Industrial and commercial user gas metering system fault in-situ intelligent diagnosis method

Abstract

The invention belongs to the technical field of gas flow meter diagnosis, and discloses an industrial and commercial user gas metering system fault in-situ intelligent diagnosis method, which specifically comprises the following steps of: 1, exporting data; step 2, data noise reduction processing; step 3, distinguishing flow states; step 4, analyzing a pressure signal; 5, analyzing the correlation between the low-frequency pressure signal and the temperature signal; 6, calculating the temperature difference of the pipeline; and 7, making a diagnosis strategy. According to the invention, on the basis of industrial and commercial user metering system data collected by a gas SCADA system, through comprehensive analysis of multi-sensor data, potential faults of a user gas metering system are found in time; compared with a traditional personnel regular inspection tour and background manual analysis method, the method can achieve in-situ intelligent diagnosis of the faults of the metering system, improve the accuracy and real-time performance of fault diagnosis, and achieve intelligent rapid diagnosis of the faults of the gas metering system.

Description

Technical Field

[0001] This invention belongs to the field of gas flow meter diagnostic technology, specifically a method for in-situ intelligent diagnosis of faults in gas metering systems for industrial and commercial users. Background Technology

[0002] As a green, low-carbon, and clean energy source, natural gas plays a bridging and supporting role in promoting my country's energy structure transformation and energy supply security. For gas companies, industrial and commercial users have high gas consumption and sales profits, making them the focus of metering management. Accurate metering of their gas consumption is crucial for reducing gas supply and sales discrepancies and ensuring the economic benefits of gas companies. In practical applications, the long-term operation of gas metering systems inevitably encounters various faults, affecting metering accuracy. Currently, fault detection in gas metering systems mainly relies on regular manual inspections or periodic meter calibration and maintenance, which cannot detect faults in a timely manner. With the development of information technology and the need for refined management in gas companies, the use of IoT technology to collect real-time gas load data (including flow, pressure, and temperature) from important industrial and commercial users and remotely transmit and store it in the company's Supervisory Control and Data Acquisition (SCADA) system, along with dedicated personnel analyzing the daily data provided by the SCADA platform to assess the metering system status, has become widely adopted. While this method can improve the speed of metering system fault diagnosis, its diagnostic accuracy is affected by personnel experience, the skill level of metering management personnel, and their sense of responsibility. At the same time, this method cannot monitor the metering operation status at all times and in all dimensions. Summary of the Invention

[0003] The purpose of this invention is to provide an in-situ intelligent diagnostic method for faults in gas metering systems for industrial and commercial users, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an in-situ intelligent diagnosis method for faults in gas metering systems for industrial and commercial users, wherein the specific steps of the intelligent diagnosis method are as follows:

[0005] Step 1: Export data;

[0006] Export the 24-hour data sequence of gas flow, pressure, and temperature for the industrial and commercial user to be diagnosed from the gas company's SCADA system; among which, the gas flow data sequence is denoted as... The pressure data series is denoted as Temperature data series denoted as

[0007] Step two, data noise reduction processing;

[0008] For the original data sequence and Noise reduction is performed to prevent data from being interfered with by impulse noise when it is transmitted to the SCADA system. Noise reduction is performed based on the physical characteristics of fluctuations in gas flow, pressure and temperature data.

[0009] Step 3: Differentiate traffic status;

[0010] Based on the flow data sequence Q s (s=1,2,…,S), the flow state of pipeline s is divided into the following three types: no flow throughout State=1, partial flow State=2, and full flow State=3;

[0011] Step 4, pressure signal analysis; analysis of pressure data sequences. Perform discrete wavelet transform;

[0012] The low-frequency pressure signal is obtained by reconstructing the approximate coefficients after wavelet transform. and high-frequency pressure fluctuation signals

[0013] Based on low-frequency pressure signals and high-frequency pressure fluctuation signals Calculate the relative coefficient of pressure fluctuation in pipeline s

[0014] Step 5: Correlation analysis of low-frequency pressure and temperature signals; calculate the low-frequency pressure and temperature signals Tem for pipe s (s=1,2,…,S). s Pearson correlation coefficient between

[0015] Step 6, Calculate the pipe temperature difference; when the number of pipes s ≥ 2, calculate the average temperature difference between pipe s and other pipes.

[0016] Step 7: Diagnostic strategy formulation; different diagnostic strategies are adopted according to the number of pipelines and their gas usage status, specifically divided into two types: pipeline number s = 1 and pipeline number s ≥ 2.

[0017] Preferably, the relative coefficient of pressure fluctuation in the pressure signal analysis in step four is... The calculation formula is:

[0018]

[0019] in denoted by and , respectively, the values ​​of the low-frequency pressure signal and the high-frequency pressure fluctuation signal at time t, where N is the total length of the sequence.

[0020] Preferably, the average temperature difference in the pipeline temperature difference calculation in step six is... The calculation formula is:

[0021]

[0022] in These represent the temperature signal data of the s-th and j-th pipelines at time t, respectively.

[0023] Preferably, in the flow status differentiation of step three, Q, which has no flow throughout, is... s The value is 0, and the flow Q in the partial flow is... s The time period >0 accounts for 20%-80% of the total time, and the flow rate Q in the total flow is... s The time period >0 accounts for more than 80% of the total time.

[0024] Preferably, in step seven, the diagnostic strategy formulation includes a diagnostic strategy for when the number of pipes s = 1, where the number of pipes s = 1 and State = 1; if the pressure and temperature correlation coefficient... It was determined that both ends of the pipeline were closed, and the metering system had no flow rate fault; if the pressure and temperature correlation coefficient... And the relative coefficient of pressure fluctuation If the fluctuation exceeds the set threshold, it is determined that there is gas consumption in the pipeline, and a metering system fault alarm signal is issued.

[0025] Preferably, in step seven, the diagnostic strategy formulation, the diagnostic strategy for a pipeline quantity s = 1 further includes, when the pipeline quantity S = 1 and State = 2, based on the flow rate Q s The situation, the pressure data P s The pressure datasets are divided into gas usage periods and non-gas usage periods. Multiple random samples are taken from the pressure datasets of gas usage periods and non-gas usage periods respectively. The T-test is used to calculate the similarity of the pressure during gas usage periods and non-gas usage periods. If the similarity is greater than the similarity threshold, a fault alarm signal for the metering system is given.

[0026] Preferably, in step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipelines s≥2 includes: if there are two or more pipelines with a flow state of State=3 among the S pipelines, calculate the average flow difference D between the full-flow pipeline and other full-flow pipelines; if the flow difference D is greater than a set threshold, give a fault alarm signal for the metering system.

[0027] Preferably, in step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipes s≥2 also includes, if among the S pipes, there are simultaneously pipes with flow state State=3 and pipes with flow state State=2, a fault alarm signal for the metering system is given.

[0028] Preferably, in step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipelines s≥2 further includes the following: if there are no pipelines with a flow state of State=3 among the s pipelines, but there are pipelines with a flow state of State=2, then according to the diagnostic strategy for the number of pipelines s=1, the flow pipelines with State=2 are diagnosed, and if the similarity is greater than the Similarity set threshold, a metering system fault alarm signal is given.

[0029] Preferably, step seven, the diagnostic strategy formulation, further includes, if the flow state of all pipes is State = 0, for any pipe s (s = 1, 2), if the pressure and temperature correlation coefficient... Pressure fluctuation relative coefficient The average temperature difference is greater than the set threshold. If the temperature difference exceeds the set threshold, it is determined that pipeline s is suspected of gas consumption, and a metering system fault alarm signal (abnormal flow) is issued.

[0030] The beneficial effects of this invention are as follows:

[0031] This invention, based on data collected from industrial and commercial user metering systems by a gas SCADA system, utilizes comprehensive analysis of multi-sensor data to promptly detect potential faults in user gas metering systems. Specifically, it includes seven steps: Step 1, data export; Step 2, data noise reduction processing; Step 3, flow state differentiation; Step 4, pressure signal analysis; Step 5, correlation analysis of low-frequency pressure and temperature signals; Step 6, pipeline temperature difference calculation; and Step 7, diagnostic strategy formulation. Compared to traditional methods involving periodic personnel inspections and manual back-end analysis, this method enables in-situ intelligent diagnosis of metering system faults, improving the accuracy and real-time nature of fault diagnosis, and achieving intelligent and rapid diagnosis of gas metering system faults. Attached Figure Description

[0032] Figure 1 This is a flowchart of the fault diagnosis method of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] like Figure 1 As shown, this embodiment of the invention provides an in-situ intelligent fault diagnosis method for gas metering systems of industrial and commercial users. The specific steps of this intelligent diagnosis method are as follows:

[0035] Step 1: Export data;

[0036] Export the 24-hour data sequence of gas flow, pressure, and temperature for the industrial and commercial user to be diagnosed from the gas company's SCADA system; among which, the gas flow data sequence is denoted as... The pressure data series is denoted as Temperature data series denoted as

[0037] Based on the non-negative property of traffic data, Sampling points were removed as outliers. Based on the pipeline type (medium pressure, low pressure) of industrial and commercial users, a distribution range (threshold) for pressure data was set. Data exceeding this range was removed as outliers. The flow and pressure data sequences after removing outliers were then subjected to nearest neighbor interpolation to obtain the denoised data. Simultaneously, based on the gradual temperature change characteristics of the pipeline, the temperature data sequence is smoothed and filtered to obtain...

[0038] Step two, data noise reduction processing;

[0039] For the original data sequence and Noise reduction is performed to prevent data from being interfered with by impulse noise when it is transmitted to the SCADA system. Noise reduction is performed based on the physical characteristics of fluctuations in gas flow, pressure and temperature data.

[0040] Step 3: Differentiate traffic status;

[0041] Based on the flow data sequence Q s (s=1,2,…,S), the flow state of pipeline s is divided into the following three types: no flow throughout State=1, partial flow State=2, and full flow State=3;

[0042] Step 4, pressure signal analysis; analysis of pressure data sequences. Perform discrete wavelet transform;

[0043] The low-frequency pressure signal is obtained by reconstructing the approximate coefficients after wavelet transform. and high-frequency pressure fluctuation signals

[0044] Based on low-frequency pressure signals and high-frequency pressure fluctuation signals Calculate the relative coefficient of pressure fluctuation in pipeline s

[0045] Step 5: Correlation analysis of low-frequency pressure and temperature signals; calculate the low-frequency pressure and temperature signals Tem for pipe s (s=1,2,…,S). s Pearson correlation coefficient between

[0046] Step 6, Calculate the pipe temperature difference; when the number of pipes s ≥ 2, calculate the average temperature difference between pipe s and other pipes.

[0047] Step 7: Diagnostic strategy formulation; different diagnostic strategies are adopted according to the number of pipelines and their gas usage status, specifically divided into two types: pipeline number s = 1 and pipeline number s ≥ 2.

[0048] Among them, the pressure fluctuation relative coefficient in step four of the pressure signal analysis The calculation formula is:

[0049]

[0050] in denoted by and , respectively, the values ​​of the low-frequency pressure signal and the high-frequency pressure fluctuation signal at time t, where N is the total length of the sequence.

[0051] In step six, the average temperature difference in the pipeline temperature difference calculation The calculation formula is:

[0052]

[0053] in These represent the temperature signal data of the s-th and j-th pipelines at time t, respectively.

[0054] In the flow status differentiation in step three, Q, which has no flow throughout, is... s The value is 0, and the flow Q in the partial flow is... s The time period >0 accounts for 20%-80% of the total time, and the flow rate Q in the total flow is... s The time period >0 accounts for more than 80% of the total time.

[0055] In step seven, the diagnostic strategy formulation includes a strategy for when the number of pipes s = 1, where the number of pipes s = 1 and State = 1 (no flow throughout); if the pressure and temperature correlation coefficient... It was determined that both ends of the pipeline were closed, and the metering system had no flow rate fault; if the pressure and temperature correlation coefficient... And the relative coefficient of pressure fluctuation If the flow rate exceeds the set fluctuation threshold, it is determined that there is gas consumption in the pipeline, and a metering system fault alarm signal (abnormal flow) is issued.

[0056] In step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipes s=1 also includes, when the number of pipes S=1 and State=2 (partial flow), based on the flow rate Q s The situation, the pressure data Ps The pressure datasets are divided into gas usage periods and non-gas usage periods. Multiple random samples are taken from the pressure datasets of the gas usage periods and non-gas usage periods respectively. The T-test is used to calculate the similarity of the pressure during the gas usage periods and the non-gas usage periods. If the similarity is greater than the similarity threshold, a fault alarm signal (abnormal flow or abnormal pressure) is given to the metering system.

[0057] In step seven, the diagnostic strategy formulation includes a diagnostic strategy for the number of pipes s≥2. If there are two or more pipes with a flow state of State=3 (total flow) among the S pipes, the average flow difference D between the total flow pipe and other total flow pipes is calculated. If the flow difference D is greater than a set threshold, a metering system fault alarm signal (filter fault) is given.

[0058] In step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipelines s≥2 also includes, if among the S pipelines, there are pipelines with a flow state of State=3 (total flow) and pipelines with a flow state of State=2 (partial flow), a fault alarm signal for the metering system is given.

[0059] In step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipelines s≥2 also includes the following: if there are no pipelines with a flow state of State=3 (full flow) among the s pipelines, but there are pipelines with a flow state of State=2 (partial flow), the pipeline with partial flow State=2 is diagnosed according to the diagnostic strategy for the number of pipelines s=1. If the similarity is greater than the Similarity set threshold, a metering system fault alarm signal (abnormal flow or abnormal pressure) is given.

[0060] The seventh step, the diagnostic strategy formulation, further includes, if the flow state of all pipelines is State = 0 (no flow throughout), for any pipeline s (s = 1, 2), if the pressure and temperature correlation coefficients... Pressure fluctuation relative coefficient The average temperature difference is greater than the set threshold. If the temperature difference exceeds the set threshold, it is determined that pipeline s is suspected of gas consumption, and a metering system fault alarm signal (abnormal flow) is issued.

[0061] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0062] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for in-situ intelligent diagnosis of faults in gas metering systems for industrial and commercial users, characterized in that: The specific steps of this intelligent diagnostic method are as follows: Step 1: Export data; Export the previous day's full-day data sequence of gas flow, pressure, and temperature for the industrial and commercial user to be diagnosed from the gas company's SCADA system; among which, the gas flow data sequence is denoted as... The pressure data series is denoted as Temperature data series denoted as Step two, data noise reduction processing; For the original data sequence and Noise reduction is performed to prevent data from being interfered with by impulse noise when it is transmitted to the SCADA system. Noise reduction is performed based on the physical characteristics of fluctuations in gas flow, pressure and temperature data. Step 3: Differentiate traffic status; Based on the flow data sequence Q s (s=1,2,…,S), the flow state of pipeline s is divided into the following three types: no flow throughout State=1, partial flow State=2, and full flow State=3; Step 4, pressure signal analysis; analysis of pressure data sequences. Perform discrete wavelet transform; The low-frequency pressure signal is obtained by reconstructing the approximate coefficients after wavelet transform. and high-frequency pressure fluctuation signals Based on low-frequency pressure signals and high-frequency pressure fluctuation signals Calculate the relative coefficient of pressure fluctuation in pipeline s Step 5: Correlation analysis of low-frequency pressure and temperature signals; calculate the low-frequency pressure and temperature signals Tem for pipe s (s=1,2,…,S). s Pearson correlation coefficient between Step 6, Calculate the pipe temperature difference; when the number of pipes s ≥ 2, calculate the average temperature difference between pipe s and other pipes. Step 7: Diagnostic strategy formulation; different diagnostic strategies are adopted according to the number of pipelines and their gas usage status, specifically divided into two types: pipeline number s = 1 and pipeline number s ≥ 2.

2. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: The pressure fluctuation relative coefficient in step four of the pressure signal analysis The calculation formula is: in denoted by and , respectively, the values ​​of the low-frequency pressure signal and the high-frequency pressure fluctuation signal at time t, where N is the total length of the sequence.

3. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: The average temperature difference in step six, pipe temperature difference calculation The calculation formula is: in These represent the temperature signal data of the s-th and j-th pipelines at time t, respectively.

4. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: In the flow status differentiation in step three, Q, which has no flow throughout, is... s The value is 0, and the flow Q in the partial flow is... s The time period >0 accounts for 20%-80% of the total time, and the flow rate Q in the total flow is... s The time period >0 accounts for more than 80% of the total time.

5. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: In step seven, the diagnostic strategy formulation includes a strategy where the number of pipes s = 1. This strategy applies when the number of pipes s = 1 and State = 1; if the pressure and temperature correlation coefficient... The pipe is determined to be closed at both ends, and the metering system has no flow fault; if the pressure and temperature correlation coefficient is... And the relative coefficient of pressure fluctuation If the fluctuation exceeds the set threshold, it is determined that there is gas consumption in the pipeline, and a metering system fault alarm signal is issued.

6. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: In step seven, the diagnostic strategy formulation, the diagnostic strategy for a pipeline quantity s = 1 also includes, when the pipeline quantity S = 1 and State = 2, based on the flow rate Q s The situation, the pressure data P s The pressure datasets are divided into gas usage periods and non-gas usage periods. Multiple random samples are taken from the pressure datasets of both periods. The T-test is used to calculate the similarity of the pressures during gas usage periods and non-gas usage periods. If the similarity is greater than the similarity threshold, a fault alarm signal for the metering system is given.

7. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: In step seven, the diagnostic strategy formulation includes the following diagnostic strategy for the number of pipelines s≥2: if there are two or more pipelines with a flow state of State=3 among the S pipelines, calculate the average flow difference D between the full-flow pipeline and other full-flow pipelines. If the flow difference D is greater than a set threshold, give a fault alarm signal for the metering system.

8. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: In step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipelines s≥2 also includes, if there are pipelines with flow state State=3 and pipelines with flow state State=2 in the S pipelines, a fault alarm signal for the metering system is given.

9. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: In step seven, the diagnostic strategy formulation, the diagnostic strategy for the number of pipelines s≥2 also includes the following: if there are no pipelines with a flow state of State=3 among the s pipelines, but there are pipelines with a flow state of State=2, then according to the diagnostic strategy for the number of pipelines s=1, the flow pipelines with State=2 are diagnosed. If the similarity is greater than the Similarity threshold, a fault alarm signal for the metering system is given.

10. The in-situ intelligent fault diagnosis method for industrial and commercial user gas metering systems according to claim 1, characterized in that: The seventh step, formulating the diagnostic strategy, also includes considering the following: if the flow state of all pipes is State = 0, for any pipe s (s = 1, 2), if the pressure and temperature correlation coefficients are... Pressure fluctuation relative coefficient The average temperature difference is greater than the set threshold. If the temperature difference exceeds the set threshold, it is determined that pipeline s is suspected of gas consumption, and a fault alarm signal for the metering system is issued.

Images