated iteratively until a preset termination condition is met. Finally, the algorithm outputs the location coordinates of the sparrow with the highest fitness value (i.e., the highest classification accuracy). These location coordinates are the second objective process parameters of the optimized second objective acceptance process.

[0171] For example, please refer to Figure 5 , Figure 5 This is a schematic diagram of a scenario in this embodiment. Specifically, it uses the acceptance task of a batch of newly received low-voltage distribution boxes as a scenario to explain in detail the complete workflow of the system from data acquisition to strategy generation and dynamic optimization.

[0172] The system can collect basic attribute information of the low-voltage distribution box to be inspected through sensors, database interfaces, and manual input. This information includes the equipment model, manufacturer, rated voltage of 380V, rated current of 200A, rated power of 100kW, service life of 5 years, and installation environment of "indoor power distribution room". It also collects corresponding acceptance standard and specification data, including relevant acceptance requirements, industry standards, and enterprise-defined supplementary acceptance standards. Furthermore, it can retrieve historical acceptance records for this equipment model, including past acceptance dates, personnel, test data for each item, acceptance results, descriptions of non-conformities, and rectification records.

[0173] After data acquisition, the system can input the preprocessed structured dataset into the adaptive learning initial matching module. The weight allocation unit within this module first employs a combined weighting strategy of the analytic hierarchy process (AHP) and entropy weighting to determine the comprehensive weight of each equipment attribute. For example, for low-voltage distribution boxes, core acceptance requirements include safety, reliability, and durability. The AHP constructs a judgment matrix based on expert experience to derive subjective weights for attributes such as rated voltage, rated current, and service life. The entropy weighting method calculates objective weights based on the dispersion of attribute values ​​in historical data. Finally, a weighted sum is performed according to the priority of the indoor distribution room installation scenario to obtain the comprehensive weight of each attribute.

[0174] The standard matching unit uses comprehensive weights as its core basis to construct a weighted feature vector of the equipment's basic attributes, and then calculates the weighted similarity between this feature vector and the attribute requirement vectors of each preset acceptance standard. The calculation shows that National Standard A has the highest matching degree with the equipment; therefore, the system selects it as the initial acceptance standard and simultaneously extracts key information such as its acceptance items, testing indicators, and pass thresholds, integrating them to form a standard subset.

[0175] The strategy generation unit automatically generates an initial acceptance process framework based on this standard subset. This initial acceptance process framework plans the testing sequence according to the priority and logical relationship of each acceptance item. For example, it first performs visual inspection, then insulation resistance testing, and finally power-on operation test. It clarifies the testing methods for each item. For example, the insulation resistance test uses a 500V megohmmeter, and the data acquisition frequency is 3 readings per item, with the average value taken. At the same time, it constructs logical rules for testing data, threshold comparison, and result judgment, such as determining whether the equipment is qualified.

[0176] During the formal acceptance process, the system activates the adaptive adjustment module to monitor three types of core data in real time: equipment status data (such as real-time operating parameters of equipment voltage fluctuations and current stability during the acceptance process), acceptance progress data (completion status of each project, testing time, data acquisition quality, etc.), and deviation data (the degree of deviation between actual test data and the pass threshold, the degree of dispersion of multiple test results for the same project, etc.).

[0177] Assuming that during insulation resistance testing, the system detects voltage fluctuations in the equipment exceeding the normal range, and the deviation between the actual test data and the pass threshold exceeds a preset threshold, the system automatically triggers a strategy adjustment process. The adjustment module first performs similarity matching between the equipment attribute vector and acceptance standard vector of the current acceptance scenario and historical acceptance scenarios. If the matching degree is high, it performs local fine-tuning based on the historical acceptance strategy with the highest matching degree and adds a retest; if the matching degree is low, it triggers an adaptive learning algorithm to re-execute the weight allocation, standard matching, and strategy generation process to generate a completely new acceptance strategy. The adjusted strategy is first applied to a small-scale pilot acceptance test to verify its feasibility and accuracy. Once the pass rate fluctuation is confirmed to be within the preset range, it is officially implemented.

[0178] The adaptive adjustment module incorporates random forest and support vector machine algorithms. By analyzing the data collected and input by the aforementioned modules and combining it with historical acceptance data, it optimizes the acceptance judgment rules and process parameters, thereby improving the accuracy and efficiency of the acceptance process. Specifically, the adaptive adjustment module uses two ensemble machine learning models to predict the acceptance time and accuracy under different testing sequences. It selects the testing sequence with the shortest time and highest accuracy as the optimized acceptance sequence and analyzes the importance of each testing item through the model, eliminating redundant testing items to shorten the acceptance process. To further improve the prediction accuracy of the machine learning model and enhance the efficiency of low-voltage equipment acceptance, this invention employs an improved sparrow search algorithm to optimize the hyperparameters of the built-in random forest and support vector machine models, thereby improving prediction accuracy and reducing prediction time, ensuring the efficient and stable operation of the proposed low-voltage equipment acceptance system.

[0179] Furthermore, the system can acquire the output acceptance result data and drive the adaptive learning module to correct the algorithm parameters based on the newly acquired data. At the same time, it can drive the machine learning module to perform incremental training. The system is connected to the adaptive learning module and the machine learning module respectively to achieve adaptive optimization of the algorithm and model.

[0180] In the intelligent machine learning module, the Random Forest algorithm is a supervised learning algorithm based on the ensemble learning concept. It uses decision trees as base learners, constructing multiple decision trees through random sampling and feature selection, and finally outputting the results through voting or averaging. It boasts advantages such as high accuracy, resistance to overfitting, and ease of interpretation, and is widely used in regression and classification problems. The Support Vector Machine (SVM) algorithm is a classic machine learning algorithm. Its core idea is to find the optimal separating hyperplane in the feature space to achieve accurate sample partitioning, especially performing well in scenarios with small samples and high-dimensional data. This invention chooses the Gaussian kernel function as the kernel function for the SVM to obtain better data capture and generalization capabilities.

[0181] Regarding the improved sparrow search algorithm, this invention employs a Piecewise chaotic mapping method to initialize the sparrow search algorithm's population, resulting in a more uniform population distribution and thus improving the algorithm's global search performance. Simultaneously, addressing the issue that linear inertial weights in traditional sparrow search algorithms cannot adequately balance local and global information during the search process, this invention introduces a nonlinear inertial weight mechanism to improve and optimize the algorithm, enhancing its global search capability and avoiding getting trapped in local optima. The combination of these two methods significantly improves the algorithm's global optimization performance.

[0182] Upon completion of acceptance testing, the feedback iteration module collects the final acceptance results data, including the test data for each project, acceptance conclusions, descriptions of non-conformities, and rectification records. This module feeds the newly acquired data back to the adaptive learning module, driving it to correct algorithm parameters; simultaneously, it drives the machine learning module to perform incremental training, continuously optimizing the prediction accuracy of the random forest model and support vector machine model, achieving continuous adaptive optimization of the algorithm and model.

[0183] In the above technical solution, on the one hand, this embodiment intelligently selects the most critical and effective second-target acceptance process from multiple first-target acceptance processes based on historical data and machine learning algorithms. This avoids unnecessary or redundant testing steps during the acceptance process, significantly improving the efficiency of automated equipment acceptance. Simultaneously, by identifying key attributes and using support vector machines for precise classification, it ensures that the selected acceptance process is more accurately tailored to the actual situation of the equipment to be accepted, thereby improving the reliability and accuracy of the acceptance results and reducing the risk of misjudgment or missed detection. On the other hand, it effectively solves the problems of low efficiency, susceptibility to local optima, and slow convergence speed inherent in traditional process parameter optimization methods. The initialization strategy based on piecewise linear chaotic mapping ensures the uniform distribution and diversity of the sparrow population in the search space, significantly enhancing the algorithm's global search capability and preventing the optimization process from prematurely falling into local optima. Meanwhile, the introduction of a non-inertial weighting mechanism allows the algorithm to dynamically adjust the search step size based on the iteration process and population fitness value. This enables thorough global exploration in the early stages of optimization and precise local development in the later stages, effectively balancing the algorithm's exploration and development capabilities and significantly improving the efficiency and accuracy of process parameter optimization. Ultimately, the second target process parameters obtained through this optimization method can more accurately and effectively guide equipment acceptance, improving the overall performance and reliability of automated acceptance.

[0184] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0185] Based on the same inventive concept, this application also provides an automated equipment acceptance system for implementing the aforementioned automated equipment acceptance method. The solution provided by this system is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the automated equipment acceptance system provided below can be found in the limitations of the automated equipment acceptance method described above, and will not be repeated here.

[0186] In one exemplary embodiment, such as Figure 6 As shown, an automated equipment acceptance system is provided, including: a data acquisition and preprocessing module, an adaptive matching module, a dynamic adjustment module, an adaptive adjustment module, and an acceptance execution module, wherein:

[0187] The data acquisition and preprocessing module is used to acquire the acceptance data set; the acceptance data set includes the attribute information of the equipment to be accepted, the acceptance standards corresponding to each attribute information, and historical acceptance records.

[0188] The adaptive matching module is used to match multiple initial acceptance processes corresponding to attribute information based on acceptance criteria and historical acceptance records, and to construct an initial acceptance strategy based on multiple initial acceptance processes.

[0189] The dynamic adjustment module is used to acquire real-time dynamic data of the equipment to be accepted and adjust the initial acceptance strategy according to the real-time dynamic data to obtain the first acceptance strategy; wherein, the first acceptance strategy includes multiple first target acceptance processes and first target process parameters corresponding to each first target acceptance process;

[0190] An adaptive adjustment module is used to filter multiple second-target acceptance processes from multiple first-target acceptance processes, and to optimize the process parameters of each second-target acceptance process to determine the second acceptance strategy; wherein, the second acceptance strategy includes multiple second-target acceptance processes and the second-target process parameters corresponding to each second-target acceptance process;

[0191] The acceptance execution module is used to perform acceptance testing on the equipment to be accepted based on the second target acceptance process and the corresponding second target process parameters, and to obtain acceptance result data.

[0192] Each module in the aforementioned automated equipment acceptance system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0193] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores acceptance data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for automated device acceptance.

[0194] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0195] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of a method for automated acceptance testing of the device.

[0196] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the steps of a method for automated acceptance testing of a device being executed by a processor.

[0197] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of a method for automated acceptance testing of a device.

[0198] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0199] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0200] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for automated acceptance testing of equipment, characterized in that, The method includes: Obtain the acceptance data set; wherein, the acceptance data set includes the attribute information of the equipment to be accepted, the acceptance criteria corresponding to each attribute information, and historical acceptance records; Based on the acceptance criteria and the historical acceptance records, multiple initial acceptance processes corresponding to the attribute information are matched, and an initial acceptance strategy is constructed based on the multiple initial acceptance processes. The real-time dynamic data of the equipment to be accepted is obtained, and the initial acceptance strategy is adjusted according to the real-time dynamic data to obtain a first acceptance strategy; wherein, the first acceptance strategy includes multiple first target acceptance processes and first target process parameters corresponding to each first target acceptance process; A second acceptance strategy is determined by filtering multiple second target acceptance processes from multiple first target acceptance processes and optimizing the process parameters of each second target acceptance process; wherein, the second acceptance strategy includes multiple second target acceptance processes and second target process parameters corresponding to each second target acceptance process; The equipment to be inspected is inspected based on the second target acceptance process and the corresponding second target process parameters to obtain acceptance result data.

2. The method according to claim 1, characterized in that, The initial acceptance strategy includes multiple initial acceptance processes and initial process parameters corresponding to each acceptance process. The step of matching multiple initial acceptance processes corresponding to the attribute information based on the acceptance criteria and the historical acceptance records, and constructing the initial acceptance strategy based on the multiple initial acceptance processes includes: A judgment matrix is ​​constructed based on the acceptance criteria to determine the subjective weight of the attribute information, and the objective weight is calculated based on the data dispersion of the data information in the historical acceptance records. The weighting coefficients of the subjective weight and the objective weight are adjusted according to the priority of the attribute information in the acceptance scenario, and the subjective weight and the objective weight are weighted and summed based on the weighting coefficients to obtain the comprehensive weight. Based on the comprehensive weight, a feature vector corresponding to the attribute information of the equipment to be inspected is constructed, and the feature vector is matched with the attribute requirement vectors of multiple preset acceptance standards to determine multiple initial acceptance processes for the equipment to be inspected. Based on the priority and logical association of the multiple initial acceptance processes, the detection sequence is planned, multiple initial acceptance processes are planned, and the qualified threshold of the detection indicators corresponding to each acceptance process is determined, so as to obtain the initial process parameters corresponding to each acceptance process.

3. The method according to claim 1, characterized in that, The steps of acquiring real-time dynamic data of the equipment to be accepted and adjusting the initial acceptance strategy accordingly to obtain the first acceptance strategy include: Acquire real-time dynamic data of the equipment to be accepted during the acceptance process; wherein, the real-time dynamic data includes equipment status data for representing the real-time operating parameters of the equipment to be accepted, acceptance progress data for representing the acceptance implementation, and deviation data for representing the actual data and standard data; If at least one of the equipment status data, the acceptance progress data, and the deviation data is in the target state, adjust each of the initial acceptance processes and the corresponding initial process parameters in the initial acceptance strategy to obtain the adjusted first target acceptance process and the corresponding adjusted first target process parameters.

4. The method according to claim 1, characterized in that, The step of selecting multiple second target acceptance processes from multiple first target acceptance processes includes: The system extracts the attribute information, status data, progress data, and deviation data of the historical acceptance equipment from the historical acceptance records as input features, and uses the historical acceptance processes that were actually triggered and marked as valid by the historical acceptance equipment as output labels to construct a sample dataset. The sample dataset is input into a random forest algorithm to calculate the importance score of each feature and sort them. Based on a preset importance score threshold, a subset of key attributes is selected; The subset of key attributes is input into a support vector machine algorithm, which maps it to a high-dimensional feature space through a kernel function to find the optimal classification hyperplane, thereby determining the second target acceptance process for multiple first target acceptance processes for equipment to be accepted.

5. The method according to claim 1 or 4, characterized in that, The steps for optimizing the process parameters of each of the second target acceptance processes include: The initial positions of the sparrow population are randomly generated based on piecewise linear chaotic mapping, so that the sparrow individuals are evenly distributed in the search space. The process parameters of the second target acceptance process are used as the position coordinates of the sparrow individuals. The fitness value of each sparrow individual is calculated with the classification accuracy as the objective function. Based on a non-inertial weighting mechanism, the search step size is adjusted according to the number of iterations and the fitness value to search the search space. The iteration is repeated to output the position coordinates corresponding to the target fitness value as the second target process parameter of the second target acceptance process.

6. The method according to claim 1, characterized in that, The steps for obtaining the acceptance data set include: Obtain basic equipment acceptance data; wherein, the basic equipment acceptance data includes at least equipment attribute data, acceptance standards and specifications, and historical acceptance records; Data cleaning is performed on the basic equipment acceptance data to remove invalid and abnormal data; and / or The equipment acceptance baseline data is deduplicated to remove duplicate acceptance data. The formats of the equipment attribute data, the acceptance standards and specifications, and the historical acceptance records are unified, and the equipment attribute data, the acceptance standards and specifications, and the historical acceptance records are converted into structured data to obtain the acceptance data set.

7. An automated equipment acceptance system, characterized in that, The system includes: The data acquisition and preprocessing module is used to acquire the acceptance data set; wherein, the acceptance data set includes the attribute information of the equipment to be accepted, the acceptance criteria corresponding to each attribute information, and historical acceptance records; An adaptive matching module is used to match multiple initial acceptance processes corresponding to the attribute information according to the acceptance criteria and the historical acceptance records, and to construct an initial acceptance strategy based on the multiple initial acceptance processes. A dynamic adjustment module is used to acquire real-time dynamic data of the equipment to be accepted and adjust the initial acceptance strategy in accordance with the real-time dynamic data to obtain a first acceptance strategy; wherein, the first acceptance strategy includes multiple first target acceptance processes and first target process parameters corresponding to each first target acceptance process; An adaptive adjustment module is used to filter multiple second target acceptance processes from multiple first target acceptance processes, and to optimize the process parameters of each second target acceptance process to determine a second acceptance strategy; wherein, the second acceptance strategy includes multiple second target acceptance processes and second target process parameters corresponding to each second target acceptance process; The acceptance execution module is used to perform acceptance testing on the equipment to be accepted based on the second target acceptance process and the corresponding second target process parameters, and obtain acceptance result data.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images