Digital supervision and traceability management method and system for pressure special equipment

Abstract

The present application relates to the technical field of data management, in particular to a pressure-bearing special equipment digital supervision and traceability management method and system, in the present application, by collecting the surface temperature signal of the pressure-bearing container body and the pressure signal at the interface, a multi-dimensional real-time running time sequence of the special equipment is constructed, and the time sequence is mapped and compared with a standard trial running time sequence, a grid path difference matrix is further constructed, the comprehensive morphological difference information corresponding to the continuous grid path is extracted, the dynamic quantitative supervision of the running state of the pressure-bearing special equipment is realized, and the real-time running deviation is converted from the discrete monitoring value into a comparable and determinable supervision result. The present application also introduces the material thickness record in the manufacturing stage, the welding temperature record in the installation stage and the environmental humidity record in the use stage, realizes the through processing from the running abnormality identification to the life cycle reason tracing of the pressure-bearing special equipment.

Description

Technical Field

[0001] This invention relates to the field of data management technology, and in particular to a method and system for digital supervision and traceability management of pressure-bearing special equipment. Background Technology

[0002] Data management technology refers to the technical system surrounding data collection, storage, processing, analysis, and application. It involves information platform construction, data standardization, data security and sharing mechanisms, etc. It is mainly used to improve the organization and utilization of information resources in various industries and is widely used in fields such as industrial manufacturing, public safety, energy management and equipment operation and maintenance.

[0003] Among them, the digital supervision and traceability management method for pressure-bearing special equipment refers to the management method that records, supervises and tracks relevant data through information technology throughout the entire life cycle of pressure-bearing special equipment such as boilers, pressure vessels and pressure pipelines, including design, manufacturing, installation, inspection, use and maintenance. It usually relies on database management technology, information collection technology and network communication technology to uniformly manage and query equipment information, so as to achieve systematic management of equipment status and historical information.

[0004] Traditional management methods for pressure-bearing special equipment, while capable of uniformly recording and querying basic, status, and historical information of equipment such as boilers, pressure vessels, and pressure pipelines using database management, information collection, and network communication technologies, suffer from several drawbacks when dealing with continuous equipment operation, comparative analysis of trial operation status, and precise determination of abnormal conditions. These include: monitoring data remaining largely at the static ledger management level; a lack of correlation analysis based on timestamps for multi-dimensional operating parameters such as temperature and pressure; a lack of quantitative comparison paths between operating sequences and standard trial operation sequences; and difficulty in establishing a comprehensive mapping between equipment anomalies and data from the manufacturing, installation, and usage stages. Traditional traceability methods, while capable of post-event investigations based on inspection records, maintenance records, or individual test results, also suffer from problems when pressure vessels exhibit thermal drift, pressure distortion, or excessive overall morphological differences. These issues include: coarse-grained anomaly identification; regulatory results remaining at the symptom level; reliance on manual review of multiple stages of records for cause identification; and fragmented and insufficiently correlated accountability criteria. Summary of the Invention

[0005] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for digital supervision and traceability management of pressure-bearing special equipment, comprising the following steps:

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] The digital supervision and traceability management method for pressure-bearing special equipment includes the following steps:

[0008] S1: Acquire the temperature and pressure signals of the pressure vessel body, determine the temperature and pressure values ​​and corresponding timestamps, and generate a multi-dimensional real-time operating time series of special equipment.

[0009] S2: Obtain the standard trial operation time series of special equipment, compare it with the multidimensional real-time operation time series of the special equipment, and construct a grid path difference matrix;

[0010] S3: Based on the grid path difference matrix, determine the comprehensive morphological difference information of special equipment, and generate digital supervision results for pressure-bearing special equipment based on the comprehensive morphological difference information of multiple types.

[0011] S4: Obtain the cause node status parameters of pressure-bearing special equipment, and determine the causal association probability of the equipment life cycle based on the cause node status parameters and the digital supervision results of the pressure-bearing special equipment;

[0012] S5: Compare the causal association probability of the equipment lifecycle with the preset responsibility tracing judgment threshold, determine the target posterior probability inference value, obtain the cause node state parameters corresponding to the target posterior probability inference value, and the source tracing and responsibility attribution result of pressure-bearing special equipment.

[0013] As a further aspect of the present invention, the multidimensional real-time operation time series of the special equipment includes real-time temperature values, real-time pressure values, and corresponding timestamps; the grid path difference matrix includes the comparison difference of the standard trial operation time series and the comparison difference of the multidimensional real-time operation time series; the digital supervision result of the pressure-bearing special equipment includes a comprehensive morphological difference judgment item and a comprehensive morphological difference type; the causal association probability of the equipment life cycle includes the association probability of the cause node state parameter and the posterior probability inference value; and the source tracing and accountability result of the pressure-bearing special equipment includes the target posterior probability inference positioning item and the corresponding cause node state parameter.

[0014] As a further aspect of the present invention, step S1 specifically comprises:

[0015] S101: Collects large amounts of voltage signals from the surface temperature sensor of the pressure vessel body during the real-time operation cycle of pressure-bearing special equipment, inputs the voltage signals into the analog-to-digital converter for conversion, outputs the temperature value, collects the current signals from the pressure transmitter at the interface of the pressure vessel body during the same real-time operation cycle, performs numerical conversion on the current signals, outputs the pressure value, and obtains multi-dimensional operating parameters of the equipment.

[0016] S102: Based on the multi-dimensional operating parameters of the equipment, extract the timestamps corresponding to the temperature and pressure values, and pair and combine the temperature and pressure values ​​corresponding to the same occurrence time based on the timestamps to generate a parameter time sequence pairing set.

[0017] S103: Arrange and splice each pair of data in the parameter time series pairing set according to the time stamp order to establish a multi-dimensional real-time operation time series of special equipment.

[0018] As a further aspect of the present invention, step S2 specifically comprises:

[0019] S201: Collect the test temperature and test pressure values ​​during the trial operation period after the initial installation of the pressure-bearing special equipment. Arrange them in chronological order to form the standard trial operation time series of the special equipment. Combined with the multi-dimensional real-time operation time series of the special equipment, calculate the ratio between the temperature value and the pressure value and the corresponding preset temperature range reference value and preset pressure range reference value to obtain the first temperature mapping ratio and the first pressure mapping ratio. At the same time, calculate the ratio between the test temperature value and the test pressure value and the corresponding preset temperature range reference value and preset pressure range reference value to obtain the second temperature mapping ratio and the second pressure mapping ratio.

[0020] S202: Calculate the absolute value of the first characteristic deviation between the first temperature mapping ratio and the second temperature mapping ratio, and simultaneously calculate the absolute value of the second characteristic deviation between the first pressure mapping ratio and the second pressure mapping ratio. Sum the absolute values ​​of the first and second characteristic deviations to obtain the absolute value of the multidimensional comprehensive characteristic deviation.

[0021] S203: Combine the absolute values ​​of all the multidimensional comprehensive feature deviations within the same time range to construct a grid path difference matrix.

[0022] As a further aspect of the present invention, step S3 specifically comprises:

[0023] S301: Extract multiple continuous grid paths from the grid path difference matrix, calculate the sum of the absolute values ​​of all the multidimensional comprehensive feature deviations of each continuous grid path, and obtain the cumulative path distance of multiple pressure vessels;

[0024] S302: Compare the numerical values ​​of the cumulative path distances of the multiple pressure vessels, select the smallest numerical value, and set it as the comprehensive morphological difference value of the special equipment;

[0025] S303: Compare the comprehensive morphological difference value of the special equipment with the preset morphological difference safety threshold. If the comprehensive morphological difference value of the special equipment is greater than the morphological difference safety threshold, extract the corresponding comprehensive morphological difference information of the special equipment and generate the digital supervision result of the pressure-bearing special equipment.

[0026] As a further aspect of the present invention, step S4 specifically comprises:

[0027] S401: Collect big data on the material thickness records of pressure vessels during the manufacturing stage, the welding temperature records of pressure vessels during the installation stage, and the humidity records of the operating environment of special equipment during the use stage. Set these data as cause node status parameters. At the same time, set the digital supervision results of the pressure vessels as result node status parameters and merge them into a set of network model node parameters.

[0028] S402: Input the set of node parameters of the network model into the Bayesian network model, and calculate the joint distribution probability of nodes between the state parameters of the cause nodes and the state parameters of the result nodes;

[0029] S403: Based on the joint distribution probability of the nodes, deduce the posterior probability inference value between the state parameters of the cause node and the state parameters of the result node, extract all posterior probability inference values ​​in the Bayesian network model, perform parameter combination operation, and establish the causal association probability of the device life cycle.

[0030] As a further aspect of the present invention, step S5 specifically comprises:

[0031] S501: Extract multiple posterior probability inference values ​​from the causal association probability of the equipment lifecycle, calculate the ratio between the sum of the multiple posterior probability inference values ​​and the total number of posterior probability inference values, obtain the average probability distribution, and set it as the responsibility tracing determination threshold;

[0032] S502: Select posterior probability inference values ​​that are greater than the responsibility tracing determination threshold from multiple posterior probability inference values ​​to obtain the target posterior probability inference value;

[0033] S503: Address and map the matching item of the cause node state parameter corresponding to the target posterior probability inference value from the cause node state parameter, extract the cause node state parameter located at the corresponding position of the target posterior probability inference value, and generate the source tracing and accountability results for pressure-bearing special equipment.

[0034] Based on the same inventive concept, a digital supervision and traceability management system for pressure-bearing special equipment is also proposed, including:

[0035] The sequence generation module acquires the temperature and pressure signals of the pressure vessel body, determines the temperature and pressure values ​​and the corresponding timestamps, and generates a multi-dimensional real-time operating time series of special equipment.

[0036] The matrix construction module obtains the standard trial operation time series of special equipment, compares it with the multidimensional real-time operation time series of the special equipment, and constructs a grid path difference matrix.

[0037] The regulatory result generation module determines the comprehensive morphological difference information of special equipment based on the grid path difference matrix, and generates digital regulatory results for pressure-bearing special equipment based on the comprehensive morphological difference information of multiple types.

[0038] The probability determination module obtains the cause node status parameters of pressure-bearing special equipment, and determines the causal association probability of the equipment life cycle based on the cause node status parameters and the digital supervision results of the pressure-bearing special equipment.

[0039] The traceability result generation module compares the causal association probability of the equipment lifecycle with the preset responsibility tracing judgment threshold, determines the target posterior probability inference value, obtains the cause node state parameters corresponding to the target posterior probability inference value, and the traceability and responsibility attribution result of pressure-bearing special equipment.

[0040] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0041] In this invention, by collecting surface temperature signals and pressure signals at interfaces of pressure vessels, a multi-dimensional real-time operating time series of special equipment is constructed. This time series is then mapped and compared with a standard trial operation time series. Furthermore, a grid path difference matrix is ​​constructed, and comprehensive morphological difference information corresponding to continuous grid paths is extracted. This achieves dynamic quantitative monitoring of the operating status of pressure-bearing special equipment, transforming real-time operating deviations from discrete monitoring values ​​into comparable and verifiable monitoring results. This invention also introduces material thickness records from the manufacturing stage, welding temperature records from the installation stage, and environmental humidity records from the usage stage. Combined with digital monitoring results, a causal correlation probability for the equipment lifecycle is established. Further filtering of target posterior probability inference values ​​and mapping corresponding cause node state parameters enables seamless processing from identifying operational anomalies to tracing the causes throughout the lifecycle of pressure-bearing special equipment. This enhances the traceability of monitoring results, improves the targeting of accountability, and increases the utilization of data linkage throughout the entire lifecycle. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the steps of the present invention;

[0044] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0045] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0046] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0047] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0048] Figure 6 This is a detailed schematic diagram of S5 of the present invention;

[0049] Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0050] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0051] Please see Figure 1 This invention provides a method for digital supervision and traceability management of pressure-bearing special equipment, including the following steps:

[0052] S1: Acquire the temperature and pressure signals of the pressure vessel body, determine the temperature and pressure values ​​and corresponding timestamps, and generate a multi-dimensional real-time operating time series of special equipment.

[0053] S2: Obtain the standard trial operation time series of special equipment, compare it with the multidimensional real-time operation time series of special equipment, and construct a grid path difference matrix;

[0054] S3: Based on the grid path difference matrix, determine the comprehensive morphological difference information of special equipment, and generate digital supervision results for pressure-bearing special equipment based on the comprehensive morphological difference information of multiple types.

[0055] S4: Obtain the cause node status parameters of pressure-bearing special equipment, and determine the causal relationship probability of the equipment life cycle based on the cause node status parameters and the digital supervision results of pressure-bearing special equipment;

[0056] S5: Compare the causal correlation probability of the equipment life cycle with the preset responsibility tracing judgment threshold, determine the target posterior probability inference value, obtain the cause node state parameters corresponding to the target posterior probability inference value, and the source tracing and responsibility attribution results for pressure-bearing special equipment.

[0057] The multidimensional real-time operation time series of special equipment includes real-time temperature values, real-time pressure values, and corresponding timestamps. The grid path difference matrix includes the comparison difference between the standard trial operation time series and the comparison difference between the multidimensional real-time operation time series. The digital supervision results of pressure-bearing special equipment include comprehensive morphological difference judgment items and comprehensive morphological difference types. The causal association probability of the equipment life cycle includes the association probability of the cause node state parameters and the posterior probability inferred value. The source tracing and accountability results of pressure-bearing special equipment include the target posterior probability inferred location items and the corresponding cause node state parameters.

[0058] Please see Figure 2 Step S1 is as follows:

[0059] S101: Collects large amounts of voltage signals from the surface temperature sensor of the pressure vessel body during the real-time operation cycle of pressure-bearing special equipment, inputs the voltage signals into the analog-to-digital converter for conversion, outputs the temperature value, collects the current signals from the pressure transmitter at the interface of the pressure vessel body during the same real-time operation cycle, performs numerical conversion on the current signals, outputs the pressure value, and obtains multi-dimensional operating parameters of the equipment.

[0060] First, connect the output of the temperature sensor installed on the outer surface of the pressure vessel to the voltage input channel of the acquisition board. Then, connect the output of the pressure transmitter installed at the interface of the pressure vessel to the current input channel of the acquisition board. Next, trigger synchronous acquisition at the start of the same operating cycle, continuously read the voltage and current signals at each sampling time according to the preset sampling frequency, and write the equipment number, measurement point number, sampling time and signal type into each original record. For the temperature channel, first call the zero-point voltage, full-scale voltage, zero-point temperature, and full-scale temperature from the sensor calibration table, and perform a conversion for each voltage sample value: Temperature value = Zero-point temperature + (Voltage sample value - Zero-point voltage) / (Full-scale voltage - Zero-point voltage) × (Full-scale temperature - Zero-point temperature); For the pressure channel, first call the 4 mA corresponding to the lower pressure limit and the 20 mA corresponding to the upper pressure limit from the pressure transmitter range table, and perform a conversion for each current sample value: Pressure value = Lower pressure limit + (Current sample value - 4 mA) / (20 mA - 4 mA) × (Upper pressure limit - Lower pressure limit). When the voltage signal is below the lower limit voltage of the temperature sensor calibration, it is recorded as a low-side out-of-bounds record. When it is between the lower and upper limits of the calibration voltage, it is converted according to the above formula. When it is above the upper limit voltage, it is recorded as a high-side out-of-bounds record. When the current signal is below 4 mA, it is recorded as a low-side abnormal record. When it is between 4 mA and 20 mA, it is converted according to the above formula. When it is above 20 mA, it is recorded as a high-side abnormal record. Both temperature and pressure values ​​retain the corresponding sampling time and are written into the operation data table as multi-dimensional operating parameters of the equipment.

[0061] S102: Based on the multi-dimensional operating parameters of the equipment, extract the timestamps corresponding to the temperature and pressure values, and pair and combine the temperature and pressure values ​​corresponding to the same occurrence time based on the timestamps to generate a parameter time sequence pairing set.

[0062] For temperature and pressure records already formed within the same operating cycle, the sampling time series is extracted separately. First, the records are sorted in ascending order by time. Then, the timestamps of each temperature record are read one by one. The pressure records are then searched for all timestamps within the same time period, and the time difference is calculated: Time difference = Pressure record timestamp - Temperature record timestamp; Absolute time deviation = |Pressure record timestamp - Temperature record timestamp|. When the absolute time deviation is 0, the temperature value is directly paired with the pressure value. When the absolute time deviation is greater than 0 and does not exceed the preset pairing tolerance, the pressure record with the smallest absolute time deviation that meets the tolerance condition is selected as the match. When two or more pressure records have the same absolute time deviation and are simultaneously the minimum, the record with a timestamp earlier than the others is selected. When the absolute time deviation exceeds the preset pairing tolerance, the temperature record is recorded as an unpaired record and not written into the parameter time series pairing set. After matching is completed, the temperature value, pressure value, pairing time identifier, equipment number, temperature measurement point number, and pressure measurement point number are written into a pairing record, and then written into the parameter time series pairing set. The pairing tolerance is set to 1 to 2 times the sampling period. Within the sampling period, the sensor output belongs to the same discrete sampling step. A time difference of 0 indicates sampling at the same moment. A time difference near the sampling period indicates a near-synchronous record formed within the same sampling step. When the time difference exceeds 2 times the sampling period, it has crossed at least two sampling steps and cannot be used as a parameter for the same moment.

[0063] S103: Arrange and splice each pair of data in the parameter time series pairing set according to the time stamp order to establish a multi-dimensional real-time operation time series of special equipment;

[0064] After extracting the time identifier column from all paired records in the parameter time series pairing set, they are sorted in ascending order by time. The first paired record after sorting is read as the starting record of the sequence, and then the second to the last paired records are read in sequence. Each subsequent record is appended to the previous record, forming a continuously unfolded multidimensional real-time operation time series of special equipment. During the appending process, the timestamp of the current record is compared with the timestamp of the previous record. If the timestamp of the current record is greater than the timestamp of the previous record, the current record is directly retained and written to the next sequence position. If the timestamp of the current record is equal to the timestamp of the previous record, the original sampling and writing order of the two records is compared. The record written to the operation data table first is retained first, and the record written to the operation data table later is retained later. If the timestamp of the current record is less than the timestamp of the previous record, the current record is returned to the sorting sequence and repositioned before being written. Each sequence position is fixed with one set of temperature and pressure values ​​written, without adding other parameters or changing the original paired content.

[0065] Please see Figure 3 Step S2 is as follows:

[0066] S201: Collect the test temperature and test pressure values ​​during the trial operation period after the initial installation of pressure-bearing special equipment. Arrange them in chronological order to form the standard trial operation time series of special equipment. Combined with the multi-dimensional real-time operation time series of special equipment, calculate the ratio between the temperature and pressure values ​​and the corresponding preset temperature range reference value and preset pressure range reference value to obtain the first temperature mapping ratio and the first pressure mapping ratio. At the same time, calculate the ratio between the test temperature and test pressure values ​​and the corresponding preset temperature range reference value and preset pressure range reference value to obtain the second temperature mapping ratio and the second pressure mapping ratio.

[0067] During the trial operation phase after the initial installation of pressure-bearing special equipment, the same temperature sensors and pressure transmitters as during the regular operation phase are used. Test temperature and pressure values ​​are collected at the same sampling frequency throughout the trial operation period and arranged in ascending order of sampling time to form a standard trial operation time series. Subsequently, data from the same time position in both the standard trial operation time series and the multidimensional real-time operation time series are read, and the real-time temperature value, real-time pressure value, test temperature value, test pressure value, preset temperature range reference value, and preset pressure range reference value are retrieved sequentially. The preset temperature range reference value is the difference between the maximum and minimum values ​​of all test temperature values ​​during the trial operation phase of the equipment; the preset temperature range reference value = maximum test temperature value - minimum test temperature value. Similarly, the preset pressure range reference value is the difference between the maximum and minimum values ​​of all test pressure values ​​during the trial operation phase of the equipment; the preset pressure range reference value = maximum test pressure value - minimum test pressure value. Subsequently, mapping ratio conversions are performed separately: First temperature mapping ratio = real-time temperature value / preset temperature range reference value; First pressure mapping ratio = real-time pressure value / preset pressure range reference value; Second temperature mapping ratio = test temperature value / preset temperature range reference value; Second pressure mapping ratio = test pressure value / preset pressure range reference value. When the preset temperature range reference value is greater than 0, temperature mapping ratio conversion is performed; when the preset temperature range reference value is equal to 0, it indicates that there is no fluctuation in temperature records during the trial operation phase, and the temperature mapping ratio at the current time position is recorded as an invalid record. Similarly, when the preset pressure range reference value is greater than 0, pressure mapping ratio conversion is performed; when the preset pressure range reference value is equal to 0, it indicates that there is no fluctuation in pressure records during the trial operation phase, and the pressure mapping ratio at the current time position is recorded as an invalid record. Here, the denominators of all ratios represent the range of the same physical quantity, and the numerators represent the same type of measured value at the corresponding time, ensuring no mismatch in dimensions.

[0068] S202: Calculate the absolute value of the first characteristic deviation between the first temperature mapping ratio and the second temperature mapping ratio, and simultaneously calculate the absolute value of the second characteristic deviation between the first pressure mapping ratio and the second pressure mapping ratio. Sum the absolute values ​​of the first and second characteristic deviations to obtain the absolute value of the multidimensional comprehensive characteristic deviation.

[0069] For each time point, simultaneously read the first temperature mapping ratio and the second temperature mapping ratio, and calculate the absolute value of the first characteristic deviation: First characteristic deviation absolute value = |first temperature mapping ratio - second temperature mapping ratio|. Then read the first pressure mapping ratio and the second pressure mapping ratio, and calculate the absolute value of the second characteristic deviation: Second characteristic deviation absolute value = |first pressure mapping ratio - second pressure mapping ratio|. Subsequently, sum the absolute values ​​of the first and second characteristic deviations at the same time point to obtain the absolute value of the multidimensional comprehensive characteristic deviation: Multidimensional comprehensive characteristic deviation absolute value = First characteristic deviation absolute value + Second characteristic deviation absolute value. Before calculation, check whether all four mapping ratios are valid records. If all four mapping ratios are valid, write them into the difference table. If any temperature mapping ratio is invalid, the absolute value of the first characteristic deviation at that time point is recorded as invalid; if any pressure mapping ratio is invalid, the absolute value of the second characteristic deviation at that time point is recorded as invalid; if any deviation absolute value is invalid, the absolute value of the multidimensional comprehensive characteristic deviation at that time point is recorded as invalid. When the absolute value of the first characteristic deviation is equal to 0, it indicates that the real-time temperature mapping ratio at that time location is consistent with the trial operation temperature mapping ratio; when it is greater than 0, it indicates that there is a temperature deviation at that time location. The absolute values ​​of the second characteristic deviation and the multi-dimensional comprehensive characteristic deviation are interpreted in the same way: a value equal to 0 indicates that there is no difference in the corresponding ratio, and a value greater than 0 indicates that there is a difference in the corresponding ratio.

[0070] S203: Combine the absolute values ​​of all multidimensional comprehensive feature deviations within the same time range to construct a grid path difference matrix;

[0071] A two-dimensional table is created by constructing all valid absolute values ​​of multidimensional comprehensive characteristic deviations within the same time range, oriented along both the standard trial run time series and the multidimensional real-time run time series. First, the position number of the standard trial run time series is used as the row index, and then the position number of the multidimensional real-time run time series is used as the column index. Subsequently, the absolute value of the multidimensional comprehensive characteristic deviation corresponding to each row and column position is read sequentially and written to the corresponding grid, forming a grid path difference matrix. When writing the matrix, the standard trial run record corresponding to a certain row index is called first, followed by the real-time run record corresponding to a certain column index. The comprehensive deviation value of the corresponding time positions is written to the current grid. If a valid comprehensive deviation value exists at a position, it is marked as a passable grid; if a position cannot be written due to invalid temperature mapping ratio, invalid pressure mapping ratio, or an empty comprehensive deviation value, it is marked as an inaccessible grid. The first row and first column of the matrix correspond to the deviation value between the trial run start position and the real-time run start position, and the last row and last column correspond to the deviation value between the trial run end position and the real-time run end position. The intermediate grids are filled continuously in row and column order without adding other data items.

[0072] Please see Figure 4 Step S3 is as follows:

[0073] S301: Extract multiple continuous grid paths from the grid path difference matrix, calculate the sum of the absolute values ​​of all multidimensional comprehensive feature deviations for each continuous grid path, and obtain the cumulative path distance of multiple pressure vessels;

[0074] When extracting continuous grid paths from the grid path difference matrix, the first row and first column are set as the starting grid, and the last row and last column are set as the ending grid. Then, adjacent passable grids are read grid by grid starting from the starting grid. Each step only allows movement from the current grid to the adjacent grid on the right, the adjacent grid below, or the adjacent grid on the lower right; leftward, upward, and skipping grids are not allowed. After reading a path that can continuously reach the ending grid from the starting grid, all grid coordinates on the path are recorded in the path order. Then, the absolute value of the multidimensional comprehensive feature deviation in each grid is extracted and accumulated. The accumulated path distance is equal to the sum of the absolute values ​​of the multidimensional comprehensive feature deviations of all grids on the path. A path is considered valid if all grids in the path are passable and the path finally reaches the ending grid; an invalid path is considered invalid if there are no subsequent passable grids in the allowed direction at any transition position in the path and the path has not yet reached the ending grid; a path is not established if the starting point is an impassable grid; a path is not established if the ending point is an impassable grid. When the cumulative path distance is equal to 0, it means that all grid deviation values ​​on the path are 0; when the cumulative path distance is greater than 0, it means that there is a cumulative deviation value on the path.

[0075] S302: Compare the numerical values ​​of the cumulative path distances of multiple pressure vessels, select the smallest numerical value, and set it as the comprehensive morphological difference value of the special equipment.

[0076] For all valid paths, a path distance comparison is performed one by one. First, a table mapping path numbers to cumulative path distances is established. Then, the cumulative path distance of the first valid path is read as the current minimum value, and its path number is recorded. Next, the cumulative path distances of the second to the last valid paths are read sequentially and compared with the current minimum value. If the newly read distance value is less than the current minimum value, the current minimum value is replaced with the newly read distance value, and the corresponding path number is replaced with the current optimal path number. If the newly read distance value is equal to the current minimum value, both path numbers are retained, and the total number of steps of the two paths is compared. The path with fewer steps is retained as the priority path. If the total number of steps is still the same, the time span from the start to the end point of the two paths is compared, and the path with the shorter time span is retained as the priority path. If the time span is still the same, both path numbers are retained. If the newly read distance value is greater than the current minimum value, the current minimum value and the current optimal path number remain unchanged. After all comparisons are completed, the minimum value parameter is recorded as the special equipment comprehensive morphological difference value, and the path number corresponding to this value is written into the results table. The value selected here is the one with the smallest cumulative comprehensive deviation along the path. There is no situation where different physical quantities are directly multiplied or spliced ​​together to obtain a result without a physical basis.

[0077] S303: Compare the comprehensive form difference value of special equipment with the preset form difference safety threshold. If the comprehensive form difference value of special equipment is greater than the form difference safety threshold, extract the corresponding comprehensive form difference information of special equipment and generate digital supervision results for pressure-bearing special equipment.

[0078] The comprehensive morphological difference information of special equipment includes the temperature and thermal drift difference parameter mapped by the continuous grid path that generates the minimum numerical parameter, the medium pressure distortion difference parameter mapped by the continuous grid path that generates the minimum numerical parameter, and the numerical out-of-bounds offset parameter corresponding to the portion of the comprehensive morphological difference value of special equipment that is greater than the morphological difference safety threshold.

[0079] After reading the comprehensive morphological difference value and the preset morphological difference safety threshold of the special equipment, first extract all grid coordinates corresponding to the continuous grid path that generates the minimum numerical parameter, and then read the temperature deviation record and pressure deviation record on the path grid by grid. The temperature thermal drift difference parameter is taken as the sequence value of the absolute value of the first characteristic deviation of each grid on the optimal path, the medium pressure distortion difference parameter is taken as the sequence value of the absolute value of the second characteristic deviation of each grid on the optimal path, and the numerical boundary offset amplitude parameter is taken as the difference between the comprehensive morphological difference value and the morphological difference safety threshold. The numerical boundary offset amplitude parameter = comprehensive morphological difference value of special equipment - morphological difference safety threshold. The morphological difference safety threshold is set as the maximum allowable value among the cumulative values ​​of the corresponding paths during the trial operation and verification operation phases. Before the equipment is put into use, select multiple sets of qualified trial operation and verification operation data, establish a set of cumulative path values ​​according to the same steps mentioned above, and then take the maximum value in the set as the morphological difference safety threshold; this value corresponds to the maximum allowable cumulative path deviation under the confirmed qualified state. When the overall morphological difference value exceeds the morphological difference safety threshold, it is determined to be in an over-threshold state, and the temperature thermodynamic drift difference parameter, the medium pressure distortion difference parameter, and the positive offset record are output. When the overall morphological difference value is equal to the morphological difference safety threshold, it is determined to be in a critical state, and the temperature thermodynamic drift difference parameter, the medium pressure distortion difference parameter, and the zero offset record are output. When the overall morphological difference value is less than the morphological difference safety threshold, it is determined to be in a state within the threshold, and the temperature thermodynamic drift difference parameter, the medium pressure distortion difference parameter, and the negative offset record are output.

[0080] Please see Figure 5 Step S4 is as follows:

[0081] S401: Collect big data on the material thickness of pressure vessels during the manufacturing stage, the welding temperature of pressure vessels during the installation stage, and the humidity of the operating environment of special equipment during the use stage. Set these data as cause node status parameters. At the same time, set the digital supervision results of pressure-bearing special equipment as result node status parameters and merge them into a set of network model node parameters.

[0082] The system reads the material thickness records of the pressure vessel during the manufacturing stage, the welding temperature records during the installation stage, and the ambient humidity records during the usage stage. Material thickness records are obtained point-by-point from the shell plates, head plates, and adjacent weld locations using a thickness measuring instrument during the manufacturing stage. Each record includes the thickness value, component location number, and measurement time. Welding temperature records are continuously measured by a welding temperature measuring device during the installation stage at each weld. Each record includes the welding temperature value, weld location number, and recording time. Ambient humidity records are continuously measured by a humidity sensor installed in the equipment's operating area during the usage stage. Each record includes the ambient humidity value, area location number, and acquisition time. Subsequently, the system reads the special equipment comprehensive morphological difference value, temperature and thermodynamic drift difference parameter, medium pressure distortion difference parameter, and numerical out-of-bounds offset amplitude parameter from the aforementioned digital monitoring results. The material thickness records, welding temperature records, and ambient humidity records are recorded as cause node state parameters, and the digital monitoring results are recorded as result node state parameters. These are then written into the network model node parameter set in a unified field order. Before writing, time alignment and state discretization are performed: The material thickness state range is determined based on the design thickness value and the manufacturing allowable deviation. A thickness value less than the lower limit of the design thickness is recorded as a low thickness state; a thickness value between the lower and upper limits of the design thickness is recorded as a normal thickness state; and a thickness value above the upper limit of the design thickness is recorded as a high thickness state. The welding temperature state range is determined based on the permissible lower and upper limits of the welding process document. A temperature below the lower limit is recorded as a low temperature state; a temperature between the lower and upper limits is recorded as a normal temperature state; and a temperature above the upper limit is recorded as a high temperature state. The environmental humidity state range is determined based on the long-term stable range in the equipment operating environment record. A humidity below the lower limit is recorded as a low humidity state; a humidity between the lower and upper limits is recorded as a normal humidity state; and a humidity above the upper limit is recorded as a high humidity state. Below the lower limit indicates the value falls below the allowable range; between the upper and lower limits indicates the value falls within the allowable range; and above the upper limit indicates the value falls above the allowable range.

[0083] S402: Input the set of node parameters of the network model into the Bayesian network model, and calculate the joint distribution probability of nodes between the state parameters of the cause nodes and the state parameters of the result nodes;

[0084] The Bayesian network model is constructed based on the network directed graph node parameters consisting of the state parameters of cause nodes and the state parameters of result nodes in the network model node parameter set, the network directed graph associated edge structure parameters pointing from the state parameters of cause nodes to the state parameters of result nodes, and the set node initial conditional probability distribution parameters.

[0085] After expanding the material thickness state, welding temperature state, environmental humidity state, and digital monitoring result state in the network model node parameter set into sample records, a node state count table is first established, followed by a co-occurrence count table of cause node states and result node states. Then, the sample records are read line by line, and the frequency of occurrence of each material thickness state, welding temperature state, and environmental humidity state is counted. The frequency of co-occurrence of each cause node state and each result node state in the same sample record is also counted. The joint distribution probability of nodes is obtained using a frequency conversion method: Joint distribution probability = Number of times a cause node state and a result node state co-occur / Total number of samples; Conditional distribution value of a result node state under a given cause state = Number of times a cause node state and a result node state co-occur / Number of times the cause node state occurs. When the total number of samples is greater than 0, the joint distribution probability conversion is performed; when the total number of samples is equal to 0, all joint distribution probabilities are considered invalid. When the frequency of occurrence of a cause node state is greater than 0, the conditional distribution value conversion is performed; when the frequency of occurrence of a cause node state is equal to 0, all conditional distribution values ​​corresponding to that cause state are considered invalid. When the number of times a certain combination of states occurs simultaneously is 0, the joint probability distribution is recorded as 0; when it is greater than 0, the corresponding probability value is written. Here, the probability value is formed by the ratio of the number of simultaneous occurrences to the total number of samples and the number of causal states. The numerator and denominator of each term come from the same batch of sample statistics, and there is no unfounded conversion.

[0086] S403: Based on the joint distribution probability of nodes, deduce the posterior probability inference value between the state parameters of the cause node and the state parameters of the result node, extract all posterior probability inference values ​​in the Bayesian network model, perform parameter combination operation, and establish the causal association probability of the device life cycle.

[0087] When further inferring the posterior probability inference values ​​between causal node state parameters and result node state parameters based on the joint distribution probability of nodes, a specific result node state is first fixed. Then, the joint distribution probability of each causal node state related to that result node state is read one by one, and the sum of the joint distribution probabilities of all causal node states corresponding to that result node state is summarized. Subsequently, a posterior probability conversion is performed: the posterior probability inference value = the joint distribution probability of a causal node state and the result node state / the sum of the joint distribution probabilities of all causal node states corresponding to that result node state. If the sum of the joint distribution probabilities of all causal node states corresponding to the result node state is greater than 0, the posterior probability conversion is performed; if the sum equals 0, all posterior probability inference values ​​corresponding to that result node state are recorded as invalid. After calculation, all posterior probability inference values ​​for material thickness state—result state, welding temperature state—result state, and environmental humidity state—result state are written into the equipment lifecycle causal association probability table in field order. This uses a step-by-step normalization process on the statistically calculated joint distribution results. The numerator is the joint probability of a certain cause state and a certain result state, and the denominator is the sum of the joint probabilities of all cause states under the same result state. The statistical caliber of the numerator and denominator is consistent.

[0088] Please see Figure 6 Step S5 is as follows:

[0089] S501: Extract multiple posterior probability inference values ​​from the causal association probability of the equipment life cycle, calculate the ratio between the sum of multiple posterior probability inference values ​​and the total number of posterior probability inference values, obtain the average probability distribution, and set it as the responsibility tracing judgment threshold;

[0090] After extracting all valid posterior probability inferences from the causal association probability table of the equipment lifecycle, the total number of posterior probability inferences is first counted. Then, all posterior probability inferences are summed item by item to obtain a probability sum. Finally, the average probability distribution is obtained by comparing the probability sum with the total number of posterior probability inferences, and this average probability distribution is set as the responsibility tracing threshold. The responsibility tracing threshold = sum of all valid posterior probability inferences / total number of valid posterior probability inferences. When the total number of posterior probability inferences is greater than 0, it is written into the responsibility tracing threshold; when the total number of posterior probability inferences is equal to 0, the responsibility tracing threshold is marked as invalid, and subsequent filtering is not performed. This threshold is set based on the average level of all valid posterior probabilities in the current equipment lifecycle sample. The value is at the center of all counted posterior probabilities, used to distinguish causal state records that are higher than the overall average level and those that are not higher than the overall average level, without introducing any other unrecorded parameters. When the responsibility tracing threshold is equal to 0, it means that all valid posterior probabilities are 0; when it is greater than 0, it means that there is at least one valid probability association between a causal state and an outcome state.

[0091] S502: Select posterior probability inference values ​​that are greater than the responsibility tracing judgment threshold from multiple posterior probability inference values, and obtain the target posterior probability inference value;

[0092] After reading all posterior probability inferred values ​​and responsibility tracing thresholds, comparisons are performed item by item, and the cause status number and result status number corresponding to each posterior probability inferred value are recorded synchronously. During comparison, the probability difference is calculated first: Probability difference = Posterior probability inferred value - Responsibility tracing threshold. When the probability difference is greater than 0, the posterior probability inferred value and its corresponding cause status number and result status number are written to the target posterior probability table; when the probability difference is equal to 0, the posterior probability inferred value and its corresponding number are written to the critical probability table; when the probability difference is less than 0, the posterior probability inferred value and its corresponding number are written to the non-target probability table. When the responsibility tracing threshold is invalid, no filtering is performed, and all posterior probability inferred values ​​are recorded as undetermined records. After all comparisons are completed, only the records in the target posterior probability table are retained as subsequent cause node addressing objects; the critical probability table and non-target probability tables are not included in the subsequent addressing process.

[0093] S503: Address and map the matching item of the cause node state parameter corresponding to the target posterior probability inference value from the cause node state parameter, extract the cause node state parameter located at the corresponding position of the target posterior probability inference value, and generate the source tracing and responsibility results for pressure-bearing special equipment;

[0094] When performing cause node state parameter addressing for each target posterior probability inference value in the target posterior probability table, the cause state number, result state number, and posterior probability value corresponding to that record are first read. Then, the corresponding material thickness state item, welding temperature state item, or environmental humidity state item is located in the cause node state parameter table according to the number. The located cause state item is then mapped one-to-one with the result state item in the current record to form the traceability and accountability results for pressure-bearing special equipment. When a target posterior probability inference value corresponds to only one cause node state parameter matching item, the cause state item is directly output. When a target posterior probability inference value corresponds to two or more cause node state parameter matching items, the posterior probability values ​​are compared first, and the record with the larger posterior probability value is ranked first. When the posterior probability values ​​are equal, they are all output after being arranged in the field order of material thickness state, welding temperature state, and environmental humidity state. When there is no target posterior probability inference value greater than the responsibility tracing judgment threshold, no target cause state item record is output. The attribution results retain the name of the cause node, the state range of the cause node, the corresponding result state, the posterior probability value, and the sorting position. Each attribution record is written into the attribution result table according to the equipment number, result state number, and time identifier, thus completing the unified summary of records related to digital supervision, source tracing and attribution, and big data management of pressure-bearing special equipment.

[0095] Please see Figure 7The digital supervision and traceability management system for pressure-bearing special equipment includes:

[0096] The sequence generation module acquires the temperature and pressure signals of the pressure vessel body, determines the temperature and pressure values ​​and the corresponding timestamps, and generates a multi-dimensional real-time operating time series of special equipment.

[0097] The matrix construction module obtains the standard trial operation time series of special equipment, compares it with the multidimensional real-time operation time series of special equipment, and constructs a grid path difference matrix.

[0098] The regulatory result generation module determines the comprehensive morphological difference information of special equipment based on the grid path difference matrix, and generates digital regulatory results for pressure-bearing special equipment based on the comprehensive morphological difference information of multiple types.

[0099] The probability determination module obtains the causal node status parameters of pressure-bearing special equipment, and determines the causal association probability of the equipment life cycle based on the causal node status parameters and the digital supervision results of pressure-bearing special equipment.

[0100] The traceability result generation module compares the causal association probability of the equipment lifecycle with the preset responsibility tracing judgment threshold, determines the target posterior probability inference value, obtains the cause node status parameters corresponding to the target posterior probability inference value, and generates the traceability and responsibility attribution results for pressure-bearing special equipment.

[0101] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for digital supervision and traceability management of pressure-bearing special equipment, characterized in that, Includes the following steps: S1: Acquire the temperature and pressure signals of the pressure vessel body, determine the temperature and pressure values ​​and corresponding timestamps, and generate a multi-dimensional real-time operating time series of special equipment. S2: Collect the test temperature and pressure values ​​during the trial operation period after the initial installation of pressure-bearing special equipment. Arrange them in chronological order to form a standard trial operation time series for special equipment. Combined with the multi-dimensional real-time operation time series of the special equipment, calculate the ratios between the temperature and pressure values ​​and the corresponding preset temperature and pressure range reference values ​​to obtain the first temperature mapping ratio and the first pressure mapping ratio. Simultaneously, calculate the ratios between the test temperature and pressure values ​​and the corresponding preset temperature and pressure range reference values ​​to obtain the second temperature mapping ratio and the second pressure mapping ratio. Calculate the absolute value of the first characteristic deviation between the first and second temperature mapping ratios, and simultaneously calculate the absolute value of the second characteristic deviation between the first and second pressure mapping ratios. Summate the absolute values ​​of the first and second characteristic deviations to obtain the absolute value of the multi-dimensional comprehensive characteristic deviation. Combine all the absolute values ​​of the multi-dimensional comprehensive characteristic deviations within the same time range to construct a grid path difference matrix. S3: Extract multiple continuous grid paths from the grid path difference matrix, calculate the sum of the absolute values ​​of all multidimensional comprehensive feature deviations for each continuous grid path, and obtain the cumulative path distances of multiple pressure vessels; compare the numerical parameters among the cumulative path distances of multiple pressure vessels, select the smallest numerical parameter, and set it as the comprehensive morphological difference value of special equipment; compare the comprehensive morphological difference value of special equipment with a preset morphological difference safety threshold, and if the comprehensive morphological difference value of special equipment is greater than the morphological difference safety threshold, extract the corresponding comprehensive morphological difference information of special equipment, and generate digital supervision results for pressure-bearing special equipment. The comprehensive morphological difference information of special equipment includes the temperature and thermal drift difference parameter mapped by the continuous grid path that generates the smallest numerical parameter, the medium pressure distortion difference parameter mapped by the continuous grid path that generates the smallest numerical parameter, and the numerical out-of-bounds offset amplitude parameter corresponding to the portion of the comprehensive morphological difference value of special equipment that is greater than the morphological difference safety threshold. S4: Collect big data on the material thickness of the pressure vessel during the manufacturing stage, the welding temperature of the pressure vessel during the installation stage, and the humidity of the operating environment of the special equipment during the use stage. Set these data as causal node status parameters. Based on the causal node status parameters and the digital supervision results of the pressure-bearing special equipment, determine the probability of causal association in the equipment life cycle. S5: Compare the causal association probability of the equipment lifecycle with the preset responsibility tracing judgment threshold, determine the target posterior probability inference value, obtain the cause node state parameters corresponding to the target posterior probability inference value, and the source tracing and responsibility attribution result of pressure-bearing special equipment.

2. The method for digital supervision and traceability management of pressure-bearing special equipment according to claim 1, characterized in that, The multidimensional real-time operating time series of the special equipment includes real-time temperature values, real-time pressure values, and corresponding timestamps. The digital supervision results of the pressure-bearing special equipment include comprehensive morphological difference judgment items and comprehensive morphological difference types. The causal association probability of the equipment life cycle includes the association probability of the cause node state parameter and the posterior probability inference value. The source tracing and accountability results of the pressure-bearing special equipment include the target posterior probability inference positioning item and the corresponding cause node state parameter.

3. The method for digital supervision and traceability management of pressure-bearing special equipment according to claim 1, characterized in that, Step S1 is as follows: S101: Collects large amounts of voltage signals from the surface temperature sensor of the pressure vessel body during the real-time operation cycle of pressure-bearing special equipment, inputs the voltage signals into the analog-to-digital converter for conversion, outputs the temperature value, collects the current signals from the pressure transmitter at the interface of the pressure vessel body during the same real-time operation cycle, performs numerical conversion on the current signals, outputs the pressure value, and obtains multi-dimensional operating parameters of the equipment. S102: Based on the multi-dimensional operating parameters of the equipment, extract the timestamps corresponding to the temperature and pressure values, and pair and combine the temperature and pressure values ​​corresponding to the same occurrence time based on the timestamps to generate a parameter time sequence pairing set. S103: Arrange and splice each pair of data in the parameter time series pairing set according to the time stamp order to establish a multi-dimensional real-time operation time series of special equipment.

4. The method for digital supervision and traceability management of pressure-bearing special equipment according to claim 1, characterized in that, Step S4 is as follows: S401: Set the digital supervision result of the pressure-bearing special equipment as the result node status parameter, and merge the cause node status parameter and the result node status parameter into a network model node parameter set; S402: Input the set of node parameters of the network model into the Bayesian network model, and calculate the joint distribution probability of nodes between the state parameters of the cause nodes and the state parameters of the result nodes; S403: Based on the joint distribution probability of the nodes, deduce the posterior probability inference value between the state parameters of the cause node and the state parameters of the result node, extract all posterior probability inference values ​​in the Bayesian network model, perform parameter combination operation, and establish the causal association probability of the device life cycle.

5. The method for digital supervision and traceability management of pressure-bearing special equipment according to claim 1, characterized in that, Step S5 is as follows: S501: Extract multiple posterior probability inference values ​​from the causal association probability of the equipment lifecycle, calculate the ratio between the sum of the multiple posterior probability inference values ​​and the total number of posterior probability inference values, obtain the average probability distribution, and set it as the responsibility tracing determination threshold; S502: Select posterior probability inference values ​​that are greater than the responsibility tracing determination threshold from multiple posterior probability inference values ​​to obtain the target posterior probability inference value; S503: Address and map the matching item of the cause node state parameter corresponding to the target posterior probability inference value from the cause node state parameter, extract the cause node state parameter located at the corresponding position of the target posterior probability inference value, and generate the source tracing and accountability results for pressure-bearing special equipment.

6. The method for digital supervision and traceability management of pressure-bearing special equipment according to claim 4, characterized in that, The Bayesian network model is constructed based on the network directed graph node parameters consisting of the state parameters of the cause nodes and the state parameters of the result nodes in the network model node parameter set, the network directed graph associated edge structure parameters pointing from the state parameters of the cause nodes to the state parameters of the result nodes, and the set initial conditional probability distribution parameters of the nodes.

7. A digital supervision and traceability management system for pressure-bearing special equipment, characterized in that: The method for implementing the digital supervision and traceability management of pressure-bearing special equipment as described in any one of claims 1-6 includes: The sequence generation module acquires the temperature and pressure signals of the pressure vessel body, determines the temperature and pressure values ​​and the corresponding timestamps, and generates a multi-dimensional real-time operating time series of special equipment. The matrix construction module obtains the standard trial operation time series of special equipment, compares it with the multidimensional real-time operation time series of the special equipment, and constructs a grid path difference matrix. The regulatory result generation module determines the comprehensive morphological difference information of special equipment based on the grid path difference matrix, and generates digital regulatory results for pressure-bearing special equipment based on the comprehensive morphological difference information of multiple types. The probability determination module obtains the cause node status parameters of pressure-bearing special equipment, and determines the causal association probability of the equipment life cycle based on the cause node status parameters and the digital supervision results of the pressure-bearing special equipment. The traceability result generation module compares the causal association probability of the equipment lifecycle with the preset responsibility tracing judgment threshold, determines the target posterior probability inference value, obtains the cause node state parameters corresponding to the target posterior probability inference value, and the traceability and responsibility attribution result of pressure-bearing special equipment.

Images