A method for evaluating performance degradation of a gas pressure regulator in service based on BX life

Abstract

The application discloses a kind of in-service gas pressure regulator performance attenuation evaluation method based on BX life, the method is processed training and identification to the running data of in-service running gas pressure regulator itself, proposes a pressure regulating work state evaluation index according to data anomaly proportion, and introduces BX life to evaluate the performance attenuation of in-service gas pressure regulator, and then give the maintenance strategy of next work cycle.The method of the application can first carry out performance analysis of gas pressure regulator without stopping gas and without removing pressure regulating device, improving the practicability and convenience of the method;Secondly, by obtaining the attenuation trend of pressure regulating performance, the performance of gas pressure regulator in subsequent operation can be more accurately predicted, which makes up for the shortcomings of traditional methods that only monitor but do not predict and only alarm when there is a fault;Finally, based on the maintenance strategy given by the method, the change of application scenario and use strategy is fully considered, which improves the effectiveness and applicability of the method.

Description

Technical Field

[0001] This invention relates to the field of urban gas pressure regulators, and in particular to a method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan. Background Technology

[0002] Currently, gas pressure regulators play a crucial role in gas transmission and distribution systems by reducing upstream pressure and stabilizing downstream pressure. As gas pressure regulators age, their internal components inevitably experience aging, wear, and malfunctions, ultimately leading to a decline in their pressure-regulating performance and threatening the gas safety of downstream users. In recent years, an increasing number of domestic gas pressure regulators have incorporated monitoring systems to monitor parameters such as inlet pressure, outlet pressure, and flow rate; however, a relatively unified evaluation method has yet to be established.

[0003] BX life is a key indicator of product reliability, representing the expected failure rate of X% of units after a specific time or mileage, while (1-X)% of units are expected to continue operating normally. For example, when X=10, it becomes B10 life. Common BX lifespans include B0.1, B1, B5, and B10. Using BX life to describe the operating time of a gas pressure regulator when the percentage of abnormal pressure regulation data reaches X% can help avoid occasional pressure regulation anomalies and accurately reflect the degradation of pressure regulation performance before regulator failure.

[0004] Currently, domestic methods for evaluating gas pressure regulator faults mostly employ on-site data processing and fault identification. Patent publication number CN113944801A, "A Method and Device for Performance Testing of Gas Pressure Regulators Based on Data Analysis," acquires characteristic data of the pressure regulator, processes this data using a distance-based clustering algorithm, calculates the regulator's steady-state pressure and shut-off pressure, and compares them with corresponding thresholds to detect and warn of faults in the regulator. Patent publication number CN108108665A presents a multivariate-based safety early warning method for gas pressure regulators. This method uses wavelet packet energy analysis to obtain the energy distribution pattern of the first frequency band of the outlet pressure signal, acquires the pressure radar map of gas pressure regulators exhibiting high-frequency or low-frequency faults, and determines whether to issue a fault warning or alarm based on the pressure radar map. The patent publication CN103454114A, entitled "Diagnosis and Safety Early Warning Device and Method for Operational Faults of Medium and Low Pressure Gas Regulators," acquires operational data online or offline via a field data logger and performs spectral analysis. The spectral analysis results are compared and identified with the spectral characteristics of typical faults obtained through prior research, and a fault diagnosis conclusion is drawn based on the comparison and identification results. However, the aforementioned control methods all require prior use of a large amount of operational data from other healthy regulators to obtain corresponding fault thresholds, without considering individual differences; and all are state-based early warnings after a gas pressure fault occurs. While these methods improve the safety of gas pressure regulators to some extent, they still cannot handle or identify performance degradation trends. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan. This method can train, identify and evaluate the operating data of existing gas pressure regulators in service, obtain the degradation trend of pressure regulation performance, and provide maintenance strategies for the next working cycle, effectively avoiding situations that threaten gas safety.

[0006] The present invention provides a method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan, comprising the following steps: Step 1: Continuously sample the operating data of the gas pressure regulators in service, dividing the sampling time into a data training phase and a real-time operation phase. Each set of operating data includes the sampling time and the gas inlet pressure of the gas pressure regulator. Instantaneous gas flow rate Q through the in-service gas regulator; gas outlet pressure of the gas regulator. Among them, the operating data of the gas pressure regulators in service, which were obtained during the data training phase, were used as sample training data; Step 2: Collect all gas inlet pressures from the sample training data. The maximum pressure in is denoted as Minimum pressure is denoted as In order to satisfy Under the condition that n is pre-set 1 pressure segment threshold , … The training data is divided into n pressure sample segment intervals based on the pressure segmentation threshold, namely: , , ..., , ; Step 3: For the sample training data in each pressure sample segment interval, establish a segmented fitting relationship between the measured value of gas outlet pressure and the measured value of instantaneous gas flow rate, and determine the healthy range of gas outlet pressure corresponding to each pressure sample segment interval based on the statistical results of the fitting deviation. Step 4: After the gas pressure regulator enters the real-time operation stage, first set the evaluation cycle of the gas pressure regulator, and use the piecewise linear fitting relationship and the healthy range of gas outlet pressure as the evaluation benchmark. During the in-service operation of the gas pressure regulator, analyze the real-time operation data in each evaluation cycle, calculate the pressure regulation anomaly ratio of the corresponding evaluation cycle, and complete the performance degradation evaluation of the in-service gas pressure regulator in the current evaluation cycle based on the pressure regulation anomaly ratio. Step 5: Complete the determination of the gas pressure regulator's pressure regulation performance status within the current evaluation period, as described in Step 4, and determine the proportion of abnormal pressure regulation within the current evaluation period. When the performance degradation rate is below the preset threshold X%, a fitting model is established based on the pressure regulation anomaly rate of the current evaluation cycle and several previous evaluation cycles to predict the pressure regulation anomaly rate of the in-service gas pressure regulator in the next evaluation cycle. To make predictions in order to predict and assess the degradation trend of the pressure regulating performance of in-service gas pressure regulators in the next evaluation cycle; Step 6: Based on the performance degradation evaluation results of the gas pressure regulators in service, formulate corresponding treatment strategies.

[0007] Compared with the prior art, the advantages of the present invention are as follows: The method of this invention firstly allows for performance analysis of gas pressure regulators without interrupting gas supply or removing the pressure regulating device, thus improving the method's practicality and convenience. Secondly, by acquiring the degradation trend of pressure regulating performance, it can more accurately predict the performance of the gas pressure regulator in subsequent operation, overcoming the shortcomings of traditional methods that only monitor but do not predict and only alarm when a fault occurs. Finally, based on the maintenance strategy provided by this method, it fully considers changes in application scenarios and usage strategies, improving the method's effectiveness and applicability. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating the performance degradation evaluation of the gas pressure regulator under in-service conditions according to the present invention. Figure 2 Training data (ingress pressure); Figure 3 Training data (traffic); Figure 4 Training data (exit pressure); Figure 5 The fitting results for segment 3 in the example; Figure 6 The deviation calculation results for segment 3 in the embodiment; Figure 7 Real-time data (ingress pressure); Figure 8 Real-time data (traffic); Figure 9 Real-time data (export pressure). Detailed Implementation

[0009] The process and working principle of the in-service gas pressure regulator performance degradation evaluation method based on BX life provided by the present invention will be further described in detail below with reference to embodiments and accompanying drawings. It should be noted that this embodiment is descriptive and not limiting, and should not be construed as limiting the scope of protection of the present invention.

[0010] To facilitate understanding of the technical solution provided in this application, the relevant content of the technical solution is explained.

[0011] The present invention provides a method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan, comprising the following steps: Step 1: Continuously sample the operating data of the gas pressure regulators in service, dividing the sampling time into a data training phase and a real-time operation phase. Each set of operating data includes the sampling time and the gas inlet pressure of the gas pressure regulator. Instantaneous gas flow rate Q through the in-service gas regulator; gas outlet pressure of the gas regulator. The operational data of in-service gas pressure regulators acquired during the data training phase were used as sample training data.

[0012] Preferably, the amount of sample training data should be sufficient to fully reflect the long-term continuous operation performance of the in-service voltage regulator, that is, the data training phase should be no less than 1 year and the sample training data should be no less than 600 sets. Step 2: Collect all gas inlet pressures from the sample training data. The maximum pressure in is denoted as Minimum pressure is denoted as In order to satisfy Under the condition that n is pre-set 1 pressure segment threshold , … The training data is divided into n pressure sample segment intervals based on the pressure segmentation threshold, namely: , , ..., , ;

[0013] The following example illustrates this: If: =0.1 MPa =0.9 MPa, n=4, pressure interval = (0.9-0.1) / 4 = 0.2 MPa, then =0.1 + 0.2 = 0.3 MPa; =0.1 + 0.4 = 0.5 MPa =0.1+0.6 =0.7 MPa.

[0014] Step 3: For the training data within each pressure sample segment interval, establish a piecewise fitting relationship between the measured gas outlet pressure and the measured instantaneous gas flow rate. Based on the statistical results of the fitting deviation, determine the healthy range of gas outlet pressure corresponding to each pressure sample segment interval. The steps are as follows: Step 301: First, for each pressure sample segment interval, extract the measured instantaneous gas flow rate from the sample training data within that pressure sample segment interval. Q and the measured value of the gas outlet pressure corresponding to the instantaneous gas flow rate. constituting data point pairs Then, a linear fit is performed on the data point pairs to establish a linear fit relationship between the gas outlet pressure and the measured instantaneous gas flow rate within the segmented interval of the pressure sample. The measured instantaneous gas flow rate is used as the horizontal axis (X-axis), and the measured gas outlet pressure is used as the vertical axis (Y-axis). Finally, the linear fit formula is used to calculate the relationship between the measured instantaneous gas flow rate and the measured gas flow rate. Q Fitted calculation value of the corresponding gas outlet pressure The formula is: Where i = 1, 2, ..., n The slope is obtained by linear fitting for the segmented interval of the i-th pressure sample. The constant term is obtained by linear fitting for the segmented interval of the i-th pressure sample; Step 302: Within each pressure sample segment interval, calculate the corresponding measured value of the gas outlet pressure for each data point in the sample training data. Comparison with fitted values Deviation between The calculation formula is: ; Step 303: For each pressure sample segment interval, calculate the deviation of each data point pair within each pressure sample segment interval. The absolute values ​​are sorted in descending order, and the samples with the highest absolute values ​​of deviation within the pressure sample interval are removed. Training data from a sample is used to eliminate the influence of outliers on the determination of the healthy range of gas outlet pressure. Preferably, ,in This represents the number of values ​​to be removed within the segment interval of the i-th pressure sample. The total amount of training data for the samples within the segmented interval of the i-th pressure sample.

[0015] Step 304: Extract the corresponding bias from the remaining training data within each pressure sample segment interval. Upper limit of deviation and lower limit of deviation And the deviation that meets the following conditions The corresponding measured value of the gas outlet pressure is determined as the healthy range of gas outlet pressure for this pressure sample segment: ; Step 4: After the gas pressure regulator enters the real-time operation stage, firstly, set the evaluation cycle of the gas pressure regulator, and use the piecewise linear fitting relationship and the healthy range of gas outlet pressure as the evaluation benchmark. During the in-service operation of the gas pressure regulator, analyze the real-time operation data in each evaluation cycle, calculate the pressure regulation anomaly ratio of the corresponding evaluation cycle, and complete the performance degradation evaluation of the in-service gas pressure regulator in the current evaluation cycle based on the pressure regulation anomaly ratio.

[0016] Preferably, an evaluation period is a certain time period (such as 1 month, 6 months, 1 year, etc.), and the amount of data in an evaluation period should be greater than 100.

[0017] The specific steps can be as follows: Step 401: Obtain the real-time operating data of the gas pressure regulator in service during the current evaluation period; where the current evaluation period is the j-th evaluation period, j=1, 2, ...

[0018] Step 402: For the real-time operating data acquired within the current evaluation period, based on the measured values ​​of the gas inlet pressure in the real-time operating data, the real-time operating data is categorized into pressure sample segment intervals as described in Step 2 using the sample segmentation method in Step 2. For each real-time operating data point within each pressure sample segment interval, the measured value of the gas outlet pressure is calculated. The deviation between the calculated gas outlet pressure and the value obtained based on the piecewise linear fitting relationship established in step three. , ,in Q s This refers to the measured value of the instantaneous gas flow rate in the real-time operating data; and These are the slope and constant term obtained by linear fitting for the corresponding pressure sample segment intervals.

[0019] Step 403: Calculate the deviations corresponding to each real-time data point within the current evaluation period. The upper limit of the deviation from the pressure sample segment interval determined in step three. and lower limit of deviation Compare the results and make a judgment based on the following rules: such as deviation or If the voltage regulation fails, the real-time data point is determined to have failed, and the data point is marked.

[0020] such as deviation or If the voltage regulation is abnormal at that real-time data point, then the data point will be marked.

[0021] such as deviation If the voltage regulation at that data point is normal, then it can be determined that the voltage regulation is normal.

[0022] Step 404: Count the number of real-time data points that are judged as voltage regulation anomalies (excluding voltage regulation failures) within the current evaluation period. And calculate the abnormal voltage regulation ratio for the current evaluation period. The calculation formula is: , Let j be the total number of real-time running data in the j-th evaluation period, where j = 1, 2, ...; Step 405, based on the abnormal voltage regulation ratio of the current evaluation period. The performance degradation of gas pressure regulators operating during the current evaluation period is evaluated. For example, the proportion of abnormal pressure regulation is assessed. If the value is ≥X%, the gas pressure regulator is determined to have experienced a decline in its pressure regulation performance. X is a preset performance decline judgment threshold, preferably 0.1, 1, 5, 10, etc.

[0023] Step 5: Complete the determination of the gas pressure regulator's pressure regulation performance status within the current evaluation period, as described in Step 4, and determine the proportion of abnormal pressure regulation within the current evaluation period. When the performance degradation rate is below the preset threshold X%, a fitting model is established based on the pressure regulation anomaly rate of the current evaluation cycle and several previous evaluation cycles to predict the pressure regulation anomaly rate of the in-service gas pressure regulator in the next evaluation cycle. Predictions are made to forecast and assess the performance degradation trend of in-service gas pressure regulators in the next evaluation cycle. The specific steps are as follows: Step 501: Select the voltage regulation anomaly ratios corresponding to the (j-3), (j-2), (j-1), and (j)th evaluation periods, and construct a fitting model of the voltage regulation anomaly ratio as a function of operating time, using the evaluation period number as the time variable. The voltage regulation anomaly ratio model adopts a failure trend function based on the Weibull distribution (an existing model; see "Reliability Design" p.44, Electronic Industry Press for details), and its expression is: ; in, λ ( t (For runtime) t The fitted value of the abnormal voltage regulation ratio, where m is the shape parameter. The scale parameter is used to adapt to the application scenario of periodic evaluation in engineering. Each evaluation cycle after the gas pressure regulator enters the real-time operation stage is regarded as a time unit, and the operation time is approximately represented by the evaluation cycle number j. t The proportion of abnormal voltage regulation during the evaluation period will be determined. The equivalent failure rate of the gas pressure regulator within the corresponding evaluation period is considered and used for failure trend modeling; by performing regression calculations on the pressure regulation anomaly ratio model, the corresponding model parameters m and are obtained. .

[0024] Step 502, using the voltage regulation anomaly proportional model parameters m and determined in step 501. Calculate the abnormal voltage regulation ratio for the next evaluation period (the (j+1)th evaluation period). The corresponding predicted value of the abnormal voltage regulation ratio The calculation formula is: ; If the voltage regulation ratio is abnormal If the value is ≥X%, the gas pressure regulator is deemed to have a risk of performance degradation in the next evaluation period.

[0025] Step Six: Based on the performance degradation evaluation results of the in-service gas pressure regulators, formulate corresponding treatment strategies. Details are as follows: Voltage regulation failure occurred: When a gas pressure regulator fails to regulate pressure during the current evaluation period, it indicates that the gas pressure regulator has some faults and can no longer meet the requirements for safe operation. At this time, the gas pressure regulator should be inspected on-site, and gas supply should be shut off or emergency repair measures should be taken according to the inspection results.

[0026] Voltage regulation performance degradation has occurred or is predicted to occur: When a gas pressure regulator is determined to have experienced a decline in its pressure regulation performance during the current evaluation period, or when a risk of performance decline is predicted for the next evaluation period, the following handling strategy will be adopted based on a comprehensive assessment of the regulator's operating status and gas demand: If the comprehensive evaluation results indicate that the pressure regulation performance degradation of the pressure regulator has not affected its normal use, then the real-time operating data marked as pressure regulation abnormality in the current evaluation period will be added to the corresponding pressure sample segment interval as new training data. Then, the updated segment interval training data will be retrained using the method in step (iii), the segment linear fitting relationship and the corresponding healthy range of gas outlet pressure will be updated, and a new real-time operation phase will be entered to achieve the in-service operation of the gas pressure regulator with reduced performance. If the comprehensive evaluation results indicate that the reduced pressure regulation performance of the gas regulator has affected its normal use, then the gas regulator shall be repaired.

[0027] Normal operation status: If the in-service gas pressure regulator does not experience pressure regulation failure during the current evaluation period, and no pressure regulation performance degradation occurs during the current evaluation period or the predicted next evaluation period, it indicates that the pressure regulation performance of the in-service gas pressure regulator can meet the current usage requirements, and no treatment measures are needed.

[0028] Handling after maintenance or resetting operating conditions: When the gas pressure regulator has been repaired, parts replaced, or the gas outlet pressure has been reset according to the gas demand, it is considered that the operating condition has changed. Steps (I) to (III) are then executed again to rebuild the corresponding training model. Based on this, a new real-time operation phase is entered to complete the post-repair operation management of the in-service gas pressure regulator.

[0029] Example 1: Step 1: Taking a gas pressure regulator in operation as an example, the nominal size is DN150, the nominal inlet pressure range is 0.8~2.4MPa, the outlet pressure setting is 200kPa, the data acquisition interval of the gas pressure regulator is 6 hours, and each set of data includes sampling time, gas inlet pressure, gas instantaneous flow rate, and gas outlet pressure.

[0030] Set 5% (X=5) as the threshold for the proportion of abnormal voltage regulation data.

[0031] The first three years of normal operation of the gas pressure regulator were used as the data training phase, with a total of 4372 training samples. See Figures 2-4 .

[0032] Step 2: In the sample training data, the maximum gas inlet pressure is 1.574 MPa, and the minimum gas inlet pressure is 1.056 MPa. Considering that there are only 242 data points within the range of 0–1.3 MPa, the inlet pressure segmentation threshold is selected as follows: =1.3MPa =1.4MPa =1.5MPa, and the training data were divided into 4 sample segments according to the gas inlet pressure. The specific sample segments and data volume are shown in Table 1.

[0033] Table 1. Pressure Segmentation Intervals of Sample Training Data

[0034] Step 3: Divide the instantaneous gas flow rate of the training data within each sample segment interval. Q With gas outlet pressure Composition of data points Perform linear fitting, such as Figure 5 The fitting results for the third sample segment interval are shown in Table 3. The fitting relationship between the gas outlet pressure and the instantaneous gas flow rate for each sample segment interval is shown in Table 3.

[0035] Table 3. Outlet pressure of each sample segment interval With instantaneous flow Q Fitting equation for change

[0036] The relationship between the instantaneous flow rate of each gas was obtained by fitting the formula. Q Calculated value of the corresponding gas outlet pressure And calculate the data points for each data point within each sample segment interval. and The deviation ε between them Figure 6 The result shown is the calculated deviation ε for the third segment interval.

[0037] The absolute values ​​of the deviations ε within each sample segment interval are sorted from largest to smallest, and the top values ​​within each sample segment interval corresponding to the deviation ε are removed. The training data consists of a sample of data. The upper limit of the deviation ε within the remaining training data in each segment interval is also considered. and lower limit of deviation The results are shown in Table 4.

[0038] Table 4. Deviation ε and maximum value for each segment and minimum value

[0039] Step 4: After entering the real-time operation phase, select an evaluation period of 12 months and a real-time data collection interval of 6 hours.

[0040] The gas pressure regulator is currently undergoing its 7th evaluation cycle, with a total of 1459 real-time data points acquired. Each data set includes sampling time, gas inlet pressure, instantaneous gas flow rate, and gas outlet pressure. (See attached data.) Figures 7-9 .

[0041] Calculate the gas outlet pressure P at each data point of the real-time data within each segment interval. s The deviation between the calculated value and the value obtained using the fitted relation , will deviation and the corresponding upper limit of deviation and lower limit of deviation The comparison results are as follows: Table 6 Deviation Comparison results

[0042] Based on the assessment, the gas pressure regulator did not show any degradation in pressure regulation performance during the current evaluation period.

[0043] Step 5: Based on the assessment, determine the proportion of abnormal voltage regulation within the current evaluation period. <5%, substituting the abnormal voltage regulation ratios of the 4th, 5th, 6th, and 7th evaluation periods and their corresponding period numbers into the abnormal voltage regulation ratio model, and through regression calculation, the constants m=3.915 and =40127. The performance degradation evaluation results for the 4th to 7th evaluation periods are shown in Table 7.

[0044] Table 7 Performance degradation evaluation results

[0045] Using formula Calculate the abnormal voltage regulation ratio for the next evaluation period. =2.84%.

[0046] Based on the judgment, <5%, indicating that the voltage regulation performance will not degrade in the next evaluation cycle.

[0047] Step 6: If it is determined that the in-service gas pressure regulator has not experienced pressure regulation failure and will not experience pressure regulation performance degradation in the current evaluation cycle or the predicted next evaluation cycle, no treatment measures are required.

[0048] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for evaluating the performance degradation of in-service gas pressure regulators based on BX life, characterized in that, Includes the following steps: Step 1: Continuously sample the operating data of the gas pressure regulators in service, dividing the sampling time into a data training phase and a real-time operation phase. Each set of operating data includes the sampling time and the gas inlet pressure of the gas pressure regulator. Instantaneous gas flow rate Q through the in-service gas regulator; gas outlet pressure of the gas regulator. Among them, the operating data of the gas pressure regulators in service, which were obtained during the data training phase, were used as sample training data; Step 2: Collect all gas inlet pressures from the sample training data. The maximum pressure in is denoted as Minimum pressure is denoted as In order to satisfy Under the condition that n is pre-set 1 pressure segment threshold , … The training data is divided into n stress sample segment intervals based on the stress segment threshold, namely: , , ..., , ; Step 3: For the sample training data in each pressure sample segment interval, establish a segmented fitting relationship between the measured value of gas outlet pressure and the measured value of instantaneous gas flow rate, and determine the healthy range of gas outlet pressure corresponding to each pressure sample segment interval based on the statistical results of the fitting deviation. Step 4: After the gas pressure regulator enters the real-time operation stage, first set the evaluation cycle of the gas pressure regulator, and use the piecewise linear fitting relationship and the healthy range of gas outlet pressure as the evaluation benchmark. During the in-service operation of the gas pressure regulator, analyze the real-time operation data in each evaluation cycle, calculate the pressure regulation anomaly ratio of the corresponding evaluation cycle, and complete the performance degradation evaluation of the in-service gas pressure regulator in the current evaluation cycle based on the pressure regulation anomaly ratio. Step 5: Complete the determination of the gas pressure regulator's pressure regulation performance status within the current evaluation period, as described in Step 4, and determine the proportion of abnormal pressure regulation within the current evaluation period. When the performance degradation rate is below the preset threshold X%, a fitting model is established based on the pressure regulation anomaly rate of the current evaluation cycle and several previous evaluation cycles to predict the pressure regulation anomaly rate of the in-service gas pressure regulator in the next evaluation cycle. To make predictions in order to predict and assess the degradation trend of the pressure regulating performance of in-service gas pressure regulators in the next evaluation cycle; Step 6: Based on the performance degradation evaluation results of the gas pressure regulators in service, formulate corresponding treatment strategies.

2. The method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan according to claim 1, characterized in that: The specific process of step three is as follows: Step 301: First, for each pressure sample segment interval, extract the measured instantaneous gas flow rate from the sample training data within that pressure sample segment interval. Q and the measured value of the gas outlet pressure corresponding to the instantaneous gas flow rate. constituting data point pairs Then, a linear fit is performed on the data point pairs to establish a linear fit relationship between the gas outlet pressure and the measured instantaneous gas flow rate within the segmented interval of the pressure sample, where the measured instantaneous gas flow rate is used as the horizontal axis and the measured gas outlet pressure is used as the vertical axis; finally, the linear fit formula is used to calculate the relationship between the measured instantaneous gas flow rate and the measured gas outlet pressure. Q Fitted calculation value of the corresponding gas outlet pressure ; Step 302: Within each pressure sample segment interval, calculate the corresponding measured value of the gas outlet pressure for each data point in the sample training data. Comparison with fitted values The deviation ε between them; Step 303: For each pressure sample segment interval, sort the absolute values ​​of the deviations ε of each data point within each pressure sample segment interval in descending order, and remove the data points with the highest absolute values ​​of deviations within that pressure sample segment interval. Training data from individual samples was used to eliminate the influence of outliers on the determination of the healthy range of gas outlet pressure. Step 304: Extract the upper limit of the deviation from the remaining training data in each pressure sample segment interval. and lower limit of deviation The measured value of the gas outlet pressure corresponding to the deviation ε that meets the following conditions is determined as the healthy range of gas outlet pressure for this pressure sample segment: .

3. The method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan according to claim 2, characterized in that: ,in This represents the number of values ​​to be removed within the segmented interval of the i-th pressure sample. The total amount of training data for the samples within the segmented interval of the i-th pressure sample.

4. The method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan according to claim 2, characterized in that: The specific process of step four is as follows: Step 401: Obtain real-time operating data of the gas pressure regulators in service during the current evaluation period; where the current evaluation period is the j-th evaluation period, j=1, 2, ... Step 402: For the real-time operating data acquired within the current evaluation period, based on the measured values ​​of the gas inlet pressure in the real-time operating data, the real-time operating data is categorized into pressure sample segment intervals as described in Step 2 using the sample segmentation method in Step 2. For each real-time operating data point within each pressure sample segment interval, the measured value of the gas outlet pressure is calculated. The deviation between the calculated gas outlet pressure and the value obtained based on the piecewise linear fitting relationship established in step three. ; Step 403: Calculate the deviations corresponding to each real-time data point within the current evaluation period. The upper limit of the deviation from the pressure sample segment interval determined in step three. and lower limit of deviation Compare the results and make a judgment based on the following rules: such as deviation or If the voltage regulation fails, the real-time data point is determined to have failed, and the data point is marked. such as deviation or If so, it is determined that a voltage regulation anomaly has occurred at the real-time data point, and the data point is marked. such as deviation If so, the voltage regulation at that data point is normal; Step 404: Count the number of real-time data points identified as voltage regulation anomalies within the current evaluation period. And calculate the abnormal voltage regulation ratio for the current evaluation period. The calculation formula is: , Let j be the total number of real-time running data in the j-th evaluation period, where j = 1, 2, ...; Step 405, based on the abnormal voltage regulation ratio of the current evaluation period. The performance degradation of gas pressure regulators operating during the current evaluation period is evaluated, such as the proportion of abnormal pressure regulation. If the value is ≥X%, then the gas pressure regulator is determined to have experienced a decline in its pressure regulating performance.

5. The method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan according to claim 4, characterized in that: X can be 0.1, 1, 5 or 10.

6. The method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan according to claim 4, characterized in that: The specific process of step five is as follows: Step 501: Select the abnormal voltage regulation ratios corresponding to the (j-3), (j-2), (j-1), and (j-2)-j evaluation periods, and construct a fitting model of the abnormal voltage regulation ratio as a function of operating time, using the evaluation period number as the time variable. The expression is: ； in, λ ( t (For runtime) t The fitted value of the abnormal voltage regulation ratio, where m is the shape parameter. For scale parameters; Step 502, using the voltage regulation anomaly proportional model parameters m and determined in step 501. Calculate the abnormal voltage regulation ratio for the next evaluation period. The corresponding predicted value of the abnormal voltage regulation ratio The calculation formula is: ； If the voltage regulation ratio is abnormal If the value is ≥X%, the gas pressure regulator is deemed to have a risk of performance degradation in the next evaluation period.

7. The method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan according to claim 6, characterized in that: The specific process of step six is ​​as follows: Voltage regulation failure occurred: When a gas pressure regulator fails to regulate pressure during the current evaluation period, it indicates that the gas pressure regulator has some faults and can no longer meet the requirements for safe operation. At this time, the gas pressure regulator should be inspected on-site, and gas supply should be stopped or emergency repair measures should be taken according to the inspection results. Voltage regulation performance degradation has occurred or is predicted to occur: When a gas pressure regulator is determined to have experienced a decline in its pressure regulation performance during the current evaluation period, or when a risk of performance decline is predicted for the next evaluation period, the following handling strategy will be adopted based on a comprehensive assessment of the regulator's operating status and gas demand: If the comprehensive evaluation results indicate that the degradation of the pressure regulating performance of the pressure regulator has not affected its normal use, then the real-time operating data marked as pressure regulation abnormality in the current evaluation period will be added to the corresponding pressure sample segment interval as new training data. Then, the updated segmented interval training data is retrained using the method in step (iii), the segmented linear fitting relationship and the corresponding pressure health range are updated, and a new real-time operation phase is entered to achieve the performance reduction of the gas pressure regulator in service. If the comprehensive evaluation results indicate that the reduced pressure regulation performance of the gas regulator has affected its normal use, then the gas regulator shall be repaired. Normal operation status: If the in-service gas pressure regulator does not experience pressure regulation failure during the current evaluation period, and no pressure regulation performance degradation occurs during the current evaluation period or the predicted next evaluation period, it indicates that the pressure regulation performance of the in-service gas pressure regulator can meet the current usage requirements, and no treatment measures are required. Handling after maintenance or resetting operating conditions: When the gas pressure regulator has been repaired, parts replaced, or the gas outlet pressure has been reset according to the gas demand, it is considered that the operating condition has changed. Steps (I) to (III) are then executed again to rebuild the corresponding training model. Based on this, a new real-time operation phase is entered to complete the post-repair operation management of the in-service gas pressure regulator.

8. The method for evaluating the performance degradation of in-service gas pressure regulators based on BX lifespan according to claim 1, characterized in that: The amount of training data should be no less than one year during the data training phase, and the number of training data samples should be no less than 600.

Images