CN113094910A - Method for determining main control factors of lost circulation without human intervention - Google Patents
Method for determining main control factors of lost circulation without human intervention Download PDFInfo
- Publication number
- CN113094910A CN113094910A CN202110409253.4A CN202110409253A CN113094910A CN 113094910 A CN113094910 A CN 113094910A CN 202110409253 A CN202110409253 A CN 202110409253A CN 113094910 A CN113094910 A CN 113094910A
- Authority
- CN
- China
- Prior art keywords
- factors
- equation
- factor
- main control
- lost circulation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Feedback Control In General (AREA)
Abstract
The method for determining the main control factors of the lost circulation without human intervention comprises the following steps: 1: determining model independent variables, 2: determining a lost circulation objective function, 3: establishing a mathematical relation equation, and substituting data into the equation to solve equation coefficients; 4: evaluating the reasonability of an equation by adopting a t test method and an F test method, 5: and sequentially screening the main control factors of the lost circulation by adopting a prime mover method, a contribution rate method and a functional method. The invention is based on field data, the element cutting method is not influenced by human subjective factors, and the result is more in line with the field reality; each factor is a quantitative parameter, has physical significance and can guide the adjustment of the field process performance.
Description
Technical Field
The invention belongs to the technical field of well drilling and completion, and particularly relates to a method for determining a lost circulation main control factor without human intervention.
Background
The lost circulation problem has become one of the bottlenecks that severely restrict oil and gas development and well completion efficiency. At present, the research on leakage mainly focuses on the aspects of a leakage stopping process, a leakage stopping system, leakage prediction and the like. The plugging process and the plugging system are the most studied contents in the field, usually take theoretical analysis and indoor experiments as the main points, firstly evaluate the formation characteristics theoretically, optimize the materials through the experiments, optimize the system formula and evaluate the plugging capability. The research mode is limited by the theoretical understanding of the stratum and the plugging material, is necessarily influenced by human subjective factors, can only be used for carrying out experiments within a certain condition range, and has the natural defect of incomplete consideration.
With the development of mathematical methods, new directions for predicting missed horizons and missed levels with mathematical algorithms have been developed in recent years. The Leson comprehensively identifies and predicts the leakage layer by analyzing geological characteristics of the leakage layer, logging technical response, leakage pressure diagnosis and other methods; and establishing a drilling fluid loss model and an invasion depth model, and respectively researching nonlinear relations among fracture width, drilling fluid loss rate and accumulated loss and influence factors of the invasion depth of the loss. A leakage fault tree constructed by Wanghai Biao analyzes the root cause and the influence factors of leakage, Li and the like predict the leakage occurrence by using ANN, SVM and RF data methods. AI-Hamededi and the like adopt a multiple linear regression method to predict the leakage; liu Biao and the like adopt a support vector regression method to predict the well leakage, Li Jian and the like judge the type of the well leakage based on the optimized ID3, Shexin and the like classify the leakage through a statistical method and correspondingly give out a leakage stopping measure.
The above-mentioned partial researches respectively predict the leakage from the angles of leakage layer position, leakage depth, leakage reason, leakage degree, etc., and no quantitative leakage main control factor which can be used for mathematical calculation is given. The other part belongs to a prediction model based on a neural network, more attention points are paid to the judgment of a leakage result, such as the leakage degree, and specific leakage main control factors are not given.
Disclosure of Invention
The invention aims to provide a method for determining the main control factor of the lost circulation without human intervention, which is based on field data, adopts a cell cutting method without being influenced by human subjective factors, and has a result more conforming to the actual field engineering; each factor in the model is a quantitative parameter, and the model has definite physical significance and has better guiding significance for field process performance adjustment.
The technical scheme adopted by the invention is as follows:
the method for determining the main control factors of the lost circulation without human intervention comprises the following steps:
step 1: determining model independent variables
Collecting geological, engineering and fluid data; and the data are collected and stored in a database or a table;
step 2: determining a lost circulation objective function
Selecting a parameter which has an original data record on site, can represent the lost circulation degree and can carry out quantitative calculation as a lost circulation objective function;
and step 3: mathematical relation equation establishment
A mathematical model between the objective function and the influencing factors (model independent variables) is established according to formula (1),
M＝a1K1+a2K2+……+anKn+b1Y1+b2Y2+……bnYn+c1Z1+c2Z2+……+cnZn+C0(1)
in the formula, a1, a2, … … an, b1, b2, … … bn, C1, C2 and … … cn are undetermined coefficients respectively, and C0 is a constant term; k1, K2 and … … Kn are geological data respectively, Y1, Y2 and … … Yn are engineering data respectively, and Z1, Z2 and … … Zn are fluid data respectively; m is a well leakage objective function; (formula (1) is a model general formula, and if the formula (1) contains a plurality of coefficients to be determined, the corresponding number of equations is needed, if k unknowns are totally available, k equations are needed, namely k known objective function values and k geological, engineering and fluid data)
Substituting the parameters collected in the step 1 and the step 2 into the formula (1), and solving the equation coefficient of the formula (1) by adopting a undetermined coefficient method; the formula in which the equation coefficients are solved is expressed as equation (1-0);
and 4, step 4: evaluation of equation rationality
The rationality of the evaluation equation comprises single-factor action credibility evaluation and multi-factor combined action credibility evaluation; and (3) adopting a t test method for evaluating the single-factor action reliability, and calculating t dispersion statistic according to a formula (2):
in the formula: x is the sample mean, μ is the population mean, σxIs the sample standard deviation, n is the sample capacity;
screening out factors with t deviation statistic larger than 40 according to the calculation result;
the credibility evaluation of the multi-factor combined action adopts an F test method, and the equation is considered to have good accuracy when the F test value is more than 90%;
and 5: well leakage master control factor screening
Sorting the terms on the right side of the equal sign of the equation (1-0) in the step 3 according to the sequence of the absolute values of the coefficients from large to small, recording as an equation (1-1), then removing the term with the maximum absolute value of the coefficient of the equation (1-1) (removing the factor with the maximum absolute value of the coefficient), recording as an equation (1-2), changing the coefficients of the factors on the right side of the equal sign of the equation (1-2) into unknown coefficients again, and recording as an equation (1-3); solving the unknown coefficients of all factors in the equation (1-3) by adopting a undetermined coefficient method, recording the unknown coefficients as an equation (1-4), sequencing the terms on the right side of the equal sign of the equation (1-4) according to the absolute value of the coefficients from large to small, recording the terms as an equation (1-5), if the sequence of all factors in the equation (1-5) is consistent with the sequence of the corresponding factors in the equation (1-2), indicating that the item removed in the equation (1-1) has little influence on the well leakage, namely is a non-main control factor, screening the factors, otherwise, the factors are main control factors and need to be reserved; then, the second largest factor is removed, and the process is repeated until all the main control factors remain (the method is called element-cutting screening).
Example (c):
y＝0.5x1+2.5x2+2x3-3x4+3.5 (1-0)
y＝-3x4+2.5x2+2x3+0.5x1+3.5 (1-1)
y＝2.5x2+2x3+0.5x1+3.5 (1-2)
y＝a1x2+a2x3+a3x1+c (1-3)
y＝2.2x2+5x3+8x1+3 (1-4)
y＝8x1+5x3+2.2x2+3 (1-5)
in this example, the order of the factors of equations (1-5) does not coincide with the order of the corresponding factors in equations (1-2), indicating that x is removed from equations (1-1)4Is a main control factor and needs to be reserved; then, the coefficient in equation (1-1) is given the second largest term (x)2) Removing, retaining x4、x3、x1The above-described method is continued (note: it is not necessary to compare whether the coefficient values of the corresponding factors are equal, but only to make the order uniform).
Further, the geological data in the step 1 comprises an X coordinate, a Y coordinate, well depth, vertical depth, well inclination angle, azimuth angle, porosity, permeability and lithology; the engineering data comprises pump pressure, rotating speed, drilling time, drilling pressure and displacement; fluid data includes funnel viscosity, initial shear force, final shear force,
Water loss amount of the drilling completion fluid, solid phase content, density of the drilling completion fluid and pH of the drilling completion fluid.
Further, the lost circulation objective function in step 2 is a loss amount or a loss rate.
Further, on the basis of the step 5, dividing the coefficient absolute value of each factor on the right side of the equation equal sign by the sum of all the coefficient absolute values to obtain the contribution rate of each factor to the target function; then sorting according to the sequence of the contribution rate from large to small; secondly, from the first factor with the largest contribution rate, the contribution rates of all the factors are accumulated one by one until the sum of the accumulated contributions exceeds 90%, and all the factors with the sum of the contribution rates larger than 90% are selected as main control factors (the method is called as the screening of the contribution rate method).
Furthermore, in the screened main control factors, the related factors are combined to form a representative parameter; the merging process comprises the following steps: firstly, selecting a representative parameter T11 according to the types and physical meanings of the factors T11, T12, T13, … … and T1n which are related to each other, and then sequentially superposing the contribution rates of the other factors T12, T13, … … and T1n to the target function on the representative parameter T11 (the method is called functional screening); similarly, according to the types and physical meanings of the factors T21, T22, T23, … … and T2w which are related to each other, selecting a representative parameter T21, and then sequentially superposing the contribution rates of the other factors T22, T23, … … and T2w to the target function on the representative parameter T21; similarly, other related factors Tm1, Tm2, Tm3, … … and Tmz are treated by the same method;
the superposition method comprises the following steps:
rewriting the equation (1-0) in step 3 with the undetermined coefficient into equation (3)
M＝k1*T11+k2*T21+……+km*Tm1+C (3)
Wherein T11, T21, … … and Tm1 represent m parameters respectively;
k1, k2 … … and km are coefficients with functional factors respectively, and satisfy the following relational expressions:
k1*T11＝k11*T11+k12*T12+k13*T13+……+k1n*T1n，
k2*T21＝k21*T21+k22*T22+k23*T23+……+k2w*T2w，
km*Tm1＝km1*Tm1+km2*Tm2+km3*Tm3+……+kmz*Tmz，
wherein the content of the first and second substances,
k11, k12, k13, … … k1n, k21, k22, k23, … … k2n, km1, km2, km3, … … kmn are respectively before being superposed
The contribution values of T11, T12, T13, … …, T1n, T21, T22, T23, … …, T2w, Tm1, Tm2, Tm3, … …, Tmz to the objective function;
c is a new constant term.
The invention has the beneficial effects that:
(1) the modeling process of the invention incorporates multi-dimensional data such as geology, well drilling and completion engineering, fluid and the like, and compared with the existing mathematical method, the modeling method has the advantages of large modeling data volume, more data types and small influence from experience, and is more in line with the characteristic of big data in the field of petroleum engineering.
(2) From the selection of a target function, the analysis, the collection and the processing of data, the establishment of a model to the screening of main control factors by adopting a prime mover method, a contribution rate method and a functional method without human intervention, the whole process is based on field data, the adopted core prime mover method is not influenced by human subjective factors, and the result is more in line with the actual field engineering; each main control factor determined in the model is a quantitative parameter, has definite physical significance and has better guiding significance for field process performance adjustment.
Detailed Description
This example is based on the on-site data of the northward oil field and specifically illustrates the method of the present invention.
The method for determining the main control factors of the lost circulation without human intervention comprises the following steps:
step 1: determining model independent variables
Respectively collecting geological, engineering and fluid data of each well; and the data are collected and stored in a database or a table;
the method requires that the selected independent variable parameters are not limited by artificial subjectivity, and when data are collected, the data integrity and quantitative calculation are taken as requirements, and all data measured on site are selected as much as possible. Independent variable parameters are divided into three major categories of geology, engineering and fluid, called layers, so as to ensure the integrity of data on the major categories. For each major class, a corresponding minor class, called a "dimension," of data records is found in the existing material.
Determining factors possibly related to the well leakage in the northward area according to the collected field data of the northward oil field: (1) geological aspect: x coordinate, Y coordinate, well depth, vertical depth, well inclination angle, azimuth angle, porosity, permeability and lithology; (2) in the aspect of engineering: pump pressure, rotation speed, drilling time, drilling pressure and displacement; (3) fluid aspect: the viscosity, initial shear force, final shear force of the funnel,
Water loss and solid of drilling and completion fluidThe phase content, the drilling completion fluid density, the drilling completion fluid pH and the like, totaling 27 items.
Step 2: determining a lost circulation objective function
The objective function value is a parameter used to characterize the extent of lost circulation and may be referred to as the dependent variable, i.e., the left side of the equation. Whether the objective function is correct or not is related to the correctness and objectivity of the model result. The leakage rate reflects the whole process of the well leakage in the operation, not only can reflect the leakage severity, but also can establish a real-time relation with engineering parameters and fluid performance, not only can be quantized, and original data can be easily obtained.
In the collected multiple wells, the leakage rate data is found in 29 wells; and the above-mentioned independent variable is 27 terms, show that 27 objective function values can solve the model. Therefore, the leakage rate can meet the physical significance of representing the well leakage and meet the modeling requirement in quantity, so the leakage rate is selected as a target function.
And step 3: mathematical relation equation establishment
The method comprises the steps of selecting a multiple regression mathematical method to establish a model; the mathematical relation equation established by the multiple regression method has an intuitive expression, and the respective variable parameters have definite physical meanings; meanwhile, the element cutting and the determination of the contribution rate of each parameter are convenient. The mathematical relationship between the leak-off rate and the influencing factors established by the method is shown in equation (1).
V(t)＝a1X1+...+anXn+b1Y1+...+bnYn+c1Z1+...+cnZn+C0
(1)
In the formula, X1......Xn，Y1......Yn，Z1......Zn-data representing three layers of geological, engineering and fluid properties, respectively; a is1......an，b1......bn，c1......cnDenotes the undetermined coefficient, C0Representing a constant term.
And selecting data of 27 wells from 29 wells with the recorded leakage rate, substituting the data into the equation (1), and solving the equation coefficient by adopting a undetermined coefficient method to obtain an equation (2). The remaining 2 wells were used for model testing.
V(t)＝131.941+37.936W1+32.468W2+19.715W3+6.852W4+4.717W5+2.285W6+2.066W7+0.318W8+0.318W9+0.284W10+0.114W11+0.103W12+0.090W13+0.046*W14+0.040W15+0.017W16+0.005W17+0.000W18+0.000W19-0.001W20-0.003W21-0.008W22-0.042W23-0.161W24-0.167W25-0.434W26-2.4W27 (2)
And 4, step 4: evaluation of equation rationality
The rationality of the evaluation equation comprises single-factor action reliability evaluation and multi-factor joint action reliability evaluation. The single-factor action reliability evaluation adopts a t-test method, and the larger the value is, the more inaccurate the information provided by the factor is; and F test method is adopted for evaluating the credibility of the multi-factor combined action, the larger the value is, the higher the model accuracy is, and the higher the credibility of the combined action of all factors on the target prediction is. According to the test method, the correlation factor data is substituted into a t dispersion statistic formula (3) for calculation.
In the formula: x is the sample mean, μ is the population mean, σxIs the sample standard deviation and n is the sample volume.
The calculation result shows that most of the t test values of the single factors are less than 40 and relatively small; the factors exceeding 40 are: phi 200-41.503, drilling completion fluid density-632.953, drilling completion fluid water loss-124.587 and displacement-94.242.
In contrast, the values of the drilling completion fluid density, the drilling completion fluid water loss and the displacement volume ttest are too large, and according to the ttest standard, the three factors need to be further eliminated.
Through calculation, the F test value of the leakage rate model is 96.557% and is larger than 95%, the F test value is larger, according to the F test standard, the significance of each factor and the target function is obvious, and therefore the equation has accuracy and can be used for screening the next main control factor.
And 5: well leakage master control factor screening
In the embodiment, the main control factor screening is divided into three progressive levels, and non-main control factors are eliminated by adopting a non-human intervention element cutting method; then, from the rest factors, the factors with the sum of the contribution rate of each factor to the leakage loss larger than 90 percent are preferably selected, namely the final main control factor; finally, in order to facilitate field application, a functional idea is adopted, and factors with the association relation in the main control factors are combined into a representative factor, which is called functional screening.
5.1 screening by the chipping method
From the 27 factors, 17 factors were selected by the above method of element reduction, see formula (4).
V(t)＝0.5574*X1+2.2605*X2+0.6678*X3+0.1783*X4+0.5876*X5+4.3681*X6+0.1184*X7-3.1583*X8-0.1017*X9-0.4847*X10-7.5184*X11-0.8003*X12-4.2287*X13-0.4192*X14-0.6284*X15-1.6597*X16-0.5639*X17+0.0913 (4)
Wherein X1 is the drilling completion fluid pH; x2 is Pump pressure; x3 is vertical depth; x4 is well depth; x5 is the rotation speed; x6 is as drilled; x7 is weight on bit; x8 is funnel viscosity; x9 is initial shear force; x10 is the final shear force; x11 is
X12 is
X13 is
X14 is
X15 is
X16 is
X17 is the Y coordinate.
The element cutting method eliminates all non-main control factors, but the factors are still too many, and the practical application operability is poor. There is a continuing need to reduce the number of factors.
5.2 screening by contribution ratio method
The contribution rate method is that a single factor coefficient is divided by the sum of all factor coefficients to obtain the contribution rate of the single factor to the leakage rate, then the contribution rates are sequenced from large to small, and all factors with the sum of the contribution rates larger than 90% are selected as main control factors. The method is used for further selecting 13 main control factors influencing the loss rate from the 17 influencing factors.
TABLE 1 leak Rate 13 contribution of the master factors (from big to small)
Of the 13 main factors, the highest contribution to leak-off rate was 29.94% while drilling, followed by a funnel viscosity of 16.84%, and the other factors contributed from 9.40% on the Y-coordinate to 1.67% of the initial force. The first two factors have larger influence, indicating that they are dominant factors and are both external factors. The contribution rate of the remaining factors is relatively low, but the influence is not negligible.
Although the contribution rate method has screened out the main control factors which greatly contribute to the leakage rate, the observation shows that some factors have close relationship, such as the viscosity of the funnel, phi 600 and phi 300, and the former two parameters are changed correspondingly. Therefore, it is necessary to combine the related factors into a representative parameter from the perspective of the oil engineering expertise, so as to facilitate the field application.
5.3 functional screening
The 13 main control factors are divided into three types according to types and physical meanings, the characteristics of each type of factors are respectively analyzed, a representative parameter is further selected for each type, and the contribution rate of other related parameters to the loss rate is superposed on the representative parameter. The method has similar idea to the functional method, so the method is called functional method.
(1) The funnel viscosity T11, phi 600T 12 and phi 300T 13 values are used for representing the rheological property of the drilling and completion fluid and are negatively related to the loss; when the former is changed, the latter two will be changed correspondingly. Therefore, funnel viscosity is taken as a rheological representative value, and the contribution rates of Φ 600 and Φ 300 to the leak-off rate are superimposed on funnel viscosity. The funnel viscosity after the superposition treatment completely keeps the contribution values of the original three parameters to the leakage rate, and the essential error of the result can not be caused; in practical application, only the viscosity of the funnel needs to be input, and the calculation is convenient.
(2) The initial shear force T32 and the final shear force T31 belong to time-dependent rheological parameters, the final shear force is selected as a representative parameter, and the contribution rate of the initial shear force is superposed with the representative parameter.
(3) The drilling time T21 represents the comprehensive result performance generated by certain fluid performance and engineering conditions under the formation conditions, so the drilling time T21 represents the drilling pressure T23, the pump pressure T22 and the rotating speed T25, the contribution rates of the three parameters to the leakage rate are superposed on the drilling time, and meanwhile, the contribution rate of the drilling fluid pH value T24 to the leakage rate is reduced.
(4) The vertical depth T42, the well depth T43 and the Y coordinate T41 belong to geological structure factors, and the factors are constant values for specific construction objects, so that a constant term C is included.
The final leak-off rate equation becomes equation (5) by functional screening:
V(t)＝k1*T11+k2*T21+k3*T31+C (5)
wherein T11 is funnel viscosity, T21 is at drill time, T31 is final shear; k1, k2, and k3 are coefficients with functional factors, and C is a new constant term.
From equation (5), the final leak-off rate model has only three representative parameters, funnel viscosity, time-to-drill, and final shear. k1, k2 and k3 are directly multiplied by actual values of respective representative parameters, but the values of the associated factors are different, such as the average value of the funnel viscosity is 60 s; while an average value of phi 300 of 36 is obviously not suitable if the contribution rate of phi 300 is directly superimposed on the coefficient k1 and multiplied by the funnel viscosity value. Therefore, a new variable is defined, called scaling factor λ, which is equal to the actual value of the representative factor divided by the actual value of the representative factor, independent of the unit dimension. The above idea of superposition is satisfied when the contribution ratio is multiplied by the scaling factor and then superposed on k 1. The scaling factors for all the represented factors were calculated and are shown in table 2.
TABLE 2 proportionality coefficients of the represented factors
The values of k1, k2, and k3 were calculated using the contribution ratios calculated in table 1 and the scaling factor data in table 2, respectively, as:
k1＝-(a1*λ1+a2*λ2+a3)＝-(6.66％*0.917+3.25％*0.6+20.84％)＝-0.28897
k3＝-(a4*λ3+a5)＝-(1.67％*0.419+3.50％)＝-0.04199
k2＝a6*λ4+a7*λ5+a8*λ6-a9*λ7+a10
＝9％*0.645+4.59％*0.6123+1.93％*0.677-2.22％*0.274+29.94％＝0.39254
the new constant term C is the sum of the products of Y coordinate, well depth and vertical depth and their respective contribution rates, and is added with the original constant term to obtain the value equal to 38.6.
Wherein, λ 1- λ 7 are respectively phi 600, phi 300, initial cutting pressure, pump pressure, drill pressure, rotation speed, and proportional coefficient of drilling completion fluid pH, a 1-a 10 are respectively phi 600, phi 300, funnel viscosity, initial cutting pressure, final cutting pressure, pump pressure, drill pressure, rotation speed, drilling completion fluid pH, and contribution rate value during drilling. Equation (5) becomes the final lost circulation model equation (6):
V(t)＝-0.28897*T11+0.39254*T21-0.04199*T31+38.6 (6)
example verification
The two inspection wells SHB1-3 and SHB1-15 completely record the leakage rate data of the leakage layer segment and have all the data requirements required by the model. And (5) substituting the data into a mathematical relation equation (6) of the leakage rate to calculate a theoretical value of the model, and obtaining a result shown in table 3.
TABLE 3 theoretical value calculation results of the model
The results in Table 3 show that the calculated values and the actual values of the leak-off rates of the two wells have a difference of-9.036 m3/h、-2.374m3The coincidence rate is relatively good. Indicating that it can be used for field prediction.
Claims (5)
1. The method for determining the main control factors of the lost circulation without human intervention comprises the following steps:
step 1: determining model independent variables
Collecting geological, engineering and fluid data; and the data are collected and stored in a database or a table;
step 2: determining a lost circulation objective function
Selecting a parameter which has an original data record on site, can represent the lost circulation degree and can carry out quantitative calculation as a lost circulation objective function;
and step 3: mathematical relation equation establishment
Establishing a mathematical model between the objective function and the influencing factors according to the formula (1),
M＝a1K1+a2K2+……+anKn+b1Y1+b2Y2+……bnYn+c1Z1+c2Z2+……+cnZn+C0(1)
in the formula, a1, a2, … … an, b1, b2, … … bn, C1, C2 and … … cn are undetermined coefficients respectively, and C0 is a constant term; k1, K2 and … … Kn are geological data respectively, Y1, Y2 and … … Yn are engineering data respectively, and Z1, Z2 and … … Zn are fluid data respectively; m is a well leakage objective function;
substituting the parameters collected in the step 1 and the step 2 into the formula (1), and solving the equation coefficient of the formula (1) by adopting a undetermined coefficient method; the formula in which the equation coefficients are solved is expressed as equation (1-0);
and 4, step 4: evaluation of equation rationality
The rationality of the evaluation equation comprises single-factor action credibility evaluation and multi-factor combined action credibility evaluation; and (3) adopting a t test method for evaluating the single-factor action reliability, and calculating t dispersion statistic according to a formula (2):
in the formula: x is the sample mean, μ is the population mean, σxIs the sample standard deviation, n is the sample capacity;
screening out factors with t deviation statistic larger than 40 according to the calculation result;
the credibility evaluation of the multi-factor combined action adopts an F test method, and the equation is considered to have good accuracy when the F test value is more than 90%;
and 5: well leakage master control factor screening
Sorting the terms on the right side of the equal sign of the equation (1-0) in the step 3 according to the sequence of the absolute values of the coefficients from large to small, recording as an equation (1-1), then removing the term with the maximum absolute value of the coefficient of the equation (1-1), recording as an equation (1-2), changing the coefficients of the factors on the right side of the equal sign of the equation (1-2) into unknown coefficients again, and recording as an equation (1-3); solving the unknown coefficients of all factors in the equation (1-3) by adopting a undetermined coefficient method, recording the unknown coefficients as an equation (1-4), sequencing the terms on the right side of the equal sign of the equation (1-4) according to the absolute value of the coefficients from large to small, recording the terms as an equation (1-5), if the sequence of all factors in the equation (1-5) is consistent with the sequence of the corresponding factors in the equation (1-2), indicating that the item removed in the equation (1-1) has little influence on the well leakage, namely is a non-main control factor, screening the factors, otherwise, the factors are main control factors and need to be reserved; and then removing the second largest factor, and repeating the steps until all the main control factors are remained.
2. The method for determining lost circulation primary factors without human intervention of claim 1, wherein the geological data of step 1 comprises X-coordinate, Y-coordinate, well depth, vertical depth, well angle, azimuth, porosity, permeability, lithology; engineering dataComprises pump pressure, rotating speed, drilling time, drilling pressure and displacement; fluid data includes funnel viscosity, initial shear force, final shear force,
Water loss amount of the drilling completion fluid, solid phase content, density of the drilling completion fluid and pH of the drilling completion fluid.
3. The method for determining the lost circulation main control factor without human intervention of claim 1, wherein the lost circulation objective function in the step 2 is a loss amount or a loss rate.
4. The method for determining the main control factor of the lost circulation without human intervention as claimed in claim 1, wherein on the basis of step 5, the coefficient absolute value of each factor on the right side of the equation equal sign is divided by the sum of all the coefficient absolute values to obtain the contribution rate of each factor to the objective function; then sorting according to the sequence of the contribution rate from large to small; and secondly, accumulating the contribution rates of all the factors one by one from the first factor with the largest contribution rate until the accumulated sum exceeds 90%, and selecting all the factors with the sum of the contribution rates being more than 90% as the main control factor.
5. The method for determining lost circulation primary factors without human intervention of claim 4, wherein the selected primary factors are combined into a representative parameter;
the merging process comprises the following steps: firstly, selecting a representative parameter T11 according to the types and physical meanings of factors T11, T12, T13, … … and T1n which have mutual correlation, and then sequentially superposing the contribution rates of the other factors T12, T13, … … and T1n to an objective function on the representative parameter T11; similarly, according to the types and physical meanings of the factors T21, T22, T23, … … and T2w which are related to each other, selecting a representative parameter T21, and then sequentially superposing the contribution rates of the other factors T22, T23, … … and T2w to the target function on the representative parameter T21; similarly, other related factors Tm1, Tm2, Tm3, … … and Tmz are treated by the same method;
the superposition method comprises the following steps:
rewriting the equation (1-0) in step 3 with the undetermined coefficient into equation (3)
M＝k1*T11+k2*T21+……+km*Tm1+C (3)
Wherein T11, T21, … … and Tm1 represent m parameters respectively;
k1, k2 … … and km are coefficients with functional factors respectively, and satisfy the following relational expressions:
k1*T11＝k11*T11+k12*T12+k13*T13+……+k1n*T1n，
k2*T21＝k21*T21+k22*T22+k23*T23+……+k2w*T2w，
km*Tm1＝km1*Tm1+km2*Tm2+km3*Tm3+……+kmz*Tmz，
wherein the content of the first and second substances,
k11, k12, k13, … … k1n, k21, k22, k23, … … k2n, km1, km2, km3, … … kmn are respectively before being superposed
The contribution values of T11, T12, T13, … …, T1n, T21, T22, T23, … …, T2w, Tm1, Tm2, Tm3, … …, Tmz to the objective function;
c is a new constant term.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110409253.4A CN113094910B (en) | 2021-04-16 | 2021-04-16 | Method for determining lost circulation main control factor by unmanned intervention method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110409253.4A CN113094910B (en) | 2021-04-16 | 2021-04-16 | Method for determining lost circulation main control factor by unmanned intervention method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113094910A true CN113094910A (en) | 2021-07-09 |
| CN113094910B CN113094910B (en) | 2023-05-09 |
Family
ID=76678101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110409253.4A Active CN113094910B (en) | 2021-04-16 | 2021-04-16 | Method for determining lost circulation main control factor by unmanned intervention method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113094910B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160370274A1 (en) * | 2015-02-20 | 2016-12-22 | Halliburton Energy Services, Inc. | Classifying particle size and shape distribution in drilling fluids |
| CN112561144A (en) * | 2020-12-07 | 2021-03-26 | 西安石油大学 | Method for evaluating main control factor of productivity of tight oil fracturing horizontal well and predicting productivity |
-
2021
- 2021-04-16 CN CN202110409253.4A patent/CN113094910B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160370274A1 (en) * | 2015-02-20 | 2016-12-22 | Halliburton Energy Services, Inc. | Classifying particle size and shape distribution in drilling fluids |
| CN112561144A (en) * | 2020-12-07 | 2021-03-26 | 西安石油大学 | Method for evaluating main control factor of productivity of tight oil fracturing horizontal well and predicting productivity |
Non-Patent Citations (5)
| Title |
|---|
| YAN ZHANG ET AL: "The environment friendliness of fuzzy-ball drilling fluids during their entire life-cycles for fragile ecosystems in coalbed methane well plants", 《JOURNAL OF HAZARDOUS MATERIALS》 * |
| 张生军;刘伟;李丽;潘登雷;王超;: "元坝超深井钻井成本影响因素分析", 内江科技 * |
| 李松;康毅力;李大奇;张杜杰;周长林;: "复杂地层钻井液漏失诊断技术系统构建", 钻井液与完井液 * |
| 肖剑;任胜利;王伟;余林锋;徐同台;: "数据科学在钻井工程防漏堵漏技术中的应用", 化工管理 * |
| 金衍;卢运虎;李再均;: "一种井漏层位钻前风险预测新方法", 石油钻采工艺 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113094910B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2951720B1 (en) | Production analysis and/or forecasting methods, apparatus, and systems | |
| CA2640725C (en) | Methods, systems, and computer-readable media for fast updating of oil and gas field production models with physical and proxy simulators | |
| RU2474846C2 (en) | Method and apparatus for multidimensional data analysis to identify rock heterogeneity | |
| CA2469957C (en) | Method to assist with decision making for the management of an oil deposit in the face of uncertain technical and economic parameters | |
| CN104533400B (en) | Method for reconstructing logging curve | |
| He et al. | Proxy-based work flow for a priori evaluation of data-acquisition programs | |
| CN107451325A (en) | Deep ＆ ultra-deep well pressure break casing failure risk real-time quantitative appraisal procedure and device | |
| WO2020215850A1 (en) | Method and system for measuring composition and property of formation fluid | |
| CN104899411B (en) | A kind of reservoir productivity prediction model method for building up and system | |
| US8788252B2 (en) | Multi-well time-lapse nodal analysis of transient production systems | |
| CN113094924B (en) | Optimization method and system for hydraulic jet fracturing construction parameters | |
| CN114444823A (en) | Method and system for predicting lost circulation risk in drilling operation process | |
| Tavares et al. | Combined effect of non-Darcy flow and formation damage on gas-well performance of dual-porosity and dual-permeability reservoirs | |
| de Holanda et al. | Probabilistically mapping well performance in unconventional reservoirs with a physics-based decline-curve model | |
| CN110671092A (en) | Oil gas productivity detection method and system | |
| US20130226672A1 (en) | Production by actual loss allocation | |
| Han et al. | Technical comparisons of simulation-based productivity prediction methodologies by means of estimation tools focusing on conventional earthmovings | |
| CN113094910A (en) | Method for determining main control factors of lost circulation without human intervention | |
| CA2760533C (en) | Method for building a flow pattern associated with capillary pressure curves and/or relative permeability curves | |
| CN112859192A (en) | Rock core saturation prediction model construction method and rock core saturation prediction method | |
| Aylor Jr | Measuring the impact of 3-D seismic on business performance | |
| CN110119522A (en) | A kind of stability ranking method excavated rock side slope and destroy risk analysis | |
| CN108647417B (en) | Simple method for determining gas saturation of shale gas reservoir | |
| CN117648573A (en) | Stratum drillability prediction method based on element logging data and comprehensive logging data | |
| CN110020398B (en) | Method and device for calculating water saturation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |