Method for analyzing the influence range of environmental faults driven by function inspection

Abstract

The present application relates to the technical field of fault analysis, and discloses a function inspection driven environmental fault influence range analysis method, which comprises the following steps: collecting result data and environmental parameter data of function inspection, and performing state quantization to obtain function health states and environmental factor states; performing correlation analysis on the function health states and the environmental factor states, determining quantitative influence relations, and constructing a function and environmental influence relation matrix; performing bidirectional influence reasoning on the function health states and the environmental factor states; through time window aggregation and weight decay, the function and environmental influence relation matrix is adaptively adjusted according to the state change of function inspection; the results of bidirectional influence reasoning analysis and preset risk level evaluation principles are fused, the influence range and severity level of environmental faults in function inspection are quantitatively output, and an influence report is generated; the present application can improve the efficiency of the function inspection driven environmental fault influence range analysis method.

Description

Technical Field

[0001] This invention relates to the field of fault analysis technology, and in particular to a method for analyzing the impact range of environmental faults driven by functional inspection. Background Technology

[0002] Existing methods for analyzing the impact range of environmental failures have significant limitations in the data processing stage. They fail to achieve efficient integration of functional inspection results data and environmental parameter data. Their analysis of key performance indicators is not comprehensive enough, and their noise filtering and interference elimination effects are poor, resulting in insufficient accuracy in the quantitative results of functional health status and environmental factor status. At the same time, they lack effective control over the temporal characteristics of the data, and the temporal alignment mechanism is imperfect, which makes it difficult for subsequent correlation analysis to have a reliable data foundation and accurately reflect the intrinsic relationship between the two.

[0003] Existing technologies have significant shortcomings in the construction and inference of impact relationships, failing to form dynamic and accurate models of functional and environmental impact correlations. Their correlation analyses are mostly limited to a single time scale, unable to integrate multi-dimensional collaborative change patterns, resulting in a lack of comprehensiveness in the construction of quantitative impact relationships. Furthermore, the lack of bidirectional impact reasoning capabilities makes it difficult to quickly locate key environmental factors when functions are abnormal, and also makes it impossible to accurately deduce the affected functional scope when the environment is abnormal. Moreover, the relationship matrix lacks adaptive adjustment capabilities and cannot be optimized in real time according to changes in inspection status, ultimately leading to ambiguous judgments of the fault impact range and biased assessments of severity levels, affecting the timeliness and effectiveness of fault handling. Therefore, improving the efficiency of environmental fault impact range analysis has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method for analyzing the impact range of environmental faults driven by functional inspection, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for analyzing the impact range of environmental faults driven by functional inspection, comprising: S1. Collect the result data and environmental parameter data of the functional inspection, and quantify the status to obtain the functional health status and environmental factor status of the functional inspection. S2. Perform a correlation analysis on the functional health status and the environmental factor status to determine the quantitative influence relationship between the functional health status and the environmental factor status, and construct the functional and environmental influence relationship matrix of the functional inspection. S3. Based on the functional-environmental impact relationship matrix, perform bidirectional impact reasoning on the functional health status and the environmental factor status: S31. When the functional health status is abnormal, locate the key environmental factors that have the greatest impact on the functional health status according to the functional-environmental influence relationship matrix, and generate the environmental factor sequence list for the functional inspection. S32. When the environmental factor state is abnormal, deduce the functional state most affected by the environmental factor state based on the functional-environmental impact relationship matrix, and generate the functional impact sequence table for the functional inspection. S4. By using time window aggregation and weight decay, the relationship matrix between the function and the environment is adaptively adjusted as the status of the function inspection changes. S5. Integrate the results of the bidirectional impact reasoning analysis with the preset risk level assessment principles, quantitatively output the impact range and severity level of the environmental faults in the functional inspection, and generate an impact report of the environmental faults.

[0006] In a preferred embodiment, the process of collecting functional inspection result data and environmental parameter data, and quantifying their states to obtain the functional health status and environmental factor status of the functional inspection includes: The result data of the functional inspection is subjected to multi-dimensional feature analysis to obtain the key performance index sequence of the functional inspection, and noise points in the key performance index sequence are removed to obtain the functional feature set of the functional inspection. The environmental parameter data is time-series aligned, and instantaneous interference is eliminated through multi-level filtering technology to obtain a stable environmental parameter sequence for the functional inspection. The indicator values ​​in the functional feature set are compared and analyzed with the preset health threshold range, and the corresponding status identifier is generated based on the comparison results to constitute the functional health status of the functional inspection. The environmental factor status of the functional inspection is generated based on the degree of deviation between the stable environmental parameter sequence and the normal operating parameters in the functional inspection.

[0007] In a preferred embodiment, the step of performing a correlation analysis between the functional health status and the environmental factor status to determine the quantitative influence relationship between the functional health status and the environmental factor status, and constructing the functional-environmental influence relationship matrix of the functional inspection, includes: The functional health status is divided into multi-scale time windows to obtain a subset of the functional statuses for functional inspection. The environmental factors are synchronously divided to obtain an environmental state subset that matches the functional state subset. At each time scale, a multi-dimensional feature comparison is performed between the functional state subset and the environmental state subset to obtain the collaborative change pattern of the functional inspection. Based on the aforementioned collaborative change pattern, a correlation analysis is performed on the functional health status and the environmental factor status to obtain the correspondence between the functional health status and the environmental factor status. By integrating the correspondences at different time scales, a quantitative influence relationship between the functional health state and the environmental factor state is constructed.

[0008] In a preferred embodiment, the step of performing a correlation analysis between the functional health status and the environmental factor status to determine the quantitative influence relationship between the functional health status and the environmental factor status, and constructing the functional-environmental influence relationship matrix of the functional inspection, includes: Based on the quantitative influence relationship, the contribution coefficient of each environmental factor state to each functional health state is obtained; Dynamic stability analysis is performed on the contribution coefficients to obtain the association confidence level between the environmental factor state and the functional health state; Based on the contribution coefficient and the correlation confidence level, a function-environment impact relationship matrix for the functional inspection is constructed, wherein the calculation formula for the function-environment impact relationship matrix is: ; in, This represents the matrix showing the relationship between the stated function and its environmental impact. This represents the contribution coefficient. This represents the coupling coefficient in the quantitative influence relationship. Indicates the current moment. Represents a time variable. This indicates the confidence level of the association. This represents the mean of the association confidence level. The standard deviation represents the confidence level of the association. This represents the preset time decay factor. This indicates the preset scale adjustment parameters. This represents the preset confidence threshold. Represents the hyperbolic tangent function. This represents the Gaussian error function.

[0009] In a preferred embodiment, when the functional health status is abnormal, the key environmental factors that have the greatest impact on the functional health status are located according to the function-environment influence relationship matrix, and a sequence list of environmental factors for the functional inspection is generated, including: Multidimensional feature analysis is performed on the abnormal states in the functional health state to obtain the abnormal state feature vector of the functional health state. The abnormal state feature vector is mapped to the function and environment influence relationship matrix to obtain the initial set of associated environmental factors for the function inspection; An impact assessment is performed on the initial set of related environmental factors to determine the relative impact of the environmental factor status on the functional abnormality status during the functional inspection. Based on the relative degree of influence, the set of associated environmental factors is sorted and filtered to obtain the environmental factor sequence list for the functional inspection.

[0010] In a preferred embodiment, when the environmental factor state is abnormal, the functional state most affected by the environmental factor state is derived based on the function-environment impact relationship matrix, and a functional impact sequence list for the function inspection is generated, including: Based on the function-environment impact relationship matrix, identify the associated function set when the state of the environmental factor is abnormal; An impact propagation path analysis is performed on the associated function set to determine the sensitivity of each function in the function inspection to the state of the environmental factors. Based on the sensitivity level, construct a functional impact assessment system for the functional inspection to assess the potential impact range of each function; Based on the potential impact range, the affected functions of the functional inspection are dynamically sorted to obtain the functional impact sequence list of the functional inspection.

[0011] In a preferred embodiment, the step of adaptively adjusting the function-environment impact matrix as the function inspection status changes through time window aggregation and weight decay includes: In different time windows, the relationship matrix between the function and the environment is sliced ​​and analyzed to obtain a snapshot of the matrix status in multiple time intervals during the function inspection. The state state snapshots of the matrix are integrated to obtain the historical state evolution trajectory of the functional inspection. Dynamic evolution pattern analysis is performed on the historical state evolution trajectory to obtain the changing trend of the functional inspection; Based on the changing trend, the structure of the functional-environmental impact relationship matrix is ​​optimized and adjusted to obtain the adjusted relationship matrix of the functional inspection. The accuracy and timeliness of the adjusted relationship matrix are ensured through a matrix update verification mechanism.

[0012] In a preferred embodiment, the step of performing dynamic evolution pattern analysis on the historical state evolution trajectory to obtain the changing trend of the functional inspection includes: State transition pattern recognition is performed on the historical state evolution trajectory to obtain the key evolutionary features of the functional inspection; Based on the key evolutionary characteristics, an evolutionary trend analysis framework for the functional inspection is constructed. Based on the aforementioned evolution trend analysis framework, the key evolutionary features are correlated and matched to obtain the regularity pattern of the functional inspection. Based on the aforementioned pattern of change, the influence relationship between the functional health status and the environmental factor status is predicted to obtain the changing trend of the functional inspection.

[0013] In a preferred embodiment, the results of the bidirectional impact reasoning analysis are integrated with preset risk level assessment principles to quantitatively output the impact range and severity level of environmental faults during functional inspections, and to generate an impact report of the environmental faults, including: By integrating the results of the bidirectional influence reasoning analysis, the basic dataset of environmental faults in the functional inspection is obtained. Multi-dimensional impact relationship analysis is performed on the basic dataset to obtain the comprehensive impact profile of the environmental fault; Based on the preset risk level assessment principles, the comprehensive impact profile is mapped to the corresponding risk level; Boundary analysis is performed on the comprehensive impact profile to determine the impact area of ​​the environmental fault; An impact report on the environmental failure is generated based on the risk level and the affected area.

[0014] In a preferred embodiment, mapping the comprehensive impact profile to a corresponding risk level based on a preset risk level assessment principle includes: By analyzing the influence relationship patterns in the comprehensive influence profile, the key influence dimensions of the environmental failure are obtained. The key impact dimensions are matched with the preset risk level assessment principles to determine the basic risk level of the environmental failure. Based on the historical evolution trajectory of the environmental fault, the risk transmission path of the basic risk level is verified to obtain the risk level of the environmental fault.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves precise quantification of functional health status and environmental factor status by performing multi-dimensional feature analysis, noise removal, time sequence alignment, and multi-level filtering on functional inspection result data and environmental parameter data. Furthermore, through multi-scale time window segmentation and collaborative change pattern analysis, a precise quantitative influence relationship matrix between the two is constructed, significantly improving the accuracy of influence relationship identification and providing a solid foundation for subsequent fault impact analysis.

[0016] 2. This invention, through the construction of a functional-environmental impact relationship matrix, achieves bidirectional impact reasoning on both functional health status and environmental factor status. Combining time window aggregation and weight decay mechanisms, the matrix is ​​adaptively adjusted, ensuring dynamic adaptability of the analysis process. Finally, by integrating the reasoning results with risk level assessment principles, the impact range and severity level of environmental faults are quantitatively output, generating an impact report. This effectively improves the efficiency and reliability of environmental fault impact range analysis, providing precise basis for fault handling. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the environmental fault impact range analysis method driven by functional inspection provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for analyzing the impact range of environmental faults driven by functional inspection. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1The diagram shown is a flowchart illustrating a method for analyzing the impact range of environmental faults driven by functional inspection, according to an embodiment of the present invention. In this embodiment, the method includes: S1. Collect the result data and environmental parameter data of the functional inspection, and quantify the status to obtain the functional health status and environmental factor status of the functional inspection. In this embodiment of the invention, the process of collecting the result data and environmental parameter data of the functional inspection, and quantifying their states to obtain the functional health status and environmental factor status of the functional inspection includes: The result data of the functional inspection is subjected to multi-dimensional feature analysis to obtain the key performance index sequence of the functional inspection, and noise points in the key performance index sequence are removed to obtain the functional feature set of the functional inspection. The environmental parameter data is time-series aligned, and instantaneous interference is eliminated through multi-level filtering technology to obtain a stable environmental parameter sequence for the functional inspection. The indicator values ​​in the functional feature set are compared and analyzed with the preset health threshold range, and the corresponding status identifier is generated based on the comparison results to constitute the functional health status of the functional inspection. The environmental factor status of the functional inspection is generated based on the degree of deviation between the stable environmental parameter sequence and the normal operating parameters in the functional inspection.

[0021] Specifically, core dimensions directly related to the equipment's operating status are extracted from the functional inspection results data, including response efficiency, operational stability, and task completion accuracy. Specific indicators that can intuitively reflect the actual performance of the function are selected under each dimension. These indicators are arranged sequentially according to the inspection time to form a sequence of key performance indicators. Then, the change range of adjacent indicator data within the same time period is compared. When the difference between a certain indicator data and its adjacent data exceeds the fluctuation range during normal equipment operation, the data is determined to be a noise point. This noise point is replaced with the average value of the adjacent normal data. After screening and correction, the functional feature set of the functional inspection is obtained.

[0022] Furthermore, using the time nodes of functional inspection as a unified benchmark, environmental parameter data acquired by different acquisition devices and at different acquisition frequencies are standardized to ensure that there are corresponding environmental parameter data at each inspection time point, thereby achieving temporal alignment of environmental parameter data. Then, multi-level filtering technology is used to process these data. First, a fixed-length time window is selected using the sliding window method, retaining the median value of the data within the window and removing abnormal data with extreme deviations. Then, adjacent data smoothing is performed on the data after window processing, and the mean of the current data and the data before and after it is replaced with the original data, thereby eliminating abnormal data caused by instantaneous voltage fluctuations, temporary signal interference, etc., and finally obtaining a stable environmental parameter sequence for functional inspection.

[0023] Furthermore, the preset health threshold range is determined based on the statistical analysis of historical data from the long-term normal operation of the equipment. The reasonable operating range of each functional indicator is clearly defined. Each indicator value in the functional feature set is compared with the corresponding health threshold range. If the indicator value is within the threshold range, a status identifier indicating normal operation is generated. If the indicator value is higher than the upper limit of the threshold range or lower than the lower limit of the threshold range, a status identifier indicating abnormal operation is generated. The status identifiers corresponding to all functional indicators are combined in sequence according to the indicator type to form the functional health status of the functional inspection.

[0024] Furthermore, the standard values ​​of each environmental parameter during normal operation of the functional inspection are first defined. These standard values ​​come from the standard parameters set by the equipment at the factory and the average parameters during long-term stable operation. Each parameter value in the stable environmental parameter sequence is compared with the corresponding standard value, and the absolute difference between the two is calculated. The deviation level is divided according to the size of the difference. A difference of zero indicates no deviation, a difference within a preset slight range indicates slight deviation, and a difference exceeding a preset severe range indicates severe deviation. The corresponding environmental factor status description is generated according to the deviation level, which constitutes the environmental factor status of the functional inspection.

[0025] In summary, by performing multi-dimensional feature analysis on the functional inspection results data, key performance indicators of core dimensions such as response efficiency and operational stability are accurately selected and formed into a sequence. Then, by replacing noise points that exceed the normal fluctuation range, a true and reliable functional feature set is obtained, which lays a data foundation for the accurate determination of the subsequent functional health status and avoids the interference of noise data on functional status analysis.

[0026] In summary, aligning environmental parameter data timelines based on functional inspection time nodes ensures consistency in the time dimension of the data. Furthermore, multi-level filtering techniques, such as using a sliding window to take the median value and smoothing adjacent data, effectively eliminate abnormal data caused by momentary interference. The resulting stable environmental parameter sequence truly reflects the actual environmental conditions of the equipment operation, providing accurate data support for environmental factor status analysis.

[0027] In summary, the health threshold range determined based on the historical normal operation data of the equipment provides a clear standard for judging the status of functional indicators. By comparing the indicator values ​​in the functional feature set with the threshold range one by one and generating corresponding status identifiers, the functional health status is presented intuitively through clear identifiers, avoiding the ambiguity and subjectivity of functional status judgment.

[0028] In summary, based on the equipment's factory standard parameters and long-term stable operating average parameters, the deviation level is classified by calculating the absolute difference between stable environmental parameters and standard parameters, thereby generating environmental factor status. This clearly defines the degree of abnormality of environmental parameters, making the description of environmental status more targeted. The entire data acquisition and status quantification process provides accurate and effective data prerequisites for subsequent functional and environmental correlation analysis and fault impact range judgment.

[0029] S2. Perform a correlation analysis on the functional health status and the environmental factor status to determine the quantitative influence relationship between the functional health status and the environmental factor status, and construct the functional and environmental influence relationship matrix of the functional inspection. In this embodiment of the invention, the step of performing correlation analysis between the functional health status and the environmental factor status to determine the quantitative influence relationship between the functional health status and the environmental factor status, and constructing the functional-environmental influence relationship matrix of the functional inspection, includes: The functional health status is divided into multi-scale time windows to obtain a subset of the functional statuses for functional inspection. The environmental factors are synchronously divided to obtain an environmental state subset that matches the functional state subset. At each time scale, a multi-dimensional feature comparison is performed between the functional state subset and the environmental state subset to obtain the collaborative change pattern of the functional inspection. Based on the aforementioned collaborative change pattern, a correlation analysis is performed on the functional health status and the environmental factor status to obtain the correspondence between the functional health status and the environmental factor status. By integrating the correspondences at different time scales, a quantitative influence relationship between the functional health state and the environmental factor state is constructed.

[0030] In this embodiment of the invention, the step of performing correlation analysis between the functional health status and the environmental factor status to determine the quantitative influence relationship between the functional health status and the environmental factor status, and constructing the functional-environmental influence relationship matrix of the functional inspection, includes: Based on the quantitative influence relationship, the contribution coefficient of each environmental factor state to each functional health state is obtained; Dynamic stability analysis is performed on the contribution coefficients to obtain the association confidence level between the environmental factor state and the functional health state; Based on the contribution coefficient and the correlation confidence level, a function-environment impact relationship matrix for the functional inspection is constructed, wherein the calculation formula for the function-environment impact relationship matrix is: ; in, This represents the matrix showing the relationship between the stated function and its environmental impact. This represents the contribution coefficient. This represents the coupling coefficient in the quantitative influence relationship. Indicates the current moment. Represents a time variable. This indicates the confidence level of the association. This represents the mean of the association confidence level. The standard deviation represents the confidence level of the association. This represents the preset time decay factor. This indicates the preset scale adjustment parameters. This represents the preset confidence threshold. Represents the hyperbolic tangent function. This represents the Gaussian error function.

[0031] Specifically, the criteria for dividing the multi-scale time windows are first determined. The division scale includes hourly, daily, and weekly levels. The division is based on the regular frequency of functional inspections. If the inspection frequency is once per hour, the hourly window length is set to 1 hour, the daily window length is set to 24 hours, and the weekly window length is set to 168 hours. Then, according to the set time window length, the sequence of status identifiers corresponding to the functional health status is segmented. Each time window contains the status identifiers of all functional indicators within that time period. The set corresponding to each segment is the functional status subset of the functional inspection.

[0032] Furthermore, based on the standard of dividing functional health status into multi-scale time windows, the deviation level sequence corresponding to the environmental factor status is synchronously segmented to ensure that the segmented time window of the environmental factor status is completely consistent with the segmented time window of the functional health status. Each identical time window contains the deviation level of all environmental parameters within that time period, and the set corresponding to each segment is the environmental state subset that matches the functional state subset.

[0033] Furthermore, at each time scale, the status identifiers of each functional indicator are extracted from the functional status subset, and the deviation levels of each environmental parameter are extracted from the corresponding environmental status subset. Then, the status identifiers of the functional indicators and the deviation levels of the environmental parameters within the same time window are compared one by one. The changes in the deviation levels of the environmental parameters when the functional indicators change from normal to abnormal are recorded, as well as the stability of the environmental parameters when the functional indicators remain normal. The correlation of these recorded changes is the collaborative change pattern of the functional inspection.

[0034] Furthermore, for the collaborative change patterns obtained at each time scale, the number of times each state of functional health and each state of environmental factors occur simultaneously in the pattern is counted. The environmental factor state that occurs most frequently is bound to the functional health state to determine the relationship between the two. The binding result of each functional health state and the corresponding environmental factor state is the correspondence between functional health state and environmental factor state.

[0035] Furthermore, the correspondence between functional health status and environmental factor status obtained at all time scales is collected, and the total frequency of each correspondence at different time scales is counted. The correlation strength of the correspondence is determined based on the total frequency, with higher frequencies indicating greater correlation strength. Subsequently, the correlation strength is transformed into a specific description of the degree of influence, such as "the correlation strength between an abnormal functional indicator and a serious deviation of an environmental parameter is high, and the degree of influence is directly caused." All these correspondences with descriptions of the degree of influence are integrated to form the quantitative influence relationship between functional health status and environmental factor status.

[0036] Specifically, the criteria for dividing the multi-scale time windows are first determined. The division scale includes hourly, daily, and weekly levels. The division is based on the regular frequency of functional inspections. If the inspection frequency is once per hour, the hourly window length is set to 1 hour, the daily window length is set to 24 hours, and the weekly window length is set to 168 hours. Then, according to the set time window length, the sequence of status identifiers corresponding to the functional health status is segmented. Each time window contains the status identifiers of all functional indicators within that time period. The set corresponding to each segment is the functional status subset of the functional inspection.

[0037] Furthermore, based on the standard of dividing functional health status into multi-scale time windows, the deviation level sequence corresponding to the environmental factor status is synchronously segmented to ensure that the segmented time window of the environmental factor status is completely consistent with the segmented time window of the functional health status. Each identical time window contains the deviation level of all environmental parameters within that time period, and the set corresponding to each segment is the environmental state subset that matches the functional state subset.

[0038] Furthermore, at each time scale, the status identifiers of each functional indicator are extracted from the functional status subset, and the deviation levels of each environmental parameter are extracted from the corresponding environmental status subset. Then, the status identifiers of the functional indicators and the deviation levels of the environmental parameters within the same time window are compared one by one. The changes in the deviation levels of the environmental parameters when the functional indicators change from normal to abnormal are recorded, as well as the stability of the environmental parameters when the functional indicators remain normal. The correlation of these recorded changes is the collaborative change pattern of the functional inspection.

[0039] Furthermore, for the collaborative change patterns obtained at each time scale, the number of times each state of functional health and each state of environmental factors occur simultaneously in the pattern is counted. The environmental factor state that occurs most frequently is bound to the functional health state to determine the relationship between the two. The binding result of each functional health state and the corresponding environmental factor state is the correspondence between functional health state and environmental factor state.

[0040] Furthermore, the correspondence between functional health status and environmental factor status obtained at all time scales is collected, and the total frequency of each correspondence at different time scales is counted. The correlation strength of the correspondence is determined based on the total frequency, with higher frequencies indicating greater correlation strength. Subsequently, the correlation strength is transformed into a specific description of the degree of influence, such as "the correlation strength between an abnormal functional indicator and a serious deviation of an environmental parameter is high, and the degree of influence is directly caused." All these correspondences with descriptions of the degree of influence are integrated to form the quantitative influence relationship between functional health status and environmental factor status.

[0041] Specifically, the contribution coefficient comes from the quantitative impact relationship obtained after the correlation analysis between functional health status and environmental factor status. It is the quantitative result of the impact of each environmental factor status on each functional health status. Furthermore, the coupling coefficient is a quantitative indicator inherent in quantitative influence relationships, used to reflect the strength of interaction between factors.

[0042] Furthermore, the association confidence score is derived by performing dynamic stability analysis on the contribution coefficient and is used to characterize the reliability of the association between environmental factor status and functional health status.

[0043] Furthermore, the mean of the association confidence score is the result of statistically averaging all the association confidence score data. The calculation is done by first adding all the association confidence score data together, and then dividing the sum by the total number of association confidence score data.

[0044] Furthermore, the standard deviation of the association confidence score is calculated by taking the square root of the average of the squared deviations of all association confidence score data from the mean. The calculation is as follows: first, the difference between each association confidence score data and the mean is calculated; then, each difference is squared; next, the average of all squared results is calculated; and finally, the square root of the average is taken.

[0045] Furthermore, the time decay factor, scale adjustment parameter, and confidence threshold are all fixed parameters that are preset during the analysis process. They are used to adjust the accuracy and adaptability of the formula calculation. The specific values ​​are determined based on the actual scenario of the functional inspection and the historical data verification results.

[0046] Furthermore, the significance of this formula is to construct the numerator by integrating the normal distribution correction results of contribution coefficient, time decay effect, and correlation confidence, and to construct the denominator by integrating the hyperbolic tangent correction results of all relevant contribution coefficients, coupling coefficients, and correlation confidence. Then, by calculating the ratio of the numerator to the denominator, the degree of influence of a certain environmental factor state on a certain functional health state is quantitatively characterized, and finally, the elements of the functional-environmental influence relationship matrix are formed.

[0047] Furthermore, as the difference between the time variable and the current moment increases, the value of the time decay term will gradually decrease, causing the integral result of the numerator to decrease accordingly. The integral calculation here is a cumulative summation operation on the relevant numerical changes of the time variable within a set interval.

[0048] Furthermore, as the correlation confidence level gradually approaches its mean, the result after Gaussian error function correction will tend to stabilize, keeping the integral result of the numerator stable. The Gaussian error function correction process involves dividing the difference between the correlation confidence level and the mean by the product of the standard deviation corresponding to the mean and the square root of 2, and then substituting the result into the Gaussian error function for calculation to obtain the correction value.

[0049] Furthermore, the larger the value of the contribution coefficient, the larger the integral result of the numerator, and the larger the value of the matrix elements. During the integral calculation, the product of the contribution coefficient, the time decay term, and the Gaussian error function correction result will be accumulated and summed over the time interval.

[0050] Furthermore, the larger the value of the coupling coefficient, the larger the integral result of the denominator, and the smaller the value of the matrix elements. The integral calculation of the denominator is to accumulate and sum the products of all relevant contribution coefficients, coupling coefficients, and hyperbolic tangent correction results over the time interval.

[0051] Furthermore, when the correlation confidence is higher than the confidence threshold, the result after hyperbolic tangent correction will increase, which will increase the integral result of the denominator and decrease the value of the matrix elements. The hyperbolic tangent correction process is to multiply the difference between the correlation confidence and the confidence threshold by the scaling parameter, and then substitute the result into the hyperbolic tangent function for calculation to obtain the correction value.

[0052] Furthermore, when the correlation confidence is lower than the confidence threshold, the result after hyperbolic tangent correction will decrease, which will reduce the integral result in the denominator and increase the value of the matrix elements. The hyperbolic tangent correction process is also to multiply the difference between the correlation confidence and the confidence threshold by the scaling parameter, and then substitute the result into the hyperbolic tangent function for calculation to obtain the corrected value.

[0053] In summary, dividing functional health status by multi-scale time windows can cover changes in functional status across different time dimensions, including short, medium, and long term, avoiding the limitations of analysis at a single time scale and ensuring that the resulting subset of functional status more comprehensively reflects the functional operation.

[0054] In summary, synchronously dividing the environmental factors into states ensures a complete match between the environmental state subset and the functional state subset in the time dimension, providing a time consistency basis for subsequent feature comparison and correlation analysis between the two, and avoiding correlation deviations caused by time misalignment.

[0055] In summary, multi-dimensional feature comparison at each time scale can accurately capture subtle changes and correlations between functional states and environmental states, making the collaborative change pattern more consistent with actual operation and providing a real basis for determining the corresponding relationship.

[0056] In summary, correlation analysis based on collaborative change patterns, which determines the correspondence by counting the number of times they occur simultaneously, ensures the objectivity and accuracy of the correspondence and avoids errors caused by subjective judgment.

[0057] In summary, integrating the correspondences across different time scales and transforming them into quantitative impact relationships can synthesize multi-dimensional information, making the degree of impact between function and environment clearer and more measurable, and providing a core quantitative basis for the subsequent construction of a matrix of function-environment impact relationships.

[0058] In summary, dividing functional health status by multi-scale time windows can cover changes in functional status across different time dimensions, including short, medium, and long term, avoiding the limitations of analysis at a single time scale and ensuring that the resulting subset of functional status more comprehensively reflects the functional operation.

[0059] In summary, synchronously dividing the environmental factors into states ensures a complete match between the environmental state subset and the functional state subset in the time dimension, providing a time consistency basis for subsequent feature comparison and correlation analysis between the two, and avoiding correlation deviations caused by time misalignment.

[0060] In summary, multi-dimensional feature comparison at each time scale can accurately capture subtle changes and correlations between functional states and environmental states, making the collaborative change pattern more consistent with actual operation and providing a real basis for determining the corresponding relationship.

[0061] In summary, correlation analysis based on collaborative change patterns, which determines the correspondence by counting the number of times they occur simultaneously, ensures the objectivity and accuracy of the correspondence and avoids errors caused by subjective judgment.

[0062] In summary, integrating the correspondences across different time scales and transforming them into quantitative impact relationships can synthesize multi-dimensional information, making the degree of impact between function and environment clearer and more measurable, and providing a core quantitative basis for the subsequent construction of a matrix of function-environment impact relationships.

[0063] S3. Based on the functional-environmental impact relationship matrix, perform bidirectional impact reasoning on the functional health status and the environmental factor status: S31. When the functional health status is abnormal, locate the key environmental factors that have the greatest impact on the functional health status according to the functional-environmental influence relationship matrix, and generate the environmental factor sequence list for the functional inspection. S32. When the environmental factor state is abnormal, deduce the functional state most affected by the environmental factor state based on the functional-environmental impact relationship matrix, and generate the functional impact sequence table for the functional inspection. In this embodiment of the invention, when the functional health status is abnormal, the key environmental factors that have the greatest impact on the functional health status are located according to the functional-environmental influence relationship matrix, and an environmental factor sequence list for the functional inspection is generated, including: Multidimensional feature analysis is performed on the abnormal states in the functional health state to obtain the abnormal state feature vector of the functional health state. The abnormal state feature vector is mapped to the function and environment influence relationship matrix to obtain the initial set of associated environmental factors for the function inspection; An impact assessment is performed on the initial set of related environmental factors to determine the relative impact of the environmental factor status on the functional abnormality status during the functional inspection. Based on the relative degree of influence, the set of associated environmental factors is sorted and filtered to obtain the environmental factor sequence list for the functional inspection.

[0064] In this embodiment of the invention, when the environmental factor state is abnormal, the function state most affected by the environmental factor state is derived based on the function-environment impact relationship matrix, and a function impact sequence table for the function inspection is generated, including: Based on the function-environment impact relationship matrix, identify the associated function set when the state of the environmental factor is abnormal; An impact propagation path analysis is performed on the associated function set to determine the sensitivity of each function in the function inspection to the state of the environmental factors. Based on the sensitivity level, construct a functional impact assessment system for the functional inspection to assess the potential impact range of each function; Based on the potential impact range, the affected functions of the functional inspection are dynamically sorted to obtain the functional impact sequence list of the functional inspection.

[0065] Specifically, first, the multidimensional feature dimensions of abnormal states in functional health status are defined, including the specific type of abnormal function, the duration of the abnormality, and the number of secondary functional failures caused by the abnormality. Then, all state entries marked as abnormal are selected from the functional health status. For each abnormal entry, the specific information of the above three dimensions is extracted one by one, such as "abnormal operation and processing function, abnormality in 3 consecutive inspection cycles, accompanied by 1 secondary functional abnormality". Then, according to the fixed order of "abnormal function type - abnormal duration - number of secondary failures", the feature information of each abnormal entry is transformed into structured feature elements. The feature elements of all abnormal entries are concatenated in the order of the first occurrence of the abnormality to form the abnormal state feature vector of functional health status.

[0066] Furthermore, first confirm the column dimension definition of the function and environmental impact relationship matrix. Each column of the matrix uniquely corresponds to a functional health state. Match each feature element in the abnormal state feature vector with the matrix column dimension to find the matrix column corresponding to each abnormal function type. For example, the third column of the matrix corresponds to "data transmission function abnormality" in the vector. Then check all rows in these matching columns where the cell value is not 0. The environmental factor status corresponding to the row label of these rows is the environmental factor associated with the functional abnormality. Collect all these associated environmental factor statuses, remove duplicate status entries, and form the initial set of associated environmental factors for functional inspection.

[0067] Furthermore, the cell values ​​of each environmental factor status in the corresponding abnormal function column are retrieved from the initial set of associated environmental factors in the function and environmental impact relationship matrix. These cell values ​​are used as the core basis to assess the degree of impact. For example, the value of the environmental factor status "insufficient network bandwidth" in the "abnormal data transmission function" column is 0.85, and the value of the environmental factor status "abnormal transmission protocol" in the same column is 0.62. Therefore, the degree of impact of "insufficient network bandwidth" on the abnormal function is higher than that of "abnormal transmission protocol". At the same time, the specific values ​​corresponding to each associated environmental factor status are recorded to clarify the relative degree of impact of each environmental factor status on the abnormal function status in the function inspection.

[0068] Furthermore, the environmental factor states in the initial set of related environmental factors are sorted in descending order of the numerical values ​​corresponding to the relative impact degree. If there are environmental factor states with the same value, they are sorted in descending order of the number of times the environmental factor state has caused functional abnormalities. Then, an impact degree screening threshold is set to remove environmental factor states with values ​​lower than the threshold. The remaining environmental factor states after screening are arranged in order of the sorting results, and the corresponding impact degree value is marked after each state, forming an ordered list containing the environmental factor state name and the impact degree value. This list is the environmental factor sequence list for functional inspection.

[0069] Specifically, first, define the row dimension of the function and environmental impact relationship matrix. Each row of the matrix uniquely corresponds to an environmental factor state. When an environmental factor state becomes abnormal, locate the row corresponding to the abnormal environmental factor state in the matrix. Then, check all the columns in that row where the cell value is not 0. The functional health status corresponding to the column labels of these columns is the function associated with the abnormal environmental factor state. Collect all these associated functional health statuses, remove duplicate entries, and form the associated function set for function inspection.

[0070] Furthermore, the impact propagation path of each function in the related function set is analyzed. First, the position of each function in the entire inspection system and its connection relationship with other functions are determined. For example, the downstream of the data acquisition function is connected to the data processing function, and the downstream of the data processing function is connected to the data storage function. When abnormal environmental factors directly affect the data acquisition function, the abnormality of the data acquisition function will indirectly affect the data processing function, and the abnormality of the data processing function will indirectly affect the data storage function. Then, the cell values ​​of each related function column in the row corresponding to the abnormal environmental factor in the function-environment impact relationship matrix are retrieved. The sensitivity is judged in combination with the impact path hierarchy. The sensitivity of functions with a first-level impact path and a cell value ≥ 0.7 is judged as "extremely high". The sensitivity of functions with a first-level impact path and a cell value between 0.4 and 0.6 or a second-level impact path and a cell value ≥ 0.7 is judged as "relatively high". The sensitivity of other related functions is judged as "moderate". This determines the sensitivity of each function to the state of the environmental factor.

[0071] Furthermore, using the sensitivity of each function as the core dimension, and combining the importance level of the function in the inspection system with the number of downstream modules associated with the function, a function impact assessment system is constructed. In the assessment system, "extremely high" sensitivity corresponds to 30 points, "relatively high" corresponds to 20 points, and "average" corresponds to 10 points; core functions correspond to 40 points, key functions correspond to 25 points, and auxiliary functions correspond to 10 points; the number of downstream modules ≥3 corresponds to 30 points, 1-2 corresponds to 20 points, and 0 corresponds to 10 points. The scores of each function in the three dimensions are added together to obtain a comprehensive score. Functions with a comprehensive score ≥80 points are judged as having a "wide" potential impact range, 60-79 points as "relatively high", 40-59 points as "moderate", and <40 points as "relatively low". This is used to assess the potential impact range of each function.

[0072] Furthermore, the affected functions in the related function set are initially sorted according to the order of their potential impact range from largest to smallest. If there are functions with the same potential impact range, they are sorted again according to their importance level from highest to lowest. Then, a screening threshold is set to remove functions with a "small" potential impact range. The remaining functions after sorting are arranged in order, and each function is labeled with its potential impact range level and importance level, forming an ordered list containing function name, potential impact range, and importance level. This list is the function impact sequence list for function inspection.

[0073] In summary, by analyzing the multidimensional features of abnormal states and constructing vectors, scattered abnormal information is transformed into a structured and ordered set of features, clearly presenting the key attributes of functional abnormalities. This provides clear feature anchors for subsequent accurate matching of associated environmental factors, avoiding omissions or misjudgments caused by fragmented abnormal information.

[0074] In summary, based on the precise mapping between matrix column dimensions and abnormal feature vectors, environmental factors related to functional abnormalities can be directly located, ensuring that the initial set of associated environmental factors can fully cover all potential influencing factors without missing any environmental triggers that may lead to functional abnormalities, thus providing a comprehensive candidate basis for subsequent impact assessment.

[0075] In summary, determining the relative impact based on matrix cell values ​​transforms the influence of environmental factors on functional abnormalities into quantifiable and comparable indicators, eliminating the uncertainty of subjective experience-based judgments and providing objective data support for the magnitude of each environmental factor's impact, thereby improving the accuracy and credibility of impact assessments.

[0076] In summary, by generating a sequence list of environmental factors through sorting and threshold filtering, the priority of key environmental factors affecting functional abnormalities can be intuitively presented. Subsequent troubleshooting can focus on the factors at the beginning of the sequence list, reducing ineffective troubleshooting steps, significantly improving the efficiency of fault location, and helping to quickly develop targeted fault solutions.

[0077] In summary, by locating abnormal environmental factors in the function-environment impact matrix and filtering non-zero value columns, all related functions can be accurately identified, ensuring that the set of related functions covers all potentially affected objects. This avoids analytical loopholes caused by neglecting implicit related functions during manual screening and provides a complete functional foundation for subsequent impact analysis.

[0078] In summary, by combining functional connections to analyze the propagation path of influence, and by using the correlation matrix to determine the sensitivity, we can not only reflect the hierarchical differences in influence, but also incorporate the quantitative intensity of influence. This allows the sensitivity determination to align with the actual operating logic of the system, avoid the subjectivity of relying solely on experience to determine sensitivity, and improve the reliability of the sensitivity results.

[0079] In summary, by taking sensitivity as the core and integrating functional importance with the number of downstream modules to build an evaluation system, the potential impact range can be comprehensively measured from multiple dimensions. This can objectively reflect the extent to which the inspection system is affected after the function is affected, avoid the one-sidedness of a single dimension judgment, and make the assessment of the potential impact range more in line with the actual failure impact scenario.

[0080] In summary, by dynamically sorting and filtering functions according to their potential impact and functional importance, the generated functional impact sequence list clearly presents the priority of affected functions. Subsequent fault handling can prioritize key functions at the top of the sequence list, reducing ineffective functional verification steps, significantly improving the targeting and efficiency of responding to environmental fault impacts, and helping to quickly control the spread of faults.

[0081] S4. By using time window aggregation and weight decay, the relationship matrix between the function and the environment is adaptively adjusted as the status of the function inspection changes. In this embodiment of the invention, the step of adaptively adjusting the function-environment influence matrix according to the state changes of the function inspection through time window aggregation and weight decay includes: In different time windows, the relationship matrix between the function and the environment is sliced ​​and analyzed to obtain a snapshot of the matrix status in multiple time intervals during the function inspection. The state state snapshots of the matrix are integrated to obtain the historical state evolution trajectory of the functional inspection. Dynamic evolution pattern analysis is performed on the historical state evolution trajectory to obtain the changing trend of the functional inspection; Based on the changing trend, the structure of the functional-environmental impact relationship matrix is ​​optimized and adjusted to obtain the adjusted relationship matrix of the functional inspection. The accuracy and timeliness of the adjusted relationship matrix are ensured through a matrix update verification mechanism.

[0082] In this embodiment of the invention, the step of performing dynamic evolution pattern analysis on the historical state evolution trajectory to obtain the changing trend of the functional inspection includes: State transition pattern recognition is performed on the historical state evolution trajectory to obtain the key evolutionary features of the functional inspection; Based on the key evolutionary characteristics, an evolutionary trend analysis framework for the functional inspection is constructed. Based on the aforementioned evolution trend analysis framework, the key evolutionary features are correlated and matched to obtain the regularity pattern of the functional inspection. Based on the aforementioned pattern of change, the influence relationship between the functional health status and the environmental factor status is predicted to obtain the changing trend of the functional inspection.

[0083] Specifically, the time window is first divided into daily time windows, meaning each time window corresponds to a 24-hour functional inspection cycle. At 24:00 every day, the functional and environmental impact relationship matrix data corresponding to the last functional inspection of that day is extracted as a matrix status snapshot for that time window. Each snapshot fully retains the row dimensions, column dimensions, and numerical information of all cells of the matrix. The matrix data of the past 30 days are extracted in chronological order to obtain 30 matrix status snapshots belonging to different time intervals, ensuring that each time window has a unique corresponding matrix status snapshot.

[0084] Furthermore, the acquired matrix state snapshots are arranged in chronological order, and the numerical differences of cells in the same row and column of adjacent snapshots are compared one by one. If the difference is less than 0.1, the value of that cell in the next snapshot is directly retained; if the difference is greater than or equal to 0.1, the average value of that cell in the first three snapshots is calculated, and the average value is used to replace the value of that cell in the next snapshot to correct the inconsistency. This correction process is repeated until the numerical differences of cells between all adjacent snapshots are less than 0.1. Then, all the corrected matrix state snapshots are concatenated in chronological order to form a historical state evolution trajectory of the function inspection that reflects the changes of the matrix over time. Each node in the trajectory is a matrix state snapshot that has been corrected for consistency.

[0085] Furthermore, for each matrix state snapshot in the historical state evolution trajectory, the numerical change data of all cells are extracted. The focus is on analyzing the rows and columns with high-frequency changes in the matrix. For example, the number of days that the values ​​of all cells in the row corresponding to a certain environmental factor state have increased continuously, or the period of continuous decrease of the values ​​of all cells in the column corresponding to a certain functional health state, are counted. The patterns of these numerical changes are summarized, such as "the value of the cell in the row corresponding to a slight deviation of environmental parameter X has increased for 10 consecutive days, with an average daily increase of 0.03" and "the value of the cell in the column corresponding to an abnormal function Y has decreased for 5 consecutive days, with an average daily decrease of 0.02". Based on these patterns, the dynamic evolution pattern of the relationship matrix between function and environmental influence is summarized, and then the trend of "the influence of environmental parameter X on each function continues to increase, and the influence of each environmental factor on function Y continues to decrease" is obtained in the functional inspection.

[0086] Furthermore, based on the obtained trend, the original functional and environmental impact relationship matrix is ​​structurally optimized and adjusted. For cells with "continuously increasing impact", the value is increased by 0.1 to enhance its impact weight; for cells with "continuously decreasing impact", the value is decreased by 0.05 to reduce its impact weight. If the trend shows the emergence of new environmental factor states or functional health states, corresponding rows or columns are added to the matrix, and the initial cell values ​​of the new rows or columns are set according to the numerical levels of similar states. After all adjustments are completed, the adjusted relationship matrix of functional inspection is obtained.

[0087] Furthermore, the latest functional inspection data from the three days in which the adjusted relationship matrix was generated is selected. The data is then input into the adjusted matrix for bidirectional impact reasoning to obtain the reasoning results on the scope and severity of environmental faults. At the same time, the actual environmental faults and their impacts in these three inspections are obtained. The degree of agreement between the reasoning results and the actual situation is compared. If the degree of agreement is higher than 90%, the adjusted relationship matrix is ​​deemed to have passed verification and can be put into use. If the degree of agreement is lower than 90%, the previous step is returned to readjust the values ​​of the matrix cells according to the actual differences until the degree of agreement between the reasoning results and the actual situation is higher than 90%. This matrix update verification mechanism ensures the accuracy and timeliness of the adjusted relationship matrix.

[0088] Specifically, we first define the types of state transition patterns, including continuously rising values, continuously falling values, stable values, and fluctuating values. Then, we iterate through all matrix state snapshots in the historical state evolution trajectory, track the value changes of each cell in chronological order, and count the state transition pattern type corresponding to each cell. At the same time, we record the frequency of different transition patterns in the rows corresponding to each environmental factor state and the columns corresponding to each functional health state. We then filter out the rows or columns corresponding to transition patterns with a frequency higher than 60% of the total time window. For example, "80% of the cells in the row corresponding to the slight deviation of environmental parameter 1 are in the continuously rising value pattern" and "70% of the cells in the column corresponding to the abnormal function 2 are in the stable value pattern". The row, column, and value change characteristics corresponding to these high-frequency transition patterns are the key evolutionary characteristics of functional inspection.

[0089] Furthermore, taking key evolutionary features as the core, an evolutionary trend analysis framework is constructed from two dimensions: the first dimension is the "environmental factor dimension," which classifies the environmental factor status in the key evolutionary features according to the type of environmental factor, and records the state transition pattern and numerical change magnitude corresponding to the type of environmental factor under each category; the second dimension is the "functional dimension," which classifies the functional health status column in the key evolutionary features according to the type of functional module, and records the state transition pattern and numerical change law corresponding to the type of function under each category. At the same time, a "correlation analysis layer" is set in the framework to establish the correspondence between the environmental factor dimension and the functional dimension features in the future, and finally form a functional inspection evolutionary trend analysis framework with a clear structure and clear classification.

[0090] Furthermore, in the "correlation analysis layer" of the evolution trend analysis framework, the key evolutionary characteristics of each type of environmental factor under the environmental factor dimension are matched one by one with the key evolutionary characteristics of each type of function under the functional dimension. The number of times the two types of characteristics appear simultaneously within the same time window is counted. For example, if the "characteristic of continuous increase in temperature environmental factor values" and the "characteristic of continuous increase in data storage module function values" appear simultaneously 15 times in 20 time windows, and the proportion of simultaneous occurrences to the total number of time windows is higher than 70%, then it is determined that the two types of characteristics have a strong correlation, and it is recorded as "strong correlation between the increase in temperature environmental factor values ​​and the increase in data storage module function values". All strongly correlated characteristics are combined and organized to form a description that reflects the changing pattern of the relationship between the environment and function. These descriptions are the changing pattern of functional inspection.

[0091] Furthermore, for each pattern of change, combined with the rate of change of the corresponding numerical value in the historical state evolution trajectory, the influence relationship between functional health status and environmental factors status within the next three time windows is predicted. For example, if a certain pattern of change is "the value of network-related environmental factors increases by 0.04 per week, and is strongly correlated with the value of data transmission module function increasing by 0.03 per week", then it is predicted that in the next three weeks, the value of the influence of network-related environmental factors on the data transmission module function will continue to increase at a rate of 0.04 and 0.03 per week, that is, "the influence of network-related environmental factors on the data transmission module function continues to strengthen". All prediction results are integrated to form a description covering the direction and magnitude of the changes in the relationship between each environmental factor and function. These descriptions are the changing trends of functional inspection.

[0092] In summary, obtaining matrix state snapshots by slicing the matrix into daily time windows can accurately capture the state characteristics of the matrix in different time intervals, avoiding the ambiguity of state information caused by large time spans. This provides basic data with appropriate granularity for subsequent analysis of the matrix's temporal evolution, ensuring that each snapshot can truly reflect the functional and environmental influences of the corresponding time period.

[0093] In summary, consistent integration of matrix state snapshots eliminates the interference of random fluctuations on the matrix state by correcting numerical differences between adjacent snapshots. This allows the historical state evolution trajectory formed by the series of snapshots to smoothly reflect the changes in the matrix, avoids deviations in subsequent evolutionary pattern analysis due to data inconsistency, and improves the reliability of the trajectory.

[0094] In summary, by analyzing the dynamic evolution pattern based on the historical state evolution trajectory and obtaining the change trend, the core change law of the relationship between function and environmental impact can be extracted from a large amount of snapshot data. This provides a clear directional basis for the adaptive adjustment of the matrix, avoids the blindness of the adjustment process, and ensures that the adjustment direction is consistent with the actual inspection state changes.

[0095] In summary, by optimizing and adjusting the matrix structure according to the changing trends, and by adding or removing cell values, or adding rows or columns, the matrix can adapt to changes in the function inspection status in real time. This breaks the limitation of traditional fixed matrices being unable to cope with dynamic changes in status and improves the matrix's adaptability to real-world scenarios.

[0096] In summary, by comparing the consistency between the inference results and the actual situation through the matrix update verification mechanism, deviations in the adjusted matrix can be detected and corrected in a timely manner, ensuring that the matrix always maintains a high level of accuracy. At the same time, by using the latest inspection data for verification, it is ensured that the matrix can reflect the current inspection status and has good timeliness, providing reliable matrix support for subsequent fault impact range analysis.

[0097] In summary, by defining clear state transition pattern types and statistically analyzing high-frequency transition features, key evolutionary features can be accurately located from massive amounts of historical state evolution trajectories. This avoids feature omissions or redundancy caused by a lack of target selection, providing focused and valuable analytical objects for subsequent framework construction and ensuring the relevance of evolutionary analysis.

[0098] In summary, the evolutionary trend analysis framework is constructed based on two dimensions: environmental factors and function. This systematically classifies scattered key evolutionary features and sets up a correlation analysis layer to lay the foundation for subsequent feature matching. This breaks the limitations of single-dimensional analysis, makes the evolutionary pattern analysis more structured, and improves the orderliness and efficiency of the analysis process.

[0099] In summary, by determining strong correlations and forming patterns of change through the frequency of simultaneous occurrence of statistical features, we can objectively uncover the inherent laws governing the relationship between environmental and functional influences, avoid biases caused by subjective assumptions about correlations, and make the summary of patterns more consistent with the actual situation reflected in the inspection data, thus providing a reliable basis for trend prediction.

[0100] In summary, combining historical numerical change rates to predict the impact relationship and obtain the trend can clarify the direction and magnitude of future changes in the relationship between function and environmental impact, avoid the ambiguity of trend prediction, provide clear target guidance for subsequent matrix adjustments, ensure that the matrix adaptive adjustment can adapt to future changes in inspection status in advance, and improve the forward-looking nature of the matrix.

[0101] S5. Integrate the results of the bidirectional impact reasoning analysis with the preset risk level assessment principles, quantitatively output the impact range and severity level of the environmental faults in the functional inspection, and generate an impact report of the environmental faults.

[0102] In this embodiment of the invention, the method of integrating the results of the bidirectional impact reasoning analysis with preset risk level assessment principles to quantitatively output the impact range and severity level of environmental faults during functional inspections, and generating an impact report of the environmental faults, includes: By integrating the results of the bidirectional influence reasoning analysis, the basic dataset of environmental faults in the functional inspection is obtained. Multi-dimensional impact relationship analysis is performed on the basic dataset to obtain the comprehensive impact profile of the environmental fault; Based on the preset risk level assessment principles, the comprehensive impact profile is mapped to the corresponding risk level; Boundary analysis is performed on the comprehensive impact profile to determine the impact area of ​​the environmental fault; An impact report on the environmental failure is generated based on the risk level and the affected area.

[0103] In this embodiment of the invention, mapping the comprehensive impact profile to a corresponding risk level based on a preset risk level assessment principle includes: By analyzing the influence relationship patterns in the comprehensive influence profile, the key influence dimensions of the environmental failure are obtained. The key impact dimensions are matched with the preset risk level assessment principles to determine the basic risk level of the environmental failure. Based on the historical evolution trajectory of the environmental fault, the risk transmission path of the basic risk level is verified to obtain the risk level of the environmental fault.

[0104] Specifically, the results of the bidirectional impact reasoning analysis are integrated. These results include a sequence list of environmental factors generated when the functional health status is abnormal and a sequence list of functional impacts generated when the environmental factors are abnormal. At the same time, basic information such as the inspection time when the fault was first detected, the number of inspection cycles during which the fault lasted, and the number of the inspection equipment involved are supplemented. All information is organized according to the structure of "basic fault information - key environmental factor information - affected functional information" to ensure that each information item has clear data support, and finally forms the basic dataset of environmental faults in functional inspection.

[0105] Furthermore, a multi-dimensional impact relationship analysis was performed on the basic dataset of environmental failures. The analysis dimensions were set as the number of affected functions, the importance level distribution of affected functions, the degree of anomaly of key environmental factors, and the duration of failure impact. The number of affected functions is the total number of affected functions. The importance level distribution is the number and proportion of core functions, key functions, and auxiliary functions. The degree of anomaly of key environmental factors is determined with reference to the deviation level. The duration of failure impact is determined based on the number of continuous inspection cycles. The results of the four-dimensional analysis are integrated into a structured expression, such as "Total number of affected functions: 8, Degree of anomaly of key environmental factors: severe deviation, Duration of failure: 4 inspection cycles". This expression is the comprehensive impact profile of the environmental failure.

[0106] Furthermore, the preset risk level assessment principle is a multi-dimensional scoring system: ≥10 affected functions receive 30 points, 5-9 receive 20 points, and <5 receive 10 points; core functions account for ≥50% of the total function population, receiving 30 points, 30%-49% receive 20 points, and <30% receive 10 points; serious deviations from key environmental factors receive 25 points, and slight deviations receive 15 points; failures lasting ≥5 inspection cycles receive 15 points, 3-4 receive 10 points, and <3 receive 5 points. The total score of 100 points corresponds to the risk level. The comprehensive impact profile data is extracted and scored, and the corresponding risk level is determined after calculating the total score. For example, the comprehensive impact profile score of 20+20+25+10=75 points corresponds to a level 2 risk.

[0107] Furthermore, boundary analysis is performed on the comprehensive impact profile. The spatial boundary is determined by the physical area to which it belongs based on the centralized inspection equipment number in the basic data, and the functional boundary is determined by the module affiliation of the affected functions. The spatial and functional boundaries are integrated into "Spatial impact area: Area 1 of Workshop A, Floor 2 of Computer Room B; Functional impact area: Data acquisition module, Data processing module". This description is the impact area of ​​the environmental failure.

[0108] Furthermore, an impact report is generated based on the risk level and affected area. The report structure includes basic information about the fault, details of the affected area, risk level and judgment basis, and response recommendations. By filling in the information in each part, a complete environmental fault impact report is finally formed.

[0109] Specifically, we first thoroughly analyze all the information in the comprehensive impact profile, including the total number of affected functions, the distribution of importance levels of affected functions, the degree of anomaly of key environmental factors, and the duration of the failure impact. Then, we analyze the impact relationship patterns between these information, such as correlation patterns like "high proportion of core functions and serious deviation of key environmental factors" and "long duration of failure and large number of affected functions". Next, based on the preset risk weight standards, we select the dimensions that play a decisive role in the risk level determination. Finally, we determine "proportion of affected core functions", "degree of anomaly of key environmental factors", and "duration of failure impact" as the key impact dimensions of environmental failures, ensuring that each key impact dimension comes from the comprehensive impact profile and is strongly correlated with the risk determination.

[0110] Furthermore, the pre-defined risk level assessment principles clearly define the score range and corresponding level threshold for each key impact dimension. For example, "30 points for ≥50% of the affected core functions, 20 points for 30%–49%, and 10 points for <30%", "25 points for serious deviation of key environmental factors, and 15 points for slight deviation", and "15 points for failure lasting ≥5 inspection cycles, 10 points for 3–4, and 5 points for <3". The total score corresponds to the risk level. The actual data of the key impact dimensions obtained from the analysis are matched with the score range in the principles. For example, "20 points for 37.5% of the affected core functions, 25 points for serious deviation of key environmental factors, and 10 points for failure lasting 4 inspection cycles". The total score is calculated, and then the basic risk level of the environmental failure is determined to be Level 3 based on the level threshold corresponding to the total score. This ensures that the determination of the basic risk level is entirely based on the matching results between the key dimensions and the pre-defined principles.

[0111] Furthermore, the historical evolution trajectory of environmental failures is retrieved, and records of similar historical failures that are consistent with the key impact dimensions of the current failure are selected. The risk transmission paths of these historical failures are analyzed, including whether they trigger secondary functional failures or cause a chain reaction of abnormal environmental factors. If the risk transmission path of the current failure is consistent with that of similar historical failures, the basic risk level is maintained. If the current failure exhibits transmission characteristics not seen in similar historical failures, 10 points are added to the basic risk level score, adjusting it to level three. By comparing and verifying the transmission path with the historical evolution trajectory, the risk level of the environmental failure is finally determined, ensuring that the level is based on both the basic matching results and the actual transmission situation.

[0112] In summary, integrating the results of bidirectional reasoning to form a basic dataset, centralizing scattered environmental and functional impact information and supplementing basic data, provides comprehensive support for subsequent analysis and avoids biases caused by information fragmentation.

[0113] In summary, multi-dimensional analysis forms a comprehensive impact profile, presenting the impact of the fault in a three-dimensional way from dimensions such as the number of functions and their importance, avoiding the one-sidedness of a single dimension, and making the impact of the fault clearer.

[0114] In summary, the pre-defined scoring system maps risk levels, transforming qualitative impacts into quantitative levels, avoiding subjective judgments, and ensuring objective and consistent risk assessment.

[0115] In summary, the structured generation of impact reports integrates analysis results with response recommendations, combining analytical and guiding aspects to help quickly develop fault handling solutions.

[0116] In summary, by analyzing the impact relationship patterns of the comprehensive impact profile and screening key impact dimensions, we can focus on the core elements that play a decisive role in risk, avoid irrelevant dimensions from interfering with risk assessment, make subsequent risk matching more targeted, and ensure that the core basis for risk level determination is clear and unambiguous.

[0117] In summary, by directly matching key impact dimensions with preset principles to calculate the total score, the qualitative impact profile is transformed into quantitative scores and levels. This eliminates the uncertainty of subjective experience-based judgments, provides a unified standard for determining basic risk levels, and improves the objectivity and consistency of risk assessment.

[0118] In summary, verifying the risk transmission path based on historical evolution trajectories can correct the basic risk level by combining the actual impact of similar failures, avoiding judgment bias caused by relying solely on current data, and ensuring that the final risk level not only matches the current failure characteristics but also refers to historical experience, thus improving the accuracy and reliability of the risk level.

[0119] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0120] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for analyzing the impact range of environmental faults driven by functional inspection, characterized in that, The method includes: S1. Collect the result data and environmental parameter data of the functional inspection, and quantify the status to obtain the functional health status and environmental factor status of the functional inspection. S2. Perform a correlation analysis on the functional health status and the environmental factor status to determine the quantitative influence relationship between the functional health status and the environmental factor status, and construct the functional and environmental influence relationship matrix of the functional inspection. S3. Based on the functional-environmental impact relationship matrix, perform bidirectional impact reasoning on the functional health status and the environmental factor status: S31. When the functional health status is abnormal, locate the key environmental factors that have the greatest impact on the functional health status according to the functional-environmental influence relationship matrix, and generate the environmental factor sequence list for the functional inspection. S32. When the environmental factor state is abnormal, deduce the functional state most affected by the environmental factor state based on the functional-environmental impact relationship matrix, and generate the functional impact sequence table for the functional inspection. S4. By using time window aggregation and weight decay, the relationship matrix between the function and the environment is adaptively adjusted as the status of the function inspection changes. S5. Integrate the results of the bidirectional impact reasoning analysis with the preset risk level assessment principles, quantitatively output the impact range and severity level of the environmental faults in the functional inspection, and generate an impact report of the environmental faults.

2. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 1, characterized in that, The system collects the results data and environmental parameter data of the functional inspection, and quantifies the status to obtain the functional health status and environmental factor status of the functional inspection, including: The result data of the functional inspection is subjected to multi-dimensional feature analysis to obtain the key performance index sequence of the functional inspection, and noise points in the key performance index sequence are removed to obtain the functional feature set of the functional inspection. The environmental parameter data is time-series aligned, and instantaneous interference is eliminated through multi-level filtering technology to obtain a stable environmental parameter sequence for the functional inspection. The indicator values ​​in the functional feature set are compared and analyzed with the preset health threshold range, and the corresponding status identifier is generated based on the comparison results to constitute the functional health status of the functional inspection. The environmental factor status of the functional inspection is generated based on the degree of deviation between the stable environmental parameter sequence and the normal operating parameters in the functional inspection.

3. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 1, characterized in that, The process of performing a correlation analysis between the functional health status and the environmental factor status to determine the quantitative impact relationship between the functional health status and the environmental factor status, and constructing the functional and environmental impact relationship matrix of the functional inspection, includes: The functional health status is divided into multi-scale time windows to obtain a subset of the functional statuses for functional inspection. The environmental factors are synchronously divided to obtain an environmental state subset that matches the functional state subset. At each time scale, a multi-dimensional feature comparison is performed between the functional state subset and the environmental state subset to obtain the collaborative change pattern of the functional inspection. Based on the aforementioned collaborative change pattern, a correlation analysis is performed on the functional health status and the environmental factor status to obtain the correspondence between the functional health status and the environmental factor status. By integrating the correspondences at different time scales, a quantitative influence relationship between the functional health state and the environmental factor state is constructed.

4. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 1, characterized in that, The process of performing a correlation analysis between the functional health status and the environmental factor status to determine the quantitative impact relationship between the functional health status and the environmental factor status, and constructing the functional and environmental impact relationship matrix of the functional inspection, includes: Based on the quantitative influence relationship, the contribution coefficient of each environmental factor state to each functional health state is obtained; Dynamic stability analysis is performed on the contribution coefficients to obtain the association confidence level between the environmental factor state and the functional health state; Based on the contribution coefficient and the correlation confidence level, a function-environment impact relationship matrix for the functional inspection is constructed, wherein the calculation formula for the function-environment impact relationship matrix is: ； in, This represents the matrix showing the relationship between the stated function and its environmental impact. This represents the contribution coefficient. This represents the coupling coefficient in the quantitative influence relationship. Indicates the current moment. Represents a time variable. This indicates the confidence level of the association. This represents the mean of the association confidence level. The standard deviation represents the confidence level of the association. This represents the preset time decay factor. This indicates the preset scale adjustment parameters. This represents the preset confidence threshold. Represents the hyperbolic tangent function. This represents the Gaussian error function.

5. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 1, characterized in that, When the functional health status is abnormal, the key environmental factors with the greatest impact on the functional health status are located according to the functional-environmental impact relationship matrix, and a sequence list of environmental factors for the functional inspection is generated, including: Multidimensional feature analysis is performed on the abnormal states in the functional health state to obtain the abnormal state feature vector of the functional health state. The abnormal state feature vector is mapped to the function and environment influence relationship matrix to obtain the initial set of associated environmental factors for the function inspection; An impact assessment is performed on the initial set of related environmental factors to determine the relative impact of the environmental factor status on the functional abnormality status during the functional inspection. Based on the relative degree of influence, the set of associated environmental factors is sorted and filtered to obtain the environmental factor sequence list for the functional inspection.

6. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 1, characterized in that, When the environmental factor state is abnormal, the functional state most affected by the environmental factor state is derived based on the function-environment impact relationship matrix, and a functional impact sequence table for the function inspection is generated, including: Based on the function-environment impact relationship matrix, identify the associated function set when the state of the environmental factor is abnormal; An impact propagation path analysis is performed on the associated function set to determine the sensitivity of each function in the function inspection to the state of the environmental factors. Based on the sensitivity level, construct a functional impact assessment system for the functional inspection to assess the potential impact range of each function; Based on the potential impact range, the affected functions of the functional inspection are dynamically sorted to obtain the functional impact sequence list of the functional inspection.

7. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 1, characterized in that, The method of using time window aggregation and weight decay to adaptively adjust the function-environment impact matrix as the function inspection status changes includes: In different time windows, the relationship matrix between the function and the environment is sliced ​​and analyzed to obtain a snapshot of the matrix status in multiple time intervals during the function inspection. The state state snapshots of the matrix are integrated to obtain the historical state evolution trajectory of the functional inspection. Dynamic evolution pattern analysis is performed on the historical state evolution trajectory to obtain the changing trend of the functional inspection; Based on the changing trend, the structure of the functional-environmental impact relationship matrix is ​​optimized and adjusted to obtain the adjusted relationship matrix of the functional inspection. The accuracy and timeliness of the adjusted relationship matrix are ensured through a matrix update verification mechanism.

8. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 7, characterized in that, The dynamic evolution pattern analysis of the historical state evolution trajectory to obtain the changing trend of the functional inspection includes: State transition pattern recognition is performed on the historical state evolution trajectory to obtain the key evolutionary features of the functional inspection; Based on the key evolutionary characteristics, an evolutionary trend analysis framework for the functional inspection is constructed. Based on the aforementioned evolution trend analysis framework, the key evolutionary features are correlated and matched to obtain the regularity pattern of the functional inspection. Based on the aforementioned pattern of change, the influence relationship between the functional health status and the environmental factor status is predicted to obtain the changing trend of the functional inspection.

9. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 1, characterized in that, The results of the combined bidirectional impact reasoning analysis and the preset risk level assessment principles are used to quantitatively output the impact range and severity level of environmental faults during functional inspections, and to generate an impact report of the environmental faults, including: By integrating the results of the bidirectional influence reasoning analysis, the basic dataset of environmental faults in the functional inspection is obtained. Multi-dimensional impact relationship analysis is performed on the basic dataset to obtain the comprehensive impact profile of the environmental fault; Based on the preset risk level assessment principles, the comprehensive impact profile is mapped to the corresponding risk level; Boundary analysis is performed on the comprehensive impact profile to determine the impact area of ​​the environmental fault; An impact report on the environmental failure is generated based on the risk level and the affected area.

10. The method for analyzing the impact range of environmental faults driven by functional inspection as described in claim 9, characterized in that, The method of mapping the comprehensive impact profile to a corresponding risk level based on the preset risk level assessment principle includes: By analyzing the influence relationship patterns in the comprehensive influence profile, the key influence dimensions of the environmental failure are obtained. The key impact dimensions are matched with the preset risk level assessment principles to determine the basic risk level of the environmental failure. Based on the historical evolution trajectory of the environmental fault, the risk transmission path of the basic risk level is verified to obtain the risk level of the environmental fault.

Images