CN107525733B

CN107525733B - Wellhead downhole corrosion rate correlation model algorithm and downhole corrosion rate online monitoring method using same

Info

Publication number: CN107525733B
Application number: CN201710675014.7A
Authority: CN
Inventors: 刘晶姝; 龙媛媛; 谭晓林; 杨为刚; 刘瑾; 刘丽; 柳言国
Original assignee: China Petroleum and Chemical Corp; Technology Inspection Center of Sinopec Shengli Oilfield Co
Current assignee: China Petroleum and Chemical Corp; Technology Inspection Center of Sinopec Shengli Oilfield Co
Priority date: 2017-08-09
Filing date: 2017-08-09
Publication date: 2020-07-03
Anticipated expiration: 2037-08-09
Also published as: CN107525733A

Abstract

The correlation between the corrosion rates of different depths in the well and the corrosion rate of the well mouth is determined through experimental research, the correlation model algorithm of the corrosion rates of the well mouth and the well mouth is established, then the corrosion rates of the well mouth and the well mouth are used for calculating the corrosion rates of different depths in the well, and a new compensation method is researched to solve the problem that the curve of the measured value of the thick-wall columnar inductance probe test piece and the original inductance probe mathematical model deviates from the linear relation, so that the online monitoring of the corrosion rate of the well mouth is realized, and the problem that the corrosion condition of underground equipment cannot be effectively monitored is solved. Has remarkable economic and social benefits and good popularization and application prospects.

Description

Wellhead downhole corrosion rate correlation model algorithm and downhole corrosion rate online monitoring method using same

Technical Field

The invention belongs to the technical field of oilfield monitoring, and particularly relates to a wellhead downhole corrosion rate correlation model algorithm and a downhole corrosion rate online monitoring method using the algorithm.

Background

CO₂Originally in 1962 as an additive for acidizing and fracturing treatments, it was used by people, and since then as a well treatment medium, its use has rapidly developed. With the deep layer containing CO₂Development of oil and gas fields and reinjection of CO in tertiary oil recovery₂Widespread use of Enhanced Oil Recovery (EOR), CO₂The corrosion problem has always plagued the development of the oil and gas industry. CO2₂Oil well produced liquid is rich in CO₂And the underground equipment of the production well can be seriously corroded when meeting water. According to the primary screening, the method is suitable for CO in the victory oil field₂The reserve of oil displacement low-permeability oil fields reaches more than 2 hundred million tons, and if CO is adopted completely₂The oil displacement development can improve the oil recovery ratio of the oil field by 10 to 15 percent, and the newly increased recoverable reserves are estimated to be 3300 to 4700 ten thousand tons. Therefore, it is necessary to research into CO through a targeted study₂The wide application of the flooding technology solves the bottleneck problem of corrosion of underground equipment of the production well, can guide the corrosion protection of the produced liquid along the process of collection and transportation, and powerfully ensures CO₂The running economy and reliability of the corrosion protection scheme of the underground driving device and the ground gathering and transportation system have obvious economic and social benefits.

The problem of corrosion control of any type is solved, firstly, the corrosion characteristics are known, and secondly, the monitoring of corrosion, the management of corrosion data and the basic flow of corrosion control are carried out. CO2₂Corrosion monitoring technique for driving well, which can be subdivided into corrosion at well headThe corrosion and the corrosion monitoring problem in the pit are solved, the traditional monitoring means mainly adopts a hanging piece (ring) technology, the technology has the characteristics of low economic investment, but the technology has the defects of long test period, large error caused by artificial influence on the test result and the like, in addition, the data of the corrosion process can not be reflected, and the technical problems can be effectively solved for the current corrosion on-line monitoring technology. The inductive corrosion monitoring technology is an on-line monitoring technology developed in recent years, has the characteristics of high measurement sensitivity and universal applicability to media due to the measurement principle of electromagnetic induction, is widely applied to the oil refining and chemical industry, and has a series of problems in corrosion monitoring of a shaft.

Disclosure of Invention

The invention aims to provide a wellhead downhole corrosion rate correlation model algorithm and a downhole corrosion rate online monitoring method using the same, the algorithm is used for calculating the downhole corrosion rate through a correlation model of the downhole corrosion rate and the wellhead corrosion rate, a new compensation method is mainly researched to solve the problem that an inductance probe measured value and an original inductance probe mathematical model curve deviate from a linear relation, the correlation of the downhole corrosion condition and the wellhead corrosion result is determined through experimental research, and the downhole corrosion rate is calculated by using the wellhead corrosion rate monitoring result, so that the corrosion condition and the corrosion development trend of downhole equipment are known.

The corrosion monitoring of oil gas downhole equipment does not have a mature and stable technology for use except for a hanging ring at present, because the corrosion environment in the pit is harsher, compared with a wellhead environment, the temperature, the pressure and the corrosivity of the corrosion monitoring all show a trend of gradually rising, the corrosion on-line monitoring of the downhole equipment is realized due to the temperature resistance of electronic components and technical condition limitations such as narrow annular space of a downhole oil sleeve, the realization is difficult at present, therefore, the idea of monitoring the corrosion rate of the wellhead by using an inductance probe is provided, the downhole corrosion rate correlation model of the wellhead is used, and the downhole corrosion rate is calculated according to the corrosion rate of the wellhead.

The technical scheme is as follows:

the wellhead downhole corrosion rate correlation model algorithm and the downhole corrosion rate online monitoring method using the algorithm are provided, and the problems are solved.

Through the experiment, the correlation model algorithm of different degree of depth corrosion rate and well head corrosion rate in the pit is researched, and then the problem that the corrosion condition of equipment in the pit can not be effectively monitored is solved by using the corrosion detection result of the well head.

The model (i.e., algorithm) of correlation of wellhead to downhole corrosion rate is based on the c.de Waard (DM model); with high CO content₂The corrosion environment is a research object, a six-factor five-level orthogonal experiment is adopted, the model is corrected by utilizing orthogonal experiment data, and well mouth and underground corrosion rate prediction models are respectively established. The temperature and CO of the well head and the well₂Three key corrosion influence factors of partial pressure and total pressure are normalized to temperature parameters, the functional relations between wellhead corrosion rates and temperatures and between the underground corrosion rates and the temperatures are respectively determined, then an empirical formula of temperature gradients existing between the underground temperatures and the wellhead temperatures is utilized to carry out substitution calculation, and finally a correlation model of the wellhead corrosion rates and the underground corrosion rates is obtained.

The corrosion rate of the wellhead is monitored on line in real time through the inductive probe, the corrosion rates of different underground depths are calculated by utilizing the correlation model of the underground corrosion rate of the wellhead, and then the corrosion condition of underground equipment is known.

The materials of the inductive probe and the hanging piece are consistent with those of the underground pipe column to be measured.

Due to CO₂The underground medium driving environment and the flow rate and flow state are complex, the underground medium driving system has the characteristics of strong corrosivity and high-pressure multiphase mixed flow, and the comprehensive corrosion and the local corrosion coexist in the system. Oil field CO₂The following bottlenecks exist for corrosion drive protection: (1) high temperature and high pressure cause severe CO₂Corrosion, and lack of a rapid and reliable corrosion on-line monitoring means. (2) The water content of the downhole fluid affecting CO₂Important factors for displacement efficiency and corrosion rate, the lack of corresponding CO₂Partial pressure, water content and other parameters. (3) And integrating various corrosion monitoring technologies and various monitoring information to carry out comprehensive corrosion state evaluation.

The new compensation method is researched to solve the problem that the curve of the measured value of the thick-wall cylindrical inductance probe test piece and the original inductance probe mathematical model deviates from the linear relation.

The advantages are that:

according to statistics, a haman oil production plant has 2731 ports of oil production wells, wherein 2085 ports of the oil production wells are opened, the problems of corrosion, breakage, eccentric wear, no oil production of the oil production wells and the like of pipe columns can occur in the production operation process of the oil production wells, production stop operation maintenance needs to be carried out, and the maintenance operation caused by underground corrosion accounts for more than 20%. Assuming that the total operation frequency per year is 0.9 (the total operation frequency of the Minam oil production plant in 2014 is 0.924), the cost of materials, labor and the like generated by each shut-in maintenance operation is 10 ten thousand yuan, and the operation cost per year is as high as 18765 ten thousand yuan. The conservative estimation can reduce 20% of maintenance and replacement of underground equipment, and the capital of the Minam oil production plant is saved by 3753 ten thousand yuan each year, so that the method has remarkable economic and social benefits. The direct economic benefit (savings) produced in 2016 is: 3753 ten thousand yuan/2085 one port x 2 is 3.6 ten thousand yuan.

The low-permeability reservoir reserves of 7.67 million tons account for 15.4 percent of the total resource amount in the victory oil zone at present, and the low-permeability newly-increased exploratory reserves are about 2000 million tons in the newly-increased exploratory reserves every year since the nine five days. According to the primary screening, in the victory oil field, the reserve volume of the low-permeability oil field suitable for carbon dioxide oil displacement reaches more than 2 hundred million tons, and if CO is adopted completely₂The oil displacement development can improve the oil recovery ratio of the oil field by 10 to 15 percent, and the newly increased recoverable reserves are estimated to be 3300 to 4700 ten thousand tons. CO2₂In the oil-displacement process, CO₂After being injected into the underground, about 50 to 60 percent of the gas is permanently sealed in the underground, and the rest 40 to 50 percent of the gas returns to the ground along with the oilfield associated gas. The carbon dioxide in the crude oil can cause serious corrosion of underground equipment of a production well, and in a high-temperature and high-pressure environment, the carbon dioxide is generated by CO₂The presence of high corrosion (7-20 mm/a) and tensile stress to carbon steel can lead to corrosive perforation or early failure (6 months failure fastest) of the downhole string.

Currently, the victory oil field only implements CO in 89 blocks of rational high with low (6%) and stable overall water content₂And (5) performing a driving pilot test. The Shengli oil field can provide 150 ten thousand square carbon dioxide gas sources each year to realize CO₂The technology of flooding is applied to pure beam cattle villages, present rivers,And (5) popularization and application of other hypotonic blocks with higher water content, such as eastern soliton and the like.

The research result of the project can monitor the corrosion of equipment and the corrosion control effect on line, and quickly and accurately judge the underground CO₂The corrosion condition and the hidden danger, the adjustment of the adding amount of the corrosion inhibitor, the maintenance and the replacement of facilities and the like are guided, and the loss caused by corrosion is reduced to the maximum extent. To CO₂The safety assessment and integrity management of the operation of the underground equipment and the ground gathering and transportation system are of great significance to avoid safety accidents. Will be CO₂The wide application of the flooding technology solves the bottleneck problem of corrosion of underground equipment of the production well, can guide the corrosion protection of the produced liquid along the process of collection and transportation, and powerfully ensures CO₂The corrosion protection scheme of the underground driving device and the ground gathering and transportation system has the advantages of high economical efficiency and reliability, remarkable economic and social benefits and good popularization and application prospects.

Drawings

FIG. 1 shows temperature and CO₂Influence of partial pressure on corrosion rate.

Fig. 2 is a graphical illustration of the inductance probe output data curve.

FIG. 3 shows the micro-topography of the surface of the coupon corresponding to the most severe corrosion condition of A3 steel.

Fig. 4 is a graph of the consistency of a3 steel wellhead coupon and inductance probe data.

Fig. 5 is a plot of the consistency of N80 steel wellhead hanger plate and induction probe data.

Fig. 6 is a plot of the consistency of N80 steel downhole hanger and inductive probe data.

Fig. 7 is a graph of the consistency of J55 steel wellhead hanger plate and inductance probe data.

Fig. 8 is a plot of the consistency of J55 steel downhole hanger and inductance probe data.

Fig. 9 shows the fitting results of the inductance probe data and the coupon data.

FIG. 10 is a graph of the corrosion rate of A3 material with different parameters.

FIG. 11 is a graph of the corrosion rate of N80 material with different parameters.

FIG. 12 is a graph of the corrosion rate of J55 material with different parameters.

FIG. 13 is a graph of erosion rate of N80 material downhole as a function of different parameters.

FIG. 14 is a graph of erosion rate of downhole J55 material with different parameters.

FIG. 15 shows technical scheme 1.

FIG. 16 shows the difference in CO at 20 ℃₂Bicarbonate-containing (saturated or supersaturated) water under pressure produces a pH of water.

FIG. 17 shows the difference in CO at 60 ℃₂Bicarbonate-containing (saturated or supersaturated) water under pressure produces a pH of water.

FIG. 18 shows different CO at 100 ℃₂Bicarbonate-containing (saturated or supersaturated) water under pressure produces a pH of water.

Fig. 19 shows measured data of three types of hanging pieces.

Fig. 20 shows the relationship between the predicted value and the measured value of the a3 model.

FIG. 21 shows the relationship between the predicted value and the measured value (45 ℃ C.) of the A3 model.

FIG. 22 shows the relationship (> 45 ℃) between the predicted value and the measured value of the A3 model.

FIG. 23 shows the relationship between the predicted model value and the actual measured value of N80 wellhead (45 ℃ or lower).

FIG. 24 is a plot (> 45 ℃) of the model predicted values versus the N80 well-head measured values.

FIG. 25 shows the relationship between the predicted model value and the measured value of J55 wellhead (45 ℃ or lower).

FIG. 26 is a plot (> 45 ℃) of model predicted values versus measured values for the J55 wellhead.

FIG. 27 is a graph of the relationship between model predicted values and N80 measured downhole values (< 45 ℃.

FIG. 28 is a plot (> 45 ℃) of model predicted values versus N80 measured downhole.

FIG. 29 is a graph of the relationship between model predicted values and J55 measured downhole values (< 45 ℃).

FIG. 30 is a plot (> 45 ℃) of model predicted values versus measured values in J55 downhole.

FIG. 31 is a plot of downhole erosion rate versus downhole temperature.

FIG. 32 is a plot of wellhead temperature versus wellhead corrosion rate.

Fig. 33 is a # 1 well indoor validation test.

FIG. 34 is 1# well A3 steel initial stabilization segment in-situ inductance probe data.

Fig. 35 shows the induction probe monitoring data of the wellhead a3 of the test well # 1.

Fig. 36 is a data graph of monitoring data of a No. 1 test well wellhead N80 inductive probe.

FIG. 37 is a graph showing the relationship between the content of iron ions and the etching rate.

Fig. 38 is a schematic structural diagram of an inductance probe that can be used in the present embodiment.

FIG. 39 is a schematic view of a 4-11x101 field shackle installation.

Fig. 40 shows the data of the 2# test well a3 material inductive probe.

FIG. 41 is a graph showing the relationship between the iron ion content and the corrosion rate of the inductive probe.

Fig. 42 shows the data of the N80 material inductive probe monitoring of test well No. 2.

Fig. 43 is a comparison before and after compensation of the inductance probe mathematical model used.

FIG. 44 is the maximum corrosion rate for well # 1.

FIG. 45 is the maximum corrosion rate for # 2 well.

FIG. 46 is the maximum erosion rate for well # 3.

Detailed Description

Example 1.

A wellhead downhole corrosion rate correlation model algorithm and a downhole corrosion rate online monitoring method using the algorithm are provided.

(di) CO₂Research on corrosion driving rules:

1. steel material for corrosion test:

the materials of the inductance probe and the hanging piece in the corrosion experiment are all general steel materials for oil field oil casings, including J55, N80 and A3 steel materials. The chemical components of the steel materials are analyzed as follows:

the chemical composition of the A3 steel in the experiment is shown in the table 2-1:

TABLE 2-1 chemical composition of A3 Steel for experiments (%)

The chemical composition of the N80 steel used in the experiment is shown in Table 2-2:

TABLE 2-2 chemical composition of experimental N80 Steel (%)

The chemical composition of the J55 steel used in the experiment is shown in tables 2-3:

tables 2-3 chemical composition of experimental J55 Steel (%)

2. And (4) experimental reagents.

Calcium chloride, sodium carbonate, sodium bicarbonate, potassium chloride, magnesium chloride, absolute ethyl alcohol, concentrated hydrochloric acid, hexamethylenetetramine, distilled water and diesel oil.

The simulated produced fluid can be prepared by adopting a plurality of the reagents.

The solution is prepared by self according to the ion mineralization range of the on-site produced liquid of 7500-75000mg/L, which is the known technology in the field, and the content of various ions is as the following table, and the unit is (mol/L).

The self-prepared simulated production fluid components are shown in tables 2-4:

tables 2-4 ion content tables (mol/L) for different degrees of mineralization

3. Experimental protocol.

In view of the six corrosion affecting factors contained herein: temperature, flow rate, degree of mineralization, water content, CO₂Partial pressure and total pressure, and if a single-factor experiment is carried out on each factor, the result has no great practical utility and reference value. So the scheme of orthogonal test method is selected in this textDesign, because the orthogonal test has the characteristics of uniformity, comparability and uniform dispersion, the regression treatment can be carried out on the result, and the corrosion model result is more convincing.

(1) Setting of temperature conditions.

According to the current situation, the temperature range of the wellhead is 20-80 ℃, the temperature rise rule of the underground temperature is that the temperature rises 3.5-4 ℃ every 100 meters of well depth, most of the temperature approaches 3.5 ℃ after relevant documents are consulted, therefore, the temperature is calculated according to 3.5 ℃, and if the well depth is 2200 meters as the maximum depth, the temperature at the depth is 97 ℃. FIG. 1 shows the temperature and CO described in the literature₂The influence of partial pressure on the corrosion speed is researched, and the temperature turning point is 100 degrees. FIG. 1 temperature and CO₂Influence of partial pressure on corrosion rate.

According to the above conditions, the influence of the temperature distribution on the corrosion speed at the well head and the well is synthesized, and the parameter setting range of the temperature is shown in tables 2 to 5.

TABLE 2-5 set ranges for well head and downhole experimental temperatures

Node point	1	2	3	4	5
						Temperature/. degree.C	35	45	60	85	105
About well depth/m	0	600	1200	1800	2400

(2) Total pressure and CO of experiment₂And setting the partial pressure.

According to the different production wells, CO thereof₂The range of partial pressures is different and therefore in order to have a generally applicable effect on the experimental results, the reference sets the experimental pressure range with respect to the measured values of the well flow pressure, as shown in tables 2-6, according to the depth range selected by the temperature.

TABLE 2-6 Total pressure setting range for downhole simulation experiment

Node point	1	2	3	4	5
						Total pressure/MPa	0.5	2	6	11	14
About well depth/m	0	600	1200	1800	2400

From CO₂Partial pressure, typically less than 0.483 × 10⁵When Pa is higher, uniform corrosion is easy to occur, and when the partial pressure is 0.483 × 10⁵～2.07×10⁵Different degrees of pitting corrosion may occur between Pa, when the partial pressure is greater than 2.07 × 10⁵At Pa, severe local corrosion occurs.

CO₂The partial pressure measurement has not been explored in detail so far; and due to CO₂The critical temperature and pressure are respectively 304k and 7.52Mpa, and are converted into supercritical fluid, CO, at 1000 m (temperature is about 333 k) in the well₂Partial pressure to CO₂Solubility in water also does not play a significant scale, so only CO will be used₂The partial pressure of (A) is taken to be the critical pressure, in which the pressure is sufficient to investigate the CO₂Influence of partial pressure on corrosion. Secondly, the selected nodes also take into account the partial pressure values stated in the literature for the different corrosion patterns to occur, and the node selection is shown in tables 1-7.

Tables 2-7 downhole simulation experiment CO₂Partial pressure parameter setting

Node point	1	2	3	4	5
						Partial pressure/MPa	0.04	0.2	4.5	6	7.5

According to the actual working condition of oil well, the pressure of wellhead is about 1Mpa basically, so that the total pressure of wellhead simulation experiment is 1Mpa and CO₂The parameters of the partial pressure are selected according to different mole ratios and the influence node values of the partial pressure on the corrosion morphology, as shown in tables 2 to 8.

TABLE 2-8 wellhead simulation experiment CO₂Partial pressure parameter setting

Node point	1	2	3	4	5
						Partial pressure/MPa	0.04	0.1	0.2	0.4	0.6

(3) And selecting the flow rate.

When the flow rate is low, the corrosion rate will accelerate with increasing flow rate; at higher flow rates, the erosion rate is completely controlled by charge transfer, at which point the change in flow rate is less important and the effect of temperature becomes a major contributor. Therefore, the flow rate of the experimental scheme is in a low flow rate section.

As previously understood, the range of motion speed of the sucker rod is about 1-3 times per minute, and the stroke is about 5-8 meters. The moving speed of the sucker rod is calculated according to the interval, the fastest moving speed of the sucker rod is set to be 3 times/minute, the stroke is 8 meters, and the moving speed is calculated to be 6 times 8/60 times 0.8 m/s. The slowest speed 2x 5/60 is 0.17m/s, and the flow rates selected for this practical case are shown in tables 2-9 below:

TABLE 2-9 wellhead and downhole flow Rate parameter ranges

Node point	1	2	3	4	5
						Flow rate of flow	0.1	0.3	0.6	0.9	1.1

(4) And setting a water content parameter.

In a production fluid system, if a water-in-oil (water/oil) emulsion is formed, the water wettability of the steel will be prevented or greatly reduced, thereby reducing the corrosion rate; conversely, if an oil-in-water (oil/water) emulsion is formed, significant water wetting occurs. In many oil pipes, the transition from a water/oil emulsion to an oil/water emulsion occurs at w (H)₂O) is 30 to 40 percent, and according to the experience, when w (H)₂O) is 30%, the corrosion rate is often significantly reduced with exceptions. The selected water content ranges are summarized in the following tables 2-10:

TABLE 2-10 wellhead and downhole moisture content parameter ranges

Node point	1	2	3	4	5
						Water content	0	15	35	60	90

The results obtained from such a selection can both prove the current conclusions and achieve the goals that are not achieved in the field at present.

(5) And (4) selecting the degree of mineralization.

According to the test data of three oil production plants provided by the winning oil field, the mineralization degree range is the minimum 9045 and the maximum 69369, and the range is very wide. In order to improve the wide applicability of the experimental results, the mineralization range covers the above range, and the parameters are selected as shown in tables 2-11 below.

TABLE 2-11 ranges of degree of mineralization for well head and downhole experiments

Node point	1	2	3	4	5
						Degree of mineralization	7500	15000	30000	50000	75000

Three orthogonal schemes need to be established for three steels A3, J55 and N80:

a3 scheme for orthogonal test of low pressure at wellhead, five factors and five levels are selected, and are shown in tables 2-12, and the total pressure is 1 MPa.

Tables 2-12A 3 factor and level tables for orthogonal testing

N80, J55 Low pressure orthogonal test protocol, five factors, five levels selected, see tables 2-13, total pressure 1 MPa.

Tables 2-13N 80, factor and level table corresponding to J55 low voltage orthogonal test

N80, J55 downhole high pressure orthogonal test protocol, six factors, five levels, are selected, see tables 2-14 for details, since N80, J55 are applied to downhole oil casings. The pressure increases with increasing well depth.

Tables 2-14N 80, factor and level table corresponding to J55 high-pressure orthogonal test

The experimental steps are as follows:

firstly, sealing the reaction kettle and checking the air tightness of the device.

Secondly, preparing a medicine solution by selecting CaCl₂，MgCl₂Anhydrous Na₂CO₃，NaHCO₃KCl, five medicines are prepared for simulating the preparation of produced liquid, and are prepared for later use one day before the test. This is known in the art and will not be repeated.

And thirdly, pre-treating the corrosion hanging piece, taking the hanging piece out of the package, wiping the hanging piece with absolute ethyl alcohol, drying the hanging piece with an air duct, and putting the hanging piece into a drying container for later use. And preparing the hanging piece device and accessories, and installing the hanging pieces for later use.

And fourthly, filling the solution into the reaction kettle according to the experiment set value (1L), and starting the heating device to raise the temperature to the preset temperature in the table.

Fifthly, connecting the inductance probe and the hanging piece with the reaction kettle, firstly filling nitrogen to a set value, and then filling CO₂To steady pressure, nitrogen pressure and CO₂The pressure reached the total pressure in the table above.

And sixthly, connecting the inductive probe collector, starting the data software and starting communication.

And seventhly, observing the data reading condition and the pressure change condition, and performing fine adjustment and fault detection.

And step eight, intercepting the data curve of the inductance probe after the experiment is maintained for 2-4 days, and taking the hanging piece out of the reaction kettle.

And ninthly, treating the inductance probe and the hanging piece, taking out corrosion products on the surfaces of the inductance probe and the hanging piece by using a cleaning agent, drying and then weighing, recording data, and waiting for subsequent calculation.

And step ten, pumping the solution in the reaction kettle by a water pump, and recovering and treating.

The eleventh step, analyze the data and calculate a single set of corrosion rates.

And step twelve, compiling a single group of experimental reports, and analyzing photos before and after reaction, inductive probe corrosion data and coupon corrosion data.

And (4) a corrosion rate calculation method.

(1) And (4) hanging piece weight loss method.

The weight loss method is the most common corrosion rate calculation method in laboratory corrosion research, and is the most intuitive and effective uniform corrosion measurement method. The hanging method has simple steps, and the hanging piece is processed before the experiment, dried for standby, weighed by a one-ten-thousandth balance and recorded. After the test, the hanging piece is treated by a cleaning agent, and the corrosion product on the surface is removed, dried and weighed.

The principle can be expressed by the following equation:

m-weight before coupon corrosion, g;

m_t-weight after coupon corrosion, g;

S₁surface area of the hanging piece, m²；

t is the time of the corrosion experiment, h;

ρ -density of the experimental metal, g/cm.

(2) Inductance probe monitoring method.

The inductive probe is a measure for indirectly measuring the corrosion rate in recent years, and has the advantages of being capable of detecting the corrosivity of a medium in real time, converting an electric signal generated by corrosion into a digital signal through data communication software, and reflecting the rate change trend of the whole corrosion process. And the corrosion rate value of any time period can be obtained through the inductive probe workstation, and is compared with the corrosion rate of the weight loss of the hanging piece, and the relation between the corrosion rate value and the corrosion rate is found out. As shown in fig. 2, the red curve represents the corrosion trend, the ordinate represents the remaining thickness of the inductance probe test piece, the abscissa represents the time, and the corrosion rate of the time period can be obtained by intercepting any one section of corrosion trend line with the purple and blue vertical lines. Fig. 2 graphical illustration of inductive probe output data.

And (5) experimental results. The a3, N80, and J55 coupons were observed to corrode to varying degrees both at the wellhead and downhole. There are also instances where the inductive probe is corroded. FIG. 3 shows the micro-topography of the surface of the coupon corresponding to the most severe corrosion condition of A3 steel.

The surface appearance observation of the hanger sheet and the inductance probe after the well mouth and the underground experiment belongs to even corrosion and has local corrosion appearance such as slight pitting corrosion.

And monitoring the corrosion rate.

The results of the A3 corrosion orthogonalization experiments under wellhead conditions (total pressure 1MPa) are shown in tables 2-15 below:

tables 2-15A 3 orthogonal experimental group number and experimental results

As can be seen from the above table, the data of the inductance probe and the hanging piece are different, and the relative average error of the inductance probe and the hanging piece is 17.9%.

The results of the N80 corrosion orthogonal experiments under wellhead conditions (total pressure of 1MPa) are shown in tables 2-16 below:

TABLE 2-16 orthogonal experimental group numbers of wellhead N80 and experimental results

As can be seen from the above table, the data of the inductance probe and the hanging piece are different, and the relative average error of the inductance probe and the hanging piece is 18.1%.

The results of orthogonal experiments on N80 corrosion under downhole high pressure conditions (total pressure 1MPa) are shown in tables 2-17 below:

TABLE 2-17 orthogonal experimental group numbers of N80 downhole and experimental results

As can be seen from the above table, the data of the inductance probe and the hanging piece are different, and the relative average error of the inductance probe and the hanging piece is 18.4%.

The results of the J55 corrosion orthogonal experiments under wellhead conditions (total pressure of 1MPa) are shown in tables 2-18 below:

TABLE 2-18 wellhead J55 orthogonal experimental group numbers and experimental results

As can be seen from the above table, the data of the inductance probe and the hanging piece are different, and the relative average error of the inductance probe and the hanging piece is 18.7%.

The results of the J55 corrosion orthogonal experiments under downhole high pressure conditions are shown in tables 2-19 below:

TABLE 2-19 orthogonal experimental group numbers of J55 downhole and experimental results

As can be seen from the above table, the data of the inductance probe and the hanging piece are different, and the relative average error of the inductance probe and the hanging piece is 18.0%.

And (6) analyzing the data.

And (3) researching the correlation of the corrosion hanging piece and the monitoring data of the inductance probe.

It can be seen from the experimental data of all 125 sets shown in tables 2-15 to tables 2-19 that there is a difference between the inductive probe and coupon data, and therefore qualitative and quantitative correlation studies were performed on the data of the corrosion coupon and inductive probe. Fig. 4 is a graph of the consistency of a3 steel wellhead coupon and inductance probe data. Fig. 5 is a plot of the consistency of N80 steel wellhead hanger plate and induction probe data. Fig. 6 is a plot of the consistency of N80 steel downhole hanger and inductive probe data. Fig. 7 is a graph of the consistency of J55 steel wellhead hanger plate and inductance probe data. Fig. 8 is a plot of the consistency of J55 steel downhole hanger and inductance probe data.

Through the above analysis of three materials, namely A3, N80 and J55, the potential consistency between the inductance probe and the hanging piece can be seen. Therefore, further study of the quantitative functional relationship between the two is required.

The SPSS software has wide application in data analysis, so the experimental coupon and inductive probe data are brought into mathematical software, 100 sets of experimental data are imported except for wellhead J55 data, and the wellhead data of J55 is used as verification data, and the process is as follows, ① imports data, and fitting ② various curve fits.

The data are fitted and preferably fitted in the form of functions of linear, quadratic, complex, cubic, exponential, power, log, etc., with the results shown in fig. 9: and fitting results of inductance probe data and coupon data (orange curve is a polynomial fitting curve, black is a linear fitting curve, and basically coincides with the orange curve).

After fitting 100 sets of inductive probe and coupon data, two fitting results are preferably selected, as shown in fig. 9. Two fitting results can be seen, namely, the two fitting results have two fitting relations of phenomena and polynomials.

Compared with the result calculated by utilizing the fitting relation, the actually measured data monitored by the J55 wellhead is compared, on one hand, whether the fitting relation is accurate or not is verified, and on the other hand, the fitting relation is compared, so that the actually measured value is closer to the actually measured value.

TABLE 2-20 comparison of fitting results with measured values

As can be seen from the above comparison results, the linear fitting result is closer to the measured value of the inductance probe, the average relative error between the fitting value and the measured value of the inductance probe is 0.7%, and the average relative error between the polynomial fitting result and the measured value is 4.4%, thus indicating that the linear relationship between the inductance probe measured value and the hanging piece measured value is as follows:

Y＝1.1877X-0.0013

in the above formula, Y is the corrosion rate monitored by the inductance probe, X is the corrosion rate monitored by the coupon, and the above functional relationship is the correction model of the inductance probe.

The corrosion data of the wellhead inductance probe is collected, and the coupon corrosion data is collected simultaneously during primary application and is used for correcting the monitoring corrosion rate value of the inductance probe.

The measurement method of the coupon and the inductance probe has certain restriction factors in data accuracy, but coupon data is an internationally recognized corrosion rate measurement method, so the actual corrosion rate needs to be based on the coupon measurement data. In the functional relation between the hanging piece data and the inductance probe data found through the analysis, it can be seen that the hanging piece data and the inductance probe data have a very good corresponding relation. Therefore, the correction model of the measurement data of the inductance probe is completed.

The purpose of researching the wellhead and underground corrosion influence factors is as follows:

and (5) investigating the wellhead and underground environment, wherein the influence of each influence factor on the corrosion rate.

(1) Results of orthogonal mean analysis of wellhead A3/N80/J55 steel: FIG. 10 is a graph showing the erosion rate of A3 material according to different parameters.

From FIG. 10, it can be judged that CO is present at a temperature of 105 ℃ C₂The corrosion is the most serious under the conditions that the partial pressure is 0.1Mpa, the flow rate is 0.1m/s, the water content is 90 percent, and the mineralization degree is 50000 mg/L.

The combination is not in the orthogonal test, in order to verify the correctness of the analysis result of the orthogonal test, the combination is specially corroded, the corrosion rate of the obtained hanging piece is 6.542mm/a, the hanging piece is at the same corrosion level as the 23 rd group in the orthogonal test, the difference of the corrosion rate and the hanging piece is different from the flow rate, but the reason that the results are similar is that under the condition of high temperature, the bonding force of corrosion products is good, and the influence of the flow rate is small.

The influence magnitude sequence of each factor from the result of the mean value is as follows: water content (poor K1 ═ 2.40)>Temperature (k3 ═ 1.2)>Degree of mineralization (K4 ═ 1.1)>Flow rate (k5 ═ 1.0)>CO₂Partial pressure (k2 ═ 0.9), the greatest corrosion affecting factor was observed to be water content, and the least was CO₂Partial pressure.

FIG. 11 is a graph showing the variation of the etching rate of N80 material with different parameters, and FIG. 12 is a graph showing the variation of the etching rate of J55 material with different parameters.

From FIGS. 11 and 12, it can be seen that both materials are at a temperature of 60 ℃ and CO₂The partial pressure is 0.6MPa, the degree of mineralization is 15000mg/L, and the oil-water corrosion is the most serious under the condition of 90 percent. Wherein the flow rate is different from the above two materials, but the difference is not large, the influence of the flow rate on the corrosion rate is not large, and the corrosion rate basically fluctuates around 1 mm/a. To ensure the quality of the screening work of the corrosion inhibitors afterwards, it is considered to select the corresponding flow rate of 0.6m/s at which the maximum corrosion rate occurs in the J55 orthogonal test. The influence magnitude sequence of each factor from the result of the mean value is as follows: water content (poor K1 ═ 2.55)>CO₂Partial pressure (k2 ═ 1.725)>Temperature (k3 ═ 1.52)>Degree of mineralization (K4 ═ 1.335)>Flow rate (k5 ═ 0.825).

(2) Results of orthogonal mean analysis of downhole N80/J55 steel: FIG. 13 is a graph showing the erosion rate of N80 material downhole as a function of different parameters. FIG. 14 is a graph of erosion rate of downhole J55 material with different parameters.

From FIG. 13, it can be seen that both materials are at a temperature of 45 ℃ and CO₂The partial pressure is 6.0MPa, the degree of mineralization is 50000mg/L, the corrosion is the most serious under the conditions that the oil-water content is 90 percent and the total pressure is 6 MPa. The above two materials differ in flow rate, but not much. Since the flow rate does not have a great influence on the corrosion rate, which basically fluctuates around 5mm/a, in order to ensure the quality of the screening work of the corrosion inhibitor, the corresponding flow at which the maximum corrosion rate occurs in the J55 orthogonal test is selectedThe speed was 0.6 m/s. The influence magnitude sequence of each factor from the result of the mean value is as follows:

n80: water content (poor K1 ═ 10.342)>Temperature (k2 ═ 6.213)>Total pressure (k3 ═ 5.397)>CO₂Partial pressure (k4 ═ 4.67)>Flow rate (k5 ═ 3.216)>Degree of mineralization (K6 ═ 2.134).

J55: water content (poor K1 ═ 10.135)>Temperature (k2 ═ 5.915)>Total pressure (k3 ═ 5.285)>CO₂Partial pressure (k4 ═ 4.42)>Flow rate (k5 ═ 3.025)>Degree of mineralization (K6 ═ 2.035).

The extreme value of the influence factor corresponding to the range (water content of 90%, temperature of 45 ℃, total pressure of 6MPa and CO) analyzed by the method₂Partial pressure of 6MPa and degree of mineralization of 50000mg/L) and the results show that the weight-loss corrosion rate of the coupon made of N80 material under the condition is 20.3412 mm/a.

In order to verify whether the numerical combination of the influence factors corresponding to the maximum corrosion rate analyzed by extreme difference is a parameter condition corresponding to the maximum corrosion rate under different well conditions. Therefore, the following verification test was performed.

Each factor was chosen at 3 points.

TABLE 2-21 Total pressure verification test for downhole N80 material

Through the verification test, namely under the condition that other conditions are not changed, the total pressure value is changed, so that the corrosion rate is changed, the total pressure has a certain influence on the underground corrosion rate, and partial literature researches show that the total pressure has no influence on the corrosion rate, and the results show that the actual influence is realized.

Tables 2-22 downhole N80 temperature validation tests

The selection of the verification test point is that two points are selected near an extreme point of the maximum corrosion rate value analyzed by the extreme value, such as the temperature of 45 ℃, and the two points are respectively 40 degrees and 55 degrees.

TABLE 2-23 downhole N80 Material flow Rate verification test

TABLE 2-24 mineralization degree verification test for downhole N80 material

Table 2-25 water content verification test for downhole N80 material

TABLE 2-26 partial pressure verification test for CO2 of N80 material downhole

All the above verification data are compared with the corrosion rate (20.3412mm/a) of the most severe test determined badly, and the verification values are all less than 20.3412mm/a, so that the test condition of single-factor combination can be determined to be the most severe test condition.

Because the medium is in the mode of turbine stirring in the reation kettle, therefore the effect of medium to inductance probe and lacing film is oblique shearing force effect, and the effect of on-the-spot extraction liquid is perpendicular shearing force effect to oil sleeve pipe, whether can influence the measuring result of corrosion rate in order to investigate this kind of difference of flow state, specially adopted dynamic simulation experimental apparatus to verify the experimental result.

The number of experimental groups was selected: downhole N80 material, group 25 (temperature 105 ℃, flow rate 1.1m/s, partial pressure 6MPa, mineralization 30000, total pressure 14 MPa, water content 15).

Tables 2-27 verification test of the Effect of flow regime on Corrosion rates

From the experimental results, it can be seen that the flow regime has some, but not significant, effect on the corrosion rate results.

The experimental conclusion of J55 and N80 are consistent.

1) The range analysis corrosion factor influence weight sequence is as follows: oil-water ratio>Temperature of>Total pressure>CO₂Partial pressure>Flow rate of flow>Degree of mineralization. 2) The most severe condition of material corrosion is as follows: the temperature is 45 ℃, the degree of mineralization is 50000mg/L, the water-oil ratio is 90 percent, the total pressure is 6MPa, and CO is₂The partial pressure is saturated.

Example 2, study of correlation between wellhead and downhole corrosion rates:

because a large number of underground corrosion rate test tests are carried out, two technical routes are adopted for underground corrosion condition prediction, and the technical advantages and the technical disadvantages are finally determined through laboratory and field verification.

Technical scheme 1: FIG. 15 is a graph showing a comparison of existing CO₂The prediction model is screened and corrected to directly predict the wellhead and underground corrosion rates, if correlation can be established through a certain corrosion influence factor, such as mineralization degree, the research purpose is achieved, and if correlation cannot be achieved, the correlation data of the corrosion influence factor can be directly applied to directly predict corrosion.

Technical scheme 2: the method avoids the influence of various influencing factors on the underground corrosion rate, directly measures the underground corrosion rate at different depths through tests, and directly performs data correlation by using the underground corrosion rate, the wellhead corrosion rate and the depth to establish the relationship between the wellhead corrosion rate and the underground corrosion rate.

Technical scheme 1:

model screening:

the corrosion model is a complex relational expression formed by various parameters, three models which are most commonly used in the world at present are selected to carry out modeling work of the project, and improvement is carried out on the basis of the existing model to obtain a corrosion model with higher practicability.

1. Norsok M506 model in Norway is an empirical model with representative significance at present, which is established according to low-temperature test data and high-temperature field data and is resistant to CO at home and abroad₂One important criterion for corrosion material selection and determination of corrosion margin design, but the disadvantage is that the influence of water content on the corrosion rate of the crude oil system is not taken into account. The research work on this project is not applicable.

2. At present, most mature prediction models are semi-empirical models, and DMW models built by De Warrd and Miliams become prediction CO₂The basis of corrosion. De Warrd2003 researches the influence of the specific gravity and the water content of the crude oil on the corrosion rate according to the interfacial tension between the oil and the water and obtains a prediction formula considering the factor of the crude oil.

S.Nesic model is mainly based on CO₂The microscopic corrosion mechanism is based on the prediction model established by combining the chemical and electrochemical reactions on the surface of the material, the mass transfer process of ions at the interface of the material and the solution, the diffusion and migration processes of the ions in the corrosion product film and the like. However, the mechanism model still mainly stagnates in the research work of the laboratory, and the corrosion data of the field test is not well combined, and the corrosion condition under the oil-water mixing condition is less considered.

Ward model introduction:

CO, most widely used in the field of corrosion research in the oil and gas industry₂The corrosion model is the de. In a recent study, the model investigator selected field corrosion data and related parameters including flow rate (calculated from yield), pipe diameter, CO, for two different API degrees of crude oil accumulated over 20 years₂Partial pressure, and HCO of sampled water_2-Concentration, water content, deviation angle of oil well pipe from vertical direction, etc. The amount of corrosion thinning is the average of the data from the field report measured by a micrometer.

The influence of crude oil on corrosion is considered in the latest corrosion model of de Waard, and the influence factors contained in the model are better consistent with the influence factors researched by the paper. Meanwhile, the Waard model was applied to an oil well without a corrosion inhibitor, and the corrosion results of this experiment were also obtained without the addition of a corrosion inhibitor. In summary, the c.de Waard comparison is consistent with the research application of this project.

Waard model is mainly for CO₂The partial pressure is less than 1MPa, so if one wants to achieve its application in this study, it must be modified for application. In order to better reflect CO under the action of various factors₂And (3) based on the C.de Waard model, the corrosion model of the paper is subjected to regression processing and analysis according to data results and further obtains a proper corrosion model aiming at specific conditions researched by the paper.

The Waard model consists essentially of:

the calculation formula of the de Waard model is as follows:

vr and Vm represent the maximum kinetic reaction rate and CO, respectively₂Influence of mass transfer process on corrosion. Vr and Vm are calculated as follows:

Log(V_r)＝5.07-+0.58log(P_CO2)–0.34(pH_a–pH_CO2) 2-3

P_CO2is CO containing fugacity coefficient₂Partial pressure in MPa; pHa is the pH value of the medium under actual working conditions; pH value_CO2Is the same CO₂The pH value of the pure water under partial pressure; t represents temperature in units of; uliq is the flow velocity, and the unit is m/s; d is the internal diameter of the tube in m.

When considering the influence of pH, temperature will influence CO₂Solubility and H of₂CO₃The ionization number of (c).

pH_CO2＝3.82+0.00384T–0.5log(P_CO2) 2-5

In general, the pH value of the medium can be obtained by measurement, but the pH value of the medium is difficult to directly measure under high temperature and high pressure, so that the pH value of the medium can be determined according to international standard ISO15156-2 instead of pHa for a high temperature and high pressure system, as shown in fig. 16-17:

1：Ca²⁺＝1000meq/L、2：Ca²⁺＝100meq/L、3：Ca²⁺＝10meq/L、4：HCO_2-＝10meq/L、5：HCO_2-＝30meq/L、6：HCO_2-＝100meq/L。

-------Ca²⁺<HCO_2-__________Ca²⁺<HCO_2-———Ca²⁺<HCO_2-。

FIG. 16 shows the difference in CO at 20 ℃₂Bicarbonate-containing (saturated or supersaturated) water under pressure produces a pH value for the water.

-------Ca²⁺<HCO_2---------Ca²⁺<HCO_2-———Ca²⁺<HCO₃。

FIG. 17 shows the difference in CO at 60 ℃₂Bicarbonate-containing (saturated or supersaturated) water under pressure produces a pH value for the water.

-------Ca²⁺<HCO_2-______Ca²⁺<HCO_2-———Ca²⁺<HCO₃。

FIG. 18 shows different CO at 100 ℃₂Bicarbonate-containing (saturated or supersaturated) water under pressure produces a pH value for the water.

(2) And (3) researching corrosion prediction models of different materials.

The models screened out above are set under the condition of a low-pressure range (pressure less than 2MPa) of A3 steel, and in order to research corrosion rate prediction models of different materials, wellhead low pressure and underground high pressure, firstly, the change rules of the corrosion rates of different materials under the same test condition are researched. Relationship of corrosion rates for N80, J55, and A3 at low wellhead pressure: see fig. 19 for three coupon measurements.

From the corrosion rate curves of the above three materials, it can be seen that under the same test conditions, the corrosion rate changes in a consistent manner, but the group 23 shows that the corrosion rate of the a3 steel is relatively abnormal, so that the group is removed in the subsequent analysis.

The prediction conditions of the prediction model are specific to corrosion of A3 steel, and the change rule of the corrosion rate of N80 and J55 materials along with the corrosion conditions is consistent with that of A3 steel, so that the corrosion rate of N80 and J55 can be predicted by taking the corrosion model of A3 steel as reference, and the rule is searched by a data fitting method. First, the agreement between the prediction model and the measured value of a3 steel was determined.

TABLE 2-28 comparison of model predicted values with A3 steel measured values

Serial number	Model prediction value	Measured value of A3 Steel	Relative error
					1	0.006320151	0.0024	62.0％
2	0.010847183	0.1035	89.5％
				3	0.022802008	0.2281	90.0％
4	0.035261281	0.6331	94.4％
				5	0.016800128	0.8454	98.0％
6	0.009610199	0.4584	97.9％
				7	0.017350934	1.2151	98.6％
8	0.028888421	0.0286	1.0％
				9	0.035569233	0.1887	81.2％
10	0.007656713	0.5274	98.5％
				11	0.041010214	0.4581	91.0％
12	0.036011797	0.8784	95.9％
				13	0.121311899	3.3863	96.4％
14	0.027021614	4.998	99.5％
				15	0.042063636	0.126	66.6％
16	0.062622185	2.8714	97.8％
				17	0.093174043	0.3592	74.1％
18	0.022875856	0.4151	94.5％
				19	0.055099124	2.1599	97.4％
20	0.078093413	2.1012	96.3％
				21	0.124389274	2.6407	95.3％
22	0.051303856	0.4073	87.4％
				24	0.079132633	0.0365	59.1％
25	0.154521112	0.5154	70.0％

As can be seen from the above table, the error between the predicted value and the measured value of the model is large because there is no CO with complete consideration factors₂A prediction model, wherein the adopted model mainly aims at an oil well with low water content and takes temperature and CO into consideration₂The partial pressure and the flow velocity influence the corrosion rate, and the research of the project aims at the oil well with high water content and high mineralization degree and comprehensively considers the temperature, the total pressure and the CO₂Partial pressure, flow rate, water content and mineralization degree have influence on the corrosion rate, so that large error is inevitable.

And determining the incidence relation between the real value and the predicted value through the actually measured A3 steel data and methods such as data fitting processing and the like, thereby obtaining a prediction model of the A3 steel under the wellhead low-pressure condition.

Firstly, directly fitting the predicted values and measured values of 25 groups, and the preferable curve result is shown in fig. 20: the relationship between the predicted value and the measured value of the A3 model.

(1) When the temperature is less than or equal to 45 ℃: see FIG. 21 for the relationship between predicted and measured values (. ltoreq.45 ℃ C.) for the A3 model.

(2) When the temperature is more than 45 ℃: see FIG. 22 for the relationship (> 45 ℃) between predicted and measured values for the A3 model. The relative error of the fitted values after fitting to the measured values of the a3 steel is shown in the following tables 2 to 29:

TABLE 2-29A 3 comparison of Corrosion Rate after fitting to actual values (. ltoreq.45 ℃ C.)

Serial number	Measured value (mm/a)	Fitting value (mm/a)	Relative error
				1	0.0024	0.0025	4.0％
2	0.4025	0.4011	0.3％
				3	2.5679	2.56	0.3％
4	0.6331	0.6331	0.0％
				5	0.8454	0.8468	0.2％
6	0.4584	0.4597	0.3％
				7	1.2151	1.2056	0.8％
8	0.1887	0.1897	0.5％
				9	0.5274	0.5199	1.4％
Mean error			0.9％

TABLE 2-30A 3 comparison of Corrosion rates after fitting to actual values (> 45 ℃ C.)

Serial number	Measured value (mm/a)	Fitting value (mm/a)	Relative error
				1	0.4581	0.4878	6.1％
2	3.3863	3.3861	0.2％
				3	0.3592	0.359	0.1％
4	0.4151	0.4141	0.2％
				5	2.6407	2.6400	0.0％
6	0.0365	0.0397	8.1％
				7	0.5154	0.5149	0.1％
Mean error			2.11％

From the comparison between the fitting value and the measured value, it can be seen that the average relative error is 0.9% when the temperature is less than or equal to 45 ℃, and 2.11% when the temperature is greater than 45 ℃, which indicates that the fitting result has a high goodness of fit with the true value.

Predictive models for uphole and downhole N80 and J55:

first, whether a data correlation exists between the original prediction result of the A3 steel and N80 and J55 is directly determined by a data fitting method.

If the correlation does not exist, the measured corrosion rate of the A3 steel is further analyzed to be correlated with the measured corrosion rates of N80 and J55, and then the prediction models of N80 and J55 are obtained through the corrected prediction model of the A3 steel.

Referring to the fitting relationship between the model values and the measured values of the a3 steel, the following fitting results of N80 and J55 were obtained.

The fitting relationship between the predicted value of the wellhead model and the measured value of N80 is as follows:

when the temperature is less than or equal to 45 ℃: see FIG. 23 for the relationship between the model predicted value and the N80 well head measured value (45 ℃ or less).

At a temperature > 45 ℃: see FIG. 24 for the relationship (> 45 ℃) between the model predicted values and the N80 well-head measured values.

TABLE 2-31N 80 comparison of the corrosion rate after wellhead fitting with the measured value (≦ 45 ℃)

Serial number	Measured value (mm/a)	Fitting value (mm/a)	Relative error
				1	0.00771	0.00773	0.3％
2	0.24865	0.24877	0.0％
				3	1.8899	1.899	0.5％
4	3.2614	3.2712	0.3％
				5	4.618	4.6234	0.1％
6	4.2439	4.25	0.1％
				7	2.7789	2.7877	0.3％
8	0.2378	0.2398	0.8％
				9	0.23148	0.2389	3.1％
Mean error			0.6％

TABLE 2 comparison of Corrosion Rate to actual values (> 45 ℃ C.) after fitting to 32N 80 Steel wellhead

Besides eliminating some abnormal data, it can be seen that the relative errors of the fitting value of N80 and the measured value of the wellhead are respectively 0.6% and 0.5%, and the data goodness of fit is better. The fitting relationship between the predicted value of the wellhead model and the measured value of J55 is as follows:

the temperature is less than or equal to 45 ℃: FIG. 25 shows the relationship between the predicted model value and the measured value of J55 wellhead (45 ℃ or lower).

The temperature is higher than 45 ℃: see FIG. 26 for the relationship (> 45 ℃) between the predicted model value and the measured value of J55 wellhead.

TABLE 2-33J 55 comparison of the corrosion rate after wellhead fitting with the measured value (≦ 45 ℃)

Serial number	Measured value (mm/a)	Fitting value (mm/a)	Relative error
				1	0.93826	0.9401	0.2％
2	1.9458	1.9465	0.0％
				3	1.3935	1.3942	0.1％
4	1.7734	1.3987	21.1％
				5	3.7925	3.6785	3.0％
6	2.098	2.099	0.0％
				7	0.1795	0.183	1.9％
8	0.58721	0.6077	3.4％
				Mean error			3.7％

TABLE 2-34J 55 comparison of Corrosion rates after fitting to actual measurements (> 45 ℃ C.)

Besides eliminating some abnormal data, it can be seen that the relative errors of the fitting value of J55 and the measured value of the wellhead are respectively 3.7% and 1.0%, and the data goodness of fit is better.

The fit relationship between the predicted value of the downhole model and the measured value of N80 is as follows:

(1) the temperature is less than or equal to 45 ℃: FIG. 27 is a graph of the relationship between model predicted values and N80 downhole measured values (< 45 ℃).

(2) The temperature is higher than 45 ℃: FIG. 28 is a plot (> 45 ℃) of model predicted values versus N80 measured downhole.

Tables 2-35N 80 comparison of Corrosion rates after downhole fitting to measured values (. ltoreq.45 ℃ C.)

Serial number	Measured value (mm/a)	Fitting value (mm/a)	Relative error
				1	0.00114	0.0011	3.5％
2	3.34273	3.789	11.8％
				3	9.81846	7.5098	23.5％
4	11.08	11.1145	0.3％
				5	4.91793	4.897	0.4％
6	16.22929	16.1986	0.2％
				7	11.6774	11.6779	0.0％
8	2.3304	2.4567	5.1％
				Mean error			5.6％

TABLE 2-36N 80 comparison of Corrosion rates after downhole fitting to measured values (> 45 ℃ C.)

Besides eliminating some abnormal data, it can be seen that the relative errors of the fitting value of N80 and the measured value are 5.6% and 6.3%, respectively, and the data goodness of fit is better.

The fit between the predicted values of the downhole model and the measured values of J55 is as follows:

the temperature is less than or equal to 45 ℃: FIG. 29 is a graph of the relationship between model predicted values and J55 measured downhole values (< 45 ℃).

The temperature is higher than 45 ℃: FIG. 30 is a plot (> 45 ℃) of model predicted values versus measured values in J55 downhole.

Tables 2-37J 55 comparison of Corrosion Rate after Down hole fitting to actual measurements (. ltoreq.45 ℃ C.)

Serial number	Measured value (mm/a)	Fitting value (mm/a)	Relative error
				1	0.0045	0.005	10.0％
2	0.6556	0.7123	8.0％
				3	6.2188	6.2976	1.3％
4	6.8913	6.6678	3.2％
				5	8.0876	8.4567	4.4％
6	2.9064	2.9346	1.0％
				7	16.6049	16.5431	0.4％
8	9.9523	9.96	0.1％
				9	1.7639	1.5098	14.4％
Mean error			4.7％

TABLE 2-38J 55 comparison of Corrosion rates after downhole fitting to measured values (> 45 ℃ C.)

Besides eliminating some abnormal data, the relative errors of the fitting value of J55 and the measured value of the underground simulation test are respectively 4.7% and 4.9%, and the data goodness of fit is good.

Technical scheme 2:

the test is supposed to be derived from the fact that the temperature of the well head and the temperature in the well and the well depth have a certain empirical relationship (estimated according to the temperature gradient of the well at 3.5 ℃):

namely: t is_Downhole＝T_{Well head}+0.035h (relation 1).

Therefore, if the function of the change of the wellhead temperature and the wellhead corrosion rate and the correlation function of the downhole temperature and the downhole corrosion rate can be found out through experiments, the relationship between the downhole corrosion rate and the wellhead corrosion rate can be calculated through the relational expression.

Because the influence factors of the corrosion rate of the wellhead and the corrosion rate of the underground are more than one temperature term, if the temperature is simply changed and other conditions are not changed, the obtained relation is greatly different from the actual situation, and therefore the project uses the temperature and CO which have larger difference between the wellhead and the underground₂Three key corrosion influence factors of partial pressure and total pressure are normalized to temperature parameters, the functional relations between the wellhead corrosion rate and the normalized temperature and the functions between the underground corrosion rate and the normalized temperature are respectively determined, and then the correlation model between the wellhead corrosion rate and the underground corrosion rate is replaced through the relation between the temperature and the well depth.

(1) Setting parameters of the underground simulation test: n80 material.

Through test measurement and mean calculation, the corresponding results of the downhole temperature and the downhole corrosion rate are shown in the following table:

table 2-39 table for setting parameters of downhole simulation test

Fitting the underground corrosion rate Y through the numerical values_DownholeWith downhole temperature T_DownholeThe functional relationship of (a) is shown in FIG. 31: FIG. 31 is a plot of downhole erosion rate versus downhole temperature.

Y_Downhole＝0.00007750T_Downhole ³-0.01784687T_Downhole ²+1.21903620T_Downhole-19.68881518

(relation 2).

(2) Setting parameters of a wellhead simulation test: n80 material.

After test measurement and mean calculation, the obtained corresponding results of the wellhead temperature and the wellhead corrosion rate are shown in the following table:

TABLE 2-40 corresponding tables of wellhead temperature and wellhead corrosion rate

By numerical fitting, a relationship graph of the wellhead temperature and the wellhead corrosion rate is obtained as shown in FIG. 32: wellhead temperature versus wellhead corrosion rate.

Namely: t is_{Well head}＝59.755Y_{Well head} ^0.4744(relational expression 3).

The downhole corrosion rate Y can be obtained by substituting the

relational expressions

1, 2 and 3_DownholeWith wellhead corrosion rate Y_{Well head}And the correlation of the downhole depth h is as follows:

Y_downhole＝0.00007750×(59.755Y_{Well head} ^0.4744+0.035h)³-0.01784687×(59.755Y_{Well head} ^0.4744+0.035h)²+1.21903620×(59.755Y_{Well head} ^0.4744+0.035h)-19.68881518

It can be seen from the above relational expression that if the online monitoring inductance probe is installed at the wellhead, the corrosion rate of the wellhead can be known in real time, and then the underground corrosion rates of different depths can be calculated.

And (3) carrying out a corrosion simulation test of a single well by using the inductive probe and a laboratory, and verifying the accuracy of the correlation result. Wherein the temperature of the well mouth is 35 ℃, the back pressure of the well mouth is 1MPa, and the carbon dioxide content of the well mouth is 40%.

The downhole test conditions are shown in the following table: tables 2-41 test conditions for different depths

The results of the simulation are shown in the following table: TABLE 2-42 comparison of predicted and measured corrosion rates

The verification result shows that the errors of the predicted value and the measured value are controlled within 10%, which indicates that the predicted value of the correlation model is accurate. Summary through the laboratory simulation study of this section, the summary is as follows:

through the analysis of the components of the on-site produced liquid and the analysis of other industrial and mining, relevant conditions of a laboratory simulation test are determined, and simulated corrosion factors including temperature, total pressure and CO of a well bottom and a well head are determined₂Partial pressure, flow rate, degree of mineralization, and water content.

An inductance monitoring system and a hanging piece weight loss method are adopted to complete a laboratory simulation high-temperature high-pressure dynamic test, and a total of 125 groups of wellhead and underground simulation tests are developed for materials such as A3/N80/J55.

Through the correlation analysis of the corrosion coupon and the inductance probe data, the result shows that the coupon and the inductance probe data of the laboratory simulation test have good consistency and a quantitative correlation relationship.

The results show that the sequence of the wellhead and underground corrosion influencing factors is the water content through the analysis of the wellhead and underground corrosion influencing factors>Temperature of>Total pressure>CO₂Partial pressure>Flow rate of flow>Degree of mineralization. The parameters corresponding to the most severe corrosion in the well are respectively: temperature of 45 ℃, water content of 90 percent, total pressure of 6MPa and CO₂The partial pressure is 6MPa, the flow rate is 0.6m/s, and the degree of mineralization is 50000 mg/L.

By aligning the well head and the wellData analysis under C.de Waard CO for reference₂The corrosion prediction model respectively determines the corrosion prediction models of A3, N80 and J55 materials under the condition of a wellhead and the materials of N80 and J55 materials under the condition of the wellhead by a data fitting method, and the fitting value is consistent with the measured value through laboratory verification, so that the relative error is small, and the accuracy of the prediction model is proved.

And determining that the corrosion rates of the wellhead and the downhole have a polynomial relation related to the depth of the downhole by using the corrosion rate of the wellhead:

Through the simulation measurement of a laboratory, the errors of the predicted value and the measured value are confirmed to be within 10 percent, and the correlation model is true and reliable.

Example 3. from here on is an application example.

CO₂And driving an online corrosion monitoring system to be applied and researched.

1. The corrosion monitoring application research of the 1# experimental well of the first mine in the south of the Yangtze province.

The experimental start time for test well # 1 was 12 months and 7 days in 2015.

(1) Installing wellhead monitoring equipment and analyzing data.

The field working condition environment of the 1# experimental well of the first mine in the south of the Yangtze province is (12 months in 2015):

well head temperature: 18 ℃ and total wellhead pressure: 0.5Mpa, sucker rod stroke: 3m, punching times of the sucker rod: 4. liquid production per day: 31.8t, daily water: 27t, crude oil water content: 85 percent.

The analysis data of the liquid and gas samples of the production fluid of the test well # 1 are as follows (analysis time: 12/23 days 2015).

Table 3-11 # test well produced fluid gas analysis results

Table 3-21 # test well produced liquid analysis results

① inductance probe monitoring data:

the field working condition of the experimental well is in the range of various influence factors set by a laboratory simulation experiment, but no coincident experimental condition point exists, so that a laboratory verification experiment is carried out at the initial installation stage according to field parameters for comparing the data difference between a laboratory and a field (inductance probe material A3 steel).

Referring to field industrial and mining, determining indoor verification test conditions as follows:

degree of mineralization: 56900mg/L, water content 85%, total pressure 0.5MPa, CO₂Partial pressure of 0.4MPa, flow rate of 0.4m/s, temperature: 16 ℃ is adopted.

and a, monitoring data and analyzing by using an A3 wellhead inductance probe. See figure 33 for a # 1 well indoor validation test.

See fig. 34 for the 1# well a3 steel initial stabilization segment in-situ inductance probe data.

Compared with the initial data on site, the stage corrosion rate of the laboratory is 1.688mm/a, the site corrosion rate is 1.699mm/a, the relative error is 0.67%, and the data goodness of fit is better.

See figure 351 # pilot well wellhead a3 inductive probe monitoring data.

The starting time for installing the inductance probe and the hanging piece in the test well # 1, wellhead A3 is from 12/7/2015 to 12/2015 and 30/2015, and the monitoring data of the inductance probe is shown in FIG. 35. As can be seen from the data curves, the corrosion reduction during installation of the A3 steel inductance probe was 6098nm, and the interval corrosion rate was 0.097 mm/a.

The following table shows the measured values of the wellhead inductance probe, the predicted values of the wellhead model A3, and the comparison results of the measured values of the wellhead hanging pieces. It can be seen that the corrosion rate calculated from the weight loss data of the hanger sheet at the same period is 0.092mm/a, and the relative error with the measured value of the inductance probe is 5.1%; the annual corrosion rate of A3 steel under the industrial and mining conditions is 0.09058mm/a, and the error between a predicted value and an actual measured value of a wellhead inductance probe is 6.6%.

TABLE 3-3A 3 comparison of inductive probe etch rate to coupon etch rate and predicted value

And b, monitoring data and analyzing by using the N80 wellhead inductance probe.

The installation start time of the inductance probe and the hanging piece made of the N80 material is 2015, 12 months and 30 days to 2016, 5 months and 5 days, monitoring data of the inductance probe are shown in FIG. 35, and as can be seen from the corrosion curve, the corrosion reduction amount during the installation of the N80 inductance probe is 25217nm, and the interval corrosion rate is 0.076 mm/a.

The calculated corrosion rate of the weight loss data of the hanger at the same time is 0.072mm/a, and the relative error is 5.2 percent.

And (4) performing an N80 material test according to the conditions of the indoor verification test, and determining the consistency of the indoor test result and the field monitoring result. The results show that the weight loss data of the hanging piece in the indoor test is 0.073mm/a, and the relative error is 1.3 percent.

According to an N80 wellhead corrosion prediction model obtained by indoor tests, the annual corrosion rate of N80 steel under the industrial and mining conditions is 0.06901mm/a, the error between a predicted value and an inductance probe wellhead measured value is 5.4%, the error between the predicted value and a hanging piece wellhead measured value is 4.1%, and the coincidence degree between the predicted value and the measured value is high.

TABLE 3-4 comparison of inductive probe corrosion rate, coupon corrosion rate and predicted value for N80 material

In the process of the N80 material test, the iron content in the produced liquid was measured in 2016, 1 month and 4 days, 2016, 2 months and 4 days, 2016, 3 months and 20 days, 2016, 5 months and 5 days, respectively, and the results are shown in the following tables 3-5.

Tables 3 to 5 results of the detection of iron ion content

The relationship between the corrosion rate value monitored by the inductive probe corresponding to the sampling time and the content of iron ions is shown in fig. 37: the iron ion content and the corrosion rate.

From the above analysis, it can be seen that, as the content of iron ions increases, the corrosion rate of the corresponding inductance probe also shows an increasing trend, and the two satisfy a polynomial relationship.

(2) Correlation of wellhead and downhole corrosion data.

The site hanging ring schematic diagram of the No. 1 test well is shown in FIG. 39: 4-11x101 field shackle installation schematic.

The first group of hanging rings 12, the 6 th hanging ring (N80), the 3 rd hanging ring (J55) and the 9 th hanging ring (A3) are arranged at the position of 600 m.

900m is provided with a second group of hanging rings 13, a 5 th hanging ring (N80), a 2 nd hanging ring (J55) and an 8 th hanging ring (A3).

At 1200m, a third set of suspension loops 14, a 4 th suspension loop (N80), a 1 st suspension loop (J55) and a 7 th suspension loop (A3) are arranged. A valve pump and pin drain were placed at 1303.56m, and a wire-wrapped screen was placed at 1312.71 m.

Tables 3-6 predicted downhole depth values and measured corrosion rate values (mm/a)

As can be seen from the above table, the measured values of the downhole J55 hanging loops are compared with the predicted values of the model, and the relative errors are respectively 4.8%, 7.5% and 6.3%, and the average error is 6.2%. Compared with the predicted value of the model, the measured value of the downhole N80 hanging ring has relative errors of 3.8%, 6.7% and 6.5% respectively, and the average error is 5.7%.

The results show that the relevance of the wellhead data and the downhole data is better, the downhole corrosion rate value calculated by utilizing the wellhead corrosion rate and downhole corrosion rate correlation model is more accurate than the value estimated by utilizing the downhole corrosion influence factor data, and the analysis reasons are as follows: the influence factors involved in the previous model cannot cover all influence factors, including the influence of ions, and in addition, the downhole temperature, pressure, partial pressure and the like are not supported by real-time field measured data, and the utilized data are generally empirical values or estimated values, so that the accuracy of the model is limited by the factors. The accuracy rate of prediction by using laboratory data is higher than that of field data because the temperature, the pressure and the like of the laboratory data are specific parameters of the test and are more accurate.

The prediction is carried out by utilizing the correlation model of the wellhead corrosion rate and the underground corrosion rate, and for a single well, the prediction avoids the influence of various complex factors, and the change of the corrosion rate is only correlated with the depth, so that the value accuracy of the predicted value is higher than that of direct prediction.

Example 4

The corrosion monitoring application research of the 2# experimental well of the Miniangyi mine.

The initial test time for the 2# test well was 2016, 1, 23.

The installation scheme of the 2# test well wellhead corrosion monitoring system is still an inductance probe and a hanging piece.

The field working condition environment of the Binnan one-mine 2# experimental well is as follows: well head temperature: 25 ℃ and wellhead back pressure: 0.8MPa, daily product liquor: 10.5t, daily water production: 7.3t, sucker rod stroke: 2.4m, punching times of the sucker rod: 5.

the liquid gas analysis data for month 4 in 2015 was as follows: degree of mineralization 72334.4mg/L, chloride ion 38995mg/L, total carbonate ion 5796.9mg/L, CO₂Content 45%, crude oil water content: 70 percent.

(1) A wellhead inductance corrosion monitoring system.

and a, analyzing wellhead monitoring data of the A3 steel inductance probe. Graph 402 # test well a3 material inductance probe monitoring data.

In test well No. 2, the initial time for installing the wellhead a3 inductance probe and the hanging strip is 2016, 1-23 days-2016, 5-3 days-2016, and the monitoring data of the inductance probe is shown in fig. 40. As can be seen from the data curves, the corrosion reduction during installation of the A3 steel inductance probe was 32381nm, with a span corrosion rate of 0.117 mm/a.

TABLE 3-7A 3 comparison of inductive probe etch rate to coupon etch rate and predicted value

The calculated weight loss data of the coupons at the same time gave a corrosion rate of 0.110mm/a with a relative error of 5.9%.

The prediction result of the laboratory model under the well condition is 0.108mm/a, the relative error with the measured value of the on-site hanging film is 1.8 percent, and the error with the measured value of the inductance probe is 7.6 percent.

In the A3 material test process, the iron content in the produced liquid was measured in 2016 (1 month, 23 days), 2016 (2 months, 15 days), 2016 (3 months, 15 days), 2016 (4 months, 15 days), 2016 (5 months, 5 days), respectively, and the results are shown in the following tables 3-8.

Tables 3 to 8 results of detection of iron ion content

The relationship between the corrosion rate value monitored by the inductance probe corresponding to the sampling time and the content of iron ions is shown in fig. 41: the corresponding relation between the iron ion content and the corrosion rate of the inductance probe.

From the above analysis, it can be seen that, as the content of iron ions increases, the corrosion rate of the corresponding inductive probe also shows an increasing trend, and the two satisfy a polynomial relationship and are consistent with the result of the test well # 1.

And b, analyzing the wellhead monitoring data of the N80 steel inductance probe. See fig. 42 for data monitored by the N80 quality inductance probe of test well No. 2.

The red curve represents the inductance probe corrosion loss value and the blue curve represents the inductance probe temperature. The interval corrosion loss value of the N80 inductance probe is 689 nm. The corrosion rate is 0.0112mm/a, and the corrosion is slight. The model prediction value under the condition is 0.0108mm/a, and the relative error of the model prediction value and the model prediction value is 3.5%.

TABLE 3-9 comparison of inductive probe corrosion rate, coupon corrosion rate and predicted value for N80 materials

(2) Correlation of wellhead and downhole suspension loop data.

The field hanging ring scheme of the No. 2 test well is the same as that of the No. 1 test well.

Tables 3-10 predicted downhole depth values and measured corrosion rate values (mm/a)

As can be seen from the above table, the measured values of the downhole J55 hanging loops are compared with the predicted values of the model, and the relative errors are 7.9%, 6.5% and 8.0%, respectively, and the average error is 7.5%. Compared with the predicted value of the model, the measured value of the downhole N80 hanging ring has relative errors of 7.4%, 7.7% and 7.4% respectively, and the average error is 7.5%.

The results show that the relevance of the wellhead data and the downhole data is better, the downhole corrosion rate value calculated by utilizing the wellhead corrosion rate and downhole corrosion rate correlation model is more accurate than the value estimated by utilizing the downhole corrosion influence factor data, and the analysis reasons are as follows:

Example 5

3. Corrosion monitoring application research of Binan Yishi 3# experimental well

Test well number: 10x 100.

Well condition conditions: stroke: 4. punching times: 3.5, well head temperature: 20 ℃, wellhead back pressure: 0.9MPa, daily product liquor: 8t, daily water production: 7.4t, water content: 97 percent.

(1) And monitoring wellhead corrosion.

The installation scheme is as follows: the installation scheme of the 3# test well wellhead corrosion monitoring system is still the inductive probe and the hanging piece all-dimensional monitoring. Test initiation time: 2016 for 10 months. Wellhead corrosion monitoring application results:

TABLE 3-11A 3 comparison of inductive probe etch rate to coupon etch rate and predicted value

TABLE 3-12 comparison of inductive probe etch rate and coupon etch rate and predicted value for N80 materials

As can be seen from the wellhead corrosion monitoring test results in the tables, the inductance probe monitoring results of the wellhead A3 and N80 materials are consistent with the actually measured results of the hanging pieces and the model prediction results, and the errors are below 6%.

(2) Correlation of wellhead and downhole suspension loop data. The field hanging scheme of the No. 3 test well is the same as that of the No. 1 test well and the No. 2 test well.

TABLE 3-13 predicted values of downhole depths and measured corrosion rate values (mm/a)

As can be seen from the above table, the measured values of the downhole J55 hanging loops are compared with the predicted values of the model, the relative errors are 5.6%, 6.9% and 6.3%, and the average error is 6.3%. Compared with the predicted value of the model, the measured value of the downhole N80 hanging ring has relative errors of 8.7%, 5.9% and 6.9%, and the average error is 7.2%.

And (4) verification result: the errors of the estimated value and the measured value are controlled within 10%, which shows that the estimated value of the correlation model is accurate. And establishing the relevance model successfully.

And (3) calculating to obtain corrosion rate change curves of different underground well depths by utilizing a well mouth and underground corrosion rate correlation model according to the actual measurement result of the well mouth inductance probe, and knowing that the maximum values of the corrosion rate of the 3 wells are respectively found at the 1# well depth 910m, the 2# well depth 980m and the 3# well depth 900 m. As shown in fig. 44, 45, 46.

Through the development and the field test of the corrosion monitoring equipment of different types in the well head and the pit, the all-round monitoring of well head and underground equipment is realized, and the following can be summarized:

(1) the wellhead inductance monitoring of the No. 1 test well completes the field test of the A3 steel material and the N80 material. The monitoring results show that the wellhead corrosion monitoring rates of the A3 steel and the N80 material are 0.097mm/a and 0.076mm/a respectively. Compared with the hanger sheet monitoring result, the relative errors are respectively 5.1 percent and 5.2 percent, and the actual measurement value of the inductance probe is compared with the prediction result of the wellhead corrosion rate, and the relative errors are respectively 6.6 percent and 5.4 percent.

(2) Wellhead testing of test well No. 2 completed field testing of a3 and N80 materials. The results show that the wellhead corrosion monitoring rates for A3 and N80 are 0.117mm/a and 0.0112mm/a, respectively. The contrast coincidence degree of the result of the method and the result of the monitoring of the hanging film is higher, and the relative errors are respectively 5.9 percent and 4.9 percent. The measured value of the inductance probe is compared with a wellhead prediction model, and the relative errors are 6.0 percent and 5.8 percent respectively.

(3) Wellhead testing of test well No. 3 completed field testing of a3 and N80 materials. The results show that the wellhead corrosion monitoring rates for A3 and N80 are 0.108mm/a and 0.096mm/a, respectively. The contrast coincidence degree of the result of the method and the result of the monitoring of the hanging film is higher, and the relative errors are respectively 5.7 percent and 4.9 percent. Compared with a wellhead prediction model, the measured value of the inductance probe has relative errors of 7.6 percent and 3.5 percent respectively.

(4) The regular iron ion content detection result of the test well is consistent with the corrosion rate change result, namely the iron ion content is correspondingly increased during the corrosion rate increasing period, and the regular iron ion content detection result and the corrosion rate change result meet the relation of a polynomial.

(5) The comparison of the measured value result of the underground hanging ring with the predicted value determined by the wellhead and underground correlation model shows that the relative error is less than 10%, which indicates that the relationship between the wellhead corrosion rate and the underground corrosion rate determined by the correlation model is reliable. Through laboratory simulation studies, CO is determined₂The corrosion rule of the well is determined, and the weight sequence of the underground corrosion influence factors is determined to be the water content>Temperature of>Total pressure>CO₂Partial pressure>Flow rate of flow>Degree of mineralization.

Through the analysis of the influence factors of the corrosion of the well head and the underground, the corresponding parameters when the corrosion of the underground is the most serious are respectively as follows: temperature of 45 ℃, water content of 90 percent, total pressure of 6MPa and CO₂The partial pressure is 7.5MPa, the flow rate is 0.6m/s, and the degree of mineralization is 50000 mg/L.

The errors of the predicted value and the measured value are confirmed to be within 10% through simulation measurement in a laboratory, and the errors of the predicted value and the measured value of the model are also confirmed to be within 10% through verification of field hanging ring data, so that the associated model is true and reliable. The project target of presuming the corrosion condition in the pit through the corrosion rate of the wellhead inductance probe is realized.

The inductance probe and the compensation method adopted in the technical scheme can be as follows: 2017103383610, and a compensation method. The name is a high-voltage inductance probe for measuring the metal corrosion rate and a compensation method thereof.

A high-voltage inductance probe for measuring metal corrosion rate comprises a six-core glass screen seal aerial plug, an O-shaped seal ring check ring, an O-shaped seal ring, a transition pipe, a pouring sealant, a lead, a probe rod, a protective cap, a compensation test piece and a measurement test piece. The aerial insertion of the six-core glass sealing screen leads out a measurement signal through an internal wire, the lead and the shell are sealed in an insulating way, the wire and the aerial insertion shell are isolated and insulated, and meanwhile, the six-core glass sealing screen can play a good sealing role under a high-temperature and high-pressure medium to prevent the medium in the pipeline from seeping out after the probe is perforated. The O-shaped sealing ring retainer ring and the O-shaped sealing ring are matched with the conical surface structure to play a role in sealing the probe and the probe outlet device.

One end of the transition pipe is connected with the aerial insertion, the other end of the transition pipe is connected with the rear end of the probe rod through threads and then is welded and fixed, and the measuring test piece is fixed at the front end of the probe rod.

The wire is sealed in the probe rod, one end of the wire is connected with the measuring test piece and the compensating test piece, and the other end of the wire is connected with the glass sealing screen aerial plug. The compensation test piece is sealed in the measuring test piece and is welded with the measuring test piece in an isolation mode, and galvanic corrosion is avoided. The protective cap and the probe rod are connected together in a threaded manner.

The transition pipe and the probe rod are connected by the thread and then welded, so that the probe rod is prevented from falling into the pipeline because the thread part is connected even if the welding opening is cracked.

O-shaped sealing ring retaining rings and O-shaped sealing rings are respectively arranged in retaining ring grooves and sealing ring grooves on the cone part of the six-core glass screen-sealing aerial plug connected with the transition pipe.

The probe rod is filled by the pouring sealant, so that looseness and breakage caused by vibration are prevented, the internal supporting effect on the measuring test piece and the probe rod is realized, the pressure resistance of the probe is improved, and once the probe is perforated, the probe is prevented from being exposed and corroded.

The protective caps are opened in a staggered mode, multiple jackscrews are added for locking after the protective caps are connected with the probe rod, meanwhile, 2 thread breaking process holes are formed in the protective caps, threads are broken at the holes, and the protective caps are guaranteed not to be separated from the probe rod.

A compensation method of a high-voltage inductance probe for measuring metal corrosion rate comprises the following steps:

in order to ensure that the actual thinning value of the probe can be accurately measured when the thickness of the probe is increased under the condition that the length of the probe is not changed, and eliminate the influence of a nonlinear section with the thickness of more than 8mm on the measurement accuracy, the invention adopts the following steps:

step one, adopting proper physical size of the test piece, and enabling the actual thinning of the measurement test piece and the model compensation to form good fitting through a temperature compensation technology and a mathematical model uninterrupted continuous compensation technology; and step two, listing all the measured theoretical values one by one through a compensation algorithm to form a table, and in the later acquisition process, applying a table look-up method to obtain the actual thinning amount after reading the actually measured resistance value.

The physical dimensions of the test piece include the outer diameter R of the tubular test piece, the inner diameter R of the tubular test piece, the wall thickness delta of the tubular test piece and the test piece length L of the tubular test piece.

The temperature compensation technology and the mathematical model uninterrupted continuous compensation technology respectively compensate the temperature compensation test piece and the measurement test piece, so that the actual thinning of the measurement test piece and the model compensation form good fitting, and the measurement result is consistent with the true value.

The table comprises all measured values of the section with the outer diameter from 9.5mm to 7.5mm of thinning and intermediate functions thereof, and the table is concise and visual and convenient to search.

The invention has the following advantages:

the corrosion is indirectly measured, the corrosion condition of the pipeline is indirectly reflected by measuring the corrosion rate of the components made of the same material, the corrosion rate is calculated by accumulating the corrosion thinning amount for a period of time, the measurement is convenient and fast, and the operability is strong.

The inductance probe can adapt to the measuring environment of high pressure, the whole maximum withstand voltage of the probe including the measuring test piece can be 60Mpa, and the inductance probe can not be damaged after being used for a long time.

The fluid impact resistance and the bending resistance are strong, the measuring test piece is protected, the conducting wire and the welding terminal are fixed through the pouring sealant process, and the probe safety factor is prevented from being multiplied due to the fact that looseness and fracture occur due to vibration.

The thickness of the measuring test piece is increased to 1mm, the service life is prolonged, and the measuring result is more accurate by a temperature compensation and mathematical model double compensation method.

The measuring test piece is isolated from the compensating test piece, so that the compensating test piece is in an absolute stable state, and the probe cannot fail after a measured medium permeates into the inner space where the measuring test piece is located.

The measuring test piece and the probe rod are made of the same material, so that galvanic corrosion is avoided.

The compensation method of the invention obtains the measured resistance value of the probe through a high-precision instrument, summarizes and analyzes the measured resistance value and the theoretical thinning depth, sets a piecewise compensation function according to the nonlinear characteristic of a theoretical curve, develops a mathematical compensation model, and compensates through actual comparison, wherein the precision is higher than 1%.

The summarized actually measured resistance value and the theoretical thinning depth are compiled into a data comparison table, and corresponding numerical values in the formula are listed together, so that under an ideal state of uniform thinning, the measured value is absolutely objective and accurate, and after every measured actual resistance value is measured, the compensated actual numerical value can be displayed through a table look-up method, so that the method is convenient and reliable.

The method of the invention can enable the small-diameter pipeline to use the inductive probe.

The corrosion result can be accurately measured under the condition that the measuring test piece is thickened, and the service life and the safety factor of the probe are prolonged.

Detailed Description

Example one

The utility model provides a measure metal corrosion rate's high-pressure inductance probe, includes that six core glass seal screen boat inserts 1, transition pipe 4, probe pole 7, compensation test block 9 and measurement test block 10, its characterized in that:

the rear end of transition pipe 4 and six core glass seal screen are inserted 1 fixed connection by plane, and welded fastening behind the front end of transition pipe 4 and the 7 threaded connection of probe pole, the rear end of measuring test piece 10 and the front end fixed connection of probe pole 7, and the insulating encapsulation of compensation test piece 9 is in measuring test piece 10. One end of the lead 6 is connected with the six-core glass screen-sealed aerial plug 1, and the other end of the lead 6 is connected with the measurement test piece 10 and the compensation test piece 9.

The lead 6 is positioned in the transition tube 4 and the probe rod 7, and glue 5 is filled between the lead 6 and the inner walls of the transition tube 4 and the probe rod 7.

The measuring test piece 10 is sleeved with a protective cap 8, the rear end of the protective cap 8 is in threaded connection with the front end of the probe rod 7 and is fixed by a plurality of jackscrews, and a plurality of holes 11 are formed in the protective cap 8.

Example two

A thick sheet inductance probe compensation method for measuring pipeline corrosion comprises the following steps:

1. obtaining the actually measured resistance value of the test piece:

the inner diameter R of the test piece is 7mm, and the outer diameter R is 9.5-7.5mm, then:

δ 0.5- (R-R)/2 … … … … … … formula 2

s＝3.14*(R²-r²) … … … … … … formula 3

ρ_MeasuringL/s … … … … … … … formula 4

h

_max1000000 δ … … … … … formula 5

△ρ＝ρ_Measuring-ρ_{Supplement device}… … … … … … … formula 6

△＝Δρ/ρ_{Supplement device}… … … … … … … formula 7

h_{Measured in fact}△ x 262144 … … … … … … formula 8

Pilot symbol interpretation:

the outer diameter R (unit: mm) of the test piece, the inner diameter R (unit: mm) of the test piece, the thickness delta (unit: mm) of the test piece, the length L (unit: mm) of the test piece, the cross-sectional area s (unit: mm) of the test piece²) Maximum theoretical etch depth h_max(unit: nm), resistance value of test piece ρ_Measuring(unit: m.OMEGA.), compensation test piece resistance rho_{Supplement device}(unit: m omega), resistance difference △ rho, difference △, and actually measured resistance h_{Measured in fact}(unit: nm).

Thereby obtaining the actual measurement resistance of the measurement test piece.

2. And (4) measuring a theoretical value after adding a compensation algorithm.

Adding a compensation algorithm according to the calculated actually measured resistance value to obtain a measured theoretical value, wherein in the actual display process, visual data needed to be seen is the measured theoretical value, and the measured theoretical value h is assumed to be_{Theory of the invention}(Single sheet)Bit: nm), it can be seen from the comparison graph before and after compensation of the mathematical model (such as fig. 43), the measured theoretical value is not linearly changed, so the compensation algorithm adopts the segmented compensation according to the curve form.

h_{Theory of the invention}＝3.75*h_{Measured in fact}+65535，h_{Measured in fact}≤-52428

h_{Theory of the invention}＝2.5*h_{Measured in fact}，-52428＜h_{Measured in fact}＜0

h_{Theory of the invention}＝h_{Measured in fact}，h_{Measured in fact}≥0

3. And drawing a table.

According to the above calculation formula, the actually measured resistance value h can be obtained_{Measured in fact}Measuring the theoretical value h_{Theory of the invention}And some important intermediate variables, by which all data were calculated during the reduction of the outer diameter R from 9.5mm to 7.5mm, wherein the inner diameter R was constant at 7 mm. After the data is calculated, the data is drawn into a table as the basis of a table look-up method.

The following table is a comparison table of compensation values of the inductance probe with a part of thick sheet drawn by a compensation algorithm.

Long and long	79.06	79.06	79.06	79.06	79.06	79.06	79.06
								Outer diameter (Wide)	8.06	8.04	8.02	8.00	7.98	7.96	7.94
Inner diameter (thickness)	7.00	7.00	7.00	7.00	7.00	7.00	7.00
								UR(δ)	(0.030)	(0.020)	(0.010)	(0.000)	0.010	0.020	0.030
Cross sectional area	12.53	12.28	12.03	11.78	11.52	11.27	11.02
								Resistance (m omega)	6.31	6.44	6.57	6.71	6.88	7.01	7.17
Resistance value and	6.3751	6.5050	6.6400	6.7804	6.9266	7.0788	7.2376
								theoretical depth	(30000)	(20000)	(10000)	(0)	10000	20000	30000
Difference in resistance	(0.4053)	(0.2754)	(0.1404)	(0.0000)	0.1462	0.2985	0.4972
								Differential value	(0.0598)	(0.0406)	(0.0207)	(0.0000)	0.0216	0.0440	0.0674
18 bit	(15669)	(10648)	(5429)	(0)	5651	11539	17678
								Now measuring the theoretical value	(39172)	(26619)	(13572)	(0)	5651	11539	17678

Claims

1. A method for establishing a wellhead downhole corrosion rate correlation model is characterized by comprising the following steps:

first step, with CO₂The corrosion environment is a research object, and a six-factor five-level orthogonal experiment is adopted to research temperature and CO₂Influence degrees of partial pressure, total pressure, mineralization degree, oil-water ratio and flow rate on wellhead and underground corrosion rate;

secondly, correcting the model by using orthogonal experimental data based on the C.de Waard DM model, and respectively establishing a wellhead corrosion rate prediction model and an underground corrosion rate prediction model;

thirdly, the temperature and CO are measured₂Normalizing the three key corrosion influence factors of partial pressure and total pressure to temperature parameters, and respectively determining the functional relations between the wellhead corrosion rate and the downhole corrosion rate and the temperature;

fourthly, performing substitution calculation by using an empirical formula of temperature gradient existing between the temperature and the well depth to obtain a correlation model of the corrosion rate of the well mouth and the underground well;

the orthogonal experimental data in the first step are obtained according to the following experiments:

firstly, sealing the reaction kettle and checking the air tightness of the device;

secondly, simulating the preparation of produced liquid, and manufacturing according to the content of corrosive anion and cation components of the water quality of the measured well; thirdly, pre-treating the corrosion coupon;

fourthly, the simulated produced liquid is filled into the reaction kettle according to the experiment set value, and the heating device is started to raise the temperature to the preset temperature;

fifthly, connecting the inductance probe and the hanging piece with the reaction kettle, firstly filling nitrogen to a set value, and then filling CO₂To a stable pressure;

sixthly, connecting an inductance probe collector, starting data software and starting communication;

seventh, observe the situation of data reading and pressure change situation;

eighthly, after the experiment is maintained for 2-4 days, intercepting a data curve of the inductance probe, and taking the hanging piece out of the reaction kettle;

ninth, the inductive probe and the hanging piece are processed, corrosion products on the surfaces of the inductive probe and the hanging piece are taken out by using a cleaning agent, the inductive probe and the hanging piece are weighed after being dried, data are recorded, and the subsequent calculation is waited;

tenth, pumping the solution in the reaction kettle by a water pump, and recycling;

eleventh, analyzing the data and calculating a single set of corrosion rates;

twelfth, compiling a single group of experiment report, and analyzing photos before and after reaction, inductance probe corrosion data and coupon corrosion data;

the prediction model obtained in the second step is as follows:

n80 material well head Y-230.92 m⁶+1147.9m⁵-2170.7m⁴+1953.8m³-833.93m²+152.52m-7.6437；

N80 material Y-0.1795 m underground⁵+2.6596m⁴-13.787m³+28.627m²-19.178m+4.0761；

J55 material wellhead Y-124.09 x⁵+462.1x⁴-563.22x³+255.23x²-22.782x+0.429；

J55 material Y-0.2492 x under well⁵+3.7408x⁴-19.728x³+42.23x²-30.637x+7.0631；

The following relationships are obtained from the third and fourth steps:

relation 1: t is_Downhole＝T_{Well head}+0.035h；

Relation 2: y is_Downhole＝0.00007750T_Downhole ³-0.01784687T_Downhole ²+1.21903620T_Downhole-19.68881518；

And relation 3: t is_{Well head}＝59.755Y_{Well head} ⁰⁴⁷⁴⁴；

Calculating to obtain the underground corrosion rate Y_DownholeWith wellhead corrosion rate Y_{Well head}And the correlation of the downhole depth h is as follows:

relation 4: y is_Downhole＝0.00007750×(59.755Y_{Well head} ^0.4744+0.035h)³-0.01784687×(59.755Y_{Well head} ^0.4744+0.035h)²+1.21903620×(59.755Y_{Well head} ^0.4744+0.035h)-19.68881518。

2. The method for establishing the wellhead downhole corrosion rate correlation model according to claim 1, characterized by comprising the following steps: the influence degree obtained in the first step is as follows: 1) and the extreme difference analysis corrosion factor influence weight sequence is as follows:oil-water ratio > temperature > total pressure > CO₂Partial pressure > flow rate > degree of mineralization; 2) the most serious conditions of material corrosion are as follows: the temperature is 45 ℃, the degree of mineralization is 50000mg/L, the water-oil ratio is 90 percent, the total pressure is 6MPa, and CO is₂The partial pressure is saturated.

3. The downhole corrosion rate on-line monitoring method using the uphole downhole corrosion rate correlation model algorithm according to claim 1, comprising the steps of:

monitoring the corrosion rate of a wellhead on line in real time through an inductance probe;

and calculating the corrosion rates of different depths in the well by using the correlation model of the corrosion rates of the well head and the well, thereby knowing the corrosion condition of the underground equipment.

4. The method of claim 3 for online monitoring of downhole corrosion rate using the uphole downhole corrosion rate correlation model algorithm of claim 1, comprising the steps of:

5. The method of claim 3 for online monitoring of downhole corrosion rate using the uphole downhole corrosion rate correlation model algorithm of claim 1, comprising the steps of:

6. The method for online monitoring of downhole corrosion rate according to claim 4 using the uphole downhole corrosion rate correlation model algorithm as defined in claim 1, wherein the modification is performed according to the following modification model:

Y＝1.1877X-0.0013

in the above formula, Y is the corrosion rate monitored by the inductance probe, and X is the corrosion rate monitored by the coupon.