CN115705512A - Method and system for identifying corrosion factors in pipe and predicting residual life - Google Patents
Method and system for identifying corrosion factors in pipe and predicting residual life Download PDFInfo
- Publication number
- CN115705512A CN115705512A CN202110922905.4A CN202110922905A CN115705512A CN 115705512 A CN115705512 A CN 115705512A CN 202110922905 A CN202110922905 A CN 202110922905A CN 115705512 A CN115705512 A CN 115705512A
- Authority
- CN
- China
- Prior art keywords
- corrosion
- pipe
- main control
- pipeline
- concentration
- 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.)
- Pending
Links
- 238000005260 corrosion Methods 0.000 title claims abstract description 199
- 230000007797 corrosion Effects 0.000 title claims abstract description 199
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000010801 machine learning Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 24
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 19
- 238000011065 in-situ storage Methods 0.000 claims description 16
- 239000000463 material Substances 0.000 claims description 13
- 239000004576 sand Substances 0.000 claims description 13
- 241000894006 Bacteria Species 0.000 claims description 11
- 238000004519 manufacturing process Methods 0.000 claims description 9
- 230000008439 repair process Effects 0.000 claims description 8
- 238000002372 labelling Methods 0.000 claims description 6
- 238000007477 logistic regression Methods 0.000 claims description 6
- 230000001580 bacterial effect Effects 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000004088 simulation Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 4
- 239000007789 gas Substances 0.000 description 10
- 239000000243 solution Substances 0.000 description 5
- 229910000831 Steel Inorganic materials 0.000 description 4
- 239000010959 steel Substances 0.000 description 4
- 230000000844 anti-bacterial effect Effects 0.000 description 2
- 239000003899 bactericide agent Substances 0.000 description 2
- 239000003795 chemical substances by application Substances 0.000 description 2
- 239000011248 coating agent Substances 0.000 description 2
- 238000000576 coating method Methods 0.000 description 2
- 238000005536 corrosion prevention Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000003112 inhibitor Substances 0.000 description 2
- 238000002347 injection Methods 0.000 description 2
- 239000007924 injection Substances 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 238000002156 mixing Methods 0.000 description 2
- 239000003921 oil Substances 0.000 description 2
- 239000010865 sewage Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000003466 welding Methods 0.000 description 2
- 241000219122 Cucurbita Species 0.000 description 1
- 235000009852 Cucurbita pepo Nutrition 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Landscapes
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
Abstract
The invention relates to a method and a system for identifying corrosion factors and predicting residual life in a pipe. The method comprises the following steps: acquiring historical failure data of the pipeline; processing the historical failure data of the pipeline to form an intermediate database; establishing a corrosion rate range prediction model based on the intermediate database; according to the corrosion rate range prediction model, predicting corrosion in the pipe; and determining the main corrosion factor and the residual life in the pipe based on the prediction result of the corrosion in the pipe. The method for identifying the corrosion factors in the pipe and predicting the residual life creatively combines a supervised machine learning technology and a corrosion failure identification technology, fully utilizes the historical failure data of the pipe, and establishes a quick and convenient method for identifying the corrosion main control factors in the pipe and predicting the residual life, thereby not only avoiding errors caused by an indoor corrosion simulation experiment, but also reducing the workload and the capital investment of failure analysis.
Description
Technical Field
The invention belongs to the field of corrosion management in steel pipes for oil field gathering and transportation, and particularly relates to a method and a system for identifying corrosion factors in pipes and predicting residual life.
Background
Along with the increasing of the exploration and development of domestic oil and gas, oil and gas gathering and transportation pipe networks are more and more huge, the problem of pipeline failure is more and more serious, and the oil field oil and gas gathering and transportation pipe network brings serious challenges to green, safe, environment-friendly and efficient production of oil fields. Statistically, more than 85% of oil and gas field pipeline failures are caused by corrosion, wherein the internal corrosion accounts for more than 60%. Oil and gas field gathering and transportation pipeline contains H 2 S、CO 2 、O 2 Isocorrosive gases, cl - 、S2 - The corrosion mechanism is stubborn, complex, difficult to identify corrosion factors, very difficult to make a targeted corrosion prevention scheme and often facing the passive situation of 'pressing down the gourd ladle to float up'. Scientifically evaluating the effect of each factor on the internal corrosion, accurately identifying the main control factor of the internal corrosion and effectively predictingThe residual service life of the pipeline is very important for ensuring the safe and stable operation of the oil field gathering and transportation pipeline.
At present, the identification of corrosion factors in the steel pipe for oil field gathering and transportation under the combined action of multiple factors is determined by indoor corrosion simulation experiments and test analysis, even by inviting experts to discuss. The time span required is often very long and the capital investment required is large. Because of the complexity of the service working condition and the conveying medium of the oil field gathering and transportation pipeline, the field corrosion condition is difficult to be really reduced in the indoor corrosion simulation experiment from the corrosion medium or the service working condition. In addition, the changes of the field process flow, the field operation and the corrosion management mode all have certain influence on the corrosion in the pipe, and the changes, the field operation and the corrosion management mode can not be simulated by an indoor corrosion simulation experiment. The factors all influence the identification precision and the identification efficiency of the internal corrosion main control factors to a certain extent, and the oil field corrosion management work cannot be guided quickly and effectively. In the aspect of residual life prediction, the uniform corrosion rate of the pipeline is generally obtained through the wall thickness reduction amount and the interval time of the pipeline measured in the last two times, or a corrosion coupon is installed to obtain the corrosivity of a conveying medium (often characterized by the uniform corrosion rate), and the residual life of the pipeline is estimated by combining the residual wall thickness of the pipeline. Because the method only takes the corrosion result as the calculation basis, the influence of the corrosion factor is not considered, the calculation precision is low, and the guidance is not strong.
Therefore, a method for identifying corrosion factors and predicting the remaining life in a pipe is needed.
Disclosure of Invention
In view of the above problems, the present invention provides a method for identifying corrosion factors and predicting remaining life in a pipe,
the method comprises the following steps:
acquiring historical failure data of the pipeline;
performing data processing on the historical failure data of the pipeline to form an intermediate database;
establishing a corrosion rate range prediction model based on the intermediate database;
predicting corrosion in the pipe according to the corrosion rate range prediction model;
and determining the main control factors and the residual life of the corrosion in the pipe based on the prediction result of the corrosion in the pipe.
Further, the air conditioner is characterized in that,
the historical pipeline failure data comprises: pipe diameter, wall thickness, production time, regional grade, pipeline material, anticorrosion measure in the pipe, conveying medium, operation temperature, operation pressure, flow rate, water content, sand content and CO 2 Content, H 2 S content, O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, pH value, SRB bacterial content, whether scale exists in the tube, failure time, failure position, failure clock direction and the latest repair time, and the total number of the data items is 28.
Further, the air conditioner is provided with a fan,
the data processing of the historical failure data of the pipeline comprises data structuring, corrosion rate calculation and marking, and H 2 S partial pressure calculation, CO 2 Partial pressure calculation and in situ pH calculation.
Further, the air conditioner is provided with a fan,
the corrosion rate is calculated according to the following formula:
corrosion rate = wall thickness/time of service;
time in service = time out of service-time in production, or
Time in service = time to failure-time to last repair.
Further, the air conditioner is provided with a fan,
the marking method of the corrosion rate comprises the following steps: the corrosion rate is divided into five grades of 'low, medium and low, medium and high'.
Further, the air conditioner is characterized in that,
said H 2 S partial pressure
And CO 2 Partial pressure
Respectively obtained by the following formulas:
in the formula:
is H 2 S partial pressure, unit is MPa; p is the pipeline operating pressure and the unit is MPa;
is H 2 S content in ppm;
is CO 2 Partial pressure in MPa;
is CO 2 The content is expressed in mol%.
Further, the air conditioner is characterized in that,
the intermediate database comprises the following 21 data items: material of pipeline, anticorrosion measure in pipeline, conveying medium, running temp, flow speed, water content, sand content and O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 Partial pressure, in-situ pH and corrosion rate after labeling;
the first 17 data items are from the acquired historical failure data of the pipeline, and the last 4 data items are data obtained by calculation.
Further, the air conditioner is provided with a fan,
the specific method for establishing the corrosion rate range prediction model comprises the following steps: calling the intermediate database, taking the first 20 data of the intermediate database as input parameters, taking the marked corrosion rate as an output result, and calling a logistic regression model through a scimit-lean data packet by using a python language to perform supervised machine learning so as to establish a corrosion rate range prediction model;
the corrosion rate range prediction model is specifically expressed as:
V corr ＝aA+bB+cC+dD+eE+fF+gG+hH+iI+jJ+kK+lL+mM+nN+oO+pP+qQ+rR+sS+tT+u
wherein Vcorr is the corrosion rate range, A-T are respectively the values of 20 input parameters, a-T are the corresponding coefficients of the values of 20 input parameters, and u is the residual value.
Further, the air conditioner is provided with a fan,
the corrosion main control factor in the pipe is determined by the following method:
sorting the corresponding coefficients of the 20 input parameter values, finding out the pipe internal corrosion factor with the maximum coefficient, and marking the pipe internal corrosion factor as a first main control factor;
finding a second large coefficient, when the value of the second large coefficient reaches 70% of the maximum coefficient value, marking the corrosion factor corresponding to the second large coefficient as a second main control factor, otherwise, stopping the search of the main control factor;
finding a third large coefficient, marking a corrosion factor corresponding to the third large coefficient as a third main control factor when the value of the third large coefficient reaches 90% of the value of the second large coefficient, and otherwise, stopping searching the main control factor;
finding a fourth large coefficient, when the value of the fourth large coefficient reaches 95% of the value of the third large coefficient, marking the corrosion factor corresponding to the fourth large coefficient as a fourth main control factor, otherwise, stopping searching the main control factor;
finding a fifth large coefficient, marking a corrosion factor corresponding to the fifth large coefficient as a fifth main control factor when the value of the fifth large coefficient reaches 95% of the value of the fourth large coefficient, and otherwise, stopping searching the main control factor;
and searching other main control factors in turn according to the method for searching the fifth main control factor.
Further, the air conditioner is provided with a fan,
the remaining life is predicted according to the following formula:
T＝(PC-PD/2δn)/(Max V corr )
wherein T is the predicted remaining life in units of y; PC is the wall thickness of the pipeline, and the unit is mm; p is the pipeline operating pressure and the unit is MPa; d is the outer diameter of the pipeline, and the unit is mm; delta is the minimum yield strength of the pipe, and the unit is MPa; n is an intensity design coefficient; max V corr To predict the maximum value of the corrosion rate obtained, the units are mm/y.
The invention also provides a system for identifying the corrosion factors in the pipe and predicting the residual life, which comprises:
the acquisition unit is used for acquiring historical failure data of the pipeline;
the processing unit is used for carrying out data processing on the historical failure data of the pipeline to form an intermediate database;
the establishing unit is used for establishing a corrosion rate range prediction model based on the intermediate database;
the prediction unit is used for predicting corrosion in the pipe according to the corrosion rate range prediction model;
and the determining unit is used for determining the main control factors and the residual life of the corrosion in the pipe based on the prediction result of the corrosion in the pipe.
Further, the air conditioner is provided with a fan,
the intermediate database comprises the following 21 data items: material of pipeline, anticorrosion measure in pipeline, conveying medium, running temp, flow speed, water content, sand content and O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 Partial pressure, in-situ pH and corrosion rate after labeling;
the first 17 data items are from the acquired historical failure data of the pipeline, and the last 4 data items are data obtained by calculation.
Further, the air conditioner is provided with a fan,
the establishing of the corrosion rate range prediction model in the establishing unit specifically comprises the following steps: calling the intermediate database, taking the first 20 data of the intermediate database as input parameters, taking the marked corrosion rate as an output result, and calling a logistic regression model through a scimit-lean data packet by using a python language to perform supervised machine learning so as to establish a corrosion rate range prediction model;
the corrosion rate range prediction model is specifically expressed as:
V corr ＝aA+bB+cC+dD+eE+fF+gG+hH+iI+jJ+kK+lL+mM+nN+oO+pP+qQ+rR+sS+tT+u
wherein Vcorr is the corrosion rate range, A-T are respectively the values of 20 input parameters, a-T are the corresponding coefficients of the values of 20 input parameters, and u is the residual value.
Further, the air conditioner is provided with a fan,
the main control factors for determining corrosion in the tube in the unit comprise:
sorting the corresponding coefficients of the 20 input parameter values, finding out the pipe internal corrosion factor with the maximum coefficient, and marking the pipe internal corrosion factor as a first main control factor;
finding a second big coefficient, marking the corrosion factor corresponding to the second big coefficient as a second main control factor when the value of the second big coefficient reaches 70% of the maximum coefficient value, and otherwise, stopping the searching of the main control factor;
finding a third large coefficient, marking a corrosion factor corresponding to the third large coefficient as a third main control factor when the value of the third large coefficient reaches 90% of the value of the second large coefficient, and otherwise, stopping searching the main control factor;
finding a fourth large coefficient, when the value of the fourth large coefficient reaches 95% of the value of the third large coefficient, marking the corrosion factor corresponding to the fourth large coefficient as a fourth main control factor, otherwise, stopping searching the main control factor;
finding a fifth large coefficient, marking a corrosion factor corresponding to the fifth large coefficient as a fifth main control factor when the value of the fifth large coefficient reaches 95% of the value of the fourth large coefficient, and otherwise, stopping searching the main control factor;
and searching other main control factors in turn according to the method for searching the fifth main control factor.
Further, the air conditioner is provided with a fan,
the remaining life is predicted according to the following formula:
T＝(PC-PD/2δn)/(Max V corr )
wherein T is the predicted remaining life in units of y; PC is the wall thickness of the pipeline, and the unit is mm; p is the pipeline operating pressure and the unit is MPa; d is the outer diameter of the pipeline, and the unit is mm; delta is the minimum yield strength of the pipe, and the unit is MPa; n is an intensity design coefficient; max Vcorr is the maximum predicted corrosion rate obtained in mm/y.
The method for identifying the corrosion factors in the pipe and predicting the residual life creatively combines a supervised machine learning technology and a corrosion failure identification technology, fully utilizes the historical failure data of the pipe, and establishes a quick and convenient method for identifying the main control factors of the internal corrosion and predicting the residual life, thereby not only avoiding the error brought by an indoor corrosion simulation experiment, but also reducing the workload and the capital investment of failure analysis. Meanwhile, an output parameter (corrosion rate) labeling concept and method are creatively provided, the technical problems of poor foundation of field data of the oil field, wide corrosion rate range and difficulty in corrosion rate prediction are solved, machine learning under poor data is realized, indoor test analysis work is not required to be carried out by a third party, and the method is of great importance for improving the analysis efficiency of corrosion failure in the gathering and transportation pipe of the oil field and improving the management level of internal corrosion.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a flow chart illustrating a method for identifying corrosion factors and predicting remaining life in a pipe according to an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
The invention provides a method and a system for identifying corrosion factors in a pipe and predicting the residual life, which can identify the corrosion factors in the pipe and predict the residual life in the pipe. The embodiment of the invention takes the oilfield gathering and transportation steel pipeline as an example for illustration, but the invention is not limited to the identification of corrosion factors and the prediction of residual life in the oilfield gathering and transportation steel pipeline, and the identification of corrosion factors and the prediction of residual life in any pipeline can be applied to the invention.
Fig. 1 shows a method for identifying corrosion factors and predicting remaining life, where fig. 1 shows a flowchart of the method for identifying corrosion factors and predicting remaining life in a pipe according to an embodiment of the present invention, and specifically includes:
acquiring historical failure data of the pipeline;
the historical failure data of the pipeline comprises pipe diameter, wall thickness, production time, regional grade, pipeline material, anticorrosion measures in the pipeline, conveying medium, operation temperature, operation pressure, flow rate, water content, sand content and CO 2 Content, H 2 S content, O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, pH value, SRB bacterial content, whether scale exists in the tube, failure time, failure position, failure clock direction and latest repair time, and the total number of the data items is 28.
Processing the historical failure data of the pipeline to form an intermediate database;
the data processing of the historical failure data of the pipeline comprises data structuring, corrosion rate calculation and marking, and H 2 S partial pressure calculation, CO 2 Partial pressure calculation and in situ pH calculation.
The data structuring method comprises the following steps: the anticorrosion measures in the pipe are assigned with values of ' no, inner coating, inner interpenetration, corrosion inhibitor, bactericide, antisludging agent ' and others ' of 1-7 respectively, the conveying medium is assigned with values of ' oil-gas-water mixed conveying, wet gas, dry gas, purified oil, water injection, water mixing, sewage, others ' of 1-8 respectively, the scale in the pipe is assigned with values of ' 0,1 ' respectively, the failure position is assigned with values of ' horizontal straight pipe section, vertical pipe, near welding line, elbow/tee joint/valve, reducing diameter, and others ' of 1-6 respectively.
The corrosion rate is calculated according to the following formula:
corrosion rate = wall thickness/time of service;
time in service = time out of service-time in production, or
Time in service = time to failure-time to last repair.
The marking method of the corrosion rate comprises the following steps: the corrosion rate is divided into five grades of 'low, medium and low, medium and high', and respectively marked as 1-5.
Said H 2 S partial pressure
And CO 2 Partial pressure
Are obtained by the following formulas:
in the formula:
is H 2 S partial pressure, unit is MPa; p is the pipeline operating pressure and the unit is MPa;
is H 2 The S content is expressed in ppm;
is CO 2 Partial pressure in MPa;
is CO 2 The content is expressed in mol%.
Illustratively, the in situ pH is based on "Ca 2+ 、Mg 2+ 、HCO 3 - 、SO 4 2- Content of "and" H 2 S partial pressure AND CO 2 The partial pressure is calculated according to a method provided by a patent of a method for calculating the pH value of the produced liquid in the gathering and transportation pipe of the oil field (the patent number is 201910549378. X).
The intermediate database comprises the following 21 data items: material of pipeline, anticorrosion measure in pipeline, conveying medium, running temp, flow speed, water content, sand content and O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 Partial pressure, in situ pH, corrosion rate after labeling. The first 17 data items from the acquired historical pipeline failure data,the latter 4 data items are data obtained by calculation.
Establishing a corrosion rate range prediction model based on the intermediate database;
the method specifically comprises the following steps: calling the intermediate database to obtain the first 20 data items (including pipeline material, anticorrosion measure in pipeline, conveying medium, operation temperature, flow rate, water content, sand content, and O content) in the intermediate database 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 Partial pressure, in-situ pH) as input parameters, and calling a logistic regression model through a scimit-lean data packet by using a python language to perform supervised machine learning by taking the marked corrosion rate as an output result, thereby establishing a corrosion rate range prediction model. One way to represent a prediction model of the corrosion rate range is:
V corr ＝aA+bB+cC+dD+eE+fF+gG+hH+iI+jJ+kK+lL+mM+nN+oO+pP+qQ+rR+sS+tT+u
in the formula: vcorr is the corrosion rate range, A-T are the values of 20 input parameters respectively, a-T are the corresponding coefficients of the values of 20 input parameters, and u is the residual value.
Predicting corrosion in the pipe according to the corrosion rate range prediction model;
inputting pipeline data to be predicted, calling the established corrosion rate range prediction model by utilizing a python language, predicting corrosion in the pipe, and determining the corrosion rate range in the pipe.
Determining a corrosion main control factor and a residual life in the pipe based on a corrosion prediction result in the pipe;
and determining the residual life according to the determined internal corrosion rate range. The following formula is adopted:
T＝(PC-PD/2δn)/(Max v corr )
wherein T is the predicted remaining life in units of y; PC is the wall thickness of the pipeline, and the unit is mm; p is the pipeline operating pressure and the unit is MPa; d is the outer diameter of the pipeline, and the unit is mm; delta isThe minimum yield strength of the pipe is MPa; n is an intensity design coefficient, and is taken as a value according to the grade of the area, wherein the first-level area is 0.72, the second-level area is 0.6, the third-level area is 0.5, and the fourth-level area is 0.4; max V corr The maximum corrosion rate obtained is predicted in mm/y.
Then determining corrosion main control factors in the pipe based on the internal corrosion prediction result;
the coefficients a-t of the 20 corrosion influencing factors are sorted, and the corrosion factor with the largest coefficient is found out (the corresponding coefficient is marked as x) 1 ) The factor is marked as the first master factor;
find the second largest coefficient x 2 If x 2 Value up to x 1 If the value is 70%, the corrosion factor corresponding to the coefficient is marked as a second main control factor, otherwise, the searching of the main control factor is stopped;
find the third largest coefficient x 3 If x 3 Value up to x 2 If the value is 90%, the corrosion factor corresponding to the coefficient is marked as a third main control factor, otherwise, the searching of the main control factor is stopped;
find the fourth largest coefficient x 4 If x 4 Value up to x 3 If the value is 95%, the corrosion factor corresponding to the coefficient is marked as a fourth main control factor, otherwise, the searching of the main control factor is stopped;
find the fifth largest coefficient x 5 If x 5 Value up to x 4 If the value is 95%, the corrosion factor corresponding to the coefficient is marked as a fifth main control factor, otherwise, the searching of the main control factor is stopped;
and searching other main control factors in sequence according to the method for searching the fifth main control factor.
Each step of the calculation method provided by the embodiment of the present invention is described below:
collecting historical failure data of 500 pipelines from the site, wherein the historical failure data of each pipeline comprises pipe diameter, wall thickness, production time, regional grade, pipeline material, anticorrosion measures in the pipeline, conveying media, operating temperature, operating pressure, flow rate, water content, sand content, CO 2 Content, H 2 The S content,O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, pH value, SRB bacterial content, whether scale exists in the tube, failure time, failure position, failure clock direction and latest repair time, 28 data items and 14000 data points in total.
And structuring the collected historical failure data. The method comprises the steps of respectively marking the evaluation values of anticorrosion measures in a pipe as no, inner coating, inner interpenetration, corrosion inhibitor, bactericide, antisludging agent and other values as 1-7, respectively marking a conveying medium as oil-gas-water mixed conveying, wet gas, dry gas, purified oil, water injection, water mixing and sewage and other values as 1-8, respectively marking whether scale exists in the pipe as 0,1, respectively marking the evaluation values of horizontal straight pipe sections, vertical pipes, positions near welding lines, elbows, tee joints and valves and reducing the diameters of the other values as 1-6.
The corrosion rate of 500 pipelines was calculated in turn according to the following formula:
corrosion rate = wall thickness/time of service;
time in service = time out of service-time in production, or
Time in service = time to failure-time to last repair.
According to the corrosion rate range, the corrosion rate is divided into five grades of ' low ', medium-high and high ', wherein the range of ' low ' is 0-0.1 mm/y, the range of ' medium-low ' is 0.1-0.25 mm/y, the range of ' medium ' is 0.25-0.5 mm/y, the range of ' medium-high ' is 0.1-1 mm/y, the range of ' high ' is more than 1mm/y, and then the ' low ', medium-high ' and high ' are respectively marked as 1-5.
The corrosion rate range of 500 pipelines obtained by calculation is 0.05-2.5 mm/y.
Sequentially calculating H of 500 pipelines according to the following formula 2 S partial pressure (P) H2S ) And CO 2 Partial pressure (P) CO2 )：
H of 500 pipelines is obtained through calculation 2 The partial pressure range of S is 0.008MPa to 0.1MPa 2 The partial pressure range is 0.03MPa to 1.3MPa.
The in-situ pH value in 500 pipes is calculated according to the method provided by the patent of the method for calculating the pH value of the produced liquid in the gathering and transportation pipe of the oil field (the patent number is 201910549378. X), and the in-situ pH range is 3.6-6.8.
The 500 pipeline data are arranged to form a pipeline material, a corrosion prevention measure in the pipeline, a conveying medium, an operation temperature, a flow rate, a water content, a sand content and an O content 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 The intermediate database of partial pressure, in-situ pH, and labeled corrosion rate, totaled 10500 data points.
Taking the first 20 data items (10000 data points in total) in the intermediate database as input, taking the marked corrosion rate (500 data points in total) as output, and calling a logistic regression model through a scipit-leann data packet by utilizing python language to perform supervised machine learning to obtain a corrosion rate range prediction model as follows:
V corr ＝0.01A+0.31B+0.12C-0.74D-1.05E-0.34F+0.35G-0.16H-0.07I+0.35J-1.1K-0.56L-0.41M+0.14N+1.09O+0.02P+0.09Q+0.01R+0.74S+0.61T+165.2393
in the formula V corr In order to reach the corrosion rate range, A to T are sequentially the material of the pipeline, the anticorrosion measures in the pipeline, the conveying medium, the operation temperature, the flow velocity, the water content, the sand content and the O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- The concentration,HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 Partial pressure and in-situ pH value.
Inputting pipeline data to be predicted, wherein the pipeline material value is 1, the in-pipe anticorrosion measure value is 1, the conveying medium value is 1, the operating temperature value is 13, the flow rate value is 0.05, the water content value is 0.7, the sand content value is 0 2 The content value is 0.005 and Mg is added 2+ The concentration value is 24,Ca 2+ The concentration value is 321, cl - The concentration value is 10494, SO 4 2- The concentration value is 456, 3 2- the concentration value is 0 3 - The concentration value is 610, the SRB bacterial content value is 10500, whether scale exists in the pipe is 0, the failure position value is 1, H is obtained 2 The S partial pressure is 0 2 Partial pressure value is 0.15, in-situ pH value is 6.1, the partial pressure value and the in-situ pH value are introduced into the prediction model, the corrosion rate range obtained by calculation is 2, namely 0.1-0.25 mm/y, and Max V corr ＝0.25mm/y。
Using the formula T = (PC-PD/2 δ n)/(Max v) corr ) Calculating to obtain the residual service life of the pipeline:
PC＝1.9mm，P＝0.25MPa，D＝119mm，δ＝245MPa，n＝0.5，Max V corr and =0.25mm/y, and T is calculated to be approximately equal to 7 years, namely the residual life is 7 years.
Looking up a python calculation report, sequencing corresponding coefficients of the 20 input parameter values, wherein the first large coefficient is 1.1, and the corresponding corrosion factor is' Cl - Concentration "; the second largest factor is 1.09 (the corresponding corrosion factor is "SRB bacteria content"), 99% of 1.1, more than 70%, so "SRB bacteria content" is the second major factor; the third greatest factor is 1.05 (corresponding to corrosion factor "flow rate"), 96% of 1.09, over 90%, so "flow rate" is the third major factor; the fourth largest factor is 0.74 (corresponding to the corrosion factor "flow rate"), 70% of 1.05, and less than 95%, thus terminating the search for the dominant factor. The main control factor for finally determining the corrosion in the pipe is' Cl - Concentration, SRB bacteria content, flow rate ".
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (15)
1. An in-pipe corrosion factor identification and remaining life prediction method, the method comprising:
acquiring historical failure data of the pipeline;
processing the historical failure data of the pipeline to form an intermediate database;
establishing a corrosion rate range prediction model based on the intermediate database;
predicting corrosion in the pipe according to the corrosion rate range prediction model;
and determining the main control factors and the residual life of the corrosion in the pipe based on the prediction result of the corrosion in the pipe.
2. The method of identifying corrosion factors and predicting remaining life in a pipe according to claim 1,
the historical pipeline failure data comprises: pipe diameter, wall thickness, production time, regional grade, pipeline material, anticorrosion measure in the pipe, conveying medium, operation temperature, operation pressure, flow rate, water content, sand content and CO 2 Content, H 2 S content, O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, pH value, SRB bacterial content, whether scale exists in the pipe or not, failure time, failure position, failure clock direction and last repair time.
3. The method of identifying corrosion factors and predicting remaining life in a pipe according to claim 1,
the pair of the pipesThe data processing of the track historical failure data comprises data structuring, corrosion rate calculation and marking, and H 2 S partial pressure calculation, CO 2 Partial pressure calculation and in situ pH calculation.
4. The method of identifying corrosion factors and predicting remaining life in a pipe according to claim 3,
the corrosion rate is calculated according to the following formula:
corrosion rate = wall thickness/time of service;
time in service = time out of service-time in production, or
Time in service = time to failure-time to last repair.
5. The method of identifying corrosion factors and predicting remaining life in a pipe according to claim 3,
the marking method of the corrosion rate comprises the following steps: the corrosion rate is divided into five grades of 'low, medium and low, medium and high'.
6. The method of identifying corrosion factors and predicting remaining life in a pipe according to claim 3,
said H 2 S partial pressure
And CO 2 Partial pressure
Are obtained by the following formulas:
in the formula:
is H 2 S partial pressure, unit is MPa; p is the pipeline operating pressure and the unit is MPa;
is H 2 S content in ppm;
is CO 2 Partial pressure in MPa;
is CO 2 The content is expressed in mol%.
7. The method of in-pipe corrosion factor identification and remaining life prediction according to claim 1,
the intermediate database comprises the following 21 data items: material of pipeline, anticorrosion measure in pipeline, conveying medium, running temp, flow speed, water content, sand content and O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 Partial pressure, in-situ pH and corrosion rate after labeling;
the first 17 data items are from the acquired historical failure data of the pipeline, and the last 4 data items are data obtained by calculation.
8. The method of in-pipe corrosion factor identification and remaining life prediction according to claim 7,
the specific method for establishing the corrosion rate range prediction model comprises the following steps: calling the intermediate database, taking the first 20 data of the intermediate database as input parameters, taking the marked corrosion rate as an output result, and calling a logistic regression model through a scimit-lean data packet by using a python language to perform supervised machine learning so as to establish a corrosion rate range prediction model;
the corrosion rate range prediction model is specifically expressed as:
V corr ＝aA+bB+cC+dD+eE+fF+gG+hH+iI+jJ+kK+lL+mM+nN+oO+pP+qQ+rR+sS+tT+u
wherein Vcorr is a corrosion rate range, A to T are respectively values of 20 input parameters, a to T are corresponding coefficients of the values of the 20 input parameters, and u is a residual value.
9. The method of in-pipe corrosion factor identification and remaining life prediction according to claim 8,
the main control factors of corrosion in the pipe are determined by the following method:
sorting the corresponding coefficients of the 20 input parameter values, finding out the pipe internal corrosion factor with the maximum coefficient, and marking as a first main control factor;
finding a second big coefficient, marking the corrosion factor corresponding to the second big coefficient as a second main control factor when the value of the second big coefficient reaches 70% of the maximum coefficient value, and otherwise, stopping the searching of the main control factor;
finding a third large coefficient, marking a corrosion factor corresponding to the third large coefficient as a third main control factor when the value of the third large coefficient reaches 90% of the value of the second large coefficient, and otherwise, stopping searching the main control factor;
finding a fourth large coefficient, marking a corrosion factor corresponding to the fourth large coefficient as a fourth main control factor when the value of the fourth large coefficient reaches 95% of the value of the third large coefficient, and otherwise, stopping searching the main control factors;
finding a fifth large coefficient, marking a corrosion factor corresponding to the fifth large coefficient as a fifth main control factor when the value of the fifth large coefficient reaches 95% of the value of the fourth large coefficient, and otherwise, stopping searching the main control factor;
and searching other main control factors in sequence according to the method for searching the fifth main control factor.
10. The method of identifying corrosion factors and predicting remaining life in a pipe according to claim 1,
the remaining life is predicted according to the following formula:
T＝(PC-PD/2δn)/(MaxV corr )
wherein T is the predicted remaining life in units of y; PC is the wall thickness of the pipeline, and the unit is mm; p is the pipeline operating pressure and the unit is MPa; d is the outer diameter of the pipeline, and the unit is mm; delta is the minimum yield strength of the pipe, and the unit is MPa; n is an intensity design coefficient; maxV corr To predict the maximum value of the corrosion rate obtained, the units are mm/y.
11. An in-pipe corrosion factor identification and remaining life prediction system, the system comprising:
the acquisition unit is used for acquiring historical failure data of the pipeline;
the processing unit is used for carrying out data processing on the historical failure data of the pipeline to form an intermediate database;
the establishing unit is used for establishing a corrosion rate range prediction model based on the intermediate database;
the prediction unit is used for predicting corrosion in the pipe according to the corrosion rate range prediction model;
and the determining unit is used for determining the main control factors and the residual life of the corrosion in the pipe based on the prediction result of the corrosion in the pipe.
12. The system for identifying corrosion factors and predicting remaining life according to claim 11,
the intermediate database comprises the following 21 data items: material of pipeline, anticorrosion measure in pipeline, conveying medium, running temp, flow speed, water content, sand content and O 2 Content, mg 2+ Concentration, ca 2+ Concentration, cl - Concentration, SO 4 2- Concentration, CO 3 2- Concentration, HCO 3 - Concentration, SRB bacteria content, presence or absence of scale in the tube, failure location, H 2 S partial pressure, CO 2 Partial pressure, in-situ pH and corrosion rate after labeling;
the first 17 data items are from the acquired historical failure data of the pipeline, and the last 4 data items are data obtained by calculation.
13. The system for identifying corrosion factors and predicting remaining life according to claim 12,
the establishing of the corrosion rate range prediction model in the establishing unit specifically comprises the following steps: calling the intermediate database, taking the first 20 data of the intermediate database as input parameters, taking the marked corrosion rate as an output result, and calling a logistic regression model through a scimit-lean data packet by utilizing a python language to perform supervised machine learning so as to establish a corrosion rate range prediction model;
the corrosion rate range prediction model is specifically expressed as:
V corr ＝aA+bB+cC+dD+eE+fF+gG+hH+iI+jJ+kK+lL+mM+nN+oO+pP+qQ+rR+sS+tT+u
wherein Vcorr is the corrosion rate range, A-T are respectively the values of 20 input parameters, a-T are the corresponding coefficients of the values of 20 input parameters, and u is the residual value.
14. The system of claim 13,
the determining unit determines the corrosion main control factors in the pipe, and comprises the following steps:
sorting the corresponding coefficients of the 20 input parameter values, finding out the pipe internal corrosion factor with the maximum coefficient, and marking the pipe internal corrosion factor as a first main control factor;
finding a second big coefficient, marking the corrosion factor corresponding to the second big coefficient as a second main control factor when the value of the second big coefficient reaches 70% of the maximum coefficient value, and otherwise, stopping the searching of the main control factor;
finding a third large coefficient, when the value of the third large coefficient reaches 90% of the second large coefficient value, marking the corrosion factor corresponding to the third large coefficient as a third main control factor, otherwise, stopping the search of the main control factor;
finding a fourth large coefficient, marking a corrosion factor corresponding to the fourth large coefficient as a fourth main control factor when the value of the fourth large coefficient reaches 95% of the value of the third large coefficient, and otherwise, stopping searching the main control factors;
finding a fifth large coefficient, marking a corrosion factor corresponding to the fifth large coefficient as a fifth main control factor when the value of the fifth large coefficient reaches 95% of the value of the fourth large coefficient, and otherwise, stopping searching the main control factor;
and searching other main control factors in turn according to the method for searching the fifth main control factor.
15. The system for identifying corrosion factors and predicting remaining life according to claim 11,
the remaining life is predicted according to the following formula:
T＝(PC-PD/2δn)/(MaxV corr )
wherein T is the predicted remaining life in units of y; PC is the wall thickness of the pipeline, and the unit is mm; p is the running pressure of the pipeline, and the unit is MPa; d is the outer diameter of the pipeline, and the unit is mm; delta is the minimum yield strength of the pipe, and the unit is MPa; n is an intensity design coefficient; maxV corr To predict the maximum value of the corrosion rate obtained, the units are mm/y.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110922905.4A CN115705512A (en) | 2021-08-12 | 2021-08-12 | Method and system for identifying corrosion factors in pipe and predicting residual life |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110922905.4A CN115705512A (en) | 2021-08-12 | 2021-08-12 | Method and system for identifying corrosion factors in pipe and predicting residual life |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115705512A true CN115705512A (en) | 2023-02-17 |
Family
ID=85180799
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110922905.4A Pending CN115705512A (en) | 2021-08-12 | 2021-08-12 | Method and system for identifying corrosion factors in pipe and predicting residual life |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115705512A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116659591A (en) * | 2023-07-31 | 2023-08-29 | 济宁长胜新材料股份有限公司 | Operation early warning system suitable for chemical distillation plant |
-
2021
- 2021-08-12 CN CN202110922905.4A patent/CN115705512A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116659591A (en) * | 2023-07-31 | 2023-08-29 | 济宁长胜新材料股份有限公司 | Operation early warning system suitable for chemical distillation plant |
| CN116659591B (en) * | 2023-07-31 | 2023-10-20 | 济宁长胜新材料股份有限公司 | Operation early warning system suitable for chemical distillation plant |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111027730B (en) | Efficient positioning method for water supply network leakage based on valve operation and online water metering | |
| CN110020399B (en) | Method for determining corrosion evaluation position in pipeline | |
| CN111475913A (en) | Operation optimization method and system for steam power system | |
| Svedeman et al. | Criteria for sizing multiphase flowlines for erosive/corrosive service | |
| CN109815595B (en) | Underground pipe column of oil and gas field and hydrogen sulfide corrosion big data analysis method of wellhead gas transmission pipeline | |
| CN112668206B (en) | Multi-factor-considered acid gas field corrosion prediction model and parameter determination method | |
| CN105069486A (en) | Extreme learning machine-based petrochemical equipment corrosion prediction method | |
| CN115705512A (en) | Method and system for identifying corrosion factors in pipe and predicting residual life | |
| CN113343498A (en) | Containing H2S/CO2Method for calculating corrosion rate of environmental pipeline steel | |
| CN112069688B (en) | Corrosion simulation analysis method in natural gas long-distance pipeline | |
| Zhang et al. | A semi-empirical model for underground gas storage injection-production string time series remaining useful life analysis in process safety operation | |
| CN115359015A (en) | Oil well pipe wall thickness logging data interpretation and corrosion defect intelligent determination method | |
| Atambo et al. | Prediction model development for sanitary sewer pipes’ condition assessment using logistic regression and neural networks | |
| GB2605042A (en) | Maintenance methods and systems for fluid containment assets | |
| CN112347414A (en) | Single-defect bent pipe limit internal pressure calculation method | |
| Mulenga et al. | Investigating the root causes of major failures of critical components–With a case study of asbestos cement pipes | |
| CN113803647B (en) | Pipeline leakage detection method based on fusion of knowledge features and hybrid model | |
| CN116164241A (en) | Intelligent detection method for leakage faults of gas extraction pipe network | |
| Li et al. | [Retracted] Prediction of Chemical Corrosion Rate and Remaining Life of Buried Oil and Gas Pipelines in Changqing Gas Field | |
| CN116227378A (en) | Method for predicting scaling thickness of pipeline in oil and gas treatment industry | |
| CN108240960B (en) | Method for predicting corrosion degree in-service long oil and gas pipeline | |
| Vishwakarma | Development Of A Performance Analysis Framework For Water Pipeline Infrastructure Using Systems Understanding | |
| Guan et al. | Application of probabilistic model in pipeline direct assessment | |
| CN221003376U (en) | Pipeline oil use information detection equipment | |
| CN113361066B (en) | Accumulated liquid and accumulated liquid degree identification method |
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 |