Pressure vessel data monitoring method based on high-precision sensor

Abstract

The invention relates to the technical field of data processing, in particular to a pressure vessel data monitoring method based on a high-precision sensor. Obtaining a data period according to data change characteristics of the monitoring data sequence; obtaining the data deviation degree of the current moment according to the data difference characteristics of the current moment and the same moment in the historical data cycle; obtaining a suspected abnormal moment according to the data deviation degree; obtaining comprehensive stability according to the data difference characteristics of the suspected abnormal moment at the same moment in the historical data cycle and the fluctuation difference characteristics of the suspected abnormal moment at the same moment in the historical data cycle; and according to the data deviation degree of the suspected abnormal moment, the comprehensive stability and the quantity characteristics of the suspected abnormal moments of other sensors at the same moment, obtaining the fault abnormity degree. According to the invention, the pressure vessel at the current moment is monitored according to the fault abnormity degree, and the accuracy of pressure vessel data monitoring is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a pressure vessel data monitoring method based on high-precision sensors. BACKGROUND

[0002] Industrial pressure vessels are often used in high-temperature and high-pressure scenes, such as chemical and energy fields; because the application scene is prone to safety accidents, there are very high safety standards for pressure vessels, and pressure vessels are usually equipped with high-precision sensors for data monitoring. Because high-precision sensors are sensitive to external disturbances, they are easily disturbed by the environment, such as high-frequency equipment and radio interference in the surrounding environment, external weather or steam conduction to the inside of the pressure vessel caused by the heat dissipation of the factory equipment, resulting in changes in temperature and humidity and pressure monitoring values, and external hammering or pump body resonance interference with pressure monitoring data. Such situations will affect the monitoring data of high-precision sensors, and the monitoring data cannot reflect the true state of the pressure vessel, reducing the accuracy of pressure vessel data monitoring. SUMMARY

[0003] In order to solve the above technical problems, the purpose of the present application is to provide a pressure vessel data monitoring method based on high-precision sensors, and the technical solution adopted is as follows: Obtain the monitoring data sequence of different types of sensors in the pressure vessel monitoring process; According to the data variation characteristics of the monitoring data sequence of any sensor, obtain the data period; according to the data difference characteristics of the same time in the historical data period and the current time, obtain the data deviation degree of the current time; according to the data deviation degree, obtain the suspected abnormal time; According to the data difference characteristics of the same time in the historical data period and the suspected abnormal time, obtain the first stability; according to the fluctuation difference characteristics of the same time in the historical data period and the suspected abnormal time, obtain the second stability; according to the first stability and the second stability, obtain the comprehensive stability; According to the data deviation degree of the suspected abnormal time, the comprehensive stability, and the number characteristics of other sensors appearing at the same time, obtain the fault abnormality degree; according to the fault abnormality degree, monitor the pressure vessel at the current time.

[0004] Further, the step of obtaining the data period according to the data variation characteristics of the monitoring data sequence of any sensor comprises: Perform STL time sequence decomposition on the monitoring data sequence to obtain a periodic term; take the reciprocal of the frequency corresponding to the maximum amplitude value in the frequency spectrum data of the periodic term as the period length; segment the monitoring data sequence according to the period length to obtain different data periods.

[0005] Furthermore, the step of obtaining the degree of data deviation at the current moment based on the data difference characteristics between the current moment and the same moment in the historical data period includes: Calculate the average of the absolute values ​​of the differences between the current moment and the same moment in all historical data periods, and normalize them to obtain the degree of data deviation at the current moment.

[0006] Furthermore, the step of obtaining the suspected abnormal moment based on the degree of data deviation includes: When the deviation of the data exceeds a preset deviation threshold, the current moment is taken as the suspected abnormal moment.

[0007] Furthermore, the step of obtaining the first stability based on the data difference characteristics of the suspected abnormal moment at the same moment in the historical data period includes: In any two adjacent historical data periods, calculate the absolute value of the numerical difference at the same moment of the suspected abnormal moment to obtain the adjacent period difference value; calculate the reciprocal of the average value of all adjacent period difference values ​​and normalize it to obtain the first stability.

[0008] Furthermore, the step of obtaining the second stability based on the fluctuation difference characteristics of the suspected abnormal moment at the same moment in the historical data period includes: The local fluctuation value is obtained by summing the absolute values ​​of the differences between the same moment and the adjacent moments of the suspected abnormal moment in the historical data period; the fluctuation difference value is obtained by calculating the absolute value of the difference between the local fluctuation values ​​corresponding to any two historical adjacent data periods; the second stability is obtained by calculating the reciprocal of the average value of all fluctuation difference values ​​and normalizing it.

[0009] Further, the step of obtaining the comprehensive stability based on the first stability and the second stability includes: The sum of the first stability and the second stability is calculated to obtain the overall stability.

[0010] Furthermore, the step of obtaining the fault anomaly degree based on the data deviation degree at the suspected anomaly moment, the overall stability, and the number of other sensors exhibiting suspected anomaly moments at the same time includes: Calculate the percentage of suspected abnormal moments in the data monitoring sequences of other sensors at the current moment to obtain the correlation degree; calculate the product of the correlation degree, the data deviation degree, and the overall stability to obtain the fault abnormality degree of the suspected abnormal moment.

[0011] Furthermore, the step of monitoring the pressure vessel at the current moment based on the degree of fault anomaly includes: When the fault anomaly exceeds a preset anomaly threshold, the pressure vessel malfunctions.

[0012] The present invention has the following beneficial effects: In this invention, acquiring the data period can determine the periodic characteristics of data changes in the monitored data sequence, thereby judging data anomalies based on the periodic characteristics and initially improving the accuracy of data monitoring. Acquiring the data deviation degree can initially determine whether the current moment is a suspected anomaly moment based on the data difference characteristics between the current moment and the same moment in the historical data period. Since the data of high-precision sensors is easily affected by external environmental interference, acquiring the first stability and the second stability can determine the degree of external environmental interference at the same moment in history as the suspected anomaly moment; acquiring the comprehensive stability can accurately characterize the degree of external environmental interference at the same moment in history, further improving the accuracy of pressure vessel data monitoring. Acquiring the fault anomaly degree can accurately reflect whether the suspected anomaly moment is caused by a real anomaly inside the pressure vessel; finally, monitoring the pressure vessel at the current moment based on the fault anomaly degree improves the data monitoring accuracy of high-precision sensors and reduces the interference of the external environment on the monitoring process. Attached Figure Description

[0013] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of a pressure vessel data monitoring method based on a high-precision sensor, provided as an embodiment of the present invention. Detailed Implementation

[0015] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a pressure vessel data monitoring method based on a high-precision sensor proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0017] The following description, in conjunction with the accompanying drawings, details a specific scheme for a pressure vessel data monitoring method based on a high-precision sensor provided by the present invention.

[0018] Please see Figure 1 The diagram illustrates a flowchart of a pressure vessel data monitoring method based on a high-precision sensor, according to an embodiment of the present invention. The method includes the following steps: Step S1: Obtain monitoring data sequences from different types of sensors during the pressure vessel monitoring process.

[0019] During the monitoring of pressure vessels, data is collected through temperature sensors, humidity sensors, pressure sensors, and vibration sensors inside the vessel. Each sensor has the same acquisition frequency, thereby obtaining different monitoring data sequences from different types of sensors.

[0020] Step S2: Obtain the data period based on the data change characteristics of the monitoring data sequence of any sensor; obtain the data deviation degree of the current moment based on the data difference characteristics between the current moment and the same moment in the historical data period; and obtain the suspected abnormal moment based on the data deviation degree.

[0021] Most media within pressure vessels undergo cycling during production reactions, such as coupled temperature and pressure changes, phase transitions from boiling to condensation, and the operation of stirring or circulating pumps, all exhibiting certain periodic characteristics. Therefore, monitoring data under normal conditions within a pressure vessel typically shows periodic fluctuations. If a moment deviates from this historical periodic pattern, it indicates a potential fault or anomaly at that moment. First, the data period is obtained based on the data change characteristics of the monitoring data sequence from any sensor. Preferably, in this embodiment, the step of obtaining the data period includes: performing STL time-series decomposition on the monitoring data sequence to obtain periodic terms. It should be noted that STL time-series decomposition is existing technology, and the specific steps are not detailed here. Periodic terms can more accurately reflect the periodic characteristics of the monitoring data sequence. The reciprocal of the frequency corresponding to the maximum amplitude in the spectral data of the periodic term is taken as the period length. The spectral data is obtained by performing a Fourier transform on the periodic term; the frequency corresponding to the maximum amplitude in the spectral data is the dominant frequency of the periodic term, so the reciprocal of the dominant frequency can be used as the period length of the periodic change characteristics of the monitoring data sequence. The monitoring data sequence is then segmented according to the period length to obtain different data periods. The data characteristics of different data periods after segmentation are similar.

[0022] Furthermore, the similarity between the data characteristics of the current moment and the same moments in historical data periods can be analyzed. Therefore, the degree of data deviation at the current moment can be obtained based on the data difference characteristics between the current moment and the same moments in historical data periods. Preferably, in this embodiment, the step of obtaining the degree of data deviation includes: calculating and normalizing the average of the absolute values ​​of the differences between the current moment and the same moments in all historical data periods to obtain the degree of data deviation at the current moment; wherein the normalization method is the ratio of this value to the maximum value in the same historical calculation process. The greater the numerical difference between the current moment and the same moments in historical data periods, the greater the degree of data deviation, meaning that the data characteristics at the current moment are less consistent with the data change patterns at the same moments in historical data periods, and the more likely an anomaly is to occur within the pressure vessel at the current moment. Thus, suspected abnormal moments can be obtained based on the degree of data deviation. Preferably, in this embodiment, when the degree of data deviation exceeds a preset deviation threshold, the current moment is considered a suspected abnormal moment. In this embodiment, the preset deviation threshold is 0.6, which can be determined by the implementer according to the implementation scenario. If the current moment is a suspected anomaly, it means that there may be an anomaly inside the pressure vessel at this moment, or that the sensor is being interfered with by external factors, causing the data to deviate; therefore, further judgment is needed on suspected anomalies to improve monitoring accuracy.

[0023] Step S3: Obtain a first stability based on the data difference characteristics of the suspected anomaly at the same time in the historical data period; obtain a second stability based on the fluctuation difference characteristics of the suspected anomaly at the same time in the historical data period; obtain a comprehensive stability based on the first stability and the second stability.

[0024] Because high-precision sensors have high monitoring sensitivity, they often introduce environmental noise, and corresponding environmental noise data may also exist in historical data, leading to inaccurate judgment of suspected anomalies. Therefore, it is necessary to analyze data from the same time in historical data periods to determine the degree of environmental interference. The more obvious the environmental interference characteristics, the more likely the suspected anomaly is caused by external environmental factors; the weaker the environmental interference characteristics, the more likely the suspected anomaly is caused by a real anomaly within the pressure vessel. Firstly, the data difference characteristics at the same time in historical data periods can be analyzed. The more similar the data, the more stable the data at that time, indicating no significant changes over multiple periods and a lower probability of environmental interference. Therefore, a first stability is obtained based on the data difference characteristics of suspected anomalies at the same time in historical data periods. Preferably, in this embodiment of the invention, the step of obtaining the first stability includes: calculating the absolute value of the numerical difference at the same time in any two adjacent historical data periods to obtain the adjacent period difference value; the smaller the adjacent period difference value, the more similar the values ​​at the same time in adjacent data periods, and the lower the probability of external environmental interference. The first stability is obtained by calculating the reciprocal of the average of all adjacent period differences and normalizing it; this normalization method is the maximum-minimum normalization method. The smaller the differences of all adjacent periods, the more similar the data at that same moment is, the less likely that the same moment in history was affected by external environmental interference, and the more likely that the suspected abnormal moment was caused by a real anomaly inside the pressure vessel.

[0025] Furthermore, during a reaction within a pressure vessel, the reaction process may remain unchanged while adjusting the fundamental reaction parameters, leading to a lower first stability, such as an overall increase in temperature or a decrease in humidity. Since the parameters are adjusted but the reaction process remains constant, the data change trend within the data period is relatively stable. Therefore, a second stability can be obtained based on the fluctuation differences between the suspected anomaly moments and the same moments in historical data periods. Preferably, in this embodiment of the invention, the step of obtaining the second stability includes: calculating the sum of the absolute values ​​of the differences between the same moment of a suspected anomaly moment and its immediate preceding and following moments in historical data periods to obtain a local fluctuation value; the local fluctuation value reflects the degree of data difference between the same moment and its immediate preceding and following moments, and the larger the local fluctuation value, the more obvious the data change characteristics at that same moment. Calculating the absolute value of the difference between the local fluctuation values ​​corresponding to any two historical adjacent data periods to obtain a fluctuation change difference value; the smaller the fluctuation change difference value, the more similar the data change characteristics at the same moment in adjacent data periods. Calculating the reciprocal of the average of all fluctuation change difference values ​​and normalizing it to obtain the second stability; this normalization method is the maximum-minimum value normalization method. The greater the second stability, the more similar the fluctuation characteristics at the same moment between different data periods are, the less the external environment interferes with that same moment, and the more likely that the suspected abnormal moment is caused by a real abnormality inside the pressure vessel.

[0026] After obtaining the first stability and the second stability, a comprehensive stability can be obtained based on the first stability and the second stability. Preferably, in this embodiment of the invention, the step of obtaining the comprehensive stability includes: calculating the sum of the first stability and the second stability to obtain the comprehensive stability. The larger the comprehensive stability, the less likely the suspected abnormal moment is to be affected by external interference at the same time in history, and the more likely the suspected abnormal moment is caused by a real abnormality inside the pressure vessel.

[0027] Step S4: Obtain the fault anomaly degree based on the data deviation degree, overall stability, and the number of other sensors showing suspected anomalies at the same time; monitor the pressure vessel at the current time based on the fault anomaly degree.

[0028] Pressure vessel failures are typically caused by structural abnormalities, media leakage, fatigue damage, and other issues, which can easily lead to simultaneous anomalies in monitoring data from various types of sensors. However, sensor data anomalies caused by external environmental interference are unlikely to cause simultaneous anomalies in all types of sensors. Therefore, analyzing the anomalies of multiple types of sensors at the same time can improve monitoring accuracy. Thus, the fault anomaly degree is obtained based on the degree of data deviation, overall stability, and the number of other sensors exhibiting suspected anomalies at the same time. Preferably, in this embodiment, the step of obtaining the fault anomaly degree includes: calculating the proportion of suspected anomaly moments in the data monitoring sequences of other sensors at the current time to obtain the correlation degree; the more other sensors exhibiting suspected anomalies at the current time, the greater the correlation degree, meaning that the simultaneous anomalies in data from multiple types of sensors are more likely to be caused by a real anomaly within the pressure vessel. The product of the correlation degree, data deviation degree, and overall stability is calculated to obtain the fault anomaly degree of the suspected anomaly moment. A higher fault anomaly degree means that the suspected anomaly moment is more likely to be caused by a real anomaly within the pressure vessel; a lower fault anomaly degree means that the suspected anomaly moment is more likely to be caused by external environmental interference with the sensor. The formulas for obtaining the fault anomaly degree include:

[0029] In the formula, R represents the degree of fault anomaly, F represents the degree of data deviation, D represents the overall stability, N represents the number of other sensors, and n represents the number of times other sensors exhibit suspected anomalies at the current moment. Indicates the degree of correlation.

[0030] Furthermore, after obtaining the fault anomaly degree corresponding to all sensors that show suspected anomalies at the current moment, the pressure vessel can be monitored based on the fault anomaly degree. Specifically, when the fault anomaly degree exceeds a preset anomaly threshold, the pressure vessel is considered to have an anomaly. In this embodiment of the invention, the preset anomaly threshold is 0.3, which can be determined by the implementer according to the implementation scenario. When the fault anomaly degree of any sensor at the current moment exceeds the preset anomaly threshold, it means that an anomaly has occurred inside the pressure vessel, requiring timely repair. Thus, by using fault anomaly degree, the accuracy of high-precision sensors in monitoring pressure vessels is improved, and the impact of external environmental interference is reduced.

[0031] In summary, this invention provides a pressure vessel data monitoring method based on high-precision sensors. It obtains the data period based on the data change characteristics of the monitored data sequence; obtains the data deviation degree of the current moment based on the data difference characteristics between the current moment and the same moment in the historical data period; identifies suspected abnormal moments based on the data deviation degree; obtains the comprehensive stability based on the data difference characteristics of the suspected abnormal moments at the same moment in the historical data period and the fluctuation difference characteristics of the suspected abnormal moments at the same moment in the historical data period; and obtains the fault anomaly degree based on the data deviation degree of the suspected abnormal moments, the comprehensive stability, and the number of suspected abnormal moments appearing in other sensors at the same time. This invention monitors the pressure vessel at the current moment based on the fault anomaly degree, improving the accuracy of pressure vessel data monitoring.

[0032] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0033] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for monitoring pressure vessel data based on high-precision sensors, characterized in that, The method includes the following steps: Acquire monitoring data sequences from different types of sensors during pressure vessel monitoring; The data period is obtained based on the data change characteristics of the monitoring data sequence of any sensor; the data deviation degree of the current moment is obtained based on the data difference characteristics between the current moment and the same moment in the historical data period; and suspected abnormal moments are obtained based on the data deviation degree. A first stability is obtained based on the data difference characteristics of the suspected anomaly moment at the same time in the historical data period; a second stability is obtained based on the fluctuation difference characteristics of the suspected anomaly moment at the same time in the historical data period; and a comprehensive stability is obtained based on the first stability and the second stability. The fault anomaly degree is obtained based on the degree of data deviation at the suspected abnormal moment, the overall stability, and the number of other sensors showing suspected abnormal moments at the same time; the pressure vessel is monitored at the current moment based on the fault anomaly degree.

2. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of obtaining the data period based on the data change characteristics of the monitoring data sequence from any sensor includes: The monitoring data sequence is decomposed using STL to obtain periodic terms; the reciprocal of the frequency corresponding to the maximum amplitude in the spectral data of the periodic terms is taken as the period length; the monitoring data sequence is divided according to the period length to obtain different data periods.

3. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of obtaining the degree of data deviation at the current moment based on the data difference characteristics between the current moment and the same moment in the historical data period includes: Calculate the average of the absolute values ​​of the differences between the current moment and the same moment in all historical data periods, and normalize them to obtain the degree of data deviation at the current moment.

4. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of obtaining the suspected abnormal moment based on the degree of data deviation includes: When the deviation of the data exceeds a preset deviation threshold, the current moment is taken as the suspected abnormal moment.

5. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of obtaining the first stability based on the data difference characteristics of the suspected abnormal time at the same time in the historical data period includes: In any two adjacent historical data periods, calculate the absolute value of the numerical difference at the same moment of the suspected abnormal moment to obtain the adjacent period difference value; calculate the reciprocal of the average value of all adjacent period difference values ​​and normalize it to obtain the first stability.

6. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of obtaining the second stability based on the fluctuation difference characteristics of the suspected abnormal moment at the same moment in the historical data period includes: The local fluctuation value is obtained by summing the absolute values ​​of the differences between the same moment and the adjacent moments of the suspected abnormal moment in the historical data period; the fluctuation difference value is obtained by calculating the absolute value of the difference between the local fluctuation values ​​corresponding to any two historical adjacent data periods; the second stability is obtained by calculating the reciprocal of the average value of all fluctuation difference values ​​and normalizing it.

7. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of obtaining the overall stability based on the first stability and the second stability includes: The sum of the first stability and the second stability is calculated to obtain the overall stability.

8. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of obtaining the fault anomaly degree based on the data deviation degree at the suspected anomaly moment, the overall stability, and the number of other sensors exhibiting suspected anomaly moments at the same time includes: Calculate the percentage of suspected abnormal moments in the data monitoring sequences of other sensors at the current moment to obtain the correlation degree; calculate the product of the correlation degree, the data deviation degree, and the overall stability to obtain the fault abnormality degree of the suspected abnormal moment.

9. The pressure vessel data monitoring method based on a high-precision sensor according to claim 1, characterized in that, The step of monitoring the pressure vessel at the current moment based on the fault anomaly degree includes: When the fault anomaly exceeds a preset anomaly threshold, the pressure vessel malfunctions.

Images