CN-121977170-A - Method and system for predicting dynamic leakage of oil and gas pipeline

CN121977170ACN 121977170 ACN121977170 ACN 121977170ACN-121977170-A

Abstract

The invention discloses a method and a system for predicting dynamic leakage of an oil and gas pipeline, which relate to the technical field of oil and gas pipeline monitoring and comprise the steps of synchronously collecting pressure, flow and orthogonal annular strain sequences of a monitoring pipeline section; calculating an equivalent elliptical sectional area based on strain, constructing an elliptical respiration nonlinear abnormality index by combining a pressure sequence, carrying out nonlinear correction on geometric volume variation of a pipe section by using the index to obtain corrected respiration storage capacity, generating a leakage precursor index based on inlet and outlet flow difference and the correction quantity, and finally judging leakage trend grade according to a steady threshold value of non-leakage historical data. The invention can obviously improve the accuracy and the false alarm resistance of leakage early warning.

Inventors

Wen Chenlin
ZHANG ZHANHAO
LI SONGDA
Xing Linghai
WU FENG
WANG XIAOYING
SHANG WEI
HOU PINGPING
LIU XIN
HUO YUWEI

Assignees

烟台港裕龙管输仓储物流有限公司

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (9)

1. A method for predicting dynamic leakage of an oil and gas pipeline, comprising: s1, synchronously acquiring a pressure sequence, an inlet volume flow sequence, an outlet volume flow sequence, a first circumferential strain sequence and a second circumferential strain sequence of a monitoring pipe section, and performing time mark alignment and denoising treatment on the acquired sequences; s2, calculating an equivalent elliptical sectional area sequence based on the first circumferential strain sequence and the second circumferential strain sequence; s3, calculating an elliptical respiration nonlinear abnormality index based on the pressure sequence and the equivalent elliptical sectional area sequence; S4, determining geometric volume variation according to the equivalent elliptical sectional area sequence, and correcting the geometric volume variation based on the elliptical respiration nonlinear abnormality index to obtain corrected respiration storage capacity; S5, generating a leakage precursor index based on the inlet volume flow sequence, the outlet volume flow sequence and the corrected respiratory storage volume; and S6, judging the leakage precursor index according to a steady threshold value generated by the history data without leakage, and obtaining the leakage trend grade.
2. The method for predicting the dynamic leakage of the oil and gas pipeline according to claim 1, wherein the denoising processing is carried out on each collected sequence, and the method is characterized in that arithmetic average calculation is carried out on the current sampling point and the preamble sampling point of each sequence through a sliding window with a set width.
3. The method of claim 1, wherein calculating the equivalent elliptical cross-sectional area comprises: Acquiring a nominal inner diameter of the pipeline, and calculating a nominal circular sectional area based on the nominal inner diameter of the pipeline; constructing a correction factor reflecting the expansion and contraction of the section main shaft according to the first circumferential strain sequence and the second circumferential strain sequence; And obtaining the equivalent elliptical sectional area according to the product of the nominal circular sectional area and the correction factor.
4. A method of predicting dynamic leakage of an oil and gas pipeline according to claim 3, wherein calculating an elliptical breathing nonlinear anomaly index based on a pressure sequence and an equivalent elliptical cross-sectional area sequence comprises: constructing a discrete derivative based on the equivalent elliptical sectional area sequence and the pressure sequence, and calculating the first-order sensitivity and the second-order sensitivity of the equivalent elliptical sectional area to the pressure; obtaining a nonlinear degree index according to the ratio of the second-order sensitivity absolute value to the first-order sensitivity absolute value; determining a last pressure round trip interval according to adjacent local maximum value points and local minimum value points of a pressure sequence, and dispersing the pressure in the pressure round trip interval into pressure grid points; Integrating the absolute value of the difference value of the equivalent elliptical sectional area corresponding to the boosting section and the equivalent elliptical sectional area corresponding to the reducing section of each pressure grid point in the pressure round trip interval, and normalizing the integration result according to the total pressure change range to obtain a hysteresis intensity index; And carrying out gain correction on the nonlinear degree index based on the hysteresis intensity index to obtain an elliptical respiration nonlinear abnormality index.
5. The method of claim 4, wherein determining the geometric volume change according to the equivalent elliptical cross-sectional area sequence and correcting the geometric volume change based on the elliptical respiratory nonlinear anomaly index to obtain the corrected respiratory storage volume comprises: Acquiring the effective length and the temperature of a monitoring pipe section; Obtaining a geometric volume increment according to the product of the equivalent elliptical sectional area difference value of adjacent sampling moments and the effective length of the monitoring pipe section; converting the geometric volume increment into geometric volume variation according to the pressure and temperature conditions in the monitored pipe section; Setting a correction coefficient, and constructing an amplification factor according to the correction coefficient and the elliptic breathing nonlinear abnormality index; and obtaining the corrected respiratory storage volume according to the product of the geometric volume change and the amplification factor.
6. The method for predicting dynamic leakage of oil and gas pipeline according to claim 5, wherein setting the correction coefficient includes: and counting median values of the elliptic breathing nonlinear abnormality indexes of the non-leakage historical normal operation data, and constructing steady statistics with values between zero and one based on the median values as correction coefficients.
7. The method of claim 5, wherein generating a leak precursor index based on the inlet volume flow sequence, the outlet volume flow sequence, and the corrected breath storage amount comprises: Calculating the difference between the inlet flow and the outlet flow, and deducting the corrected respiratory storage capacity to obtain a volume balance residual error; Accumulating the volume balance residual errors in a set observation time window to obtain an accumulated unbalance amount; Obtaining the nominal total volume of the monitoring pipe section according to the product of the effective length of the monitoring pipe section and the nominal circular sectional area; And carrying out normalization processing on the absolute value of the accumulated unbalance according to the nominal total volume, and multiplying the normalization result by the elliptical respiration nonlinear abnormality index to obtain the leakage precursor index.
8. The method of claim 1, wherein determining a robust threshold for a leakage precursor index over non-leakage history data, determining a leakage trend level based on the robust threshold, comprises: constructing a baseline set based on leakage precursor indices of the leakage-free history data; Calculating the median and the median absolute deviation of the baseline set, weighting the median absolute deviation, and summing the median and the weighted median absolute deviation to obtain a robust threshold of the leakage precursor index; The leakage trend level includes a normal trend, leakage precursor concern, and leakage precursor high risk, wherein the output trend is normal when the leakage precursor index is less than the robust threshold, the leakage precursor concern is output when the leakage precursor index is not less than the robust threshold and less than twice the robust threshold, and the leakage precursor high risk is output when the leakage precursor index is not less than twice the robust threshold.
9. An oil and gas pipeline dynamic leak prediction system, the system comprising: The data preprocessing module is used for synchronously acquiring a pressure sequence, an inlet volume flow sequence, an outlet volume flow sequence, a first circumferential strain sequence and a second circumferential strain sequence of the monitoring pipe section, and performing time mark alignment and denoising processing on the acquired sequences; the section calculation module is used for calculating an equivalent elliptical sectional area sequence based on the first circumferential strain sequence and the second circumferential strain sequence; The abnormality index module is used for calculating an elliptical respiration nonlinear abnormality index based on the pressure sequence and the equivalent elliptical sectional area sequence; The correction module is used for determining geometric volume variation according to the equivalent elliptical sectional area sequence, correcting the geometric volume variation based on the elliptical respiration nonlinear abnormality index and obtaining corrected respiration storage quantity; a precursor generation module for generating a leakage precursor index based on the inlet volume flow sequence, the outlet volume flow sequence, and the corrected breath storage amount; and the trend judging module is used for judging the leakage precursor index according to a robust threshold value generated by the leakage-free historical data to obtain a leakage trend grade.

Description

Method and system for predicting dynamic leakage of oil and gas pipeline Technical Field The invention relates to the technical field of oil and gas pipeline monitoring, in particular to a method and a system for predicting dynamic leakage of an oil and gas pipeline. Background In the actual running process, the pipeline often faces dynamic operations such as start-stop switching, valve adjustment, pump station working condition disturbance and the like, so that the pressure and flow in the pipeline show remarkable fluctuation change and are influenced by external factors such as foundation settlement in a buried environment, soil extrusion or limited deformation of the pipeline, the flow section of the pipeline often is not perfectly round, but shows a certain degree of ovalization form, when the pressure in the pipeline dynamically fluctuates along with the working condition, the ovalization section can generate rounding deformation or rebound under the action of the internal pressure, namely a so-called breathing phenomenon, and the effect can directly interfere the balance calculation of inlet and outlet flow, so that a volume imbalance signal caused by leakage is covered or confused by the dynamic change of breathing storage quantity, and early pipeline leakage is difficult to accurately identify. The existing dynamic leakage monitoring technology generally adopts a mass balance method or a transient model method, calculates the density change of medium in a pipe mainly by monitoring the pressure and temperature change, and corrects the stock of the pipe, however, most of the existing methods are based on idealized assumptions, namely that the section of the pipe is considered to be regular round and keeps unchanged shape under the action of pressure, or only linear elastic expansion of pipe materials is considered, in fact, the buried pipe is influenced by soil external load or construction factors, the section of the buried pipe is often subjected to a certain degree of initial ovalization, when the pressure in the pipe greatly fluctuates, the section of the pipe tends to be round respiratory deformation, the deformation not only causes nonlinear change of the flow sectional area, but also tends to represent a hysteresis phenomenon of path misalignment in the processes of pressure rise and pressure reduction, so that the theoretical stock calculated in the dynamic regulation process has larger deviation from the actual physical process, and the deviation is extremely easy to be misjudged as a leakage signal, thereby causing misinformation or leakage of monitoring. Disclosure of Invention The invention aims to solve the problem that in the prior art, nonlinear volume change and hysteresis effect caused by elliptical respiration of a pipeline section are ignored under a dynamic working condition, so that apparent unbalance caused by the respiration effect and continuous unbalance caused by real leakage are difficult to distinguish, and provides a method and a system for predicting dynamic leakage of an oil and gas pipeline. In order to solve the problems existing in the prior art, the invention adopts the following technical scheme: a method of oil and gas pipeline dynamic leak prediction comprising: s1, synchronously acquiring a pressure sequence, an inlet volume flow sequence, an outlet volume flow sequence, a first circumferential strain sequence and a second circumferential strain sequence of a monitoring pipe section, and performing time mark alignment and denoising treatment on the acquired sequences; s2, calculating an equivalent elliptical sectional area sequence based on the first circumferential strain sequence and the second circumferential strain sequence; s3, calculating an elliptical respiration nonlinear abnormality index based on the pressure sequence and the equivalent elliptical sectional area sequence; S4, determining geometric volume variation according to the equivalent elliptical sectional area sequence, and correcting the geometric volume variation based on the elliptical respiration nonlinear abnormality index to obtain corrected respiration storage capacity; S5, generating a leakage precursor index based on the inlet volume flow sequence, the outlet volume flow sequence and the corrected respiratory storage volume; and S6, judging the leakage precursor index according to a steady threshold value generated by the history data without leakage, and obtaining the leakage trend grade. Preferably, denoising each acquired sequence comprises arithmetic average calculation of a current sampling point and a preamble sampling point of each sequence through a sliding window with a set width. Preferably, calculating the equivalent elliptical cross-sectional area includes: Acquiring a nominal inner diameter of the pipeline, and calculating a nominal circular sectional area based on the nominal inner diameter of the pipeline; constructing a correction factor reflecting the expansion and contr