CN115982641B - System and method for detecting carbon dioxide corrosion of oilfield tubular column - Google Patents
System and method for detecting carbon dioxide corrosion of oilfield tubular column Download PDFInfo
- Publication number
- CN115982641B CN115982641B CN202211664997.1A CN202211664997A CN115982641B CN 115982641 B CN115982641 B CN 115982641B CN 202211664997 A CN202211664997 A CN 202211664997A CN 115982641 B CN115982641 B CN 115982641B
- Authority
- CN
- China
- Prior art keywords
- flaw detection
- detection point
- damage
- oil pipe
- sleeve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 title claims abstract description 26
- 230000007797 corrosion Effects 0.000 title claims abstract description 22
- 238000005260 corrosion Methods 0.000 title claims abstract description 22
- 229910002092 carbon dioxide Inorganic materials 0.000 title claims abstract description 13
- 239000001569 carbon dioxide Substances 0.000 title claims abstract description 13
- 238000001514 detection method Methods 0.000 claims abstract description 429
- 238000003062 neural network model Methods 0.000 claims abstract description 70
- 238000003384 imaging method Methods 0.000 claims abstract description 18
- 239000000463 material Substances 0.000 claims description 41
- 238000005070 sampling Methods 0.000 claims description 38
- 230000015654 memory Effects 0.000 claims description 36
- 238000012360 testing method Methods 0.000 claims description 18
- 230000007787 long-term memory Effects 0.000 claims description 6
- 230000006403 short-term memory Effects 0.000 claims description 6
- 230000007774 longterm Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000002378 acidificating effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/70—Combining sequestration of CO2 and exploitation of hydrocarbons by injecting CO2 or carbonated water in oil wells
Landscapes
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention relates to a carbon dioxide corrosion detection system and method for an oilfield tubular column, which are characterized in that related damage points possibly damaged and sleeve damage points are obtained through a neural network model and a similarity model, an electromagnetic flaw detection logging instrument is utilized to detect the damage points of an oil pipe and a sleeve, the damage points obtained according to the neural network model and the similarity model and the damage points detected by the electromagnetic flaw detection logging instrument are compared, an imaging detection instrument is started according to a comparison result, and inconsistent damage points and damage types are confirmed; the invention can improve the accuracy of damage detection of the oil pipe and the sleeve and the processing efficiency of damage detection.
Description
Technical Field
The invention relates to the field of oilfield tubular column corrosion detection, in particular to a system and a method for detecting carbon dioxide corrosion of an oilfield tubular column.
Background
The oil pipe and the casing of the oil field are important pipelines in an oil conveying device, and the quality of the oil pipe and the casing directly influences the stability of oil conveying engineering. However, over time, tubing and casing are subject to manufacturing, transportation, and subterranean environments, such as high temperature and high pressure, acidic materials, other corrosive materials, etc., where the probability and extent of corrosion will progressively increase. When the corrosion situation is severe beyond the allowable conditions, serious safety accidents are likely to occur. At present, there are many detection techniques and devices for oil pipes and casings in the market, such as a common ultrasonic detection method and an electromagnetic flaw detection method, however, the methods have limitations. For example, electromagnetic flaw detection can meet the requirement of damage detection of a multilayer pipeline, but the accuracy of damage detection is to be improved; while ultrasonic detection is relatively accurate in detecting damage to a pipeline, it does not allow detection of a multilayer pipeline. Meanwhile, the method needs to carry out detection for many times and a large amount of data processing when detecting the damage of the pipeline, and particularly has low detection efficiency aiming at the pipeline with long length.
Aiming at the problems, the invention provides a system and a method for detecting carbon dioxide corrosion of an oilfield tubular column, which are used for improving the accuracy of damage detection of oil pipes and casing pipes and improving the treatment efficiency of damage detection.
Disclosure of Invention
Aiming at the problems that the accuracy of the current damage detection of oil pipes and casing pipes is to be improved and the detection efficiency is to be improved, the invention provides a method for detecting the corrosion of carbon dioxide in an oilfield tubular column. The method comprises the following steps:
S1, calculating and obtaining damage points and damage types of oil pipes and sleeves by using a neural network model and a similarity model;
S2, starting an electromagnetic flaw detection logging instrument, and detecting damage points and damage types of the oil pipe and the sleeve downwards from an oil pipe wellhead;
s3, setting sampling points of the electromagnetic flaw detection logging instrument according to the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1;
S4, detecting the oil pipe and the sleeve downwards from the wellhead of the oil pipe by using the electromagnetic flaw detection logging instrument according to the sampling points set in the step S3, and obtaining damage points and damage types of the oil pipe and the sleeve;
s5, judging whether the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1 are consistent with the damage points and the damage types obtained in the step S4, starting an ultrasonic imaging detector, and confirming inconsistent damage points and damage types;
S6, optimizing the neural network model in the step S1 according to the data of the damage points and the damage types detected by the ultrasonic imaging in the step S5.
The step S1 is to calculate and obtain damage points and damage types of the oil pipe and the sleeve by using a neural network model and a similarity model, and specifically comprises the following steps:
s11, constructing an oil pipe long-term and short-term memory neural network model and a sleeve long-term and short-term neural network model;
s12, obtaining flaw detection point samples of the oil pipe and the sleeve in a historical database, and forming an oil pipe flaw detection point sample set and a sleeve flaw detection sample set; taking 70% of oil pipe flaw detection point sample sets as oil pipe training samples and 30% as oil pipe test samples; taking 70% of the sleeve flaw detection point sample set as sleeve training samples and 30% as sleeve test samples;
The oil pipe flaw detection sample comprises: flaw detection point position, flaw detection point damage type, flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data;
The oil pipe flaw detection sample comprises: flaw detection point position, flaw detection point damage type, flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data;
The history database comprises flaw detection point sample data of the oil pipe and the sleeve;
The flaw detection point damage types comprise: grooves, holes, flakes and no damage;
S13, training the oil pipe long-short-term memory neural network model according to the oil pipe training sample obtained in the step S12, testing the oil pipe long-short-term memory neural network model by adopting an oil pipe test sample, and iteratively optimizing the oil pipe long-short-term memory neural network model to obtain the trained oil pipe long-short-term memory neural network model; training the sleeve long-short-term memory neural network model according to the sleeve training sample obtained in the step S12, testing the sleeve long-short-term memory neural network model by adopting the sleeve test sample, and iteratively optimizing the sleeve long-short-term memory neural network model to obtain a trained sleeve long-short-term memory neural network model;
The data input of the oil pipe long-term and short-term memory neural network model in the training process is flaw detection point positions of the oil pipe, a geological environment parameter sequence of the flaw detection points, flaw detection point thickness and flaw detection point material data, and the data is output as flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe;
The data input of the sleeve long-short term memory neural network model in the training process is the flaw detection point position of the sleeve, the geological environment parameter sequence of the flaw detection point, the thickness of the flaw detection point and the material data of the flaw detection point, and the data is output as the flaw detection point damage type corresponding to the flaw detection point position of the sleeve;
S14, obtaining a plurality of flaw detection points of the oil pipe and a plurality of flaw detection points of the sleeve according to preset distance intervals; the method comprises the steps of obtaining the flaw detection point position of a current oil pipe and the flaw detection point position of a current sleeve pipe, inputting the flaw detection point position of the current oil pipe, the geological environment parameter sequence data of the flaw detection point, the thickness data of the flaw detection point and the material data of the flaw detection point into an oil pipe long-short-period memory neural network model, and obtaining the flaw detection point damage type corresponding to the flaw detection point position of the current oil pipe;
Inputting the flaw detection point position, the geological environment parameter sequence data, the flaw detection point thickness data and the flaw detection point material data of the current sleeve into a sleeve long-and-short-term memory neural network model to obtain the flaw detection point damage type corresponding to the flaw detection point position of the current sleeve;
The number and the positions of the flaw detection points of the current oil pipe and the flaw detection points of the current sleeve are the same;
S15, extracting flaw detection point positions of grooves, holes and sheet corrosion of flaw detection points in the oil pipe and the sleeve, taking each flaw detection point position, the pipe column type corresponding to the flaw detection point and the flaw detection point damage type corresponding to the flaw detection point position as a record of damage points, damage point pipe column types and damage point damage types, and forming a damage point set S1 with a plurality of records;
The damage point pipe column is of an oil pipe and a casing pipe;
The flaw detection points are corresponding to and identical to the damage points, the pipe column types corresponding to the flaw detection points are corresponding to and identical to the damage point pipe column types, and the flaw detection point damage types corresponding to the flaw detection points are corresponding to and identical to the damage point damage types;
S16, calculating flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe and the sleeve by using the similarity model by adopting the flaw detection point position of the current oil pipe and the flaw detection point position of the current sleeve obtained in the step S14;
S17, adjusting the damage point set S1 obtained in the step S15 according to the flaw detection point positions obtained in the step S16 and flaw detection point damage types corresponding to the flaw detection point positions, and finally obtaining damage points and damage types of the oil pipe and the sleeve.
Step S16 adopts the current flaw detection point position of the oil pipe and the current flaw detection point position of the sleeve pipe obtained in step S14, calculates flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe and the sleeve pipe by using a similarity model, and specifically comprises the following steps:
S161, acquiring a flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data corresponding to the flaw detection point position of the current oil pipe by adopting the flaw detection point position of the current oil pipe acquired in the step S14; adopting the flaw detection point position of the current sleeve obtained in the step S14 to obtain a geological environment parameter sequence of the flaw detection point, the thickness of the flaw detection point and material data of the flaw detection point corresponding to the flaw detection point position of the current sleeve;
S162, comparing the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to the flaw detection point position of the current oil pipe obtained in the step S161 with the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to each flaw detection point of the oil pipe in a historical experience database, if the similarity is smaller than a preset threshold value, taking the flaw detection point damage type corresponding to the flaw detection point position with the similarity smaller than the preset threshold value in the historical experience database as the flaw detection point damage type of the current oil pipe flaw detection point position;
Performing similarity comparison on the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to the flaw detection point position of the current sleeve obtained in the step S161 and the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to each flaw detection point position of the sleeve in the historical experience database, and taking the flaw detection point damage type corresponding to the flaw detection point position with the similarity smaller than the preset threshold in the historical experience database as the flaw detection point damage type of the flaw detection point position of the current sleeve;
S163, extracting flaw detection point positions of grooves, holes and chip corrosion of flaw detection point damage types in the oil pipe and the sleeve obtained in the step S162, and recording the flaw detection point positions, the pipe column types corresponding to the flaw detection points and the flaw detection point damage types corresponding to the flaw detection point positions as a damage point, a damage point pipe column type and a damage point damage type to form a damage point set S2 with a plurality of records;
The damage point pipe column is of an oil pipe and a casing pipe;
The flaw detection points are corresponding to and identical to the damage points, the pipe column types corresponding to the flaw detection points are corresponding to and identical to the damage point pipe column types, and the flaw detection point damage types corresponding to the flaw detection points are corresponding to and identical to the damage point damage types;
Step S17 adjusts the damage point set S1 obtained in step S15 according to the flaw detection point position obtained in step S16 and the flaw detection point damage type corresponding to the flaw detection point position, and finally obtains damage points and damage types of the oil pipe and the casing, and specifically includes:
S171, taking a union set of the sets S2 and S1, and finally obtaining damage points and damage types of the oil pipe and the sleeve. Step S3 is to set sampling points of the electromagnetic flaw detection logging instrument according to the damage points and the damage types of the oil pipe and the casing obtained in step S1, and specifically includes:
S31, obtaining position data of damage points of the oil pipe and the sleeve, which are obtained in the step S171; the position data is the depth of the damage point from the wellhead;
S32, when the distance between two damage points W1 and W2 is larger than a preset threshold D, acquiring the occurrence times of W1 and W2 in the set S1 and the set S2;
S33, when W1 and W2 appear in the set S1 and the set S2, setting a sampling point between the damage points W1 and W2 according to a first sampling frequency; when one damage point in the W1 and the W2 is only in one of the set S1 and the set S2, setting a sampling point between the damage points W1 and W2 according to a second sampling frequency; when W1 and W2 are only in one of the set S1 and the set S2, setting a sampling point between the damage points W1 and W2 according to a third sampling frequency;
the first sampling frequency is greater than the second sampling frequency, which is greater than the third sampling frequency;
the sampling points between the damage points W1 and W2 comprise damage points W1 and W2;
S34, repeating the steps S32-S33 until sampling points among all damage points are set.
Step S5 judges whether the damage points and the damage types of the oil pipe and the sleeve obtained in step S1 are consistent with the damage points and the damage types obtained in step S4, and starts an ultrasonic imaging detector to confirm inconsistent damage points and damage types, and the method specifically comprises the following steps:
S51, when the positions of the damage points obtained in the step S4 are the same as those of the damage points obtained in the step S1, and the damage types corresponding to the positions of the damage points are the same, taking the damage points and the damage types as detection results;
S52, detecting the position of the damaged point by adopting an ultrasonic imaging detector when the position of the damaged point obtained in the step S4 is the same as the damaged point obtained in the step S1 and the damage type corresponding to the damaged point is different, so as to obtain a detection result;
The detection results comprise: the damage point, the damage point tubular column type and the damage point damage type;
S53, detecting the damaged point by an ultrasonic imaging detector when the position of the damaged point obtained in the step S4 is not present in the damaged point position obtained in the step S1 or the damaged point position obtained in the step S1 is not present in the damaged point position obtained in the step S4, so as to obtain a detection result;
The detection results comprise: the damage point, the damage point tubular column type and the damage point damage type;
S54, repeating the steps S51 to S54 until all the damage points obtained in the step S4 and the step S1 are processed, and obtaining the final detection result of the oil pipe and the sleeve.
The step S6 optimizes the neural network model in the step S1 according to the data of the damage point and the damage type detected by the ultrasonic imaging in the step S5, and specifically includes:
s61, recording the confirmed damage points and the confirmed damage types into a history database in the step S12;
S62, updating the oil pipe flaw detection point sample and the sleeve flaw detection point sample in the step S12, and retraining the oil pipe long-short-period memory neural network model and the sleeve long-short-period neural network model.
The beneficial effects of the invention are as follows:
1. According to the invention, the damage points and the damage types are obtained according to the neural network model and the similarity model of the oil pipe and the sleeve, and the possible damage point positions are provided for the electromagnetic flaw detection logging instrument, so that the method can adjust the sampling point frequency of the magnetic flaw detection logging instrument according to the damage points and the damage types, the magnetic flaw detection logging instrument can detect more pertinently, and the overall efficiency of the system is improved;
2. according to the invention, the sampling point frequency of the magnetic flaw detection logging instrument is dynamically adjusted, so that the magnetic flaw detection logging instrument performs key detection on a heavy point area, and the accuracy of pipeline flaw detection is improved;
3. According to the invention, the damage points and the damage types obtained by the neural network model are adjusted by utilizing the damage points and the damage types obtained by the similarity model, so that the accuracy and the comprehensiveness of the positioning of the damage points are improved;
4. The invention carries out comparison analysis on the damage points and the damage types obtained by the magnetic flaw detection logging instrument and the damage points and the damage types obtained by the neural network model and the similarity model so as to judge whether to start the ultrasonic detector, thereby further improving the accuracy of damage detection and the overall detection efficiency of the method.
The foregoing description is only an overview of the present invention, and is intended to be more clearly understood as the present invention, as it is embodied in the following description, and is intended to be more clearly understood as the following description of the preferred embodiments, given in detail, of the present invention, along with other objects, features and advantages of the present invention.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
FIG. 1 is a flow chart of a method for detecting carbon dioxide corrosion of an oilfield tubular string.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
A method for detecting carbon dioxide corrosion of an oilfield tubular column is provided. The method comprises the following steps:
S1, calculating and obtaining damage points and damage types of oil pipes and sleeves by using a neural network model and a similarity model;
the step S1 is to calculate and obtain damage points and damage types of the oil pipe and the sleeve by using a neural network model and a similarity model, and specifically comprises the following steps:
s11, constructing an oil pipe long-term and short-term memory neural network model and a sleeve long-term and short-term neural network model;
s12, obtaining flaw detection point samples of the oil pipe and the sleeve in a historical database, and forming an oil pipe flaw detection point sample set and a sleeve flaw detection sample set; taking 70% of oil pipe flaw detection point sample sets as oil pipe training samples and 30% as oil pipe test samples; taking 70% of the sleeve flaw detection point sample set as sleeve training samples and 30% as sleeve test samples;
The oil pipe flaw detection sample comprises: flaw detection point position, flaw detection point damage type, flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data;
The oil pipe flaw detection sample comprises: flaw detection point position, flaw detection point damage type, flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data;
The history database comprises flaw detection point sample data of the oil pipe and the sleeve;
The flaw detection point damage types comprise: grooves, holes, flakes and no damage;
S13, training the oil pipe long-short-term memory neural network model according to the oil pipe training sample obtained in the step S12, testing the oil pipe long-short-term memory neural network model by adopting an oil pipe test sample, and iteratively optimizing the oil pipe long-short-term memory neural network model to obtain the trained oil pipe long-short-term memory neural network model; training the sleeve long-short-term memory neural network model according to the sleeve training sample obtained in the step S12, testing the sleeve long-short-term memory neural network model by adopting the sleeve test sample, and iteratively optimizing the sleeve long-short-term memory neural network model to obtain a trained sleeve long-short-term memory neural network model;
The data input of the oil pipe long-term and short-term memory neural network model in the training process is flaw detection point positions of the oil pipe, a geological environment parameter sequence of the flaw detection points, flaw detection point thickness and flaw detection point material data, and the data is output as flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe;
The data input of the sleeve long-short term memory neural network model in the training process is the flaw detection point position of the sleeve, the geological environment parameter sequence of the flaw detection point, the thickness of the flaw detection point and the material data of the flaw detection point, and the data is output as the flaw detection point damage type corresponding to the flaw detection point position of the sleeve;
S14, obtaining a plurality of flaw detection points of the oil pipe and a plurality of flaw detection points of the sleeve according to preset distance intervals; the method comprises the steps of obtaining the flaw detection point position of a current oil pipe and the flaw detection point position of a current sleeve pipe, inputting the flaw detection point position of the current oil pipe, the geological environment parameter sequence data of the flaw detection point, the thickness data of the flaw detection point and the material data of the flaw detection point into an oil pipe long-short-period memory neural network model, and obtaining the flaw detection point damage type corresponding to the flaw detection point position of the current oil pipe;
Inputting the flaw detection point position, the geological environment parameter sequence data, the flaw detection point thickness data and the flaw detection point material data of the current sleeve into a sleeve long-and-short-term memory neural network model to obtain the flaw detection point damage type corresponding to the flaw detection point position of the current sleeve;
The number and the positions of the flaw detection points of the current oil pipe and the flaw detection points of the current sleeve are the same;
S15, extracting flaw detection point positions of grooves, holes and sheet corrosion of flaw detection points in the oil pipe and the sleeve, taking each flaw detection point position, the pipe column type corresponding to the flaw detection point and the flaw detection point damage type corresponding to the flaw detection point position as a record of damage points, damage point pipe column types and damage point damage types, and forming a damage point set S1 with a plurality of records;
The damage point pipe column is of an oil pipe and a casing pipe;
The flaw detection points are corresponding to and identical to the damage points, the pipe column types corresponding to the flaw detection points are corresponding to and identical to the damage point pipe column types, and the flaw detection point damage types corresponding to the flaw detection points are corresponding to and identical to the damage point damage types;
S16, calculating flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe and the sleeve by using the similarity model by adopting the flaw detection point position of the current oil pipe and the flaw detection point position of the current sleeve obtained in the step S14;
Step S16 adopts the current flaw detection point position of the oil pipe and the current flaw detection point position of the sleeve pipe obtained in step S14, calculates flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe and the sleeve pipe by using a similarity model, and specifically comprises the following steps:
S161, acquiring a flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data corresponding to the flaw detection point position of the current oil pipe by adopting the flaw detection point position of the current oil pipe acquired in the step S14; adopting the flaw detection point position of the current sleeve obtained in the step S14 to obtain a geological environment parameter sequence of the flaw detection point, the thickness of the flaw detection point and material data of the flaw detection point corresponding to the flaw detection point position of the current sleeve;
S162, comparing the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to the flaw detection point position of the current oil pipe obtained in the step S161 with the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to each flaw detection point of the oil pipe in a historical experience database, if the similarity is smaller than a preset threshold value, taking the flaw detection point damage type corresponding to the flaw detection point position with the similarity smaller than the preset threshold value in the historical experience database as the flaw detection point damage type of the current oil pipe flaw detection point position;
Performing similarity comparison on the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to the flaw detection point position of the current sleeve obtained in the step S161 and the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to each flaw detection point position of the sleeve in the historical experience database, and taking the flaw detection point damage type corresponding to the flaw detection point position with the similarity smaller than the preset threshold in the historical experience database as the flaw detection point damage type of the flaw detection point position of the current sleeve;
The calculation method of the similarity comparison comprises the following steps: respectively calculating a flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data corresponding to the current position of the flaw detection point of the oil pipe, and similarity r1, r2 and r3 of the flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data corresponding to the position of the flaw detection point of the oil pipe in a historical experience database;
similarity r1 of the geological environment parameter sequence is calculated by adopting a numerical value corresponding to the parameter category and the parameter;
the thickness similarity r2 of the flaw detection points is obtained by calculation according to the difference value of the thickness values of the two flaw detection points and the average value of the thickness values of the two flaw detection points, and the specific calculation mode adopts a common calculation mode in the prior art;
The similarity r3 of the flaw detection point material data is obtained according to a material similarity corresponding table stored in advance;
S163, extracting flaw detection point positions of grooves, holes and chip corrosion of flaw detection point damage types in the oil pipe and the sleeve obtained in the step S162, and recording the flaw detection point positions, the pipe column types corresponding to the flaw detection points and the flaw detection point damage types corresponding to the flaw detection point positions as a damage point, a damage point pipe column type and a damage point damage type to form a damage point set S2 with a plurality of records;
The damage point pipe column is of an oil pipe and a casing pipe;
The flaw detection points are corresponding to and identical to the damage points, the pipe column types corresponding to the flaw detection points are corresponding to and identical to the damage point pipe column types, and the flaw detection point damage types corresponding to the flaw detection points are corresponding to and identical to the damage point damage types;
S17, adjusting the damage point set S1 obtained in the step S15 according to the flaw detection point positions obtained in the step S16 and flaw detection point damage types corresponding to the flaw detection point positions, and finally obtaining damage points and damage types of the oil pipe and the sleeve.
Step S17 adjusts the damage point set S1 obtained in step S15 according to the flaw detection point position obtained in step S16 and the flaw detection point damage type corresponding to the flaw detection point position, and finally obtains damage points and damage types of the oil pipe and the casing, and specifically includes:
S171, taking a union set of the sets S2 and S1, and finally obtaining damage points and damage types of the oil pipe and the sleeve.
S2, starting an electromagnetic flaw detection logging instrument, and detecting damage points and damage types of the oil pipe and the sleeve downwards from an oil pipe wellhead;
s3, setting sampling points of the electromagnetic flaw detection logging instrument according to the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1;
S4, detecting the oil pipe and the sleeve downwards from the wellhead of the oil pipe by using the electromagnetic flaw detection logging instrument according to the sampling points set in the step S3, and obtaining damage points and damage types of the oil pipe and the sleeve;
Step S3 is to set sampling points of the electromagnetic flaw detection logging instrument according to the damage points and the damage types of the oil pipe and the casing obtained in step S1, and specifically includes:
S31, obtaining position data of damage points of the oil pipe and the sleeve, which are obtained in the step S171; the position data is the depth of the damage point from the wellhead;
S32, when the distance between two damage points W1 and W2 is larger than a preset threshold D, acquiring the occurrence times of W1 and W2 in the set S1 and the set S2;
S33, when W1 and W2 appear in the set S1 and the set S2, setting a sampling point between the damage points W1 and W2 according to a first sampling frequency; when one damage point in the W1 and the W2 is only in one of the set S1 and the set S2, setting a sampling point between the damage points W1 and W2 according to a second sampling frequency; when W1 and W2 are only in one of the set S1 and the set S2, setting a sampling point between the damage points W1 and W2 according to a third sampling frequency;
the first sampling frequency is greater than the second sampling frequency, which is greater than the third sampling frequency;
the specific way of setting the sampling point by using the sampling frequency between the two damage points adopts a common way in the prior art, and is not described herein again;
the sampling points between the damage points W1 and W2 comprise damage points W1 and W2;
S34, repeating the steps S32-S33 until sampling points among all damage points are set.
S5, judging whether the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1 are consistent with the damage points and the damage types obtained in the step S4, starting an ultrasonic imaging detector, and confirming inconsistent damage points and damage types;
step S5 judges whether the damage points and the damage types of the oil pipe and the sleeve obtained in step S1 are consistent with the damage points and the damage types obtained in step S4, and starts an ultrasonic imaging detector to confirm inconsistent damage points and damage types, and the method specifically comprises the following steps:
S51, when the positions of the damage points obtained in the step S4 are the same as those of the damage points obtained in the step S1, and the damage types corresponding to the positions of the damage points are the same, taking the damage points and the damage types as detection results;
S52, detecting the position of the damaged point by adopting an ultrasonic imaging detector when the position of the damaged point obtained in the step S4 is the same as the damaged point obtained in the step S1 and the damage type corresponding to the damaged point is different, so as to obtain a detection result;
The detection results comprise: the damage point, the damage point tubular column type and the damage point damage type;
S53, detecting the damaged point by an ultrasonic imaging detector when the position of the damaged point obtained in the step S4 is not present in the damaged point position obtained in the step S1 or the damaged point position obtained in the step S1 is not present in the damaged point position obtained in the step S4, so as to obtain a detection result;
The detection results comprise: the damage point, the damage point tubular column type and the damage point damage type;
S54, repeating the steps S51 to S54 until all the damage points obtained in the step S4 and the step S1 are processed, and obtaining the final detection result of the oil pipe and the sleeve.
S6, optimizing the neural network model in the step S1 according to the data of the damage points and the damage types detected by the ultrasonic imaging in the step S5.
The step S6 optimizes the neural network model in the step S1 according to the data of the damage point and the damage type detected by the ultrasonic imaging in the step S5, and specifically includes:
s61, recording the confirmed damage points and the confirmed damage types into a history database in the step S12;
S62, updating the oil pipe flaw detection point sample and the sleeve flaw detection point sample in the step S12, and retraining the oil pipe long-short-period memory neural network model and the sleeve long-short-period neural network model.
The invention also provides an oilfield tubular column carbon dioxide corrosion detection system, which is used for executing the following method:
S1, calculating and obtaining damage points and damage types of oil pipes and sleeves by using a neural network model and a similarity model;
S2, starting an electromagnetic flaw detection logging instrument, and detecting damage points and damage types of the oil pipe and the sleeve downwards from an oil pipe wellhead;
s3, setting sampling points of the electromagnetic flaw detection logging instrument according to the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1;
S4, detecting the oil pipe and the sleeve downwards from the wellhead of the oil pipe by using the electromagnetic flaw detection logging instrument according to the sampling points set in the step S3, and obtaining damage points and damage types of the oil pipe and the sleeve;
s5, judging whether the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1 are consistent with the damage points and the damage types obtained in the step S4, starting an ultrasonic imaging detector, and confirming inconsistent damage points and damage types;
S6, optimizing the neural network model in the step S1 according to the data of the damage points and the damage types detected by the ultrasonic imaging in the step S5.
The invention also provides a storage medium for storing computer instruction code, which is called by a processor to execute the method.
The beneficial effects of the invention are as follows:
1. According to the invention, the damage points and the damage types are obtained according to the neural network model and the similarity model of the oil pipe and the sleeve, and the possible damage point positions are provided for the electromagnetic flaw detection logging instrument, so that the method can adjust the sampling point frequency of the magnetic flaw detection logging instrument according to the damage points and the damage types, the magnetic flaw detection logging instrument can detect more pertinently, and the overall efficiency of the system is improved;
2. according to the invention, the sampling point frequency of the magnetic flaw detection logging instrument is dynamically adjusted, so that the magnetic flaw detection logging instrument performs key detection on a heavy point area, and the accuracy of pipeline flaw detection is improved;
3. According to the invention, the damage points and the damage types obtained by the neural network model are adjusted by utilizing the damage points and the damage types obtained by the similarity model, so that the accuracy and the comprehensiveness of the positioning of the damage points are improved;
4. The invention carries out comparison analysis on the damage points and the damage types obtained by the magnetic flaw detection logging instrument and the damage points and the damage types obtained by the neural network model and the similarity model so as to judge whether to start the ultrasonic detector, thereby further improving the accuracy of damage detection and the overall detection efficiency of the method.
The present invention is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present invention are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (5)
1. The method for detecting the carbon dioxide corrosion of the oilfield tubular column is characterized by comprising the following steps of:
S1, calculating and obtaining damage points and damage types of oil pipes and sleeves by using a neural network model and a similarity model;
S2, starting an electromagnetic flaw detection logging instrument, and detecting damage points and damage types of the oil pipe and the sleeve downwards from an oil pipe wellhead;
s3, setting sampling points of the electromagnetic flaw detection logging instrument according to the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1;
S4, detecting the oil pipe and the sleeve downwards from the wellhead of the oil pipe by using the electromagnetic flaw detection logging instrument according to the sampling points set in the step S3, and obtaining damage points and damage types of the oil pipe and the sleeve;
s5, judging whether the damage points and the damage types of the oil pipe and the sleeve obtained in the step S1 are consistent with the damage points and the damage types obtained in the step S4, starting an ultrasonic imaging detector, and confirming inconsistent damage points and damage types;
S6, optimizing the neural network model in the step S1 according to the data of the damage points and the damage types confirmed by ultrasonic imaging detection in the step S5;
the step S1 is to calculate and obtain damage points and damage types of the oil pipe and the sleeve by using a neural network model and a similarity model, and specifically comprises the following steps:
s11, constructing an oil pipe long-term and short-term memory neural network model and a sleeve long-term and short-term neural network model;
s12, obtaining flaw detection point samples of the oil pipe and the sleeve in a historical database, and forming an oil pipe flaw detection point sample set and a sleeve flaw detection sample set; taking 70% of oil pipe flaw detection point sample sets as oil pipe training samples and 30% as oil pipe test samples; taking 70% of the sleeve flaw detection point sample set as sleeve training samples and 30% as sleeve test samples;
The oil pipe flaw detection sample comprises: flaw detection point position, flaw detection point damage type, flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data;
The history database comprises flaw detection point sample data of the oil pipe and the sleeve;
The flaw detection point damage types comprise: grooves, holes, flakes and no damage;
S13, training the oil pipe long-short-term memory neural network model according to the oil pipe training sample obtained in the step S12, testing the oil pipe long-short-term memory neural network model by adopting an oil pipe test sample, and iteratively optimizing the oil pipe long-short-term memory neural network model to obtain the trained oil pipe long-short-term memory neural network model; training the sleeve long-short-term memory neural network model according to the sleeve training sample obtained in the step S12, testing the sleeve long-short-term memory neural network model by adopting the sleeve test sample, and iteratively optimizing the sleeve long-short-term memory neural network model to obtain a trained sleeve long-short-term memory neural network model;
The data input of the oil pipe long-term and short-term memory neural network model in the training process is flaw detection point positions of the oil pipe, a geological environment parameter sequence of the flaw detection points, flaw detection point thickness and flaw detection point material data, and the data is output as flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe;
The data input of the sleeve long-short term memory neural network model in the training process is the flaw detection point position of the sleeve, the geological environment parameter sequence of the flaw detection point, the thickness of the flaw detection point and the material data of the flaw detection point, and the data is output as the flaw detection point damage type corresponding to the flaw detection point position of the sleeve;
S14, obtaining a plurality of flaw detection points of the oil pipe and a plurality of flaw detection points of the sleeve according to preset distance intervals; the method comprises the steps of obtaining the flaw detection point position of a current oil pipe and the flaw detection point position of a current sleeve pipe, inputting the flaw detection point position of the current oil pipe, the geological environment parameter sequence data of the flaw detection point, the thickness data of the flaw detection point and the material data of the flaw detection point into an oil pipe long-short-period memory neural network model, and obtaining the flaw detection point damage type corresponding to the flaw detection point position of the current oil pipe;
Inputting the flaw detection point position, the geological environment parameter sequence data, the flaw detection point thickness data and the flaw detection point material data of the current sleeve into a sleeve long-and-short-term memory neural network model to obtain the flaw detection point damage type corresponding to the flaw detection point position of the current sleeve;
The number and the positions of the flaw detection points of the current oil pipe and the flaw detection points of the current sleeve are the same;
S15, extracting flaw detection point positions of grooves, holes and sheet corrosion of flaw detection points in the oil pipe and the sleeve, taking each flaw detection point position, the pipe column type corresponding to the flaw detection point and the flaw detection point damage type corresponding to the flaw detection point position as a record of damage points, damage point pipe column types and damage point damage types, and forming a damage point set P1 with a plurality of records;
The damage point pipe column is of an oil pipe and a casing pipe;
The flaw detection points are corresponding to and identical to the damage points, the pipe column types corresponding to the flaw detection points are corresponding to and identical to the damage point pipe column types, and the flaw detection point damage types corresponding to the flaw detection points are corresponding to and identical to the damage point damage types;
S16, calculating flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe and the sleeve by using the similarity model by adopting the flaw detection point position of the current oil pipe and the flaw detection point position of the current sleeve obtained in the step S14;
S17, adjusting the damage point set P1 obtained in the step S15 according to the flaw detection point position obtained in the step S16 and the flaw detection point damage type corresponding to the flaw detection point position, and finally obtaining damage points and damage types of the oil pipe and the sleeve.
2. The method for detecting carbon dioxide corrosion of an oilfield tubular string according to claim 1, wherein the step S16 uses the current flaw detection point position of the oil pipe and the current flaw detection point position of the casing obtained in the step S14, and calculates flaw detection point damage types corresponding to the flaw detection point positions of the oil pipe and the casing by using a similarity model, and specifically includes:
S161, acquiring a flaw detection point geological environment parameter sequence, flaw detection point thickness and flaw detection point material data corresponding to the flaw detection point position of the current oil pipe by adopting the flaw detection point position of the current oil pipe acquired in the step S14; adopting the flaw detection point position of the current sleeve obtained in the step S14 to obtain a geological environment parameter sequence of the flaw detection point, the thickness of the flaw detection point and material data of the flaw detection point corresponding to the flaw detection point position of the current sleeve;
S162, comparing the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to the flaw detection point position of the current oil pipe obtained in the step S161 with the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to each flaw detection point of the oil pipe in a historical experience database, if the similarity is smaller than a preset threshold value, taking the flaw detection point damage type corresponding to the flaw detection point position with the similarity smaller than the preset threshold value in the historical experience database as the flaw detection point damage type of the current oil pipe flaw detection point position;
Performing similarity comparison on the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to the flaw detection point position of the current sleeve obtained in the step S161 and the flaw detection point geological environment parameter sequence, the flaw detection point thickness and the flaw detection point material data corresponding to each flaw detection point position of the sleeve in the historical experience database, and taking the flaw detection point damage type corresponding to the flaw detection point position with the similarity smaller than the preset threshold in the historical experience database as the flaw detection point damage type of the flaw detection point position of the current sleeve;
s163, extracting flaw detection point positions of grooves, holes and chip corrosion of flaw detection point damage types in the oil pipe and the sleeve obtained in the step S162, and recording the flaw detection point positions, the pipe column types corresponding to the flaw detection points and the flaw detection point damage types corresponding to the flaw detection point positions as a damage point, a damage point pipe column type and a damage point damage type to form a damage point set P2 with a plurality of records;
The damage point pipe column is of an oil pipe and a casing pipe;
The flaw detection points correspond to and are the same as the damage points, the pipe column types corresponding to the flaw detection points correspond to and are the same as the damage points, and the flaw detection point damage types corresponding to the flaw detection points correspond to and are the same as the damage points.
3. The method for detecting carbon dioxide corrosion of an oilfield tubular string according to claim 2, wherein the step S17 adjusts the damage point set P1 obtained in the step S15 according to the flaw detection point position obtained in the step S16 and the flaw detection point damage type corresponding to the flaw detection point position, and finally obtains damage points and damage types of the oil pipe and the casing, and specifically includes:
S171, merging the collection P2 and the collection P1 to finally obtain damage points and damage types of the oil pipe and the sleeve.
4. The method for detecting carbon dioxide corrosion of an oilfield tubular string according to claim 3, wherein the step S6 optimizes the neural network model in the step S1 according to the data of the damage point and the damage type detected by the ultrasonic imaging in the step S5, specifically comprising:
s61, recording the confirmed damage points and the confirmed damage types into a history database in the step S12;
S62, updating the oil pipe flaw detection point sample and the sleeve flaw detection point sample in the step S12, and retraining the oil pipe long-short-period memory neural network model and the sleeve long-short-period neural network model.
5. A storage medium storing computer instruction code, the computer instruction code being invoked by a processor to perform the method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211664997.1A CN115982641B (en) | 2022-12-22 | 2022-12-22 | System and method for detecting carbon dioxide corrosion of oilfield tubular column |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211664997.1A CN115982641B (en) | 2022-12-22 | 2022-12-22 | System and method for detecting carbon dioxide corrosion of oilfield tubular column |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115982641A CN115982641A (en) | 2023-04-18 |
| CN115982641B true CN115982641B (en) | 2024-05-24 |
Family
ID=85957445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211664997.1A Active CN115982641B (en) | 2022-12-22 | 2022-12-22 | System and method for detecting carbon dioxide corrosion of oilfield tubular column |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115982641B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102313772A (en) * | 2011-05-27 | 2012-01-11 | 中国石油集团川庆钻探工程有限公司 | Oil-gas field oil casing damage detection and evaluation method |
| CN104847335A (en) * | 2015-06-01 | 2015-08-19 | 甘肃瀚海石油科技有限公司 | Transient electromagnetic flaw detector of oil-water well |
| CN106194158A (en) * | 2016-09-28 | 2016-12-07 | 北京捷威思特科技有限公司 | The comprehensive fault detection system of casing tube |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10125602B2 (en) * | 2016-03-24 | 2018-11-13 | King Fahd University Of Petroleum And Minerals | Method for downhole leak detection |
-
2022
- 2022-12-22 CN CN202211664997.1A patent/CN115982641B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102313772A (en) * | 2011-05-27 | 2012-01-11 | 中国石油集团川庆钻探工程有限公司 | Oil-gas field oil casing damage detection and evaluation method |
| CN104847335A (en) * | 2015-06-01 | 2015-08-19 | 甘肃瀚海石油科技有限公司 | Transient electromagnetic flaw detector of oil-water well |
| CN106194158A (en) * | 2016-09-28 | 2016-12-07 | 北京捷威思特科技有限公司 | The comprehensive fault detection system of casing tube |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115982641A (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3140506B1 (en) | Detecting defects in non-nested tubings and casings using calibrated data and time thresholds | |
| EP3478931B1 (en) | Method for in-situ calibration of electromagnetic corrosion detection tools | |
| US20180106764A1 (en) | Multi-point in situ calibration of electromagnetic pipe inspection tools | |
| Martin et al. | New high-definition frequency tool for tubing and multiple casing corrosion detection | |
| US20150377012A1 (en) | Anomaly Recognition System And Methodology | |
| WO2017100387A1 (en) | Fatigue life assessment | |
| CN111626377B (en) | Lithology recognition method, device, equipment and storage medium | |
| CN115982641B (en) | System and method for detecting carbon dioxide corrosion of oilfield tubular column | |
| US20180259671A1 (en) | Determining permeablility based on collar responses | |
| CN112329590B (en) | Pipeline assembly detection system and detection method | |
| CN114154539A (en) | High-sulfur-content gas well casing defect identification method based on direct-current magnetic field and integrated learning | |
| US10662760B2 (en) | Eddy-current responses in nested pipes | |
| CN111912897A (en) | Method, device and equipment for acquiring pipeline defect information and storage medium | |
| CN113550741A (en) | Method for detecting minimum inner diameter of casing | |
| CN108805147B (en) | A kind of tubing and casing shaft sleeve damage characteristics of image mode identification method | |
| US4736298A (en) | Method of detecting collars using computerized pattern recognition | |
| EP3362643B1 (en) | System and method for detecting material loss in a tubular | |
| CN112943224A (en) | Method for calculating dynamic liquid level of heavy oil well | |
| Poort et al. | Data-Driven Detection of Well Events in Mature Gas Fields | |
| Santos et al. | Probabilistic robotics applied to self-localization inside oil wells of an autonomous system | |
| Fathi et al. | High-Precision Single-Leak Detection and Localization in Single-Phase Liquid Pipelines Using the Negative Pressure Wave Technique: An Application in a Real-Field Case Study | |
| Assanelli et al. | Effect of measurement procedures on estimating geometrical parameters of pipes | |
| CN117725514B (en) | Overflow identification processing method and overflow identification processing device | |
| CN115949388B (en) | Sleeve anti-collision ranging early warning method and measuring unit | |
| CN112630853B (en) | Method and device for correcting gas saturation based on unimpeded flow |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |