Pressure vessel data quality evaluation method and system based on big data

Abstract

The invention relates to the technical field of data processing, and discloses a pressure vessel data quality evaluation method and system based on big data. The method comprises the following steps: generating a mapping data set through physical constraint preprocessing and working condition synchronous processing according to internal pressure, wall temperature and stress strain data of a pressure vessel; and calculating physical constraints, working condition stability and data coordination indexes to obtain a feature set. And analyzing the coupling relationship by using a deep neural network driven by physical knowledge to obtain anomaly type discrimination. And constructing a scoring matrix and dynamically calibrating to obtain an evaluation factor, and finally obtaining an equipment data quality evaluation result through multi-scale fuzzy evaluation. According to the method and the device, dual consistency of data statistical characteristics and physical constraints is realized, so that the reliability and the interpretability of an evaluation result are improved. Physical knowledge is introduced into the whole data quality evaluation process, a data-physics double-layer evaluation system is established, and it is ensured that the evaluation result meets the data distribution rule and meets the physical constraint requirement.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a pressure vessel data quality assessment method and system based on big data. Background Art

[0002] During the operation of pressure vessels, data quality assessment is an important part of ensuring the safe operation of equipment. Existing technologies mainly use data quality assessment methods based on statistical characteristics, which analyze the integrity, consistency and accuracy of collected data to achieve quantitative assessment of data quality. This type of method usually relies on mathematical statistics models, combined with a set evaluation index system, to analyze the statistical characteristics of the data, including the distribution characteristics, fluctuation patterns and trend changes of the data. At the same time, some evaluation methods also introduce machine learning algorithms to achieve automatic assessment of data quality by establishing a mapping relationship between data characteristics and quality levels. These methods have been widely used in industrial practice and provide important support for equipment operation monitoring and fault diagnosis.

[0003] However, existing data quality assessment methods have obvious shortcomings. These methods rely too much on the statistical characteristics of the data and ignore the physical constraints during the operation of the pressure vessel, resulting in the lack of physical meaning in the assessment results. Secondly, the coupling relationship between multi-source sensor data is not fully considered during the assessment process, making the assessment results unable to reflect the true operating status of the equipment. Thirdly, the existing methods mainly identify abnormal data based on statistical thresholds, lack verification of physical laws, and are prone to misjudgment. Finally, the evaluation results are not interpretable enough to provide effective guidance for operation and maintenance decisions. Summary of the invention

[0004] This application provides a pressure vessel data quality assessment method and system based on big data, which is used to achieve dual consistency of data statistical characteristics and physical constraints to improve the reliability and interpretability of the assessment results. By introducing physical knowledge into the entire process of data quality assessment, a "data-physics" two-layer assessment system is established to ensure that the assessment results conform to both the data distribution law and the physical constraint requirements.

[0005] In a first aspect, the present application provides a pressure vessel data quality assessment method based on big data, and the pressure vessel data quality assessment method based on big data includes: according to the operating data collected from the pressure vessel, including the vessel internal pressure data, the wall temperature data and the stress-strain data, a physical operating condition mapping data set is obtained through physical constraint preprocessing and operating condition synchronization processing; according to the physical operating condition mapping data set, a physical operating condition feature set is obtained by calculating the physical constraint index, the operating condition stability index and the data coordination index; according to the physical operating condition feature set, a physical coupling relationship and a change trend between the physical operating condition mapping data sets are analyzed through a deep neural network driven by physical knowledge to obtain an abnormal operating condition type discrimination result; according to the abnormal operating condition type discrimination result, a scoring matrix is ​​constructed through physical constraint scoring, and the scoring matrix is ​​dynamically calibrated to obtain a physical operating condition evaluation factor; according to the physical operating condition evaluation factor and the abnormal operating condition type discrimination result, the operating quality grade of the pressure vessel is calculated through multi-scale fuzzy evaluation to obtain an equipment data quality assessment result.

[0006] In a second aspect, the present application provides a pressure vessel data quality assessment system based on big data, and the pressure vessel data quality assessment system based on big data includes: A synchronization module is used to obtain a physical working condition mapping data set through physical constraint preprocessing and working condition synchronization processing based on working condition data collected from the pressure vessel, including vessel internal pressure data, wall temperature data and stress and strain data; A calculation module, used for obtaining a physical working condition feature set by calculating a physical constraint index, a working condition stability index and a data coordination index according to the physical working condition mapping data set; An analysis module is used to analyze the physical coupling relationship and change trend between the physical working condition mapping data sets through a deep neural network driven by physical knowledge according to the physical working condition feature set, so as to obtain a result of distinguishing the type of working condition abnormality; A construction module is used to construct a scoring matrix through physical constraint scoring according to the abnormal working condition type discrimination result, and dynamically calibrate the scoring matrix to obtain a physical working condition evaluation factor; The evaluation module is used to calculate the operation quality grade of the pressure vessel through multi-scale fuzzy evaluation according to the physical operating condition evaluation factor and the operating condition abnormality type discrimination result, so as to obtain the equipment data quality evaluation result.

[0007] The third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the computer device executes the above-mentioned pressure vessel data quality assessment method based on big data.

[0008] A fourth aspect of the present invention provides a computer-readable storage medium, in which instructions are stored, which, when executed on a computer, enable the computer to execute the above-mentioned pressure vessel data quality assessment method based on big data.

[0009] In the technical solution provided by the present application, the physical mapping relationship of multi-source data is constructed by performing physical constraint preprocessing and working condition synchronization processing on the working condition data collected by the pressure vessel, which effectively avoids the deviation caused by relying solely on the statistical characteristics of the data; a complete set of physical working condition characteristics is established by calculating the physical constraint index, working condition stability index and data coordination index, so that the evaluation process fully considers the constraints of physical laws; a deep neural network driven by physical knowledge is used to analyze the physical coupling relationship and change trend between the physical working condition mapping data sets. The network innovatively adopts a dual-stream structure, which processes the timing features and physical constraint features respectively through the timing stream branch structure and the physical stream branch structure, and realizes the dynamic fusion of features through the attention mechanism, which significantly improves the accuracy and interpretability of anomaly recognition, and fully reflects the deep integration of artificial intelligence algorithms and physical knowledge; the scoring matrix is ​​constructed by physical constraint scoring and dynamically calibrated, which realizes the real-time update and adaptive adjustment of the evaluation results, and provides The timeliness and adaptability of the evaluation are improved; multi-scale fuzzy evaluation is used to calculate the operating quality level of pressure vessels, and the operating status of equipment is analyzed at different time scales, which enhances the comprehensiveness and reliability of the evaluation results; at the algorithm level, the dual-stream structure design of the deep neural network makes full use of the advantages of the long short-term memory network in time series data processing and the characteristics of the graph convolutional network in extracting physical constraint relationships. The introduction of the attention mechanism realizes the intelligent fusion of time series features and physical constraint features. This algorithm architecture innovation effectively solves the problems of insufficient feature extraction and unclear expression of physical constraints in traditional methods; in terms of the evaluation index system, through the organic combination of physical constraint indicators, operating condition stability indicators and data coordination indicators, an evaluation system that conforms to physical laws and meets the actual needs of the project is constructed, so that the evaluation results have stronger physical significance and practical guidance value, and can accurately identify various abnormal conditions during the operation of pressure vessels, significantly improving the level of safety management of pressure vessel operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0011] Figure 1 This is a schematic diagram of an embodiment of a method for evaluating the data quality of a pressure vessel based on big data in an embodiment of the present application; Figure 2 This is a timing diagram of physical constraint preprocessing and working condition synchronization processing in an embodiment of the present application; Figure 3 This is a data processing flow chart of the deep neural network in the embodiment of this application; Figure 4 This is a schematic diagram of an embodiment of a pressure vessel data quality assessment system based on big data in an embodiment of the present application; Figure 5 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The embodiments of the present application provide a method and system for assessing the data quality of pressure vessels based on big data. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the pressure vessel data quality assessment method based on big data includes: Step S101, obtaining a physical working condition mapping data set through physical constraint preprocessing and working condition synchronization processing according to working condition data collected from the pressure vessel, including vessel internal pressure data, wall temperature data and stress-strain data; Step S102, according to the physical working condition mapping data set, by calculating the physical constraint index, the working condition stability index and the data coordination index, a physical working condition feature set is obtained; Step S103: According to the physical working condition feature set, a physical coupling relationship and a change trend between the physical working condition mapping data sets are analyzed by a deep neural network driven by physical knowledge to obtain a result of distinguishing the abnormal working condition type; Step S104: construct a scoring matrix through physical constraint scoring according to the abnormal working condition type discrimination result, and dynamically calibrate the scoring matrix to obtain a physical working condition evaluation factor; Step S105: Calculate the operation quality grade of the pressure vessel through multi-scale fuzzy evaluation according to the physical operating condition assessment factor and the abnormal operating condition type discrimination result to obtain the equipment data quality assessment result.

[0014] It is understandable that the execution subject of the present application can be a pressure vessel data quality assessment system based on big data, or a terminal or a server, which is not limited here. The present application embodiment is described by taking the server as the execution subject as an example.

[0015] Specifically, a sensor network is deployed at key locations of the pressure vessel to collect the internal pressure data, wall temperature data, and stress and strain data of the vessel. For the internal pressure data of the vessel, pressure sensors are used to collect data at the top, middle, and bottom of the vessel, with a sampling frequency of 10 Hz, and the pressure changes are recorded in real time. For the wall temperature data, the temperature sensor array is used to collect data at different heights of the vessel wall, with a sampling frequency of 1 Hz, to monitor the temperature distribution. The stress and strain data are collected in the stress concentration area of ​​the vessel through strain gauges, with a sampling frequency of 100 Hz, reflecting the degree of local deformation. The collected raw data undergoes physical constraint preprocessing and working condition synchronization processing. Physical constraint preprocessing includes two links: data cleaning and physical rule verification. During the data cleaning process, the internal pressure data of the vessel is median filtered to remove mutation points, the wall temperature data is smoothed to eliminate noise, and the stress and strain data are normalized to achieve dimensional unification. During the physical rule verification process, based on the pressure vessel design specifications, it is verified whether the data meets the physical constraint requirements. Working condition synchronization processing is to align the data collected by different sensors according to a unified time base to construct a physical working condition mapping data set.

[0016] Based on the physical working condition mapping data set, the physical working condition feature set is formed by calculating the physical constraint index, working condition stability index and data coordination index. The physical constraint index reflects the degree of compliance of the data with the physical law, including the correlation characteristics of pressure and temperature, and the coupling characteristics of temperature and stress. The working condition stability index describes the degree of fluctuation of the operating state, including the pressure fluctuation rate, the temperature change rate, and the strain growth rate. The data coordination index reflects the consistency between different types of data, including the temperature and pressure coordination degree and the thermal stress coordination degree. Based on the physical working condition feature set, the physical coupling relationship and change trend between the physical working condition mapping data sets are analyzed through the deep neural network driven by physical knowledge. The network adopts a dual-stream structure. The time series branch uses the long short-term memory network to process the time series features, and the physical branch uses the graph convolution network to process the physical constraint features. The feature fusion is realized through the attention mechanism. The network input includes the time series data and physical constraint relationship of pressure, temperature, stress and strain, and outputs the result of the abnormal type discrimination of the working condition.

[0017] According to the results of the abnormal working condition type discrimination, a scoring matrix is ​​constructed and dynamically calibrated. The discrimination results are divided into three categories: pressure anomaly, temperature anomaly and strain anomaly. Pressure anomaly includes overpressure anomaly and underpressure anomaly, temperature anomaly includes overtemperature anomaly and temperature fluctuation anomaly, and strain anomaly includes strain overlimit anomaly and strain mutation anomaly. Different weights are assigned to different types of anomalies. The weight range of pressure anomaly is 0.4-0.5, the weight range of temperature anomaly is 0.3-0.4, and the weight range of strain anomaly is 0.2-0.3. The degree of anomaly is quantified into risk level through scoring mapping to form an initial scoring matrix. The scoring matrix is ​​dynamically calibrated considering the duration of the anomaly and the degree of anomaly to obtain the physical working condition assessment factor.

[0018] Based on the physical working condition evaluation factors and the results of the abnormal working condition type discrimination, the multi-scale fuzzy evaluation method is used to calculate the operating quality level of the pressure vessel. The evaluation process is divided into three time scales: seconds, minutes and hours. At the second level, the frequency of abnormal occurrence is counted, and less than 5 times / second is divided into the low frequency domain, 5-10 times / second is divided into the medium frequency domain, and more than 10 times / second is divided into the high frequency domain. At the minute level, the persistence of abnormalities is analyzed, and the duration of less than 1 minute is divided into the instantaneous abnormal domain, 1-5 minutes is divided into the short-term abnormal domain, and more than 5 minutes is divided into the continuous abnormal domain. At the hour level, the periodicity of abnormalities is studied, and the period of less than 2 hours is divided into the high frequency abnormal domain, 2-6 hours is divided into the medium frequency abnormal domain, and more than 6 hours is divided into the low frequency abnormal domain.

[0019] The data quality assessment of an ethylene tank in a chemical enterprise is used as an example. The tank has a capacity of 1,000 cubic meters, a design pressure of 0.6 MPa, and an operating temperature range of -30°C to 40°C. One week of operation data, including pressure data, temperature data, and stress-strain data, is collected through a sensor network. In the data preprocessing stage, it was found that the pressure data had three mutations, the temperature data contained random noise, and some periods of stress-strain data were lost. Physical constraint verification showed that some data did not meet the requirements of the temperature-pressure relationship. The valid data after cleaning and verification constituted the physical working condition mapping data set. The calculated working condition characteristics showed that the pressure stability index was good, but the temperature fluctuated greatly, and there was a coordination deviation with the stress change. Deep neural network analysis identified multiple abnormal temperature fluctuations and abnormal strain mutations. The physical working condition evaluation factor after scoring and calibration was 0.82. The multi-scale fuzzy evaluation results showed that the operating quality level of the tank was in the normal operating range, but there were hidden dangers in temperature control and structural monitoring.

[0020] In the embodiments of the present application, by performing physical constraint preprocessing and working condition synchronization processing on the working condition data collected by the pressure vessel, the physical mapping relationship of multi-source data is constructed, which effectively avoids the deviation caused by relying solely on the statistical characteristics of the data; by calculating the physical constraint index, the working condition stability index and the data coordination index, a complete set of physical working condition characteristics is established, so that the evaluation process fully considers the constraints of physical laws; a deep neural network driven by physical knowledge is used to analyze the physical coupling relationship and change trend between the physical working condition mapping data sets. The network innovatively adopts a dual-stream structure, which processes the timing features and the physical constraint features respectively through the timing stream branch structure and the physical stream branch structure, and realizes the dynamic fusion of features through the attention mechanism, which significantly improves the accuracy and interpretability of anomaly recognition, and fully reflects the deep integration of artificial intelligence algorithms and physical knowledge; by constructing a scoring matrix through physical constraint scoring and performing dynamic calibration, real-time updating and adaptive adjustment of the evaluation results are realized, which improves The timeliness and adaptability of the evaluation; the multi-scale fuzzy evaluation is used to calculate the operating quality level of the pressure vessel, and the equipment operating status is analyzed at different time scales, which enhances the comprehensiveness and reliability of the evaluation results; at the algorithm level, the dual-stream structure design of the deep neural network makes full use of the advantages of the long short-term memory network in time series data processing and the characteristics of the graph convolutional network in extracting physical constraint relationships. The introduction of the attention mechanism realizes the intelligent fusion of time series features and physical constraint features. This algorithm architecture innovation effectively solves the problems of insufficient feature extraction and unclear physical constraint expression in traditional methods; in terms of the evaluation index system, through the organic combination of physical constraint indicators, operating condition stability indicators and data coordination indicators, an evaluation system that conforms to physical laws and meets the actual needs of the project is constructed, so that the evaluation results have stronger physical significance and practical guidance value, and can accurately identify various abnormal conditions during the operation of pressure vessels, significantly improving the level of safety management of pressure vessel operation.

[0021] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Convert the fluctuation points in the internal pressure data of the container into a pressure trend sequence, convert the temperature points in the wall temperature data into a thermal series, and convert the strain points in the stress-strain data into a deformation series; (2) Segmented sampling of the pressure trend sequence, thermal sequence, and deformation sequence is performed through a time window to obtain operating condition data segments; (3) Correlate and pair the pressure trend sequence and thermal sequence in the operating condition data segment to generate a temperature-pressure mapping group; (4) Correlate and pair the thermal sequence and deformation sequence in the working condition data segment to generate a thermal stress mapping group; (5) Coupling the temperature-pressure mapping group with the thermal stress mapping group to obtain the working condition correlation characteristics; (6) Physical constraint compensation and working condition synchronization calibration are performed according to the working condition correlation characteristics to obtain a physical working condition mapping data set.

[0022] Specifically, Figure 2 As shown, it is a timing diagram of physical constraint preprocessing and working condition synchronization processing in an embodiment of the present application, wherein the original collected data is subjected to fluctuation point conversion processing. For the internal pressure data of the container, the fluctuation point is identified by analyzing the pressure change amplitude between adjacent sampling points, and is marked as a fluctuation point when the pressure change amplitude exceeds the preset threshold. These fluctuation points are connected in chronological order to form a pressure trend sequence, which reflects the dynamic change trend of the pressure. Similarly, the temperature measurement points in the wall temperature data are screened and connected to construct a thermal sequence reflecting the temperature change law. After the stress-strain data is extracted by the strain point, a deformation sequence describing the deformation state of the container is generated. In order to achieve unified processing of different types of data, a time window is used for segmented sampling. The time window size is determined according to the physical properties of the pressure vessel and the data sampling frequency, and the typical value is 10 seconds. In each time window, the pressure trend sequence, thermal sequence and deformation sequence are sampled respectively to obtain a working condition data segment with the same time span. This segmented sampling method not only retains the temporal characteristics of the data, but also facilitates subsequent correlation analysis.

[0023] After obtaining the working condition data segment, the correlation between different physical quantities is established. The pressure trend sequence and the thermal sequence in the working condition data segment are correlated and paired. Based on the principle of the gas state equation, the mutual influence between pressure and temperature is analyzed to generate a temperature-pressure mapping group. This mapping group reflects the correspondence between pressure change and temperature change. Secondly, the thermal sequence is correlated and paired with the deformation sequence. According to the thermal stress theory, the stress-strain response caused by temperature change is studied to generate a thermal stress mapping group.

[0024] The key link of data fusion is to couple the temperature and pressure mapping group with the thermal stress mapping group. By analyzing the interaction between the temperature and pressure relationship and the thermal stress relationship, a coupling model between pressure, temperature and stress strain is established to obtain the working condition correlation characteristics. These characteristics contain the physical state information of the pressure vessel under different working conditions. Physical constraint compensation and working condition synchronization calibration are performed based on the working condition correlation characteristics. Physical constraint compensation refers to correcting the data according to the design parameters and material properties of the pressure vessel to meet the requirements of physical laws. Working condition synchronization calibration ensures the consistency of different types of data in time and eliminates the influence of factors such as sampling delay. After these processes, a complete physical working condition mapping data set is formed.

[0025] Take a liquefied natural gas storage tank as an example. The tank is equipped with multiple pressure sensors, temperature sensors and strain sensors, with sampling frequencies of 10Hz, 1Hz and 100Hz respectively. In the data fluctuation point conversion stage, the pressure data is analyzed and multiple pressure fluctuation points are found. These fluctuation points are related to the liquid level changes and gasification fluctuations of the tank. By connecting these fluctuation points, a trend sequence reflecting the law of pressure changes is formed. The temperature data shows obvious stratification, and the temperature sensor data at different heights form a unique thermal sequence. The stress-strain data reflects the local deformation of the tank wall, especially the strain value in the area where the temperature changes drastically.

[0026] The three types of sequences were sampled in segments using a 10-second time window to obtain a large number of operating condition data segments. During the correlation pairing process, it was found that the pressure increased as the temperature increased, but there was a time lag between the two. The thermal stress mapping group showed that there was a certain deviation between the thermal strain caused by temperature change and the measured strain, which was related to the structural characteristics and constraints of the tank. Through operating condition coupling processing, the quantitative relationship between increased pressure, decreased temperature and increased strain under rapid gasification conditions was clarified. Finally, after physical constraint compensation and operating condition synchronization calibration, a physical operating condition mapping data set reflecting the actual operating status of the tank was established.

[0027] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) The physical condition mapping data set is divided into condition period groups according to the time series, and the fluctuation value of the data in the condition period group is calculated to obtain the physical constraint index; (2) Calculate the change rate of the internal pressure data of the container in the operating period group, determine the stable operating range according to the pressure change trend, and obtain the operating stability index; (3) Perform correlation analysis on the wall temperature data and stress-strain data in the working period group, generate thermal stress correlation, and obtain data coordination index; (4) Combining the physical constraint index, working condition stability index, and data coordination index to construct a feature vector; (5) Perform normalized mapping on the feature vectors to generate standardized feature sets of physical constraint indicators, working condition stability indicators, and data coordination indicators; (6) Combine and associate the indicators in the standardized feature set, construct the operating condition feature mapping relationship, and obtain the physical operating condition feature set.

[0028] Specifically, the physical condition mapping data set is divided into time series. According to the equipment operation characteristics, the data set is divided into multiple condition time groups, each of which is 1 hour long. Within each condition time group, the fluctuation value is calculated based on the pressure, temperature and strain data. The fluctuation value is calculated using the coefficient of variation method:

[0029] in, is the fluctuation value, is the data standard deviation, The fluctuation value of each physical quantity is calculated and compared with the preset threshold to generate the physical constraint index.

[0030] For the internal pressure data of the container, the pressure change rate is calculated by the pressure difference at consecutive time points. The calculation formula for the pressure change rate is:

[0031] in, is the pressure change rate (MPa / s), is the pressure change (MPa), is the time interval (s). According to the magnitude of the pressure change rate, the stable interval of the working condition is determined. When the pressure change rate is lower than the set threshold, it is determined to be a stable interval, and the duration of the stable interval is counted to obtain the working condition stability index. For the correlation analysis between the wall temperature data and the stress-strain data, the thermal stress correlation calculation method is used:

[0032] in, is the thermal stress correlation, is the temperature value (℃), is the strain value (με), is the average temperature, is the average strain value, and n is the number of data points. The correlation reflects the consistency of temperature change and strain response, and constitutes a data coordination index.

[0033] The physical constraint index, working condition stability index and data coordination index are combined to construct a feature vector:

[0034] in, is the feature vector, is the physical constraint index, is the working condition stability index, Is the data coordination index. Normalize the feature vector:

[0035] in, is the normalized eigenvector, is the minimum value of each indicator, is the maximum value of each indicator. Normalization makes indicators of different dimensions comparable and forms a standardized feature set.

[0036] Finally, the indicators in the standardized feature set are combined and associated to construct the working condition feature mapping relationship. Through the weighted combination method:

[0037] in, is the mapping value, , , are the weight coefficients of the physical constraint index, working condition stability index and data coordination index, , , is the corresponding normalized index value. According to the distribution characteristics of the mapping value, the physical working condition feature set is obtained.

[0038] Taking a liquefied natural gas storage tank as an example, within a working period group, the standard deviation of the pressure data is 0.02MPa, the mean is 0.5MPa, and the calculated fluctuation value is 0.04. The maximum pressure change rate occurs during the unloading process, reaching 0.005MPa / s. The correlation coefficient between temperature and strain data is 0.85, indicating that the thermal stress response is good. After normalization and weighted combination, these indicators form a feature set that reflects the operating status of the equipment during this period.

[0039] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) The deep neural network has a dual-stream structure, including a temporal stream branch structure and a physical stream branch structure. The temporal stream branch structure uses a long short-term memory network to process temporal features, and the physical stream branch structure uses a graph convolutional network to process physical constraint features. The outputs of the two branches are fused through the attention mechanism to obtain a deep neural network driven by physical knowledge. (2) Input the physical condition feature set into the long short-term memory network, extract the time series change features, and obtain the time series feature vector; (3) Input the physical working condition mapping data set into the graph convolutional network to extract the physical constraint relationship and obtain the physical constraint vector; (4) The time series feature vector and the physical constraint vector are weighted fused through the attention mechanism to obtain a fused feature matrix; (5) Calculate the anomaly probability distribution based on the fused feature matrix and generate an anomaly mapping vector; (6) Perform type classification analysis on the abnormal mapping vector to obtain the abnormal type discrimination result of the working condition.

[0040] Specifically, Figure 3As shown, it is a data processing flow chart of the deep neural network in the embodiment of the present application. The construction of the deep neural network adopts a dual-stream structure, which is divided into a time series flow branch and a physical flow branch. The time series flow branch uses a long short-term memory network (LSTM) to process time series features. The network has three control units: an input gate, a forget gate, and an output gate, which can capture the long-term dependencies of the data. The physical flow branch uses a graph convolutional network (GCN) to process physical constraint features, and represents the physical parameter relationship of the pressure vessel as a graph structure. The nodes represent different physical quantities, and the edges represent the constraint relationship between the physical quantities. After the physical working condition feature set is input into the LSTM network, the time series data is feature extracted through the memory unit. The input gate determines the new information that needs to be retained at the current moment, the forget gate controls the degree of forgetting of historical information, and the output gate determines the output content of the current state. After multi-layer LSTM processing, a time series feature vector reflecting the time evolution characteristics of parameters such as pressure, temperature, and strain is obtained. At the same time, the physical working condition mapping data set is input into the GCN network, and the physical constraint relationship is feature extracted through the graph convolution layer. Each graph convolution layer aggregates the information of adjacent nodes to realize the feature learning of the constraint relationship between physical quantities, and finally obtains the physical constraint vector.

[0041] The attention mechanism is used to fuse the temporal feature vector and the physical constraint vector. The mechanism calculates the correlation weight between the two types of feature vectors, assigning a larger weight to the important features and a smaller weight to the minor features. The fused feature matrix is ​​obtained by weighted summation, which contains both the temporal change information of the data and the physical constraint relationship. Based on the fused feature matrix, the softmax classifier is used to calculate the probability distribution of different anomaly types and generate anomaly mapping vectors. Finally, the anomaly mapping vectors are subjected to threshold judgment and type attribution analysis to determine the specific type of abnormal working condition. The application of this method is illustrated by taking a propylene storage tank of a petrochemical enterprise as an example. The storage tank is equipped with sensors such as pressure, temperature, and strain, and the operating data is collected for one week. The LSTM network of the temporal flow branch extracts temporal features such as pressure fluctuation, temperature gradient, and strain mutation, and the GCN network of the physical flow branch identifies physical constraint features such as abnormal temperature-pressure relationship and thermal stress response deviation. The attention mechanism focuses on the violation of physical constraints at the moment of pressure mutation, weakening the general fluctuations under normal working conditions. Through abnormal probability calculation and type discrimination, multiple pressure abnormality events caused by improper temperature control are successfully identified. These anomalies show obvious characteristics in both timing features and physical constraint features, which confirms the effectiveness of the two-stream network structure. In practical applications, each branch of the two-stream neural network specifically processes different types of features. The timing stream branch focuses on analyzing the temporal correlation of the data and capturing the timing pattern of anomalies. The physical stream branch focuses on the degree of compliance with physical laws and discovers anomalies that violate physical constraints. The two branches are organically combined through the attention mechanism, which not only ensures the continuity of the timing, but also satisfies the physical rationality, thereby accurately identifying various abnormal conditions during the operation of the pressure vessel.

[0042] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) The results of the abnormal working condition type discrimination are divided into three types: pressure abnormality, temperature abnormality and strain abnormality. The pressure abnormality is divided into overpressure abnormality and underpressure abnormality, the temperature abnormality is divided into overtemperature abnormality and temperature fluctuation abnormality, and the strain abnormality is divided into strain overlimit abnormality and strain mutation abnormality to generate an abnormal feature group; (2) According to the safety constraint criteria of the pressure vessel, the abnormal feature groups are divided into weight intervals, where the weight interval of pressure anomaly is 0.4 to 0.5, the weight interval of temperature anomaly is 0.3 to 0.4, and the weight interval of strain anomaly is 0.2 to 0.3, and the feature weight matrix is ​​obtained; (3) Score mapping is performed on the feature weight matrix, mapping overpressure anomalies and strain overlimit anomalies into high-risk score intervals, mapping temperature fluctuation anomalies and strain mutation anomalies into medium-risk score intervals, and mapping underpressure anomalies into low-risk score intervals, to construct an initial score matrix; (4) Dynamically update the score values ​​in the initial scoring matrix according to the duration and degree of abnormality in the abnormality type identification results to obtain a calibrated scoring matrix; (5) The scoring values ​​in the calibration scoring matrix are weighted and combined according to the hazard level of the abnormality type to generate the working condition assessment value; (6) The working condition evaluation value is normalized to obtain the physical working condition evaluation factor.

[0043] Specifically, based on physical characteristics, the operating anomalies are divided into three categories: pressure anomalies, temperature anomalies, and strain anomalies. Among them, pressure anomalies are further divided into overpressure anomalies and underpressure anomalies, temperature anomalies are further divided into overtemperature anomalies and temperature fluctuation anomalies, and strain anomalies are further divided into strain overlimit anomalies and strain mutation anomalies. Through this multi-level anomaly classification system, a complete abnormal feature group is constructed. When determining the degree of danger of the anomaly type, a weight allocation method based on the safety constraint criterion of pressure vessels is adopted. Pressure, as the factor that most directly affects the safety of the container, has an abnormal weight range of 0.4 to 0.5; temperature anomalies have a weight range of 0.3 to 0.4 due to their influence on material properties and pressure changes; strain anomalies reflect structural responses, and the weight range is set to 0.2 to 0.3. This hierarchical weight is expressed in matrix form:

[0044] in , Represent the overvoltage and undervoltage abnormal weights respectively, , They represent the abnormal weights of overtemperature and temperature fluctuation, , They represent the weights of strain over-limit and strain mutation anomalies respectively. This matrix structure clearly shows the weight distribution relationship of different types of anomalies.

[0045] In the scoring mapping process, the anomaly type is divided into three intervals according to the risk level. Based on the harmfulness of the anomaly, a risk scoring vector is constructed:

[0046] in For the comprehensive risk score, is the risk level coefficient (high, medium and low risks are 0.9, 0.6 and 0.3 respectively), is the abnormal index value corresponding to the risk level. Taking the overpressure abnormality that occurs during the operation of a pressure vessel as an example, when the pressure value exceeds 1.1 times the design pressure, the abnormality is classified into the high-risk interval, and its α value is 0.9.

[0047] Considering the dynamic characteristics of anomalies, the time factor is introduced to calibrate the score. The calibration coefficient is calculated using:

[0048] in is the calibration factor, is the time impact factor, is the abnormal duration (hours), is the amplitude influencing factor, For example, a storage tank has a 2-hour temperature fluctuation that exceeds the normal range by 20%. =0.1, =0.2, the corresponding calibration coefficient can be calculated.

[0049] The operating condition assessment value is obtained by weighted combination of each abnormality type:

[0050] in is the working condition evaluation value, is the anomaly type weight, For the abnormality score, is the corresponding calibration coefficient. Taking the data of a pressure vessel in an operating cycle as an example, when an overpressure anomaly (λ=0.45) is detected for 1.5 hours and exceeds the limit by 15%, a complete operating condition evaluation value can be obtained by combining the evaluation results of other abnormal types.

[0051] Finally, the physical condition evaluation factor is obtained through normalization:

[0052] in is the normalized evaluation factor, , These are the theoretical minimum and maximum values ​​of the evaluation value. In practical applications, such as the operation monitoring of a LNG storage tank, this evaluation system can accurately reflect the operating status and potential risks of the equipment.

[0053] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) The physical working condition evaluation factors are divided into time scales, and the second-level evaluation domain, minute-level evaluation domain, and hour-level evaluation domain are constructed to obtain multi-scale evaluation indicators; (2) The abnormality type discrimination results of the working condition are used to perform abnormality frequency statistics in the second-level evaluation domain, abnormal persistence analysis in the minute-level evaluation domain, and abnormal periodicity analysis in the hour-level evaluation domain to obtain the time scale characteristic value; (3) A fuzzy membership function is established based on the time scale eigenvalues, and the operating quality of the pressure vessel is divided into a high-quality operating range, a normal operating range, an abnormal operating range, and a dangerous operating range to obtain a fuzzy evaluation space; (4) Map the multi-scale evaluation index to the fuzzy evaluation space, calculate the membership value of each operating interval, and obtain the fuzzy evaluation vector; (5) Perform multi-scale weighted fusion on the fuzzy evaluation vector to generate a quality grade score; (6) The quality grade score values ​​are classified according to the set threshold to obtain the equipment data quality assessment results.

[0054] Specifically, in the final stage of pressure vessel data quality assessment, multi-scale time analysis is performed on the physical condition assessment factors. Three time scale evaluation domains are established: second-level evaluation domain (1-60 seconds), minute-level evaluation domain (1-60 minutes) and hour-level evaluation domain (1-24 hours). For evaluation indicators at different time scales, the scale decomposition function is used:

[0055] in is the scale decomposition value, is the time scale weight coefficient, It is an evaluation index corresponding to the time scale. For example, in the operation monitoring of a LNG storage tank, the second-level data reflects the instantaneous change, the minute-level data reflects the process trend, and the hour-level data shows the long-term characteristics.

[0056] Different statistical methods are used to analyze abnormal characteristics at different time scales. The calculation of the frequency of abnormalities at the second level is:

[0057] in is the abnormal frequency (times / second), is the number of abnormal occurrences, is the statistical time period (seconds). Taking the data of the pressure vessel during startup as an example, when three pressure fluctuations are detected within 10 seconds, the abnormal frequency can be calculated to be 0.3 times / second.

[0058] For minute-level abnormal persistence analysis, the duration ratio is introduced:

[0059] in is the duration ratio, is the abnormal weight factor, is the duration of a single abnormality (minutes), is the total observation time (minutes), and m is the number of abnormal events. For example, if a storage tank has two temperature anomalies within an hour, and the durations are 3 minutes and 5 minutes respectively, its persistence index can be calculated.

[0060] For hourly abnormal periodicity analysis, the periodic characteristic function is used:

[0061] in is the periodic eigenvalue, is the period component weight, is the cycle length (hours), and n is the number of cycle components. This analysis method is suitable for discovering regular anomalies in equipment operation, such as pressure fluctuations caused by production cycles.

[0062] Establish fuzzy membership function based on time scale characteristics:

[0063] in is the quality grade membership, is the evaluation index value, , is the membership function parameter. This fuzzy evaluation method divides the equipment operation status into different intervals, achieving accurate evaluation of the operation quality. The fusion of multi-scale evaluation indicators adopts the weighted summation method:

[0064] in For the final quality rating, is the interval weight coefficient, is the evaluation value of each operating interval. For the actual operating data of a pressure vessel, when the proportion of high-quality operation time reaches 85% and the proportion of abnormal intervals is less than 5%, it can be determined that the equipment is in good operating condition. The scoring results are graded by setting different threshold intervals. The classification standard is: 90 points or more for high-quality operation, 80-90 points for normal operation, 60-80 points for abnormal operation, and below 60 points for dangerous operation. In practical applications, this multi-scale evaluation system can not only reflect the immediate status of the equipment, but also predict potential operating risks.

[0065] In a specific embodiment, the process of dividing the operating quality of the pressure vessel into a high-quality operating range, a normal operating range, an abnormal operating range, and a dangerous operating range may specifically include the following steps: (1) Perform statistical analysis on the second-level abnormal frequency in the time scale characteristic value, classify the abnormal frequency less than 5 times / second into the low frequency range, the abnormal frequency between 5 and 10 times / second into the medium frequency range, and the abnormal frequency greater than 10 times / second into the high frequency range, and obtain the frequency distribution characteristics; (2) Statistical analysis is performed on the minute-level anomaly persistence in the time scale characteristic values. Anomalies with a duration of less than 1 minute are classified as instantaneous anomalies, anomalies with a duration of 1-5 minutes are classified as short-term anomalies, and anomalies with a duration of more than 5 minutes are classified as continuous anomalies, thus obtaining the persistence characteristics. (3) Statistical analysis is performed on the hourly anomaly periodicity in the time scale characteristic values. The anomaly period of less than 2 hours is divided into the high-frequency anomaly domain, the anomaly period of 2-6 hours is divided into the medium-frequency anomaly domain, and the anomaly period of more than 6 hours is divided into the low-frequency anomaly domain, thus obtaining the periodic characteristics. (4) Establish membership function rules based on frequency distribution characteristics, persistence characteristics, and periodic characteristics to obtain interval membership relationships; (5) Map the interval membership relationship into the operation quality interval, construct the membership boundaries of the high-quality operation interval, normal operation interval, abnormal operation interval, and dangerous operation interval, and obtain the evaluation interval matrix; (6) Perform spatial mapping transformation on the evaluation interval matrix to obtain the fuzzy evaluation space.

[0066] Specifically, in the final stage of pressure vessel data quality assessment, the abnormal characteristics at different time scales are refined and graded. The frequency density function is introduced to quantitatively analyze the frequency of second-level abnormalities:

[0067] in is the abnormal frequency density value (times / second), is the abnormal event weight coefficient, is the number of occurrences of a single abnormal event, is the observation time window (seconds). Taking the pressure fluctuation of a LNG tank during the unloading process as an example, when the system detects 45 small fluctuations within 10 seconds, the abnormal frequency density can be calculated to be 4.5 times / second, which belongs to the low frequency range.

[0068] For minute-level abnormal persistence analysis, build a continuous feature evaluation function:

[0069] in is the persistent eigenvalue, is the duration weight factor, is the characteristic time constant (minutes), and m is the number of abnormal events. In practical applications, if a pressure vessel has a temperature anomaly that lasts for 3.5 minutes, it can be determined that it belongs to the short-term anomaly domain based on this function.

[0070] For hourly abnormal periodicity analysis, a periodic feature recognition function is introduced:

[0071] in is a periodic characteristic indicator, is the amplitude of the periodic component, is the characteristic period (hours), is the phase factor, For example, if a storage tank has regular pressure fluctuations in a 24-hour operation cycle with a period of about 4 hours, it belongs to the medium-frequency anomaly domain.

[0072] Based on the characteristic analysis results of the above three time scales, a multidimensional membership function is constructed:

[0073] in is the membership matrix, , , Represent the frequency, persistence and periodicity eigenvalues ​​respectively, is the corresponding membership coefficient. This matrix structure realizes the comprehensive evaluation of the three time scale characteristics.

[0074] The spatial transformation is performed by running the mass interval mapping function:

[0075] in, is the quality interval mapping value, is the interval weight coefficient, is the interval characteristic function, is the interval reference value, and z is the comprehensive characteristic vector. In specific applications, such as the evaluation of the operating status of a pressure vessel, when the frequency belongs to the low value domain (<5 times / second), the duration belongs to the instantaneous abnormal domain (<1 minute), and the period belongs to the low frequency abnormal domain (>6 hours), after the mapping function calculation, the operating condition will be judged as a high-quality operating range.

[0076] The above describes the pressure vessel data quality assessment method based on big data in the embodiment of the present application. The following describes the pressure vessel data quality assessment system based on big data in the embodiment of the present application. Figure 4 In the embodiment of the present application, an embodiment of the pressure vessel data quality assessment system based on big data includes: A synchronization module is used to obtain a physical working condition mapping data set through physical constraint preprocessing and working condition synchronization processing based on working condition data collected from the pressure vessel, including vessel internal pressure data, wall temperature data and stress and strain data; A calculation module, used for obtaining a physical working condition feature set by calculating a physical constraint index, a working condition stability index and a data coordination index according to the physical working condition mapping data set; An analysis module is used to analyze the physical coupling relationship and change trend between the physical working condition mapping data sets through a deep neural network driven by physical knowledge according to the physical working condition feature set, so as to obtain a result of distinguishing the type of working condition abnormality; A construction module is used to construct a scoring matrix through physical constraint scoring according to the abnormal working condition type discrimination result, and dynamically calibrate the scoring matrix to obtain a physical working condition evaluation factor; The evaluation module is used to calculate the operation quality grade of the pressure vessel through multi-scale fuzzy evaluation according to the physical operating condition evaluation factor and the operating condition abnormality type discrimination result, so as to obtain the equipment data quality evaluation result.

[0077] Through the coordinated cooperation of the above-mentioned components, the physical constraint preprocessing and synchronous processing of the working condition data collected by the pressure vessel are carried out to realize the construction of the physical mapping relationship of multi-source data, which effectively avoids the deviation caused by relying solely on the statistical characteristics of the data; by calculating the physical constraint index, working condition stability index and data coordination index, a complete set of physical working condition characteristics is established, so that the evaluation process fully considers the constraints of physical laws; a deep neural network driven by physical knowledge is used to analyze the physical coupling relationship and change trend between the physical working condition mapping data sets. The network innovatively adopts a dual-stream structure, which processes the timing features and physical constraint features respectively through the timing stream branch structure and the physical stream branch structure, and realizes the dynamic fusion of features through the attention mechanism, which significantly improves the accuracy and interpretability of anomaly recognition, and fully reflects the deep integration of artificial intelligence algorithms and physical knowledge; through the physical constraint scoring, the scoring matrix is ​​constructed and dynamically calibrated, which realizes the real-time update and adaptive adjustment of the evaluation results. , which improves the timeliness and adaptability of the evaluation; multi-scale fuzzy evaluation is used to calculate the operating quality level of pressure vessels, and the operating status of equipment is analyzed at different time scales, which enhances the comprehensiveness and reliability of the evaluation results; at the algorithm level, the dual-stream structure design of the deep neural network makes full use of the advantages of the long short-term memory network in time series data processing and the characteristics of the graph convolutional network in extracting physical constraint relationships. The introduction of the attention mechanism realizes the intelligent fusion of time series features and physical constraint features. This algorithm architecture innovation effectively solves the problems of insufficient feature extraction and unclear expression of physical constraints in traditional methods; in terms of the evaluation index system, through the organic combination of physical constraint indicators, operating condition stability indicators and data coordination indicators, an evaluation system that conforms to physical laws and meets the actual needs of the project is constructed, so that the evaluation results have stronger physical significance and practical guidance value, and can accurately identify various abnormal conditions during the operation of pressure vessels, significantly improving the level of safety management of pressure vessel operation.

[0078] Reference Figure 5 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0079] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0080] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0081] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0082] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0083] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0084] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A pressure vessel data quality assessment method based on big data, characterized in that: The pressure vessel data quality assessment method based on big data includes: According to the working condition data collected from the pressure vessel, including the internal pressure data of the vessel, the wall temperature data and the stress-strain data, a physical working condition mapping data set is obtained through physical constraint preprocessing and working condition synchronization processing; According to the physical working condition mapping data set, a physical working condition feature set is obtained by calculating a physical constraint index, a working condition stability index and a data coordination index; According to the physical working condition feature set, a physical coupling relationship and a change trend between the physical working condition mapping data sets are analyzed by a deep neural network driven by physical knowledge to obtain a result of distinguishing the type of working condition abnormality; According to the abnormal working condition type discrimination result, a scoring matrix is ​​constructed through physical constraint scoring, and the scoring matrix is ​​dynamically calibrated to obtain a physical working condition evaluation factor; According to the physical operating condition evaluation factor and the abnormal operating condition type identification result, the operation quality grade of the pressure vessel is calculated through multi-scale fuzzy evaluation to obtain the equipment data quality evaluation result.

2. The pressure vessel data quality assessment method based on big data according to claim 1 is characterized in that: The physical condition mapping data set is obtained by performing physical constraint preprocessing and condition synchronization processing based on the condition data collected from the pressure vessel, including the vessel internal pressure data, wall temperature data and stress and strain data, including: Converting the fluctuation points in the container internal pressure data into a pressure trend sequence, converting the temperature points in the wall temperature data into a thermal sequence, and converting the strain points in the stress-strain data into a deformation sequence; Segmented sampling of the pressure trend sequence, the thermal sequence and the deformation sequence is performed through a time window to obtain working condition data segments; Associating and pairing the pressure trend sequence and the thermal sequence in the operating condition data segment to generate a temperature-pressure mapping group; Associating and pairing the thermal sequence and the deformation sequence in the operating condition data segment to generate a thermal stress mapping group; Performing working condition coupling processing on the temperature-pressure mapping group and the thermal stress mapping group to obtain working condition correlation characteristics; Physical constraint compensation and working condition synchronization calibration are performed according to the working condition association characteristics to obtain the physical working condition mapping data set.

3. The pressure vessel data quality assessment method based on big data according to claim 1 is characterized in that: The physical working condition feature set is obtained by calculating the physical constraint index, the working condition stability index and the data coordination index according to the physical working condition mapping data set, including: Dividing the physical working condition mapping data set into working condition time period groups according to a time series, calculating fluctuation values ​​for the data in the working condition time period groups, and obtaining physical constraint indicators; Calculating the change rate of the internal pressure data of the container in the working condition period group, determining the working condition stable interval according to the pressure change trend, and obtaining the working condition stability index; Performing correlation analysis on the wall temperature data and stress-strain data in the working condition period group, generating a thermal stress correlation, and obtaining a data coordination index; Combining the physical constraint index, the working condition stability index and the data coordination index to construct a feature vector; Performing normalized mapping on the feature vector to generate a standardized feature set of the physical constraint index, the working condition stability index, and the data coordination index; The indicators in the standardized feature set are combined and associated to construct a working condition feature mapping relationship to obtain the physical working condition feature set.

4. The pressure vessel data quality assessment method based on big data according to claim 1 is characterized in that: The physical coupling relationship and change trend between the physical working condition mapping data sets are analyzed by a deep neural network driven by physical knowledge according to the physical working condition feature set to obtain a working condition abnormality type discrimination result, including: The deep neural network is a dual-stream structure, including a temporal stream branch structure and a physical stream branch structure. The temporal stream branch structure uses a long short-term memory network to process temporal features, and the physical stream branch structure uses a graph convolutional network to process physical constraint features. The outputs of the two branches are fused through an attention mechanism to obtain a physical knowledge-driven deep neural network. Inputting the physical working condition feature set into the long short-term memory network, extracting the time series change features, and obtaining a time series feature vector; Inputting the physical working condition mapping data set into the graph convolutional network, extracting the physical constraint relationship, and obtaining a physical constraint vector; Performing weight fusion on the temporal feature vector and the physical constraint vector through the attention mechanism to obtain a fused feature matrix; Calculate the abnormal probability distribution according to the fusion feature matrix and generate an abnormal mapping vector; The abnormal mapping vector is subjected to type attribution analysis to obtain the abnormal type discrimination result of the operating condition.

5. The pressure vessel data quality assessment method based on big data according to claim 1 is characterized in that: According to the abnormal working condition type discrimination result, a scoring matrix is ​​constructed through physical constraint scoring, and the scoring matrix is ​​dynamically calibrated to obtain a physical working condition evaluation factor, including: The abnormal working condition type discrimination result is divided into three types: pressure abnormality, temperature abnormality and strain abnormality, and the pressure abnormality is divided into overpressure abnormality and underpressure abnormality, the temperature abnormality is divided into overtemperature abnormality and temperature fluctuation abnormality, and the strain abnormality is divided into strain overlimit abnormality and strain mutation abnormality, to generate an abnormality feature group; Dividing the abnormal feature group into weight intervals according to the safety constraint criteria of the pressure vessel, wherein the weight interval of the pressure anomaly is 0.4 to 0.5, the weight interval of the temperature anomaly is 0.3 to 0.4, and the weight interval of the strain anomaly is 0.2 to 0.3, to obtain a feature weight matrix; Score mapping is performed on the feature weight matrix, the overpressure anomaly and the strain overlimit anomaly are mapped into high-risk score intervals, the temperature fluctuation anomaly and the strain mutation anomaly are mapped into medium-risk score intervals, and the underpressure anomaly is mapped into a low-risk score interval, so as to construct an initial score matrix; Dynamically updating the score values ​​in the initial score matrix according to the abnormality duration and the abnormality degree in the abnormality type discrimination result of the operating condition to obtain a calibration score matrix; The scoring values ​​in the calibration scoring matrix are weighted and combined according to the degree of hazard of the abnormal type to generate a working condition assessment value; The operating condition evaluation value is normalized and converted to obtain the physical operating condition evaluation factor.

6. The pressure vessel data quality assessment method based on big data according to claim 1 is characterized in that: The operation quality grade of the pressure vessel is calculated by multi-scale fuzzy evaluation according to the physical operating condition evaluation factor and the abnormal operating condition type identification result to obtain the equipment data quality evaluation result, including: Dividing the physical working condition evaluation factor into time scales, constructing a second-level evaluation domain, a minute-level evaluation domain, and an hour-level evaluation domain, and obtaining a multi-scale evaluation index; The abnormality type discrimination result of the working condition is used to perform abnormality frequency statistics in the second-level evaluation domain, to perform abnormality persistence analysis in the minute-level evaluation domain, and to perform abnormality periodicity analysis in the hour-level evaluation domain to obtain a time scale characteristic value; A fuzzy membership function is established according to the time scale characteristic value, and the operation quality of the pressure vessel is divided into a high-quality operation range, a normal operation range, an abnormal operation range and a dangerous operation range to obtain a fuzzy evaluation space; Mapping the multi-scale evaluation index to the fuzzy evaluation space, calculating the membership value of each operating interval, and obtaining a fuzzy evaluation vector; Performing multi-scale weighted fusion on the fuzzy evaluation vector to generate a quality grade score value; The quality grade score values ​​are graded and divided according to a set threshold to obtain the equipment data quality assessment result.

7. The pressure vessel data quality assessment method based on big data according to claim 6 is characterized in that: The fuzzy membership function is established according to the time scale characteristic value, and the operation quality of the pressure vessel is divided into a high-quality operation range, a normal operation range, an abnormal operation range and a dangerous operation range, and a fuzzy evaluation space is obtained, including: Performing statistical analysis on the second-level abnormal frequency in the time scale characteristic value, dividing the abnormal frequency less than 5 times / second into the low frequency value range, the abnormal frequency between 5 and 10 times / second into the medium frequency value range, and the abnormal frequency greater than 10 times / second into the high frequency value range, to obtain the frequency distribution characteristics; Statistically analyzing the minute-level abnormal persistence in the time scale characteristic value, dividing the abnormal duration of less than 1 minute into an instantaneous abnormal domain, the abnormal duration of 1-5 minutes into a short-term abnormal domain, and the abnormal duration of more than 5 minutes into a continuous abnormal domain, to obtain a persistence feature; Performing statistical analysis on the hourly abnormal periodicity in the time scale characteristic value, dividing the abnormal period less than 2 hours into a high-frequency abnormal domain, the abnormal period between 2 and 6 hours into a medium-frequency abnormal domain, and the abnormal period greater than 6 hours into a low-frequency abnormal domain, to obtain periodic characteristics; Establishing a membership function rule according to the frequency distribution feature, the persistence feature and the period feature to obtain an interval membership relationship; Mapping the interval membership relationship into an operation quality interval, constructing the membership boundaries of a high-quality operation interval, a normal operation interval, an abnormal operation interval, and a dangerous operation interval, and obtaining an evaluation interval matrix; The evaluation interval matrix is ​​subjected to space mapping transformation to obtain the fuzzy evaluation space.

8. A pressure vessel data quality assessment system based on big data, used to implement the pressure vessel data quality assessment method based on big data as described in any one of claims 1 to 7, characterized in that: The pressure vessel data quality assessment system based on big data includes: A synchronization module is used to obtain a physical working condition mapping data set through physical constraint preprocessing and working condition synchronization processing based on working condition data collected from the pressure vessel, including vessel internal pressure data, wall temperature data and stress-strain data; A calculation module, used for obtaining a physical working condition feature set by calculating a physical constraint index, a working condition stability index and a data coordination index according to the physical working condition mapping data set; An analysis module is used to analyze the physical coupling relationship and change trend between the physical working condition mapping data sets through a deep neural network driven by physical knowledge according to the physical working condition feature set, so as to obtain a result of distinguishing the type of working condition abnormality; A construction module is used to construct a scoring matrix through physical constraint scoring according to the abnormal working condition type discrimination result, and dynamically calibrate the scoring matrix to obtain a physical working condition evaluation factor; The evaluation module is used to calculate the operation quality grade of the pressure vessel through multi-scale fuzzy evaluation according to the physical operating condition evaluation factor and the abnormal operating condition type discrimination result, so as to obtain the equipment data quality evaluation result.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor implements the pressure vessel data quality assessment method based on big data as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the processor is enabled to execute the pressure vessel data quality assessment method based on big data as claimed in any one of claims 1 to 7.

Images