Fault-tolerant diagnosis method adopting dual-channel pressure sensor

Abstract

The invention provides a fault-tolerant diagnosis method adopting a dual-channel pressure sensor, and relates to the technical field of intelligent sensors, and the method comprises the steps: arranging the dual-channel pressure sensor on a target natural gas pipeline, and synchronously collecting and obtaining a first pressure signal sequence and a second pressure signal sequence; when any pressure signal reaches a preset pressure threshold value, signal deviation analysis is carried out, and a pressure signal deviation sequence is obtained; when the deviation value at any moment reaches a preset deviation threshold value, fault detection is carried out on the dual-channel pressure sensor, and a sensor detection result is obtained; when the deviation value reaching the preset deviation threshold value does not appear in the preset time interval, pipeline fault recognition and positioning are conducted, and a pipeline fault recognition result is obtained. The technical problems that in the prior art, pressure monitoring depends on a single sensor, effective fault diagnosis cannot be carried out or accurate data cannot be continuously provided when pressure abnormity occurs, pipeline faults cannot be found in time, and normal operation of a pipeline is affected are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent sensors, and particularly to a fault tolerance diagnosis method using a dual-channel pressure sensor. Background Art

[0002] As an important infrastructure for energy transportation, natural gas pipeline systems are widely used around the world. To ensure the safety and efficiency of natural gas transportation, it is necessary to monitor the pressure state inside the pipeline in real time and promptly detect possible fault problems. As one of the key components for pipeline monitoring, pressure sensors play a crucial role in pipeline safety monitoring. Traditional pressure monitoring methods usually rely on a single sensor. Although this method can meet the requirements to a certain extent, with the increasing complexity of the pipeline operating environment, this technology faces many challenges. Specifically, the existing technology relies on a single sensor for pressure monitoring, resulting in the inability to perform effective fault diagnosis or continue to provide accurate data when the sensor fails, thereby causing pipeline faults to go undetected in a timely manner and affecting the normal operation and safety of the pipeline. Summary of the Invention

[0003] This application provides a fault tolerance diagnosis method using a dual-channel pressure sensor, aiming to solve the technical problem that the existing technology relies on a single sensor for pressure monitoring, and when a pressure anomaly occurs, it is unable to perform effective fault diagnosis or continue to provide accurate data, thereby causing pipeline faults to go undetected in a timely manner and affecting the normal operation and safety of the pipeline.

[0004] A fault tolerance diagnosis method using a dual-channel pressure sensor disclosed in this application includes: arranging dual-channel pressure sensors at preset points of a target natural gas pipeline, synchronously collecting pressure data to obtain a first pressure signal sequence and a second pressure signal sequence; when any pressure signal at any moment in the first pressure signal sequence or the second pressure signal sequence reaches a preset pressure threshold, marking the first abnormal time node and the first abnormal pressure signal, and taking the first abnormal time node as the first time series zero point, performing signal deviation analysis on the first pressure signal sequence and the second pressure signal sequence to obtain a pressure signal deviation sequence; when the deviation value at any moment in the pressure signal deviation sequence reaches a preset deviation threshold, marking the second abnormal time node, and starting a sensor fault monitoring unit to perform fault detection on the dual-channel pressure sensor to obtain a sensor detection result; when no deviation value reaching the preset deviation threshold appears within a preset time interval, starting a natural gas pipeline fault monitoring unit, and performing pipeline fault identification and positioning based on the first abnormal pressure signal to obtain a pipeline fault identification result.

[0005] One or more technical solutions provided in this application have at least the following beneficial effects: By deploying dual-channel pressure sensors at preset points on the target natural gas pipeline and synchronously collecting pressure data to obtain two sets of pressure signal sequences, this dual-channel design provides a basis for fault tolerance. Even if one sensor fails, the other sensor can still provide valid data, improving the accuracy and reliability during pressure monitoring. When the pressure signal reaches the preset pressure threshold, by marking the first abnormal time node and the first abnormal pressure signal, the moment of abnormality can be accurately located and identified, which provides a time reference for subsequent fault analysis and location. The pressure signal deviation analysis provides an effective diagnostic means for sensor faults. By comparing the pressure signals of the dual-channel pressure sensors, when the deviation reaches the preset threshold, the second abnormal time node can be marked, and the sensor fault monitoring unit is activated to detect the dual-channel pressure sensors. Through the detection of the sensor fault monitoring unit, the faulty sensor and the fault type can be quickly and accurately identified. This dual-channel sensor and deviation analysis method can determine whether there is a sensor fault by comparing the differences between the two signals, thus avoiding diagnostic errors caused by the failure of a single sensor. If no deviation value reaching the preset deviation threshold appears within the preset time interval, it indicates that the sensor is normal. At this time, the natural gas pipeline fault monitoring unit is activated, which means that the system can avoid false alarms when there is no sensor fault. This fault tolerance mechanism makes the system more stable and reliable. After the natural gas pipeline fault monitoring unit is activated, through the analysis based on the first abnormal pressure signal, the pipeline fault can be accurately identified and located to ensure timely diagnosis and location when a real pipeline fault occurs.

[0006] The above description is only an overview of the technical solution of this application. In order to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the specific embodiments of this application are specifically given below. Brief Description of the Drawings

[0007] Figure 1 This is a schematic flow chart of a fault tolerance diagnosis method using a dual-channel pressure sensor provided by an embodiment of this application.

[0008] Figure 2 This is a schematic flow chart of obtaining the sensor detection result in a fault tolerance diagnosis method using a dual-channel pressure sensor provided by an embodiment of this application. Detailed Description of the Embodiment

[0009] By providing a fault-tolerant diagnosis method using a dual-channel pressure sensor in an embodiment of the present application, the technical problem in the prior art that relies on a single sensor for pressure monitoring and is unable to perform effective fault diagnosis or continue to provide accurate data when pressure anomalies occur, thereby resulting in the inability to detect pipeline faults in a timely manner and affecting the normal operation and safety of the pipeline is solved.

[0010] After introducing the basic principle of the present application, various non-limiting implementation manners of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification.

[0011] As Figure 1 shown, an embodiment of the present application provides a fault-tolerant diagnosis method using a dual-channel pressure sensor, and the method includes: Deploy dual-channel pressure sensors at preset points on the target natural gas pipeline, synchronously collect pressure data, and obtain a first pressure signal sequence and a second pressure signal sequence.

[0012] At preset points on the target natural gas pipeline, dual-channel pressure sensors are deployed to monitor the pressure inside the pipeline. The preset points are selected according to parameters such as the operating state of the pipeline and pressure changes, and involve different regions or key control points of the pipeline. The use of dual-channel pressure sensors, that is, two pressure sensors are deployed at one pipeline point, is for fault tolerance and comparative analysis. These sensors are respectively used to obtain two different pressure signal sequences to ensure the reliability and fault tolerance of the data.

[0013] The dual-channel pressure sensors simultaneously collect pressure data. The pressure data collected by one sensor is the first pressure signal sequence, and the pressure data collected by the other sensor is the second pressure signal sequence. These data are synchronous in time for subsequent signal comparison and analysis. Synchronously collecting data is to ensure the consistency of these two groups of signals in the time dimension for more effective comparison and analysis.

[0014] When any pressure signal at any moment in the first pressure signal sequence or the second pressure signal sequence reaches a preset pressure threshold, mark the first abnormal time node and the first abnormal pressure signal, and use the first abnormal time node as the first time sequence zero point to perform signal deviation analysis on the first pressure signal sequence and the second pressure signal sequence to obtain a pressure signal deviation sequence.

[0015] Set a preset pressure threshold. When any collected pressure signal (regardless of whether it comes from the first sensor or the second sensor) exceeds or is lower than this threshold at a certain moment, it is considered that an anomaly has occurred. This pressure threshold is set based on the normal working pressure range of the pipeline, historical data, expected operating pressure, etc. If the signal value exceeds this range, it means abnormal situations such as equipment failure and excessive pressure fluctuations.

[0016] Once any pressure signal exceeds the preset threshold, mark this moment as the first abnormal time node and record the pressure value at this time as the first abnormal pressure signal, so as to determine when an abnormality occurs and collect key data. Set the first abnormal time node as the first time sequence zero point, that is, the time reference point, and all subsequent signal analyses will start based on this time point. Next, for the first pressure signal sequence and the second pressure signal sequence, perform signal deviation analysis. The purpose of signal deviation analysis is to compare the data collected by the two sensors and check the differences between them. Under normal circumstances, the pressure signals of the two sensors should be very close or consistent. If their differences are large, it indicates a sensor failure and further analysis must be carried out.

[0017] Through comparative analysis, calculate and generate a pressure signal deviation sequence. This sequence represents the pressure signal difference between the two sensors at each moment. If the deviation is large, it indicates that there is a problem with a certain sensor. Therefore, the analysis of the deviation sequence is an important basis for judging sensor failures.

[0018] When the deviation value at any moment in the pressure signal deviation sequence reaches the preset deviation threshold, mark the second abnormal time node and start the sensor fault monitoring unit to perform fault detection on the dual-channel pressure sensor to obtain the sensor detection result.

[0019] Check the deviation sequence according to the preset deviation threshold. If the deviation value at a certain moment exceeds the preset threshold, it is considered that there is an abnormality at this time. This deviation threshold is determined by the rules set by the system or historical data. Deviations exceeding this value are considered significantly abnormal. Once a deviation value exceeding the threshold is detected, immediately record this time point and mark it as the second abnormal time node. This time point is associated with the first abnormal time node and is used for further fault location and analysis.

[0020] After the deviation exceeds the standard, automatically start the sensor fault monitoring unit. The task of this monitoring unit is to conduct in-depth analysis on the dual-channel pressure sensor to detect faults. The fault monitoring unit analyzes multiple dimensions such as the working state, output signal, and data consistency of the sensor, thereby judging whether the sensor has failed and finally obtaining the sensor detection result. The sensor detection result includes the specific situation of the faulty sensor, such as whether the sensor is damaged, its performance has degraded, or other types of faults.

[0021] When no deviation value reaching the preset deviation threshold appears within the preset time interval, start the natural gas pipeline fault monitoring unit to identify and locate pipeline faults based on the first abnormal pressure signal to obtain the pipeline fault identification result.

[0022] If within a preset time interval after the first abnormal time node, which is usually a relatively short time period, no deviation value is detected exceeding the preset deviation threshold, it indicates that the pressure difference between the dual-channel pressure sensors remains within the normal range. In this case, the sensor is considered normal.

[0023] In the absence of sensor failure, it is considered that the problem lies in the pipeline itself. Therefore, the natural gas pipeline fault monitoring unit is activated. This unit will diagnose and locate pipeline faults based on the first abnormal pressure signal, which is the abnormal data recorded at the first abnormal time node and provides key information about pipeline pressure anomalies. Pipeline fault identification involves multiple aspects, such as pipeline rupture, leakage, blockage, or other structural problems. Based on the abnormal situation of the pressure signal, the specific fault type is inferred. Through the analysis of the pipeline fault monitoring unit, a pipeline fault identification result is generated, which will provide information about the location, type, and possible causes of the pipeline fault. This analysis result can help operators quickly identify pipeline problems and take repair measures to ensure the normal operation and safety of the natural gas pipeline.

[0024] Overall, through the above methods, it is possible to effectively distinguish whether it is a sensor fault or a pipeline problem, thereby conducting targeted fault diagnosis and treatment.

[0025] Furthermore, as Figure 2 shown, the method of activating the sensor fault monitoring unit to perform fault detection on the dual-channel pressure sensor and obtaining the sensor detection result includes: Constructing a fault detection time window; taking the second abnormal time node as the second time zero point, and analyzing the transient response and steady-state characteristics of the pressure signal deviation sequence through the fault detection time window to obtain the time characteristics of the pressure signal deviation; based on the time characteristics of the pressure signal deviation, performing fault analysis on the dual-channel pressure sensor to obtain the faulty sensor and the detected fault type as the sensor detection result.

[0026] Using the already marked second abnormal time node as a new time reference, that is, the second time zero point. Based on this time zero point, a fault detection time window is constructed. This time window is a relatively short time period used to focus on the area where the abnormality occurs. The size of the fault detection time window can be adjusted according to specific application scenarios and can be set based on factors such as the operating characteristics of the pipeline and the pressure change speed. For example, the time window can be selected as 1 second or 30 seconds, mainly to capture the pressure signal changes related to faults within a short time.

[0027] Within the fault detection time window, two different analyses are performed on the pressure signal deviation sequence. The transient response refers to the transient change of the pressure signal when an anomaly or interference occurs. The transient response analysis aims at the rapid response characteristics of the sensor to external interference or instantaneous faults; the steady-state characteristic analysis, on the other hand, studies the steady state that the pressure signal reaches within a certain time after an anomaly occurs. The steady-state characteristics can help identify the performance of the system under long-term or continuous load, usually manifested as a stable pressure signal fluctuation.

[0028] The combination of these two analyses can determine how the sensor and the pipeline system respond after a fault occurs and identify the nature of the fault. The transient response characteristics are generally related to short-term faults (such as electrical faults or short-term system interferences), while the steady-state characteristics are mostly used to evaluate long-term performance degradation.

[0029] Through the transient response and steady-state characteristic analyses, time characteristics of the pressure signal deviation are generated. These characteristics describe the dynamic changes of the pressure signal during the fault occurrence process and usually cover parameters such as the signal change rate, fluctuation amplitude, and frequency characteristics. These time characteristics provide key information for subsequent fault analysis and help identify the root cause and type of the fault.

[0030] The time characteristics of the pressure signal deviation are used for the fault analysis of the dual-channel pressure sensor. The purpose of the fault analysis is to determine which pressure sensor has a fault and the type of the fault. For example, the fault types include sensor damage, performance degradation, etc. The fault analysis is carried out by comparing the obtained time characteristics with known fault patterns and using pattern recognition for judgment. Through the analysis of the deviation time characteristics, it is possible to identify which sensor has a fault and determine the specific type of the fault, which is used as the sensor detection result.

[0031] Furthermore, based on the time characteristics of the pressure signal deviation, the fault analysis of the dual-channel pressure sensor is carried out to obtain the faulty sensor and the detected fault type. The method includes: Retrieve the fault history records of the dual-channel pressure sensor to obtain the fault sample set of the dual-channel pressure sensor and the corresponding deviation time characteristic sample set; based on the deviation time characteristic sample set, classify the fault sample set to obtain multiple identified fault types; traverse the deviation time characteristic sample set with the time characteristics of the pressure signal deviation and calculate the feature similarity, and use the identified fault types corresponding to multiple deviation time characteristic samples whose feature similarity meets the feature similarity threshold as the detected fault type; mark the first pressure sensor or the second pressure sensor corresponding to the first abnormal pressure signal as the faulty sensor.

[0032] Retrieve the dual-channel pressure sensor fault data in the retrieval history. These data contain the fault samples that occurred in the past, including information such as the type, location, occurrence time of the sensor fault, and the deviation of the relevant pressure signal. The fault history record is stored in the device maintenance database, which contains various possible fault situations and the corresponding sensor data.

[0033] Extract from the fault history record the set of fault samples and the set of deviation time feature samples related to each fault. The set of fault samples contains the detailed data of all the sensor faults that have occurred. Each sample includes specific fault descriptions, fault types, environmental conditions at the time of occurrence, etc.; the set of deviation time feature samples is the set of time features associated with the fault samples. Each sample records the deviation time series features of the pressure signal at the time of the fault, usually including key parameters such as transient response and steady-state characteristics.

[0034] Based on the obtained set of deviation time feature samples, classify the set of fault samples. Associate the historical fault samples with the deviation time features. In this way, when similar time features appear, the corresponding fault type can be identified. The goal of classification is to divide the fault samples into multiple known fault types, including sensor damage, electrical faults, sensor performance degradation, etc.

[0035] Compare the deviation time features of the currently collected pressure signal with the set of deviation time feature samples in the historical data. Specifically, traverse all the historical deviation time feature samples and match them one by one with the current deviation features. To compare the similarity between the current deviation time features and the historical samples, use feature similarity calculation methods such as Euclidean distance, cosine similarity, or other measurement criteria. The purpose of this process is to determine the matching degree between the currently collected features and the historical fault samples. The calculated similarity value is used to evaluate the similarity between the fault mode and the historical fault samples, and thus identify the type of the current fault.

[0036] Set a feature similarity threshold. When the similarity between the current deviation time features and the historical samples is greater than this threshold, it is considered that the current fault is similar to these historical samples. At this time, use the identified fault type corresponding to the historical fault samples similar to the current feature as the current detected fault type to quickly locate the type of the current fault.

[0037] According to the detected fault type, determine the sensor corresponding to the first abnormal pressure signal, that is, one of the first sensor or the second sensor, as the faulty sensor. Marking the faulty sensor helps subsequent isolation, maintenance, and repair operations.

[0038] Furthermore, performing transient response and steady-state characteristic analysis on the pressure signal deviation sequence through the fault detection time window to obtain the time characteristics of the pressure signal deviation, the method includes: Obtain the fault detection time window, where the fault detection time window includes a transient response detection time window and a steady-state characteristic detection time window; perform transient response analysis on the pressure signal deviation sequence based on the transient response detection time window to obtain transient response deviation time characteristics; perform steady-state characteristic analysis on the pressure signal deviation sequence based on the steady-state characteristic detection time window to obtain steady-state characteristic deviation time characteristics; integrate the transient response deviation time characteristics and the steady-state characteristic deviation time characteristics to obtain the pressure signal deviation time characteristics.

[0039] Determine the time window for fault detection, mainly based on the previously marked second abnormal time node as a reference point. This time window covers a period after the occurrence of the fault and is used to observe the changes in the pressure signal in detail. The fault detection time window includes two sub-windows. The transient response detection time window is used to observe the rapid changes in the pressure signal after the occurrence of the fault, that is, the transient response, which is manifested as a sharp rise or fall in the pressure signal. The steady-state characteristic detection time window is used to observe the steady-state characteristics of the pressure signal after the occurrence of the fault, and the steady-state characteristics are manifested as the changes in the pressure signal over a period of time, including fluctuations in pressure values, trends, etc.

[0040] Within the transient response detection time window, analyze the pressure signal deviation sequence, especially focusing on the rapid changes in the initial stage after the occurrence of the fault. This process is to detect the dynamic response after the fault, such as severe fluctuations and instantaneous jumps in the pressure signal. Transient response characteristics are usually used to identify short-term faults such as electrical faults, external shocks, and sensor damage. By identifying these characteristics, the preliminary nature of the fault can be determined. Through transient response analysis, the corresponding transient response deviation time characteristics are extracted, and these characteristics include important parameters such as the signal mutation time, maximum amplitude change, and response time.

[0041] Within the steady-state characteristic detection time window, perform steady-state characteristic analysis on the pressure signal deviation sequence. This analysis mainly focuses on the long-term stable state after the occurrence of the fault and observes the steady-state characteristics such as fluctuations and trend changes in the pressure signal. Steady-state characteristic analysis helps to identify long-term faults such as sensor performance degradation, long-term pipeline leakage, and pressure fluctuations. Through steady-state characteristic analysis, the steady-state characteristic deviation time characteristics are extracted, and these characteristics include parameters such as the average value of the pressure signal, long-term fluctuation amplitude, and frequency.

[0042] After obtaining the transient response deviation time feature and the steady-state characteristic deviation time feature, these two parts of features are integrated to form a complete pressure signal deviation time feature, which is used for further fault analysis to help determine the severity, type, and possible causes of the fault.

[0043] Furthermore, based on the deviation time feature sample set, fault classification is performed on the fault sample set to obtain multiple identified fault types. The method includes: Extracting a transient response deviation time feature sample set and a steady-state characteristic deviation time feature sample set from the deviation time feature sample set; marking the fault samples corresponding to the transient response deviation time feature samples that meet the preset transient response feature constraints as circuit faults; marking the fault samples corresponding to the steady-state characteristic deviation time feature samples that meet the preset steady-state characteristic feature constraints as sensor performance degradation faults; performing one-to-one correspondence and time alignment on the transient response deviation time feature sample set and the steady-state characteristic deviation time feature sample set, and performing feature fusion analysis, and marking the fault samples that meet the preset fusion feature constraints in the feature fusion result as sensor hardware damage or electrical faults.

[0044] From the existing deviation time feature sample set, extract the transient response deviation time feature sample set related to the transient response. These transient response feature samples include the rate of signal change, the maximum value of the amplitude, the instantaneous jump, etc. These features are used to identify short-term faults such as circuit faults. Similarly, extract the steady-state characteristic deviation time feature sample set related to the steady-state characteristics from the deviation time feature sample set. The steady-state characteristic deviation time feature samples include the fluctuation amplitude, period, average value, etc. of the pressure signal. These features help to identify long-term faults such as sensor performance degradation.

[0045] The preset transient response feature constraints are obtained based on historical data and fault mode analysis, and are used to describe which features represent circuit faults. For example, a circuit fault usually causes a sudden change in voltage, and the pressure signal may suddenly show a large fluctuation or an irregular jump. Therefore, set these features (such as the amplitude and frequency of the signal jump) as the signs of circuit faults. When the transient response deviation time feature samples meet these preset constraint conditions, mark the faults corresponding to these samples as circuit faults. Through this process, the anomalies caused by circuit faults can be identified according to the transient change characteristics of the pressure signal.

[0046] The preset steady-state characteristic feature constraints are used to determine which features represent sensor performance degradation. Sensor performance degradation usually leads to long-term signal fluctuations or instability without drastic changes. For example, features such as the gradual shift of the steady-state signal, the deterioration of signal stability, and continuous low-frequency fluctuations are signs of sensor aging or performance degradation. When the steady-state characteristic deviation time feature samples meet these preset constraints, the faults corresponding to these samples are marked as sensor performance degradation faults, which means that faults caused by sensor performance decline are identified based on the long-term changes and stability of the pressure signal.

[0047] The transient response deviation time feature sample set and the steady-state characteristic deviation time feature sample set are put into one-to-one correspondence and time alignment. Time alignment means synchronizing the time axes of the two types of feature samples so that their time points can be compared under the same reference framework. After the two types of feature sample sets are time-aligned, feature fusion analysis is carried out. At this step, the transient response features and the steady-state characteristic features are combined to form a comprehensive feature description to identify some hybrid or complex fault modes that involve both transient and steady-state characteristics. For example, an electrical fault can cause the pressure signal to fluctuate rapidly in a short period of time.

[0048] After feature fusion, according to the fused feature results, the type of the fault is judged. If the fused features meet the preset fused feature constraints, the fault samples are marked as sensor hardware damage or electrical faults. This process can more accurately identify complex types of faults such as hardware damage and electronic circuit faults, which are manifested in both transient and steady-state characteristics.

[0049] Furthermore, when the multiple deviation time feature samples correspond to more than one identified fault type, the identified fault type corresponding to the deviation time feature sample with the highest feature similarity is taken as the detected fault type.

[0050] In some cases, there may be a situation where multiple deviation time feature samples correspond to more than one identified fault type. Under the matching of multiple fault types, the fault type with the highest similarity to the current deviation time feature sample is selected as the final detected fault type. This fault type will be used as the final diagnosis result for subsequent fault handling, maintenance planning, or alarm prompts. For example, if the currently selected multiple deviation time feature samples are highly similar to both circuit faults and sensor performance degradation faults at the same time, the fault type with higher similarity will be selected as the final judgment result.

[0051] Furthermore, the method further includes: When the detection result of the sensor shows that there is a faulty sensor, isolate the faulty sensor and perform fault maintenance on the faulty sensor based on the detected fault type; after the fault maintenance is completed, cancel the isolation operation of the faulty sensor and continue to collect synchronous pressure data of the dual-channel pressure sensor.

[0052] When it is confirmed through analysis that there is a faulty sensor, perform an isolation operation on the faulty sensor. The isolation operation refers to removing the faulty sensor from real-time monitoring and data collection. The isolation operation involves disconnecting the signal of the faulty sensor, removing it from the data stream, or shutting down the operation of the faulty sensor through the control system. After isolation, quickly perform maintenance operations while using the other sensor to continue data collection.

[0053] After isolating the faulty sensor, perform fault maintenance according to the fault type, including: sensor repair: if the fault is caused by hardware damage, the sensor needs to be repaired or replaced; sensor calibration: if the fault is caused by performance degradation or calibration error, the sensor needs to be recalibrated; electrical inspection: if the fault is caused by electrical problems, the circuit, wiring, etc. need to be inspected and repaired. Fault maintenance is usually carried out by professional maintenance personnel to ensure that the sensor resumes normal function after repair and meets the requirements of the system again.

[0054] When the maintenance work of the faulty sensor is completed and it is confirmed that the faulty sensor has resumed normal operation, cancel the previous isolation operation, including reconnecting the sensor signal to the data stream of the system, re-enabling the real-time monitoring function of the sensor, and allowing the sensor to be used again for normal data synchronization and analysis. After the faulty sensor resumes normal operation, continue to collect synchronous pressure data of the dual-channel pressure sensor, which means that the two sensors work simultaneously and continue to collect real-time pressure data. Synchronous data collection helps to continuously monitor the pressure changes in the pipeline and ensure that the system continues to operate normally after the faulty sensor is maintained.

[0055] Furthermore, the method further includes: According to the historical fault maintenance records of the dual-channel pressure sensor, conduct fault maintenance statistics to obtain O fault maintenance frequency coefficients, P fault maintenance effect coefficients, and Q fault maintenance difficulty coefficients; Perform weight distribution on the O fault maintenance frequency coefficients, P fault maintenance effect coefficients, and Q fault maintenance difficulty coefficients, and perform corresponding weighted summation calculations. Obtain the fault maintenance stability coefficient according to the calculation results; When the fault maintenance stability coefficient reaches the preset stability coefficient threshold, perform replacement processing on the dual-channel pressure sensor.

[0056] Retrieve the historical failure maintenance records of the dual-channel pressure sensor. These records contain detailed information about sensor failures, including data such as the cause of each failure, the repair process, the maintenance methods used, and the post-repair effects.

[0057] Statistically analyze the frequency of occurrence of each type of failure (such as circuit failure, sensor damage, performance degradation, etc.). By calculating the number of occurrences of different types of failures, the corresponding failure maintenance frequency coefficient can be obtained. For example, if a certain type of failure occurs frequently, its frequency coefficient will be higher, indicating that this type of failure is more common in actual operation. Statistically analyze the effect after each failure repair. The effect coefficient can be calculated by comparing the performance of the sensor before and after the failure repair. A higher effect coefficient indicates a good repair effect, and the sensor can return to normal operation and maintain for a long time after repair. A lower effect coefficient indicates a poor repair effect. Statistically analyze the workload and complexity required to repair each type of failure. Calculate the difficulty coefficient based on factors such as the repair time, required professional skills, and materials used for different failures. Failures with a higher difficulty coefficient require more repair time, more professional technical support, or a more complex repair process.

[0058] Assign a weight to each coefficient. The weight assignment is based on actual needs and business goals. For example, if frequently occurring failures are more likely to affect equipment stability, a higher weight is assigned to the frequency coefficient; failures with good repair effects are considered to have a greater impact on stability, so a higher weight can be assigned to the effect coefficient; failures with greater repair difficulty may result in higher costs or time consumption, leading to system instability, so a higher weight is required. When the weight of each coefficient is determined, perform a weighted sum of the O, P, and Q coefficients. Specifically, calculate the weighted sum of the frequency coefficient, effect coefficient, and difficulty coefficient to obtain a comprehensive score, that is, the failure maintenance stability coefficient, which reflects the overall stability of the sensor's failure repair and maintenance during long-term use. The higher the value of the failure maintenance stability coefficient, the better the sensor's failure repair history, the excellent repair effect, and the relatively easy maintenance process; a lower value indicates frequent occurrence of difficult failures or poor repair effects.

[0059] Set a preset stability coefficient threshold for judging whether the sensor should be replaced. This threshold is usually set based on factors such as the expected service life of the device, repair frequency, repair effect, and maintenance cost. If the fault maintenance stability coefficient is lower than this threshold, it indicates that the maintenance and repair history of the sensor is poor, and long-term use may lead to frequent failures, thus affecting the stability of the overall system. When the stability coefficient is lower than the preset threshold, the replacement process of the sensor will be initiated, including disassembling the existing dual-channel pressure sensor from the system and installing a new dual-channel pressure sensor to ensure that it meets the specifications and requirements of the system. Replacing the sensor can effectively avoid system instability caused by long-term failures and ensure the reliability and accuracy of the pipeline pressure monitoring system.

[0060] In summary, the fault tolerance diagnosis method using a dual-channel pressure sensor provided by the embodiment of the present application has the following technical effects: By deploying dual-channel pressure sensors at preset points on the target natural gas pipeline and synchronously collecting pressure data to obtain two groups of pressure signal sequences, this dual-channel design provides a basis for fault tolerance. Even if one of the sensors fails, the other sensor can still provide valid data, improving the accuracy and reliability during pressure monitoring; when the pressure signal reaches the preset pressure threshold, by marking the first abnormal time node and the first abnormal pressure signal, the moment of abnormality can be accurately located and identified, which provides a time reference for subsequent fault analysis and location; the pressure signal deviation analysis provides an effective diagnostic means for sensor faults. By comparing the pressure signals of the dual-channel pressure sensors, the second abnormal time node can be marked when the deviation reaches the preset threshold, and the sensor fault monitoring unit is activated to detect the dual-channel pressure sensor. Through the detection of the sensor fault monitoring unit, the faulty sensor and the fault type can be quickly and accurately identified. This dual-channel sensor and deviation analysis method can judge whether there is a sensor fault by comparing the differences between the two signals, thus avoiding diagnostic errors caused by the failure of a single sensor; if no deviation value reaching the preset deviation threshold appears within the preset time interval, it indicates that the sensor is normal. At this time, the natural gas pipeline fault monitoring unit is activated, which means that the system can avoid false alarms when there is no sensor fault. This fault tolerance mechanism makes the system have higher stability and reliability. After the natural gas pipeline fault monitoring unit is activated, through the analysis based on the first abnormal pressure signal, the pipeline fault can be accurately identified and located to ensure that when a real pipeline fault occurs, it can be diagnosed and located in time.

[0061] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fault-tolerant diagnostic method using a dual-channel pressure sensor, characterized in that: The method comprises: Deploy dual-channel pressure sensors at preset points of the target natural gas pipeline, collect pressure data synchronously, and obtain a first pressure signal sequence and a second pressure signal sequence; When any pressure signal at any time in the first pressure signal sequence or the second pressure signal sequence reaches a preset pressure threshold, a first abnormal time node and a first abnormal pressure signal are marked, and the first abnormal time node is taken as a first timing zero point, and a signal deviation analysis is performed on the first pressure signal sequence and the second pressure signal sequence to obtain a pressure signal deviation sequence; When the deviation value at any time in the pressure signal deviation sequence reaches a preset deviation threshold, a second abnormal time node is marked, and a sensor fault monitoring unit is started to perform fault detection on the dual-channel pressure sensor to obtain a sensor detection result; When the deviation value reaching the preset deviation threshold value does not appear within the preset time interval, the natural gas pipeline fault monitoring unit is started to identify and locate the pipeline fault based on the first abnormal pressure signal to obtain a pipeline fault identification result.

2. A fault-tolerant diagnostic method using a dual-channel pressure sensor as claimed in claim 1, characterized in that: The method of starting the sensor fault monitoring unit to perform fault detection on the dual-channel pressure sensor to obtain a sensor detection result includes: Construct fault detection time window; Taking the second abnormal time node as the second timing zero point, performing transient response and steady-state characteristic analysis on the pressure signal deviation sequence through the fault detection time window to obtain a pressure signal deviation time feature; Based on the pressure signal deviation time characteristics, a fault analysis of the dual-channel pressure sensor is performed to obtain a faulty sensor and detect a fault type as the sensor detection result.

3. A fault-tolerant diagnostic method using a dual-channel pressure sensor as claimed in claim 2, characterized in that: The method of performing fault analysis on the dual-channel pressure sensor based on the pressure signal deviation time characteristic, obtaining the fault sensor and detecting the fault type includes: Retrieve the fault history records of the dual-channel pressure sensor to obtain a fault sample set and a corresponding deviation time feature sample set of the dual-channel pressure sensor; Based on the deviation time feature sample set, performing fault classification on the fault sample set to obtain a plurality of identified fault types; Traversing the deviation time feature sample set with the pressure signal deviation time feature, and calculating feature similarity, taking the identified fault types corresponding to the multiple deviation time feature samples whose feature similarities meet the feature similarity threshold as the detected fault type; The first pressure sensor or the second pressure sensor corresponding to the first abnormal pressure signal is marked as the faulty sensor.

4. A fault-tolerant diagnostic method using a dual-channel pressure sensor as claimed in claim 2, characterized in that: The method of performing transient response and steady-state characteristic analysis on the pressure signal deviation sequence through the fault detection time window to obtain the pressure signal deviation time feature includes: Acquire the fault detection time window, wherein the fault detection time window includes a transient response detection time window and a steady-state characteristic detection time window; Performing transient response analysis on the pressure signal deviation sequence based on the transient response detection time window to obtain a transient response deviation time feature; Performing steady-state characteristic analysis on the pressure signal deviation sequence based on the steady-state characteristic detection time window to obtain a steady-state characteristic deviation time feature; The transient response deviation time characteristic and the steady-state characteristic deviation time characteristic are integrated to obtain the pressure signal deviation time characteristic.

5. A fault-tolerant diagnostic method using a dual-channel pressure sensor as claimed in claim 3, characterized in that: The method of classifying the fault sample set based on the deviation time feature sample set to obtain a plurality of identified fault types includes: Extracting a transient response deviation time feature sample set and a steady-state characteristic deviation time feature sample set based on the deviation time feature sample set; Marking the fault samples corresponding to the transient response deviation time feature samples that meet the preset transient response feature constraints as circuit faults; Marking the fault samples corresponding to the steady-state characteristic deviation time characteristic samples that meet the preset steady-state characteristic constraint as sensor performance degradation faults; The transient response deviation time feature sample set and the steady-state characteristic deviation time feature sample set are matched one to one and time-aligned, and feature fusion analysis is performed, and the fault samples that meet the preset fusion feature constraints in the feature fusion results are marked as sensor hardware damage or electrical fault.

6. A fault-tolerant diagnostic method using a dual-channel pressure sensor as claimed in claim 3, characterized in that: When the multiple deviation time feature samples correspond to more than one identified fault type, the identified fault type corresponding to the deviation time feature sample with the highest feature similarity is taken as the detected fault type.

7. A fault-tolerant diagnostic method using a dual-channel pressure sensor as claimed in claim 1, characterized in that: The method further comprises: When the sensor detection result shows that there is a faulty sensor, the faulty sensor is isolated and the fault maintenance of the faulty sensor is performed based on the detected fault type; When the fault maintenance is completed, the isolation operation of the faulty sensor is canceled, and the synchronous pressure data collection of the dual-channel pressure sensor is continued.

8. A fault-tolerant diagnostic method using a dual-channel pressure sensor as claimed in claim 7, characterized in that: The method further comprises: According to the historical fault maintenance records of the dual-channel pressure sensor, fault maintenance statistics are performed to obtain O fault maintenance frequency coefficients, P fault maintenance effect coefficients, and Q fault maintenance difficulty coefficients; Performing weight distribution on the O fault maintenance frequency coefficients, the P fault maintenance effect coefficients, and the Q fault maintenance difficulty coefficients, and performing corresponding weighted sum calculations, and obtaining the fault maintenance stability coefficient according to the calculation results; When the fault maintenance stability coefficient reaches a preset stability coefficient threshold, the dual-channel pressure sensor is replaced.

Images