A data processing method and system for digital twin oil and gas pipeline

Abstract

The present invention relates to the technical field of pipeline data processing, and discloses a data processing method and system for digital twin oil and gas pipelines, wherein the system comprises: an acquisition and recording module, which acquires equipment operation data according to the number of equipment and data acquisition interval, records the fault type and fault time, and acquires fault symptom data before the fault time; a similarity judgment module, which calculates a first similarity by comparing the similarity between the operation data and the fault symptom data, and calculates a similarity threshold according to the severity of the equipment fault type; a cycle adjustment module, which adjusts the data acquisition interval; a fault confirmation module, which calculates a second similarity; a risk index is calculated based on an association rule, and if the risk index is greater than a preset risk threshold, it is judged that there is a potential fault; and a fault warning module, which sends a warning signal and matches the fault type. The system improves the accuracy and automation of oil and gas pipeline fault prediction and warning. The present invention improves the fault prediction and warning capabilities.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline data processing, and in particular to a data processing method and system for digital twin oil and gas pipelines. Background Art

[0002] Digital twin technology has been widely used in the operation and management of oil and gas pipelines. It collects real-time operating data of oil and gas pipelines (such as pressure, temperature, flow, vibration, etc.) through a sensor network, and combines it with the physical model and simulation algorithm of the pipeline to reflect the dynamic operating status of the pipeline in real time.

[0003] Currently, digital twin models are mainly used for equipment operation status monitoring and fault visualization analysis. With the help of digital twin models, operation and maintenance personnel can intuitively understand the operation status of equipment, analyze the causes of faults, and evaluate the long-term operation risks of pipelines.

[0004] Although the digital twin model has the function of real-time visualization, there are still technical limitations in practical applications: the current digital twin model is more used to monitor the existing status, lacks the ability to early warn of potential failures, and cannot actively identify and predict possible failure risks. In addition, potential failures rely on manual analysis and experience judgment, and lack the ability of automated intelligent judgment.

[0005] Therefore, how to improve the fault prediction capability of the digital twin model of oil and gas pipelines and achieve early warning of potential faults has become an important issue that needs to be solved urgently. Summary of the invention

[0006] In view of this, the present invention proposes a data processing method and system for a digital twin oil and gas pipeline, aiming to solve the problem of low fault prediction ability of the digital twin model of the oil and gas pipeline in the current technology.

[0007] The present invention proposes a data processing system for a digital twin oil and gas pipeline, comprising:

[0008] The acquisition and recording module is configured to determine the number of devices in the digital twin model and the data acquisition interval, and to collect the operation data of the devices based on the time series; when a device fails, the fault type and the time of the failure are recorded, and the operation data before the time of the failure is collected based on the time series and recorded as the fault symptom data;

[0009] A similarity judgment module is configured to continuously select the operation data, judge the similarity between the operation data and the fault symptom data, and obtain a first similarity; calculate a similarity threshold according to the severity of the current equipment fault type, and compare the first similarity with the similarity threshold;

[0010] a period adjustment module, configured to obtain a comparison result between the first similarity and a similarity threshold, and when the first similarity is greater than or equal to the similarity threshold, adjust the data acquisition interval;

[0011] a fault confirmation module, configured to continue to collect the operation data based on the adjusted data acquisition interval, and continue to determine the similarity between the operation data and the fault symptom data to obtain a second similarity;

[0012] When the second similarity is greater than or equal to the similarity threshold, calculating the risk index of the current device based on the association rule, and if the risk index is greater than or equal to a preset risk threshold, determining that there is a potential fault;

[0013] The fault warning module is configured to obtain the fault type matching the fault symptom data and issue a warning signal when it is determined that there is a potential fault.

[0014] Furthermore, the acquisition and recording module is further configured to determine the acquisition length of the fault symptom data before acquiring the fault symptom data based on the time series;

[0015] The acquisition length is determined by the following method:

[0016] The operation data of the equipment during normal operation is selected, and the standard deviation in the sliding window is calculated; when there are three consecutive standard deviations of the operation data acquired in real time that are greater than or equal to the standard deviation in the sliding window, the first operation data in the time series is determined to be the first fault symptom data; the acquisition length is the operation data from the first fault symptom data to the time before the fault;

[0017] The standard deviation within the sliding window is calculated as follows:

[0018] ;

[0019] is the standard deviation within the sliding window, n is the window size, For the The running data of the test, is the mean of the window.

[0020] Furthermore, when determining the similarity between the operation data and the fault symptom data, it includes:

[0021] The operation data is continuously selected, and the operation data and the fault symptom data are calculated according to the following relationship to obtain a first similarity:

[0022] ;

[0023] in, is the first similarity, is the average value of the running data, is the average value of the fault symptom data, is the fault symptom data detected for the i-th time.

[0024] Furthermore, when calculating the similarity threshold according to the severity of the current equipment fault type, it includes:

[0025] Calculate the impact value of the impact range of the fault type, the impact value of the fault frequency, the impact value of the repair difficulty and the impact value of the production impact to obtain a comprehensive severity index of the fault type; pre-set an initial similarity threshold, and calculate the similarity threshold based on the comprehensive severity index and the initial similarity threshold;

[0026] The comprehensive severity index is obtained through the following relationship:

[0027] ;

[0028] ;

[0029] ;

[0030] ;

[0031] ;

[0032] in, is the comprehensive severity index of the current fault type, =4, indicating the number of factors that affect the severity of the fault type, For each fault type, is the impact value of the i-th factor on the time series; is the impact value of the impact range, is the average value of the pipeline length affected by the current fault type, is the maximum value of the pipeline length affected by the current fault type, is the number of devices connected to the current device, is the impact value of the failure frequency, F is the number of equipment failures within a week, is the total number of equipment failures, is the impact value of the repair difficulty, T is the average fault repair time under the current fault type, is the historical maximum fault repair time for the current fault type, is the impact value of production impact, is the average economic loss value under the previous fault type, The historical maximum economic loss value for the current fault type.

[0033] Furthermore, an initial similarity threshold is preset, and the similarity threshold is calculated based on the comprehensive severity index and the initial similarity threshold, and the similarity threshold is obtained through the following relationship:

[0034] ;

[0035] in, is the similarity threshold, is the initial similarity threshold, r is the adjustment coefficient, which is used to adjust the influence of the comprehensive severity index on the similarity threshold, and 0.1≤r≤0.3.

[0036] Further, when the first similarity is greater than or equal to the similarity threshold, adjusting the data acquisition interval includes:

[0037] The data acquisition interval is adjusted according to the comprehensive severity index, the similarity threshold and the first similarity:

[0038] ;

[0039] Where T is the adjusted data acquisition interval, Get the interval for the initial data.

[0040] Furthermore, the risk index of the current device is calculated based on the association rules, including:

[0041] The operating data under each of the fault types is obtained and the operating data after the fault moment is recorded as fault data, key features are extracted and a historical association rule base is established, rule matching is performed on the fault symptom data, and the risk index is calculated according to the matching degree between the fault symptom data and the data in the historical association rule base.

[0042] Furthermore, the risk index is calculated by the following relationship:

[0043] ;

[0044] in, is the risk index; is the matching degree of the jth rule, indicating the similarity between the current fault symptom data and the historical rules; is the confidence of the jth rule; is the support of the jth rule, indicating the frequency of occurrence of the rule in historical data; is the total number of rules.

[0045] Furthermore, before the fault warning module sends out a warning signal, a warning level is determined according to the risk index, and the value of the risk index is in direct proportion to the warning level.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The standard deviation is calculated through the sliding window method, and the change of the standard deviation is used to accurately identify the starting point of the fault symptom data. Through this mechanism, the automatic collection of fault symptom data can be realized without relying on manual judgment, which improves the system's automation level and real-time response capability.

[0048] Based on the similarity judgment method, the real-time collected operation data is compared with the historical fault symptom data, the similarity is calculated, and the similarity threshold is calculated according to the severity of the equipment fault type, realizing the function of dynamically adjusting the similarity threshold. The system can dynamically adjust the sensitivity of fault warning according to the actual situation, improving the accuracy and timeliness of the prediction.

[0049] By building a historical association rule base, extracting key features of historical fault data, and combining it with the fault symptom data of the current equipment for rule matching, the risk index is calculated using the matching degree. This can better quantify the risk of faults, avoid the defects of over-reliance on experience and manual judgment in traditional fault prediction methods, and make risk assessment more accurate and automated.

[0050] By combining the risk index with the preset risk threshold, the present invention can intelligently send out warning signals and automatically adjust the warning level according to the value of the risk index. The warning level is directly proportional to the risk index, thus being more flexible and accurate.

[0051] On the other hand, the present invention also provides a data processing method for a digital twin oil and gas pipeline, which is applied to the above system and includes:

[0052] S1: Determine the number of devices in the digital twin model and the data acquisition interval, and collect the operating data of the devices based on the time series; when a device fails, record the type of failure and the time of the failure, and collect the operating data before the time of the failure based on the time series, which is recorded as the fault symptom data;

[0053] S2: continuously selecting the operation data, determining the similarity between the operation data and the fault symptom data, and obtaining a first similarity; calculating a similarity threshold according to the severity of the current equipment fault type, and comparing the first similarity with the similarity threshold;

[0054] S3: obtaining a comparison result between the first similarity and a similarity threshold, and when the first similarity is greater than or equal to the similarity threshold, adjusting the data acquisition interval;

[0055] S4: continuing to collect the operating data based on the adjusted data acquisition interval, and continuing to determine the similarity between the operating data and the fault symptom data to obtain a second similarity;

[0056] When the second similarity is greater than or equal to the similarity threshold, calculating the risk index of the current device based on the association rule, and if the risk index is greater than or equal to a preset risk threshold, determining that there is a potential fault;

[0057] S5: When it is determined that there is a potential fault, the fault type matching the fault symptom data is obtained, and a warning signal is issued.

[0058] It can be understood that the above-mentioned data processing method and system for digital twin oil and gas pipelines have the same beneficial effects and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0060] Figure 1 This is a functional framework diagram of a data processing system for a digital twin oil and gas pipeline according to an embodiment of the present invention;

[0061] Figure 2 This is a flow chart of a data processing system for a digital twin oil and gas pipeline according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0063] See also Figure 1 As shown, an embodiment of the present invention provides a data processing system for a digital twin oil and gas pipeline, comprising:

[0064] The collection and recording module is configured to determine the number of devices in the digital twin model and the data acquisition interval, and collect the operation data of the devices based on the time series; when a device fails, the fault type and the time of the failure are recorded, and the operation data before the time of the failure is collected based on the time series and recorded as the fault symptom data;

[0065] The fault type can be obtained by analyzing the data after the fault occurs through a fault diagnosis system or method, or the user can directly add the fault type. This application is aimed at the analysis and early warning before the fault occurs, so it will not be repeated.

[0066] The similarity judgment module is configured to continuously select operation data, judge the similarity between the operation data and the fault symptom data, and obtain a first similarity; calculate a similarity threshold according to the severity of the current equipment fault type, and compare the first similarity with the similarity threshold;

[0067] a period adjustment module, configured to obtain a comparison result between the first similarity and the similarity threshold, and adjust the data acquisition interval when the first similarity is greater than or equal to the similarity threshold;

[0068] A fault confirmation module is configured to continue to collect operation data based on the adjusted data acquisition interval, and continue to determine the similarity between the operation data and the fault symptom data to obtain a second similarity;

[0069] When the second similarity is greater than or equal to the similarity threshold, the risk index of the current device is calculated based on the association rule, and if the risk index is greater than or equal to the preset risk threshold, it is determined that there is a potential fault;

[0070] The fault warning module is configured to obtain a fault type that matches the fault symptom data and issue a warning signal when it is determined that there is a potential fault.

[0071] It should be noted that by collecting and recording the operating data of the equipment in real time and recording the time and type of the fault, comprehensive control of the equipment status is effectively achieved.

[0072] The similarity judgment module can identify abnormal performance of equipment in advance by judging the similarity between operating data and fault symptom data, avoiding the limitations of traditional reliance on manual experience.

[0073] The cycle adjustment module can dynamically adjust the frequency of data collection according to the status of the equipment, improve the timeliness and accuracy of the data, and optimize the utilization of monitoring resources.

[0074] The fault confirmation module ensures accurate fault confirmation and avoids false alarms or missed alarms through continuous data collection and similarity comparison.

[0075] The risk calculation module combines historical data and association rules to quantitatively evaluate the failure risk of equipment, provide more accurate failure warnings, and reduce potential risks.

[0076] The fault warning module helps operation and maintenance personnel take countermeasures in advance by issuing warning signals, thus avoiding equipment failures and improving the safety and reliability of oil and gas pipelines.

[0077] It should be noted that operating data refers to:

[0078] Pressure: The pressure of a fluid in a pipeline, including the pressure of a pipeline that transports gas or liquid.

[0079] Temperature: The temperature of the fluid in the pipe.

[0080] Flow: The flow rate or volume of a fluid (such as natural gas or oil) in a pipeline.

[0081] Vibration: Pipeline vibration data to monitor possible mechanical failure or corrosion.

[0082] Stress / strain: The stress on the pipeline under different working conditions, monitoring whether the pipeline is in an overloaded state.

[0083] Pipe bending / displacement: Changes in the geometric shape of the pipeline to detect whether the pipeline is deformed or damaged.

[0084] Equipment operation data

[0085] Current / Voltage: Changes in current and voltage of driving equipment (such as pumps, valves, compressors, etc.) are used to monitor the status of electrical equipment.

[0086] Motor load: The load condition of the driving device, reflecting whether the operating load of the equipment is abnormal.

[0087] Equipment temperature: The temperature of the drive equipment or important mechanical parts (such as motors, pumps, compressors, etc.) to determine whether the equipment is overloaded or overheated.

[0088] Vibration signal: The vibration frequency and amplitude of key equipment such as motors, compressors, and pumps. Abnormal vibration signals are usually a precursor to equipment failure.

[0089] Environmental monitoring data

[0090] External Temperature: The temperature of the environment surrounding the pipeline, which may affect the state of the fluid in the pipeline and the corrosion resistance of the pipeline.

[0091] Humidity: The humidity of the environment affects the corrosion rate and equipment operation.

[0092] Atmospheric pressure: The impact on gas transportation, especially in high altitude areas, data such as pressure and flow in the pipeline may be affected by changes in atmospheric pressure.

[0093] It should be noted that the data should be divided into the following categories in terms of time: normal operation data, fault symptom data, and fault data after a device failure.

[0094] In some embodiments of the present application, the collection and recording module is further configured to determine the collection length of the fault symptom data before collecting the fault symptom data based on the time series;

[0095] The acquisition length is determined by:

[0096] Select the operation data of the equipment during normal operation and calculate the standard deviation in the sliding window; when the standard deviation of the real-time operation data is greater than or equal to the standard deviation in the sliding window for three consecutive times, determine that the first operation data in the time series is the first fault symptom data; collect the operation data from the first fault symptom data to the time of the fault;

[0097] The standard deviation within the sliding window is calculated as follows:

[0098] ;

[0099] is the standard deviation within the sliding window, n is the window size, For the The running data of the test, is the mean of the window.

[0100] It should be noted that the data during the normal operation of the equipment is selected: first, a certain period of time is selected from the normal operation data of the equipment. This data segment represents the operation of the equipment when no failure occurs.

[0101] Calculate the standard deviation of the sliding window: In the selected normal operation data, the standard deviation is calculated step by step using the sliding window technique. The sliding window is a fixed-size time window, and the volatility of the data is measured by calculating the standard deviation of the data in the window.

[0102] Determine the relationship between the standard deviation of real-time data and the standard deviation of the window: When the standard deviation of the real-time collected operating data is greater than or equal to the standard deviation in the sliding window for three consecutive times, it is determined that the equipment operation status is abnormal. At this time, the current real-time data can be regarded as the "first fault sign data". The key to this judgment method is to monitor the volatility of the equipment operation data. Once the fluctuation of the data exceeds the normal range, the initial signs of fault can be discovered in time.

[0103] Determine the acquisition length: When the "first fault sign data" is confirmed, the acquisition length starts from this data point and continues until the equipment fails. Determining the acquisition length ensures that the acquired data can reflect the dynamic changes before the equipment fails and provide necessary data support for subsequent fault analysis.

[0104] Identify signs of failure in advance: By monitoring the volatility of equipment operation data, it is possible to identify signs of potential equipment failure in advance. This method does not rely on traditional manual experience or simple static thresholds, and can more accurately capture subtle changes in equipment status.

[0105] Improve the accuracy of data collection: Sliding window technology combined with standard deviation calculation makes data collection more flexible and accurate, avoiding the problem of collecting fault symptom data too early or too late, and can more effectively capture dynamic changes related to faults.

[0106] Optimize the accuracy of fault prediction: By reasonably determining the collection length of fault symptom data, it is ensured that the collected data can fully reflect the operating status of the equipment, and improve the accuracy of subsequent fault analysis and risk prediction.

[0107] Enhanced real-time and adaptability: This judgment method based on real-time standard deviation can dynamically adapt to changes in the operating status of the equipment. Compared with the traditional static threshold method, it is more real-time and adaptable.

[0108] In some embodiments of the present application, when determining the similarity between the operation data and the fault symptom data, the following steps are included:

[0109] Continuously select the operation data, and calculate the operation data and the fault symptom data according to the following relationship to obtain the first similarity:

[0110] ;

[0111] in, is the first similarity, is the average value of the running data, is the average value of the fault symptom data, is the fault symptom data detected for the i-th time.

[0112] It should be noted that the above formula can effectively measure the numerical correlation between operating data and fault symptom data, providing a quantitative way to evaluate whether fault symptoms have occurred, which helps to reduce false alarm rates and missed alarm rates. Similarity calculation is not only applicable to static data, but also can evaluate the dynamic changes of operating data and fault symptom data in real time, providing comprehensive support for monitoring equipment status. Similarity calculation eliminates the influence of the average value, making it independent of the specific magnitude of the data, and is applicable to sensor data of different types or magnitudes. The similarity index provides a quantitative reference, which can help the system quickly determine the degree of match between operating data and historical fault symptom data, and improve the accuracy of fault prediction.

[0113] In some embodiments of the present application, when calculating the similarity threshold according to the severity of the current device fault type, it includes:

[0114] Calculate the impact value of the fault type's impact range, the impact value of the fault frequency, the impact value of the repair difficulty, and the impact value of the production impact to obtain a comprehensive severity index of the fault type; pre-set an initial similarity threshold, and calculate based on the comprehensive severity index and the initial similarity threshold to obtain a similarity threshold;

[0115] The comprehensive severity index is obtained through the following relationship:

[0116] ;

[0117] ;

[0118] ;

[0119] ;

[0120] ;

[0121] in, is the comprehensive severity index of the current fault type, =4, indicating the number of factors that affect the severity of the fault type, For each fault type, is the impact value of the i-th factor on the time series; is the impact value of the impact range, is the average value of the pipeline length affected by the current fault type, is the maximum value of the pipeline length affected by the current fault type, is the number of devices connected to the current device, is the impact value of the failure frequency, F is the number of equipment failures within a week, is the total number of equipment failures, is the impact value of the repair difficulty, T is the average fault repair time under the current fault type, is the historical maximum fault repair time for the current fault type, is the impact value of production impact, is the average economic loss value under the previous fault type, The historical maximum economic loss value for the current fault type.

[0122] It should be noted that the method of calculating similarity thresholds based on the severity of the current equipment fault type takes into account multiple influencing factors, including the impact range, fault frequency, repair difficulty, and production impact. Through the quantitative analysis of these factors, the comprehensive severity of the fault can be fully reflected, ensuring that the threshold setting is more scientific and applicable.

[0123] The calculation of the comprehensive severity index adopts the weighted average method, which superimposes the contribution of different factors to the severity of the fault according to the weight. Through the analysis of time series data, the impact value of each factor can dynamically reflect the fault characteristics, thereby improving the accuracy of the calculation.

[0124] The impact value of impact range takes into account the impact of the fault on the length of the pipeline and the number of connected devices. This method can quantify the extent of the fault's impact within the physical scope. The impact value of fault frequency measures the commonness of the fault by the frequency of equipment failure. The impact value of repair difficulty is based on the statistical data of repair time, reflecting the resources and time required to deal with the fault. The impact value of production impact combines economic loss assessment to enable the model to strike a balance between safety and economy.

[0125] Dynamically adjusting the initial similarity threshold based on the comprehensive severity index ensures the flexibility of the system in different scenarios. The higher the severity index, the higher the similarity threshold, and the sensitivity of the system is correspondingly enhanced, thereby achieving early response to high-risk fault types. This method improves the accuracy and reliability of fault monitoring while taking into account system operation efficiency and economic benefits.

[0126] In some embodiments of the present application, an initial similarity threshold is preset, and the similarity threshold is calculated based on the comprehensive severity index and the initial similarity threshold, and the similarity threshold is obtained by the following relationship:

[0127] ;

[0128] in, is the similarity threshold, is the initial similarity threshold, r is the adjustment coefficient, which is used to adjust the influence of the comprehensive severity index on the similarity threshold, and 0.1≤r≤0.3.

[0129] It should be noted that the calculation formula of the similarity threshold is based on the comprehensive severity index and the initial similarity threshold, and the threshold change range is dynamically adjusted by introducing an adjustment coefficient. This design takes into account both flexibility and stability, allowing the system to adjust the response mechanism according to the severity of different fault types.

[0130] The initial similarity threshold is a pre-set fixed reference value, which is used as a basic standard to represent the default fault detection sensitivity when there is no additional information. The comprehensive severity index provides a differentiated weight reference for each fault type based on the multi-factor analysis results described above, so that the dynamic adjustment of the threshold has a reasonable basis.

[0131] The introduction of the adjustment coefficient further enhances the adaptability of the model. By limiting the value range of rrr to between 0.1 and 0.3, it ensures that the threshold adjustment amplitude will not be too large or too small. This design avoids the system being overly sensitive or unresponsive due to fluctuations in the comprehensive severity index, thereby maintaining the stability of the monitoring system.

[0132] The final similarity threshold is the result of combining the initial threshold and the severity index. Its dynamic adjustment method can improve the monitoring sensitivity of high-risk faults under complex working conditions, while avoiding resource waste or misjudgment caused by excessive early warning. This mechanism effectively improves the reliability and accuracy of the system and lays the foundation for intelligent fault detection of digital twin oil and gas pipelines.

[0133] In some embodiments of the present application, when the first similarity is greater than or equal to the similarity threshold, adjusting the data acquisition interval includes:

[0134] Adjust the data acquisition interval according to the comprehensive severity index, similarity threshold and first similarity:

[0135] ;

[0136] Where T is the adjusted data acquisition interval, Get the interval for the initial data.

[0137] It should be noted that the calculation method of the second similarity is the same as that of the first similarity.

[0138] Dynamic reflection of fault risk: When the first similarity approaches or exceeds the similarity threshold, it indicates that the equipment operation data may gradually deviate from the normal state and there is a potential fault risk. At this time, by shortening the data acquisition interval, the system can monitor the equipment status more intensively, thereby capturing more fine-grained change trends.

[0139] Balance monitoring efficiency and resource usage: If the first similarity is low, it means that the similarity between the equipment operating status and the fault symptoms is small and the risk is low. The system will appropriately extend the data acquisition interval and reduce the monitoring frequency to optimize resource usage and reduce system load.

[0140] Real-time adjustment capability: The proportional relationship in the formula enables the data acquisition interval to change dynamically according to the real-time calculation results, achieving rapid response. This real-time performance is particularly critical for oil and gas pipeline monitoring in complex operating environments.

[0141] Through the above mechanism, the system increases monitoring efforts when the risk of failure is high and reduces resource consumption when the risk is low, thereby improving resource utilization efficiency and overall operating performance while ensuring fault warning capabilities.

[0142] In some embodiments of the present application, the risk index of the current device is calculated based on the association rule, including:

[0143] The operating data under each fault type is obtained and the operating data after the fault moment is recorded as the fault data. The key features are extracted and a historical association rule base is established. The fault symptom data is matched with rules, and the risk index is calculated according to the matching degree between the fault symptom data and the data in the historical association rule base.

[0144] It is understandable that in the method of calculating the equipment risk index based on association rules, it is first necessary to extract the time series data of each fault type from the equipment operation data. When a device fails, the corresponding fault type and fault time are recorded, and the operation data within a period of time after the failure is marked as fault data. This step provides a labeled data basis for subsequent analysis.

[0145] Next, the labeled fault data is analyzed to extract key variables that can characterize the fault characteristics. For example, indicators such as temperature, pressure, and vibration amplitude may show abnormal trends. These characteristic variables are screened out through signal processing or machine learning algorithms, laying the foundation for subsequent rule mining.

[0146] Then, by analyzing the characteristics of the fault data and its fault symptom data, a historical association rule base is generated. These rules reflect the common operating mode changes of the equipment before the fault occurs. For example, a rule may be described as "when the pressure change rate is greater than a certain value, it may cause a specific type of fault." These rules can be automatically generated by the association rule mining algorithm and optimized and classified according to the actual scenario.

[0147] In the real-time monitoring phase, the fault symptom data of the current device is matched with the rules in the historical association rule base. The matching process evaluates the similarity between the current data and each rule, usually by calculating the similarity or matching degree. The matching result is used to determine the closeness of the current device's operating status to the fault mode.

[0148] Finally, the risk index of the device is calculated based on the matching degree. The risk index is calculated by weighting the matching degree and the weight of the corresponding rule to comprehensively reflect the failure risk of the current operating state. The weight is determined by the importance of the rule and the impact of the failure, and the risk index is used to quantify the threat level of potential failures.

[0149] In some embodiments of the present application, the risk index is calculated by the following relationship:

[0150] ;

[0151] in, is the risk index; is the matching degree of the jth rule, indicating the similarity between the current fault symptom data and the historical rules; is the confidence of the jth rule; is the support of the jth rule, indicating the frequency of occurrence of the rule in historical data; is the total number of rules.

[0152] It should be noted that the calculation of the risk index depends on several key factors, such as rule matching, rule confidence and support. Each rule can reflect the correlation strength between certain fault symptom data and historical data, and the risk index combines these correlation strengths to quantify the potential fault risk.

[0153] The calculation steps are as follows:

[0154] Matching degree: For each rule (assuming it is the jth rule), firstly, the matching degree is calculated by comparing it with the fault symptom data of the current device. The matching degree reflects the similarity between the current data and the rule in terms of features, patterns, etc. A high matching degree indicates that the current device operation status is very similar to the historical fault symptoms, and there may be a higher risk.

[0155] Confidence: Confidence is a measure of the reliability of a rule, indicating the frequency of occurrence of the rule in historical data. Specifically, the higher the confidence of a rule, the more fault symptom data and corresponding faults have appeared in the past, and the stronger the predictive ability of the rule. A high confidence indicates that the rule has a strong reliability in predicting future faults.

[0156] Support: Support measures the frequency of a rule appearing in historical data, usually expressed as the proportion of the failure mode described by the rule in all historical data. Rules with high support are usually more universal and can be applied to more devices or failure types.

[0157] Risk Index: Ultimately, the risk index combines the match, confidence, and support in a weighted sum to produce a comprehensive risk score. The total number of rules affects the final value of the risk index, but the weight of each rule (match, confidence, support) determines its impact on the final result.

[0158] By combining the matching degree, confidence and support of the rules in the historical data, the failure risk of the current equipment can be assessed more accurately. This calculation method avoids judgments based solely on data changes and improves the accuracy of predictions.

[0159] As equipment operation data is continuously updated, the matching degree, confidence level and support level of the rules will continue to change, and the risk index can be adjusted at any time to promptly reflect the changing trend of potential failures and ensure the real-time and accuracy of the early warning mechanism.

[0160] By comprehensively considering multiple dimensions of the rules, false positives and false negatives can be reduced. Rules with high matching degree, low confidence and support will contribute less to the risk index, thus improving the reliability and effectiveness of the judgment.

[0161] In some embodiments of the present application, before the fault warning module sends out a warning signal, the warning level is determined according to the risk index, and the value of the risk index is directly proportional to the warning level.

[0162] It is understandable that the corresponding relationship between the risk index and the warning level is: the higher the risk index value, the greater the possibility of equipment failure and the higher the severity of the failure. In this case, the system will convert the risk index into the corresponding warning level according to the set proportional relationship. For example, if the risk index exceeds a certain threshold, a higher warning level will be triggered, indicating that immediate processing or intervention is required.

[0163] Classification of warning levels: Usually, warning levels can be divided into multiple levels, such as "low risk", "medium risk" and "high risk" or "green", "yellow" and "red" levels. Each level corresponds to a different response strategy. Low risk levels may only prompt equipment monitoring, medium risk levels may require further inspection, and high risk levels require immediate action to avoid potential failures.

[0164] Issuance of early warning signals: After determining the early warning level, the system will issue corresponding early warning signals according to the set strategy. Low risk levels may only issue warnings, while high risk levels may issue emergency alarms, requiring equipment maintenance personnel to immediately check the equipment and take necessary preventive measures.

[0165] Timely response: By directly linking the risk index with the warning level, the system can promptly issue corresponding warning signals based on the risk status of the equipment, thereby avoiding the impact of potential equipment failures on production.

[0166] Dynamically adjust emergency response: Based on the risk index calculated in real time, the warning level can reflect the urgency of different faults, helping operation and maintenance personnel to respond to different situations in a targeted manner, thereby optimizing resource allocation and emergency measures.

[0167] Accurate fault diagnosis and processing: The system can adjust the warning level according to the risk index, thereby helping equipment maintenance personnel determine whether emergency repairs, inspections or other processing measures are needed to ensure that equipment problems are handled accurately and in a timely manner.

[0168] See also Figure 2 As shown, an embodiment of the present invention provides a data processing method for a digital twin oil and gas pipeline, comprising:

[0169] S1: Determine the number of devices in the digital twin model and the data acquisition interval, and collect the operating data of the devices based on the time series; when a device fails, record the fault type and time, and collect the operating data before the fault time based on the time series, which is recorded as fault symptom data;

[0170] S2: continuously selecting operation data, determining the similarity between the operation data and the fault symptom data, and obtaining a first similarity; calculating a similarity threshold according to the severity of the current equipment fault type, and comparing the first similarity with the similarity threshold;

[0171] S3: Obtain a comparison result between the first similarity and the similarity threshold, and when the first similarity is greater than or equal to the similarity threshold, adjust the data acquisition interval;

[0172] S4: continue to collect the operating data based on the adjusted data acquisition interval, and continue to determine the similarity between the operating data and the fault symptom data to obtain a second similarity;

[0173] When the second similarity is greater than or equal to the similarity threshold, the risk index of the current device is calculated based on the association rule, and if the risk index is greater than or equal to the preset risk threshold, it is determined that there is a potential fault;

[0174] S5: When it is determined that there is a potential fault, the fault type matching the fault symptom data is obtained and a warning signal is issued.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A data processing system for digital twin oil and gas pipelines, characterized in that: include: A collection and recording module is configured to determine the number of devices in the digital twin model and the data acquisition interval, and collect the operation data of the devices based on the time series; When a device fails, the fault type and time are recorded, and the operating data before the fault time is collected based on the time series and recorded as fault symptom data; A similarity judgment module is configured to continuously select the operation data, judge the similarity between the operation data and the fault symptom data, and obtain a first similarity; calculate a similarity threshold according to the severity of the current equipment fault type, and compare the first similarity with the similarity threshold; a period adjustment module, configured to obtain a comparison result between the first similarity and a similarity threshold, and when the first similarity is greater than or equal to the similarity threshold, adjust the data acquisition interval; a fault confirmation module, configured to continue to collect the operation data based on the adjusted data acquisition interval, and continue to determine the similarity between the operation data and the fault symptom data to obtain a second similarity; When the second similarity is greater than or equal to the similarity threshold, calculating the risk index of the current device based on the association rule, and if the risk index is greater than or equal to a preset risk threshold, determining that there is a potential fault; A fault warning module is configured to obtain the fault type matching the fault symptom data and issue a warning signal when it is determined that there is a potential fault; Wherein, the acquisition and recording module is further configured to determine the acquisition length of the fault symptom data before acquiring the fault symptom data based on the time series; The acquisition length is determined by the following method: The operation data of the equipment during normal operation is selected, and the standard deviation in the sliding window is calculated; when there are three consecutive standard deviations of the operation data acquired in real time that are greater than or equal to the standard deviation in the sliding window, the first operation data in the time series is determined to be the first fault symptom data; the acquisition length is the operation data from the first fault symptom data to the time before the fault; The standard deviation within the sliding window is calculated as follows: ； is the standard deviation within the sliding window, n is the window size, For the The running data of the test, is the mean of the window; When calculating similarity thresholds based on the severity of the current equipment fault type, including: Calculate the impact value of the impact range of the fault type, the impact value of the fault frequency, the impact value of the repair difficulty and the impact value of the production impact to obtain a comprehensive severity index of the fault type; pre-set an initial similarity threshold, and calculate the similarity threshold based on the comprehensive severity index and the initial similarity threshold; The comprehensive severity index is obtained through the following relationship: ； ； ； ； ； in, is the comprehensive severity index of the current fault type, =4, indicating the number of factors that affect the severity of the fault type, For each fault type, is the impact value of the i-th factor on the time series; is the impact value of the impact range, is the average value of the pipeline length affected by the current fault type, is the maximum value of the pipeline length affected by the current fault type, is the number of devices connected to the current device, is the impact value of the failure frequency, F is the number of equipment failures within a week, is the total number of equipment failures, is the impact value of the repair difficulty, T is the average fault repair time under the current fault type, is the historical maximum fault repair time for the current fault type, is the impact value of production impact, is the average economic loss value under the previous fault type, The historical maximum economic loss value for the current fault type.

2. The data processing system for digital twin oil and gas pipeline according to claim 1, characterized in that: When judging the similarity between the operation data and the fault symptom data, it includes: The operation data is continuously selected, and the operation data and the fault symptom data are calculated according to the following relationship to obtain a first similarity: ； in, is the first similarity, is the average value of the running data, is the average value of the fault symptom data, is the fault symptom data detected for the i-th time.

3. The data processing system for digital twin oil and gas pipeline according to claim 1, characterized in that: An initial similarity threshold is preset, and the similarity threshold is calculated based on the comprehensive severity index and the initial similarity threshold, and the similarity threshold is obtained through the following relationship: ； in, is the similarity threshold, is the initial similarity threshold, r is the adjustment coefficient, which is used to adjust the influence of the comprehensive severity index on the similarity threshold, and 0.1≤r≤0.

3.

4. The data processing system for digital twin oil and gas pipeline according to claim 3 is characterized in that: When the first similarity is greater than or equal to the similarity threshold, adjusting the data acquisition interval includes: The data acquisition interval is adjusted according to the comprehensive severity index, the similarity threshold and the first similarity: ； Where T is the adjusted data acquisition interval, Get the interval for the initial data.

5. The data processing system for digital twin oil and gas pipeline according to claim 4, characterized in that: Calculate the risk index of the current device based on association rules, including: The operating data under each of the fault types is obtained and the operating data after the fault moment is recorded as fault data, key features are extracted and a historical association rule base is established, rule matching is performed on the fault symptom data, and the risk index is calculated according to the matching degree between the fault symptom data and the data in the historical association rule base.

6. The data processing system for digital twin oil and gas pipeline according to claim 5, characterized in that: The risk index is calculated by the following relationship: ； in, is the risk index; is the matching degree of the jth rule, indicating the similarity between the current fault symptom data and the historical rules; is the confidence of the jth rule; is the support of the jth rule, indicating the frequency of occurrence of the rule in historical data; is the total number of rules.

7. The data processing system for digital twin oil and gas pipeline according to claim 6, characterized in that: Before the fault warning module sends out a warning signal, the warning level is determined according to the risk index, and the value of the risk index is in direct proportion to the warning level.

8. A data processing method for a digital twin oil and gas pipeline, applied to the data processing system for a digital twin oil and gas pipeline according to any one of claims 1 to 7, characterized in that: include: S1: Determine the number of devices in the digital twin model and the data acquisition interval, and collect the operating data of the devices based on the time series; When a device fails, the fault type and time are recorded, and the operating data before the fault time is collected based on the time series and recorded as fault symptom data; S2: continuously selecting the operation data, determining the similarity between the operation data and the fault symptom data, and obtaining a first similarity; calculating a similarity threshold according to the severity of the current equipment fault type, and comparing the first similarity with the similarity threshold; S3: obtaining a comparison result between the first similarity and a similarity threshold, and when the first similarity is greater than or equal to the similarity threshold, adjusting the data acquisition interval; S4: continuing to collect the operating data based on the adjusted data acquisition interval, and continuing to determine the similarity between the operating data and the fault symptom data to obtain a second similarity; When the second similarity is greater than or equal to the similarity threshold, calculating the risk index of the current device based on the association rule, and if the risk index is greater than or equal to a preset risk threshold, determining that there is a potential fault; S5: When it is determined that there is a potential fault, the fault type matching the fault symptom data is obtained, and a warning signal is issued.

Images