CN115062479A - Vertical well annulus rock debris concentration correction method based on Bayesian network - Google Patents
Vertical well annulus rock debris concentration correction method based on Bayesian network Download PDFInfo
- Publication number
- CN115062479A CN115062479A CN202210721312.6A CN202210721312A CN115062479A CN 115062479 A CN115062479 A CN 115062479A CN 202210721312 A CN202210721312 A CN 202210721312A CN 115062479 A CN115062479 A CN 115062479A
- Authority
- CN
- China
- Prior art keywords
- rock debris
- concentration
- annular
- equal
- less
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000011435 rock Substances 0.000 title claims abstract description 130
- 238000012937 correction Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000005553 drilling Methods 0.000 claims abstract description 34
- 238000004364 calculation method Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 13
- 239000012530 fluid Substances 0.000 claims description 26
- 238000005520 cutting process Methods 0.000 claims description 17
- 239000002245 particle Substances 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000012806 monitoring device Methods 0.000 claims description 4
- 230000035515 penetration Effects 0.000 claims description 4
- 229960001484 edetic acid Drugs 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 239000003344 environmental pollutant Substances 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 231100000719 pollutant Toxicity 0.000 description 2
- KCXVZYZYPLLWCC-UHFFFAOYSA-N EDTA Chemical compound OC(=O)CN(CC(O)=O)CCN(CC(O)=O)CC(O)=O KCXVZYZYPLLWCC-UHFFFAOYSA-N 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Earth Drilling (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Animal Husbandry (AREA)
- Mining & Mineral Resources (AREA)
- Agronomy & Crop Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Marine Sciences & Fisheries (AREA)
Abstract
The invention provides a Bayesian network-based vertical well annulus rock debris concentration correction method, which comprises the following steps of: step S1: training an error correction model; step S2: obtaining the influence factors of the current annular rock debris concentration; step S3: inputting the influence factors of the concentration of the annular rock debris into an error correction model after discrete processing to obtain an error predicted value of the concentration of the annular rock debris; step S4: and correcting the theoretical calculation value of the concentration of the annular rock debris by using the error predicted value of the concentration of the annular rock debris to obtain the corrected value of the concentration of the annular rock debris. The method can correct the theoretical calculation value of the concentration of the annular rock debris to achieve the purpose of reducing errors, and can repair data distortion or data loss of the concentration of the annular rock debris to ensure the safe operation of drilling operation.
Description
Technical Field
The invention relates to the technical field of drilling construction, in particular to a vertical well annulus rock debris concentration correction method based on a Bayesian network.
Background
The annular cuttings concentration is an important parameter affecting the drilling operation. The annular cuttings concentration refers to the concentration of cuttings contained in the drilling fluid in the annulus, expressed as a percentage. In drilling, if the content of the rock debris in the drilling fluid is greatly increased, a series of downhole complications are often caused. For example, the content of rock debris in the drilling fluid in the annulus is increased, the pressure of a liquid column is increased, and the drilling speed is reduced; the annular space is easily blocked up to the too big annular rock debris concentration, and equivalent density increases, and flow resistance increases, also increases to the pressure of the wall of a well, easily takes place the lost circulation. And a large amount of rock debris is easy to adhere to the well wall, so that the diameter is reduced, the drilling resistance is caused, various underground complex influences are generated, and even underground accidents occur.
The method for monitoring the concentration of the annular rock debris generally comprises two methods, wherein the first method is to obtain the theoretical rock debris concentration through a series of logging parameters, and the second method is to obtain the actual annular rock debris concentration through the volume of the returned rock debris and the discharge capacity of mud measured by a wellhead device. However, on one hand, the theoretical calculation model usually assumes that solid-liquid two phases in the annulus flow stably, the rock debris has the same size and shape, and the density of the drilling fluid and the hydraulic characteristics are unchanged, so that the calculation result only can approximately reflect the concentration of the rock debris in the annulus, and the accuracy is not good enough. On the other hand, the rock debris concentration measured by the rock debris return on-line monitoring device arranged on the wellhead can be relatively accurate to react the rock debris concentration in the annular space, but the method can be influenced by the delay time and cannot react the rock debris concentration in the well in real time.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a vertical well annulus rock debris concentration correction method based on a Bayesian network.
A vertical well annulus rock debris concentration correction method based on a Bayesian network comprises the following steps:
step S1: training an error correction model;
step S2: obtaining the influence factors of the current annular rock debris concentration;
step S3: inputting the influence factors of the concentration of the annular rock debris into an error correction model after discrete processing to obtain an error predicted value of the concentration of the annular rock debris;
step S4: and correcting the theoretical calculation value of the concentration of the annular rock debris by using the predicted value of the concentration error of the annular rock debris to obtain a corrected value of the concentration of the annular rock debris.
Further, in the method for correcting the concentration of the debris in the vertical well annulus based on the bayesian network, the step S4 includes:
and (3) the corrected value of the concentration of the annular rock debris is the theoretical calculated value of the concentration of the annular rock debris, namely the predicted value of the concentration error of the annular rock debris.
Further, in the method for correcting the concentration of cuttings in the vertical well annulus based on the bayesian network as described above, the step S1 includes:
s11, acquiring known sample data, wherein the known sample data is a set of items to be classified of a known class;
s12, carrying out discretization processing on the known sample data;
s13, building a naive Bayes classifier in an MATLAB environment, and training the built naive Bayes classifier according to known sample data after discretization to obtain an error correction model; the error correction model is used for obtaining the probability of each category appearing in the known sample data and the conditional probability estimation of each category by each characteristic attribute partition.
Further, in the method for correcting the concentration of the cuttings in the vertical well based on the bayesian network, in step S2, the influence factors of the concentration of the cuttings in the annulus include: the fluidity coefficient of the drilling fluid Y1, the consistency coefficient of the drilling fluid Y2, the annular back flow Y3, the rock debris particle size Y4 and the annular temperature Y5.
Further, in the method for correcting the concentration of the cuttings in the vertical well based on the bayesian network, the step S2 of performing discrete processing on the influence factors of the concentration of the cuttings in the annulus includes:
y1 was classified into 3 classes as follows: y11 is less than or equal to 0.3, Y12 is more than 0.3 and less than or equal to 0.4, and Y13 is more than 0.4; then Y1 ∈ { Y11, Y12, Y13 };
y2 was classified into 3 categories as follows: y21 is less than or equal to 0.2, Y22 is more than 0.2 and less than or equal to 0.3, and Y23 is more than 0.3; then Y1 ∈ { Y21, Y22, Y23 };
y3 was classified into 3 classes as follows: y31 is less than or equal to 1.5mm, Y32 is more than 1.5mm and less than or equal to 2.5mm, and Y33 is more than 2.5 mm; then Y1 ∈ { Y31, Y32, Y33 };
y4 was classified into 3 classes as follows: y41 is less than or equal to 1.5mm, Y42 is more than 1.5mm and less than or equal to 2.5mm, and Y43 is more than 2.5 mm; then Y1 ∈ { Y41, Y42, Y43 };
y5 was classified into 3 classes as follows: y51 is less than or equal to 30 ℃, Y52 is less than or equal to 60 ℃ at the temperature of 30 ℃, and Y53 is less than 60 ℃; then Y1 ∈ { Y51, Y52, Y53 }.
Further, according to the vertical well annulus rock debris concentration correction method based on the bayesian network, the annulus rock debris concentration deviation calculation method includes:
concentration deviation of annular rock debris equal to C a Actual monitored value
Wherein, the concentration of the annular rock debris C a The calculation method comprises the following steps:
in the formula:
C a -annular cuttings concentration; ROP-rate of penetration; d b -a drill diameter; d h -a borehole diameter; d p -a drill rod outer diameter; v a -drilling fluid annulus return velocity; v s -end rate of cuttings slip; and the actual monitoring value is acquired by a rock debris return online monitoring device installed at a wellhead.
Further, in the method for correcting the concentration of the cuttings in the vertical well based on the bayesian network, the discrete processing of the deviation of the concentration of the cuttings in the annulus in the step S2 includes:
dividing the annular rock debris concentration leakage difference D into 3 types according to the following modes: d1 is less than or equal to 10 thousandths, D2 is less than or equal to 20 thousandths, and D3 is more than 30 thousandths.
Has the beneficial effects that:
according to the Bayesian network-based vertical well annulus rock debris concentration correction method provided by the invention, the error prediction result of the influence factors on the annulus rock debris concentration is obtained through the error correction model, so that the error prediction result is more accurate and closer to the actual situation, and the actual value of the target pollutant concentration is calculated by using the error prediction result and the theoretical calculation value of the annulus rock debris concentration. By the method, the theoretical calculation value of the concentration of the annular rock debris can be corrected, the purpose of reducing errors is achieved, data distortion or data loss existing in the concentration of the annular rock debris can be repaired, and the safe drilling operation is guaranteed.
Drawings
Fig. 1 is a flowchart of a method for correcting the concentration of cuttings in a vertical well annulus based on a bayesian network according to an embodiment of the present invention;
FIG. 2 is a flow chart of the training of the error correction model.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention are described clearly and completely below, and it is obvious that the described embodiments are some, not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart of a method for correcting a rock debris concentration in a vertical well annulus based on a bayesian network according to an embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
step S01: training an error correction model;
step S02: and acquiring the influence factors of the current annular rock debris concentration.
Specifically, the influencing factors are the fluidity coefficient of the drilling fluid, the consistency coefficient of the drilling fluid, the annular back velocity, the rock debris particle size and the annular temperature.
Step S03: inputting the influence factors of the concentration of the annular rock debris into an error correction model after discrete processing to obtain an error predicted value of the concentration of the annular rock debris;
step S04: and correcting the theoretical calculation value of the concentration of the annular rock debris by using the error predicted value of the concentration of the annular rock debris to obtain the corrected value of the concentration of the annular rock debris.
According to the Bayesian network-based vertical well annulus rock debris concentration correction method provided by the invention, the error prediction result of the influence factors on the annulus rock debris concentration is obtained through the error correction model, so that the error prediction result is more accurate and closer to the actual situation, and the actual value of the target pollutant concentration is calculated by using the error prediction result and the theoretical calculation value of the annulus rock debris concentration. By the method, the theoretical calculation value of the concentration of the annular rock debris can be corrected, the purpose of reducing errors is achieved, data distortion or data loss existing in the concentration of the annular rock debris can be repaired, and the safe drilling operation is guaranteed.
How to train the error correction model in step S01 is explained below:
the step S01 includes the steps of:
s11, acquiring known sample data, wherein the known sample is a set of items to be classified of a known class;
s12, discretizing the known sample data
S13, building a naive Bayes classifier in an MATLAB environment, and training the built naive Bayes classifier according to known sample data after discretization to obtain an error correction model; the error correction model is used for obtaining the probability of each category appearing in the known sample data and the conditional probability estimation of each category by each characteristic attribute partition.
The process of building the model is a process of calculating probability, and the main formula of the model is as follows:
according to the Bayesian network-based vertical well annulus rock debris concentration correction method, the error correction model is obtained by building a naive Bayes classifier for training, and the trained error correction model can be obtained efficiently due to the simple logic of the Bayesian algorithm and the convenient training of the model.
Fig. 2 is a training flowchart of the error correction model, and as shown in fig. 2, the specific process of training the error correction model is as follows:
step 21: obtaining known sample data as training set
Specifically, the known sample data is composed of two parts, namely, parameters affecting the concentration of the annular rock debris and the deviation of the concentration of the annular rock debris corresponding to the parameters.
The parameters influencing the concentration of the rock debris in the annulus are the fluidity coefficient of the drilling fluid, the consistency coefficient of the drilling fluid, the annulus return speed, the particle size of the rock debris and the annulus temperature.
The fluidity coefficient of the drilling fluid is represented by Y1, the consistency coefficient of the drilling fluid is represented by Y2, the annular back velocity is represented by Y3, the particle size of rock debris is represented by Y4, and the annular temperature is represented by Y5.
Let Y (Y1, Y2, Y3, Y4, Y5, Y6, Y7) be an item to be classified, and Yi (i ═ 1,2, …, 7) be a characteristic attribute of Y.
The deviation of the concentration of the annular rock debris is the deviation of a theoretical calculation value of the concentration of the annular rock debris relative to an actual monitoring value.
Specifically, the deviation calculation method comprises the following steps:
concentration deviation of annular rock debris equal to C a -an actual monitored value;
further, the concentration of the annular rock debris C a The calculation method comprises the following steps:
in the formula:
C a -annular debris concentration
ROP-rate of penetration
D b Drill diameter
D h Borehole diameter
D p Outside diameter of drill rod
V a Drilling fluid annulus return velocity
V s End speed of rock debris slip
Furthermore, the actual monitoring value is obtained by a rock debris return-out online monitoring device installed at the wellhead.
One sample of the training set was formed from Y, D above { Y, D }.
Step 22: discretizing known sample data
Specifically, Y1 represents the fluidity coefficient of the drilling fluid, and Y1 is classified into 3 types as follows: y11 is less than or equal to 0.3, Y12 is more than 0.3 and less than or equal to 0.4, and Y13 is more than 0.4; then Y1 ∈ { Y11, Y12, Y13}
Y2 represents the consistency factor of the drilling fluid, and Y2 is classified into 3 categories as follows: y21 is less than or equal to 0.2, Y22 is more than 0.2 and less than or equal to 0.3, and Y23 is more than 0.3; then Y1 ∈ { Y21, Y22, Y23}
Y3 denotes annulus return velocity, and Y3 is classified into 3 types as follows: y31 is less than or equal to 1.5mm, Y32 is more than 1.5mm and less than or equal to 2.5mm, and Y33 is more than 2.5 mm; then Y1 ∈ { Y31, Y32, Y33}
Y4 represents the particle size of the rock debris, and Y4 was classified into 3 types as follows: y41 is less than or equal to 1.5mm, Y42 is more than 1.5mm and less than or equal to 2.5mm, and Y43 is more than 2.5 mm; then Y1 ∈ { Y41, Y42, Y43}
Y5 represents the annulus temperature, and Y5 is classified into 3 categories as follows: y51 is less than or equal to 30 ℃, Y52 is less than or equal to 60 ℃ at the temperature of 30 ℃, and Y53 is less than 60 ℃; then Y1 e { Y51, Y52, Y53}
D represents the concentration deviation of the annular rock debris, and D is classified into 3 types according to the following modes: d1 is less than or equal to 10 per thousand, D2 is less than or equal to 20 per thousand, and D3 is less than 30 per thousand; then D e { D1, D2, D3}
The discretization processing method can also be expressed as follows:
TABLE 1-1 Dispersion treatment
Step 23: and building a naive Bayes classifier in an MATLAB environment, and training the naive Bayes classifier by using known sample data after discretization.
The training process is to calculate the occurrence frequency of each class in the training sample and the conditional probability of each class divided by each characteristic attribute. The method specifically comprises the following steps:
and according to the characteristic values and the classes of the known samples, counting to obtain the conditional probability estimation of each characteristic attribute under each class. I.e. calculating P (D-D1), P (D-D2), P (D-D3); p (Y1 ═ Y11), P (Y1 ═ Y12), P (Y1 ═ Y13), P (Y2 ═ Y21), P (Y2 ═ Y22), P (Y2 ═ Y23), … …, P (Y5 ═ Y51), P (Y5 ═ Y52), P (Y5 ═ Y53); p (Y1| D ═ D1), P (Y2| D ═ D1), … …, P (Y5| D ═ D1); p (Y1| D ═ D2), P (Y2| D ═ D2), … …, P (Y5| D ═ D2); p (Y1| D ═ D3), P (Y2| D ═ D3), … …, P (Y5| D ═ D3).
According to the Bayesian network-based vertical well annulus rock debris concentration correction method, because various parameters influencing the annulus rock debris concentration and the annulus rock debris concentration deviation corresponding to the parameters are continuous values and are not suitable for a common Bayesian network, the method adopts the principle of equally dividing the occurrence frequency of the attribute values in each category, discretizes the related attributes according to the table, so that the subsequent calculation is more convenient, and the training efficiency of the model is improved.
Further, the step of inputting the influence factors of the concentration of the annular rock debris into an error correction model after discrete processing to obtain the error prediction value of the concentration of the annular rock debris comprises the following steps:
step 31: discretizing the influence factors of the concentration of the annular rock debris, wherein the processing mode is the same as the mode of processing the training sample, and specifically the following table is shown:
TABLE 1-1 Dispersion treatment
Step 32: inputting an item to be classified into an error correction model, namely the influence factor Y of the concentration of the annular rock debris is (Y1, Y2, Y3, Y4 and Y5), and the classification is carried out by Bayes theorem
P (Di | Y) is calculated respectively, where i is 1,2, 3.
Step 33: comparing the sizes of P (D-D1-Y), P (D-D2-Y), and P (D-D3-Y), obtaining a result P (Di-Y) max { P (D-D1-Y), P (D-D2-Y), and P (D-D3-Y), obtaining an error correction model output result Di, where i is 1,2, and 3.
Y ═ Y1, Y2, Y3, Y4, Y5 are one item to be classified, Y1, Y2, Y3, Y4, Y5 are the 5 characteristic attributes of Y.
By the Bayesian theorem, the method for the detection of the biological characteristic,
the posterior probability can be derived from the prior probability. Inputting an item to be classified into an error correction model, wherein the error correction model obtains an error correction model output result Di by calculating P (D-Y) ═ max { P (D-D1-Y), P (D-D2-Y) and P (D-D3-Y), and wherein i is 1,2 and 3.
The Bayesian network-based vertical well annulus rock debris concentration correction method provided by the invention has the following beneficial effects that through the steps 31-33:
(1) the method comprises the steps of classifying pre-divided annular rock debris concentration deviation values by a Bayesian classification method, carrying out decision classification according to the maximum posterior probability, judging the sample to be of the type with the maximum probability value, and correcting the theoretically calculated annular rock debris concentration by the predicted deviation value, so that errors caused by theoretical calculation can be reduced, and the accuracy is improved.
(2) The extracted characteristics comprise drilling fluid properties, annular return velocity, rock debris particle size and annular temperature characteristics, the influence of all factors on the annular rock debris concentration is comprehensively and systematically disclosed, the classification effect is good, and the annular rock debris concentration deviation under various working conditions can be classified and identified; wherein, the characteristics (fluidity coefficient and consistency coefficient) of the drilling fluid are adopted, and the influence of the rock carrying capacity of the drilling fluid on the concentration of the annular rock debris is considered; the characteristics of annular back velocity are adopted, and the influence of the flow velocity of the drilling fluid on the rock debris concentration in the annular space is considered; the particle size characteristics of the rock debris are adopted, and the influence of the geometrical properties of the rock debris on the concentration of the rock debris in the annular space is considered; by adopting the annular temperature characteristic, the property of the drilling fluid is influenced by considering the temperature, and the rock carrying capacity of the drilling fluid is changed, so that the concentration of rock debris in the annular space is influenced.
(3) The vertical well annular rock debris concentration correction method based on the Bayesian network is simple in structure, strong in anti-interference capability and good in robustness, provides a new method for improving the accuracy of the calculated value of the annular rock debris concentration, and has certain theoretical significance and practical value in the fields of well cleaning state research, well cleaning quantification realization, well wall instability early monitoring and the like.
Further, the obtaining of the corrected value of the concentration of the annular rock debris by using the error predicted value of the concentration of the annular rock debris to correct the theoretical calculated value of the concentration of the annular rock debris comprises:
the theoretical calculation method of the concentration of the annular rock debris comprises the following steps:
in the formula:
C a -annular cuttings concentration; ROP-rate of penetration; d b -a drill diameter; d h -a borehole diameter; d p -a drill rod outer diameter; v a -drilling fluid annulus return velocity; v s The end rate of cuttings slip.
Further, calculating a correction value C 'of the concentration of the annular rock debris' a ：
C′ a ＝C a -Di；
Wherein i is 1,2, 3; di represents the predicted value of the annular rock debris concentration error.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (7)
1. A vertical well annulus rock debris concentration correction method based on a Bayesian network is characterized by comprising the following steps:
step S1: training an error correction model;
step S2: obtaining the influence factors of the current annular rock debris concentration;
step S3: performing discrete processing on the influence factors of the concentration of the annular rock debris and inputting the influence factors into an error correction model to obtain an error predicted value of the concentration of the annular rock debris;
step S4: and correcting the theoretical calculation value of the concentration of the annular rock debris by using the predicted value of the concentration error of the annular rock debris to obtain a corrected value of the concentration of the annular rock debris.
2. The bayesian-network-based vertical well annulus rock debris concentration correction method according to claim 1, wherein the step S4 comprises:
and (3) the corrected value of the concentration of the annular rock debris is the theoretical calculated value of the concentration of the annular rock debris, namely the predicted value of the concentration error of the annular rock debris.
3. The bayesian network based vertical well annulus rock debris concentration correction method according to claim 1, wherein the step S1 comprises:
s11, acquiring known sample data, wherein the known sample data is a set of items to be classified of a known class;
s12, carrying out discretization processing on the known sample data;
s13, building a naive Bayes classifier in an MATLAB environment, and training the built naive Bayes classifier according to known sample data after discretization to obtain an error correction model; the error correction model is used for obtaining the probability of each category appearing in the known sample data and the conditional probability estimation of each category by each characteristic attribute partition.
4. The Bayesian network-based vertical well annulus rock debris concentration correction method according to claim 1, wherein in step S2, the influence factors of the annulus rock debris concentration include: the fluidity coefficient of the drilling fluid Y1, the consistency coefficient of the drilling fluid Y2, the annular back flow Y3, the rock debris particle size Y4 and the annular temperature Y5.
5. The Bayesian network-based vertical well annulus rock debris concentration correction method according to claim 4, wherein the discrete processing of the influence factors of the annulus rock debris concentration in the step S2 comprises:
y1 was classified into 3 classes as follows: y11 is less than or equal to 0.3, Y12 is more than 0.3 and less than or equal to 0.4, and Y13 is more than 0.4; then Y1 ∈ { Y11, Y12, Y13 };
y2 was classified into 3 classes as follows: y21 is less than or equal to 0.2, Y22 is more than 0.2 and less than or equal to 0.3, and Y23 is more than 0.3; then Y1 ∈ { Y21, Y22, Y23 };
y3 was classified into 3 categories as follows: y31 is less than or equal to 1.5mm, Y32 is more than 1.5mm and less than or equal to 2.5mm, and Y33 is more than 2.5 mm; then Y1 ∈ { Y31, Y32, Y33 };
y4 was classified into 3 classes as follows: y41 is less than or equal to 1.5mm, Y42 is more than 1.5mm and less than or equal to 2.5mm, and Y43 is more than 2.5 mm; then Y1 ∈ { Y41, Y42, Y43 };
y5 was classified into 3 classes as follows: y51 is less than or equal to 30 ℃, Y52 is less than or equal to 60 ℃ at the temperature of 30 ℃, and Y53 is less than 60 ℃; then Y1 ∈ { Y51, Y52, Y53 }.
6. The Bayesian network-based vertical well annulus rock debris concentration correction method according to claim 5, wherein the annulus rock debris concentration deviation calculation method comprises the following steps:
concentration deviation of annular rock debris equal to C a Actual monitored value
Wherein, the concentration of the annular rock debris C a The calculation method comprises the following steps:
in the formula:
C a -annular cuttings concentration; ROP-rate of penetration; d b -a drill diameter; d h Borehole diameter；D p -a drill rod outer diameter; v a -drilling fluid annulus return velocity; v s -end rate of cuttings slip; and the actual monitoring value is acquired by a rock debris return online monitoring device installed at a wellhead.
7. The Bayesian network-based vertical well annulus rock debris concentration correction method according to claim 6, wherein the discrete processing of the annulus rock debris concentration deviation in step S2 comprises:
dividing the annular rock debris concentration deviation D into 3 types according to the following modes: d1 is less than or equal to 10 per thousand, D2 is less than or equal to 20 per thousand, and D3 is greater than 30 per thousand.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210721312.6A CN115062479A (en) | 2022-06-24 | 2022-06-24 | Vertical well annulus rock debris concentration correction method based on Bayesian network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210721312.6A CN115062479A (en) | 2022-06-24 | 2022-06-24 | Vertical well annulus rock debris concentration correction method based on Bayesian network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115062479A true CN115062479A (en) | 2022-09-16 |
Family
ID=83202917
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210721312.6A Pending CN115062479A (en) | 2022-06-24 | 2022-06-24 | Vertical well annulus rock debris concentration correction method based on Bayesian network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115062479A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104950808A (en) * | 2015-07-20 | 2015-09-30 | 攀枝花学院 | Machine tool thermal error compensation method based on augmented naive Bayes network |
| CN112529240A (en) * | 2020-09-14 | 2021-03-19 | 桂林电子科技大学 | Prediction method, system, device and storage medium of atmospheric environment data |
| CN112529341A (en) * | 2021-02-09 | 2021-03-19 | 西南石油大学 | Drilling well leakage probability prediction method based on naive Bayesian algorithm |
| CN113266344A (en) * | 2021-05-21 | 2021-08-17 | 西南石油大学 | Method for predicting rock carrying efficiency of drilling fluid of horizontal well |
-
2022
- 2022-06-24 CN CN202210721312.6A patent/CN115062479A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104950808A (en) * | 2015-07-20 | 2015-09-30 | 攀枝花学院 | Machine tool thermal error compensation method based on augmented naive Bayes network |
| CN112529240A (en) * | 2020-09-14 | 2021-03-19 | 桂林电子科技大学 | Prediction method, system, device and storage medium of atmospheric environment data |
| CN112529341A (en) * | 2021-02-09 | 2021-03-19 | 西南石油大学 | Drilling well leakage probability prediction method based on naive Bayesian algorithm |
| CN113266344A (en) * | 2021-05-21 | 2021-08-17 | 西南石油大学 | Method for predicting rock carrying efficiency of drilling fluid of horizontal well |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108388921B (en) | Overflow loss real-time identification method based on random forest | |
| CN105089620B (en) | Monitoring system, the method and device of bit freezing | |
| Pournazari et al. | Enhanced kick detection with low-cost rig sensors through automated pattern recognition and real-time sensor calibration | |
| CN109829561B (en) | Accident prediction method based on smoothing processing and network model machine learning | |
| CN112561211A (en) | Drilling circulation pressure consumption prediction method based on neural network deep learning | |
| CN115238861A (en) | Safe drilling fluid density determination method based on borehole wall collapse degree constraint | |
| CN115062479A (en) | Vertical well annulus rock debris concentration correction method based on Bayesian network | |
| CN112990580A (en) | Drilling overflow early warning method and system based on Bayesian algorithm and storage medium | |
| CN112096374B (en) | Dynamic measurement error compensation method for measuring oil holdup by flow splitting method | |
| CN113685166B (en) | Drilling accident early warning method and system | |
| CN109763810A (en) | A kind of reservoir gas drilling risk recognition methods | |
| CN116065988A (en) | Wellbore monitoring method, device, equipment and medium | |
| CN114417536B (en) | Drilling parameter optimization method in oil and gas well drilling process | |
| CN113536691A (en) | Deepwater drilling parameter optimization method and system | |
| CN113868951A (en) | Method for intelligently judging optimal drilling short-start time | |
| CN112598061A (en) | Tunnel surrounding rock clustering and grading method | |
| US20240003246A1 (en) | Systems and methods for detecting abnormal flowback | |
| CN117927219A (en) | Wellbore stability determining method, device, equipment and medium | |
| US12025763B2 (en) | Multi-sensor data assimilation and predictive analytics for optimizing well operations | |
| Wen et al. | Intelligent prediction and key factor analysis to lost circulation from drilling data based on machine learning | |
| Sebbeh-Newton et al. | Prediction of rock mass rating using adaptive neuro-fuzzy inference system (ANFIS) for NATM-3 of Pahang-Selangor raw water transfer tunnel (PSRWT) project, Malaysia | |
| CN117932446A (en) | Drilling overflow identification and prediction method, device, equipment and medium | |
| Wang et al. | Downhole mud loss transient simulation and detection with downhole dual measurement points | |
| US20230236334A1 (en) | Multi-sensor data assimilation and predictive analytics for optimizing well operations | |
| US20220349301A1 (en) | Determining a risk of stuck pipes during well drilling operations |
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 |