A method for calculating a probability of a well deviation degree distribution

Abstract

The application discloses a method for calculating well-seismic deviation degree distribution probability, which comprises the following specific steps: step one, establishing a drilling seismic source data matrix of horizontal wells in a research block; step two, recombining and expanding the drilling seismic source data matrix into five matrices with the same size, i.e., an expanded and recombined horizontal coordinate matrix, an expanded and recombined vertical coordinate matrix, an expanded and recombined drilling distance matrix, an expanded and recombined point depth matrix and an expanded and recombined layer position matrix; step three, extracting a well-seismic deviation degree matrix; step four, establishing a score anchoring relationship matrix and determining an anchoring score; and step five, calculating a deviation degree distribution probability matrix. The application extracts well-seismic deviation degree by jointly processing the actually drilled trajectory of the drilled horizontal well in the research block and the depth domain seismic interpretation data, and makes a probability inference, thereby providing a feasible solution for obtaining the probability information of the well-seismic data deviation degree distribution.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas exploration and development, and particularly relates to a method for calculating the distribution probability of well-seismic deviation degree. BACKGROUND

[0002] At present, in the field construction of shale oil and gas horizontal wells, seismic data has been widely applied to the guiding process of horizontal wells. The processing and interpretation results of seismic data provide lateral information of the horizon, and also provide source data support for three-dimensional geological modeling in geosteering. In the process of horizontal well drilling, there is a certain probability that the interpretation results of seismic data deviate from the actual drilling of the target layer, and not all the process is consistent.

[0003] Chinese patent document with publication number CN109960897A and publication date of July 2, 2019 discloses a shale oil horizontal well trajectory design and field tracking adjustment research method, which includes shale oil sweet spot identification and evaluation, horizontal well optimization and trajectory design, shale oil horizontal well field tracking adjustment, and provides a set of technical methods for horizontal well site optimization design research and drilling dynamic tracking analysis in shale oil development area.

[0004] The above-mentioned CN109960897A patent document is a representative of the technical invention, which focuses on the design of the horizontal well guiding process and the adjustment of the well trajectory, and cannot solve the current problem that the present application should deal with, that is, how to find the deviation between the actual drilling trajectory of the drilled horizontal well in the studied block and the seismic interpretation horizon, and further understand the probability size and distribution of the deviation. SUMMARY

[0005] The purpose of the present application is to provide a method for calculating the distribution probability of well-seismic deviation degree, which solves the problem of how to find the deviation between the actual drilling trajectory of the drilled horizontal well in the studied block and the seismic interpretation horizon, and further understand the probability size and distribution of the deviation. This method jointly processes the actual drilling trajectory of the drilled horizontal well in the studied block and the depth domain seismic interpretation data, extracts the well-seismic deviation degree for probability inference, and provides a set of solutions for obtaining the probability information of the well-seismic data deviation degree distribution.

[0006] In order to achieve the above-mentioned purpose, the present application provides a technical scheme of a method for calculating the distribution probability of well-seismic deviation degree as follows:

[0007] A method for calculating the distribution probability of well-seismic deviation degree, comprising the following specific steps:

[0008] Step 1: Establishing the drilling-seismic source data matrix of the horizontal well in the studied block M S .

[0009] Step one instruction 1: the drilling seismic source data matrix has 5 columns, the first column is composed of the horizontal coordinates of each point of the drilled horizontal well trajectory, the second column is composed of the vertical coordinates of each point of the drilled horizontal well trajectory, the third column is composed of the horizontal section lengths of each point of the drilled horizontal well trajectory, the fourth column is composed of the depths of each point of the drilled horizontal well trajectory, and the fifth column is composed of the depth values of the corresponding depth domain structure interpretation horizon of each point of the drilled horizontal well trajectory.

[0010] Step one instruction 2: each row of data in the drilling seismic source data matrix is in one-to-one correspondence.

[0011] In the above technical solution, in step one, the matrix is a concept defined in mathematics.

[0012] Step two: the drilling seismic source data matrix M S is recombined and expanded into five matrices of the same size: the expanded and recombined horizontal coordinate matrix M X , the expanded and recombined vertical coordinate matrix M Y , the expanded and recombined drilling distance matrix M L , the expanded and recombined point depth matrix M D , and the expanded and recombined horizon matrix M H . The step two includes five sub-steps:

[0013] 2.1 Calculate the point distance vector V D1 .

[0014] Step 2.1 instruction 1: the vector V D1 is a column vector, the element number of the vector V D1 is equal to the row number of the drilling seismic source data matrix M S , and the first element value of the vector V D1 is 0.

[0015] Step 2.1 instruction 2: the calculation method of the i-th element (i is an integer not equal to 1) of the vector V D1 is as follows: taking the first and second columns of the drilling seismic source data matrix M S as a new matrix, denoted as matrix M 1 , transposing the matrix M 1 , and obtaining the matrix M 1The ith column elements minus the matrix M 1 The ith-1 column elements, the resulting column vector is reoriented by vector modulus operation, and the value is the vector V D1 The ith element value.

[0016] 2.2 Determine the augmented recombination matrix by calculation M X , M Y , M L , M D , M H The number of rows n .

[0017] Step 2.2 Explanation 1: If the eighth of the minimum value of the elements in the vector V D1 is not less than 100, the number of rows n is 100.

[0018] Step 2.2 Explanation 2: If the eighth of the minimum value of the elements in the vector V D1 is less than 100, divide the minimum value of the vector V D1 by 16, and round it to an integer, and the resulting value is the starting term value, and construct an arithmetic sequence with a tolerance of 2 and a termination term value not exceeding 100 S 1 ; if the starting point of the sequence is less than 1, set the starting point to 1. Take each term of the sequence S 1 as the group interval of the statistical frequency distribution, respectively find the grouping point value corresponding to the maximum frequency of each frequency distribution, and combine them into a new sequence A 1 . Remove the terms with value 0 in A 1 , and then find the arithmetic mean and median of the sequence A 1 , take the minimum of the mean and the median, and round it to an integer, which is the value of the number of rows n .

[0019] 2.3 Calculate the cumulative interval vector V D2 .

[0020] Step 2.3 Explanation 1: The vector V D2The numerical calculation method of the i-th element (i is an integer not equal to 1) of the vector is: the vector V D1 The i-th element is equal to the sum of all elements before the element.

[0021] Step 2.3 Explanation 2: The vector V D2 The first element of the vector is equal to the vector V D1 The first element.

[0022] 2.4 The drill seismic source data matrix M S Interpolation, get the extended vector V X , V Y , V L , V D , V H .

[0023] Step 2.4 Explanation: First, linear interpolation is performed on the cumulative interval vector V D2 , without limiting the interpolation step, to get the interpolated cumulative interval vector V D3 . According to the vector V D2 , the vector V D3 , the first, second, third, fourth and fifth columns of the drill seismic source data matrix M S are respectively interpolated by the same interpolation method, to get the 5-column vector: V X , V Y , V L , V D , V H .

[0024] 2.5 Re-combine the extended vector V X , V Y , V L , V D , V H , to get the extended recombined horizontal coordinate matrix M X , the extended recombined vertical coordinate matrixM Y , extended recombination drilling distance matrix M L , extended recombination point depth matrix M D , extended recombination horizon matrix M H .

[0025] Step 2.5 description: the recombination rules are as follows: define the initial sampling range as [1,N] , the value of which is equal to the number of rows of the extended matrix N ; define the sampling step, the value of which is equal to M divided by 2 and rounded to the nearest integer; loop sampling: each time the loop process is, the vector N X each element sequence number if in the sampling range, then the elements are taken out in order and combined into a column, and incorporated into the matrix V X , and the sampling range is added to the step distance, and the sampling range is updated; in this way, until the last element of the vector V X is sampled, and the loop ends. In the same way, the vector M X is recombined into the matrix V X , the vector V Y is recombined into the matrix M Y , the vector V L is recombined into the matrix M L , the vector V D is recombined into the matrix M D , and the vector V H is recombined into the matrix V H .

[0026] In the above technical solution, in the sub-steps of step two, elements, vectors, column vectors, transposition, vector modulus operation, sequences, arithmetic sequences, frequency distribution, group distance, frequency, grouping points, linear interpolation are mathematical concepts that have been defined.

[0027] Step three: extract well-seismic deviation degree matrix M m . Wherein, step three contains 4 sub-steps:

[0028] 3.1 Calculate the depth rate of change vector V G , azimuth rate of change vectorV O .

[0029] Step 3.1 Illustration 1: Take the first row of matrix M X to be row vector V 1 , matrix M X Take the last row of matrix V 2 to be row vector V 2 Subtract row vector V 1 from row vector V 3 to get row vector M Y Take the first row of matrix V 4 to be row vector M Y Take the last row of matrix V 5 to be row vector V 5 Subtract row vector V 4 from row vector V 6 to get row vector M L Take the first row of matrix V 7 to be row vector M L Take the last row of matrix V 8 to be row vector V 8 Subtract row vector V 7 from row vector V 9 to get row vector M H Take the first row of matrix V 10 to be row vector M H Take the last row of matrix V 11 to be row vector V 11 Subtract row vector V 10 from row vector V 12 .

[0030] Step 3.1 Illustration 2: TakeV 3 and V 6 After squaring all elements, sum them together and take the square root to obtain the row vector. V 13 .

[0031] Step 3.1 Explanation 3: Vector V 12 Each element divided by the vector V 9 For each element, the depth change rate vector is obtained. V G .vector V 12 Each element divided by the vector V 13 Corresponding elements yield the azimuth rate of change vector. V O . Row vector V G Transpose, transform the row vector V O Transpose.

[0032] 3.2 Calculate the matching degree vector V V , will vector V V Transpose.

[0033] Step 3.2 Explanation: Vector V V The method for calculating the value of the i-th element (i is an integer not less than 1): Extract the matrix... M D For each column, perform a loop operation: transform the matrix M D Extracting from the i-th column (where i is any value from the first to the last column), denoted as a column vector. V d_i ; Column vector V d_i Subtracting the minimum value from each element of this vector and then adding 1, the resulting column vector is denoted as vector. V d2_i ; Retrieve the matrix M H The corresponding i-th column is denoted as the column vector. V h_i ,Will V h_i minus V d2_i And find the minimum value, then transform the vector V h_i Subtracting this minimum value from each element, the resulting column vector is denoted as . Vh2_i ;Will V h_i minus V d2_i And taking the absolute value, the resulting column vector is denoted as . V hd_i ; to vector V hd_i Add all elements together, multiply the result by 2, and subtract the vector. V hd_i The result of taking the first and last element values ​​is the vector. V V The value of the i-th element.

[0034] 3.3 Vector V V The numerical range is mapped to 0 to 1.

[0035] Step 3.3 Explanation: The mapping operation method is to convert the vector... V V Divide each element by the vector V V Find the maximum value of the vector to obtain a new vector. Replace the elements of the original vector with the elements of the new vector. V V Each element, i.e., the completed vector V V Numerical range mapping.

[0036] 3.4 Merging V G , V O , V V For matrix M m .

[0037] Step 3.4 Explanation: Matrix M m There are 3 columns in total, the first column is V G The second column is V O The third column is V V .

[0038] In the above technical solution, in the steps of step three, row vectors are a mathematically defined concept.

[0039] Step 4: Establish a score anchoring relationship matrix M a Determine the anchor score. Step four includes four sub-steps:

[0040] 4.1 Calculating VectorsV a1 .

[0041] Step 4.1 Explanation: Matrix M m The 3rd column of matrix is taken out and sorted in ascending order. The resulting column vector is vector V a1 .

[0042] 4.2 Calculate vector V a2 .

[0043] Step 4.2 Explanation: Vector V a1 Data standardization (data standardization is a known method in mathematics, generally using Z-Score standardization). The resulting result is recorded as V 14 Subtract the minimum value in V 14 from each element to get vector V 14 . V 15 Take the reciprocal of each element in V 15 and multiply by 200 to get vector V 16 . V 16 . V a2 .

[0044] 4.3 Merge V a1 , V a2 into matrix M a .

[0045] Step 4.3 Explanation: V a1 The 1st column of matrix M a , V a2 The 2nd column of matrix M a .

[0046] 4.4 Determine the anchor score according to matrix M a .

[0047] Step 4.4 Explanation 1: Matrix M aColumn 2 is the score value of bias scoring, the value range is 0 to 100. 100 points represent complete agreement without deviation. The study researchers specify the threshold score according to the matrix M a Column 2 specifies the acceptable threshold score.

[0048] Step 4.4 Description 2: The row corresponding to the threshold score in the matrix M a Column 1 element value is the anchor score value.

[0049] In the above technical solution, in the sub-step of step four, data standardization is a known mathematical method, generally Z-Score standardization is adopted.

[0050] Step five: calculate the bias degree distribution probability matrix M I . Wherein, step five contains 8 sub-steps:

[0051] 5.1 Matrix M I The size of the matrix

[0052] Step 5.1 Description: The study researchers specify the number of division intervals, which is an integer greater than 1, and the size of the matrix M I The number of rows and columns of the matrix

[0053] 5.2 Construct interval number sequence S d .

[0054] Step 5.2 Description: The interval number sequence S d Is an arithmetic sequence, the starting term is 0, the terminal term is 1, and the common difference is obtained by dividing 1 by the number of division intervals.

[0055] 5.3 Calculate the frequency distribution matrix M f .

[0056] Step 5.3 Description 1: The matrix M f The size of the matrix M I .

[0057] Step 5.3 Description 2: The matrix M f The calculation method of the element value of the ith row and jth column (i and j value range is 1 to k, k is the number of rows of the matrix M I ): Take out the first column of the bias degree matrix M m Recorded as V17 The sequence whose elements are not greater than the interval number. S d Set the (i+1)th element to 1, and the rest to 0; retrieve the matrix. M m The second column is denoted as V 18 The sequence whose elements are not greater than the interval number. S d Set the (j+1)th term to 1, and the rest to 0; let V 17 and V 18 Add them together, count the number of values ​​greater than 1, and divide the result by the matrix. M m The number of rows is the matrix. M f The value in the i-th row and j-th column.

[0058] 5.4 Determining the number of iterations and the size of the iteration.

[0059] Step 5.4 Explanation: The researchers specify the number of iterations and the iteration size, with the iteration size ranging from 0 to 100%.

[0060] 5.5 Cyclic Sample Total Matrix M n The calculation.

[0061] Step 5.5 Explanation 1: Matrix M n With matrix M f They are all the same size.

[0062] Step 5.5 Explanation 2: The calculation method is as follows: Subtract the matrix from 1. M f For each element, multiply each element of the resulting matrix by the loop size, and then multiply by the matrix size. M m The number of rows is calculated to obtain a new matrix. Then, each element of the new matrix is ​​rounded to the nearest integer, which gives the total number of cyclic samples. M n .

[0063] 5.6 Well-seismic deviation matrix M m Update.

[0064] Step 5.6 Explanation: Extract the deviation matrix M m In the third column, the value is compared with the anchor score. Elements smaller than the anchor score are set to 1, and the rest are set to 0. The resulting new vector replaces the original deviation matrix. Mm Column 3, update is completed.

[0065] 5.7 One-time calculation matrix M p Operation.

[0066] Step 5.7 Explanation 1: Matrix M p And matrix M I The size is consistent.

[0067] Step 5.7 Explanation 2: Matrix M p The calculation method of the value of the element in the ith row and jth column (i and j are in the range of 1 to k, and k is the number of rows of matrix M I ): Take the first column of the deviation degree matrix M m , denoted as V 19 , and set the elements not greater than the interval number sequence S d The ith+1 item to 1, and the rest to 0; Take the second column of matrix M m , denoted as V 20 , and set the elements not greater than the interval number sequence S d The jth+1 item to 1, and the rest to 0; Let V 19 And V 20 Add, set the elements less than 2 to 0, and the rest to 1, to get vector V 21 . Take the third column of matrix M m , denoted as vector V 22 ; The row where the element value of vector V 21 is 1, the corresponding V 22 Element is selected to form a new vector V 23 . Randomly select elements in vector V 23 , and the number of times of selection is equal to the total number of cyclic sampling matrix M n The value of the element in the ith row and jth column; The selected elements form a new vector V 24 . Count the vector V 24The number of elements with a value of 1 is divided by the number of cycles, and then divided by the number of matrix M m The number of rows of the matrix M p The value of the i-th row and the j-th column.

[0068] 5.8 Deviation degree distribution probability matrix M I The calculation.

[0069] Step 5.8 Description: Repeat step 5.7, the number of repetitions is equal to the number of cycles; then all the one-time calculation matrix M p is added up to get a new matrix; divide each element of the new matrix by the number of cycles, and then subtract 1 from each element, which is the deviation degree distribution probability matrix M I .

[0070] The beneficial effects of the present application are as follows:

[0071] The present application proposes a method of making a probability inference on well-seismic deviation degree, which is a practical solution to obtain the deviation degree distribution probability information of well-seismic data, solving the current problem.

[0072] The present application establishes a horizontal well drilling seismic source data matrix in the research block, expands and reorganizes the source data set, extracts a well-seismic deviation degree matrix therefrom, establishes a score anchoring relationship, and then calculates a deviation degree distribution probability matrix, thereby providing a solution for subsequent similar problems; wherein the construction method of the drilling seismic source data matrix, the calculation process of the expanded and reorganized matrix, the calculation method of the well-seismic deviation degree matrix, the calculation method of the score anchoring relationship matrix, and the calculation method of the deviation degree distribution probability matrix are extended to have guiding significance for subsequent research. BRIEF DESCRIPTION OF DRAWINGS

[0073] Figure 1 The technical flowchart of the present application;

[0074] Figure 2 The data flow diagram of the well-seismic deviation degree matrix extracted by the embodiment of the present application;

[0075] Figure 3 The data flow diagram of the score anchoring relationship matrix established by the embodiment of the present application;

[0076] Figure 4 The data flow diagram of the deviation degree distribution probability matrix calculated by the embodiment of the present application;

[0077] Figure 5 The data structure diagram of the deviation degree distribution probability matrix of the embodiment of the present application. IMPLEMENTATION

[0078] In order to make the technical solutions and beneficial effects of the present application more clear and understandable, the preferred embodiments of the present application will be described in detail below in combination with Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5 . It should be noted that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0079] Taking the shale gas block in southeast Chongqing as an example, the well-seismic deviation degree is extracted and probability is inferred by using the data of 26 drilled horizontal wells in the DS area and the depth domain structure interpretation data, and the step process is shown in Figure 1 .

[0080] Step 101, establishing the drilling-seismic source data matrix of the horizontal wells in the study block M S .

[0081] The data in the matrix M S comes from the data of 26 horizontal wells and the depth domain structure interpretation data of seismic data, which has 629 rows and 5 columns. The first column is composed of the horizontal coordinates of each point of the drilled horizontal well trajectory, the second column is composed of the vertical coordinates of each point of the drilled horizontal well trajectory, the third column is composed of the length of the horizontal section of each point of the drilled horizontal well trajectory, the fourth column is composed of the depth of each point of the drilled horizontal well trajectory, and the fifth column is composed of the depth of the corresponding depth domain structure interpretation horizon of each point of the drilled horizontal well trajectory.

[0082] Step 102, recombining and expanding the drilling-seismic source data matrix M S into five matrices with the same size: the expanded and recombined horizontal coordinate matrix M X , the expanded and recombined vertical coordinate matrix M Y , the expanded and recombined drilling distance matrix M L , the expanded and recombined point depth matrix M D , and the expanded and recombined horizon matrix M H .

[0083] The point distance vector V D1 is calculated. The result is a column vector with a size of 629x1.

[0084] The expanded matrix M X , M Y , ML , M D , M H number of rows n The number of rows of these five augmented matrices n is 19.

[0085] Calculate the accumulated interval vector V D2 The result is a column vector of size 629 x 1.

[0086] Interpolate the matrix M S to get the augmented vector V X , V Y , V L , V D , V H The result is five column vectors of size 2002 x 1.

[0087] Recombine the augmented vectors V X , V Y , V L , V D , V H to get the augmented recombined horizontal coordinate matrix M X , the augmented recombined vertical coordinate matrix M Y , the augmented recombined drilling distance matrix M L , the augmented recombined point depth matrix M D , and the augmented recombined horizon matrix M H The result is five matrices of size 19 x 4661.

[0088] Step 103, extract the well-seismic deviation matrix M m .

[0089] Calculate the depth rate of change vector V G and the azimuth rate of change vector V O The result is two column vectors of size 4661 x 1.

[0090] Calculate the matching degree vectorV V The result is a column vector of size 4661 x 1.

[0091] The vector V V The numerical range is mapped into 0 to 1. The updated vector V V The maximum is 1 and the minimum is 0.

[0092] The merging V G , V O , V V is a matrix M m The result is a matrix of size 4661 x 3.

[0093] Step 104, the establishment of the score anchoring relationship matrix M a , the determination of the anchoring score.

[0094] The vector V a1 The result is a column vector of size 4661 x 1.

[0095] The vector V a2 The result is a column vector of size 4661 x 1.

[0096] The merging V a1 , V a2 is a matrix M a The result is a matrix of size 4661 x 2.

[0097] The anchoring score is determined according to the matrix M a The anchoring score is 0.01348 and the acceptable threshold score is 90.

[0098] Step 105, the calculation of the deviation degree distribution probability matrix M I .

[0099] The matrix M I The determination of the size. The result is a size of 10 x 10 and the number of interval intervals is 10.

[0100] The interval number sequence S d is constructed. The interval number sequence S dThe starting term is 0, the ending term is 1, the tolerance is 0.1, and the number of terms is 11.

[0101] Frequency distribution matrix M f The calculation of the matrix. The result is a matrix of size 10 x 10.

[0102] M f [0.377, 0.380, 0.380, 0.380, 0.380, 0.380, 0.380, 0.380, 0.380, 0.380; 0.377, 0.634, 0.641, 0.641, 0.641, 0.641, 0.641, 0.641, 0.641, 0.641; 0.377, 0.634, 0.738, 0.776, 0.776, 0.776, 0.776, 0.776, 0.776, 0.776; 0.377, 0.634, 0.738, 0.873, 0.910, 0.911, 0.911, 0.911, 0.911, 0.911; 0.377, 0.634, 0.738, 0.873, 0.932, 0.954, 0.955, 0.955, 0.955, 0.955; 0.377, 0.634, 0.738, 0.873, 0.932, 0.959, 0.985, 0.985, 0.985, 0.985; 0.377, 0.634, 0.738, 0.873, 0.932, 0.959, 0.989, 0.995, 0.996, 0.996; 0.377, 0.634, 0.738, 0.873, 0.932, 0.959, 0.989, 0.996, 0.998, 1.000; 0.377, 0.634, 0.738, 0.873, 0.932, 0.959, 0.989, 0.996, 0.998, 1.000; 0.377, 0.634, 0.738, 0.873, 0.932, 0.959, 0.989, 0.996, 0.998, 1.000]

[0103] The number of cycles, the number of cycles, is 1000, and the number of cycles is 40%.

[0104] The total number of cycles is 1000, and the number of cycles is 40%. M n The calculation of the matrix. The result is a matrix of size 10 x 10.

[0105] M n =[1153, 1148, 1148, 1148, 1148, 1148, 1148, 1148, 1148, 1148; 1153, 677, 665, 665, 665, 665, 665, 665, 665, 665; 1153, 677, 485, 415, 415, 415, 415, 415, 415, 415; 1153, 677, 485, 235, 166, 164, 164, 164, 164, 164; 1153, 677, 485, 235, 125, 85, 84, 84, 84, 84; 1153, 677, 485, 235, 125, 76, 28, 28, 28, 28; 1153, 677, 485, 235, 125, 76, 20, 8, 8, 7; 1153, 677, 485, 235, 125, 76, 20, 7, 4, 0; 1153, 677, 485, 235, 125, 76, 20, 7, 4, 0; 1153, 677, 485, 235, 125, 76, 20, 7, 4, 0]

[0106] Updating the well-seismic misfit matrix M m The matrix M m In the third column, the elements smaller than the anchoring score 0.01348 are set to 1, and the rest are set to 0.

[0107] The matrix of one calculation M p The result of each loop calculation is a matrix of size 10x10.

[0108] The misfit distribution probability matrix M I The result is a matrix of size 10x10.

[0109] M I =[0.598, 0.603, 0.603, 0.603, 0.603, 0.603, 0.603, 0.603, 0.603, 0.603; 0.598, 0.778, 0.782, 0.782, 0.782, 0.782, 0.782, 0.782, 0.782, 0.782; 0.598, 0.778, 0.841, 0.865, 0.865, 0.865, 0.865, 0.865, 0.865, 0.866; 0.598, 0.778, 0.841, 0.926, 0.948, 0.949, 0.949, 0.948, 0.948, 0.949; 0.598, 0.778, 0.841, 0.926, 0.961, 0.974, 0.974, 0.974, 0.974, 0.974; 0.598, 0.778, 0.841, 0.926, 0.961, 0.977, 0.992, 0.992, 0.992, 0.992; 0.598, 0.778, 0.841, 0.926, 0.961, 0.976, 0.994, 0.998, 0.998, 0.998; 0.599, 0.778, 0.841, 0.926, 0.961, 0.977, 0.994, 0.998, 0.999, 1.000; 0.598, 0.778, 0.841, 0.926, 0.961, 0.977, 0.994, 0.998, 0.999, 1.000; 0.599, 0.778, 0.841, 0.926, 0.961, 0.977, 0.994, 0.998, 0.999, 1.000]

[0110] The above description is only the preferred embodiment of the application, and is not intended to limit the application. Any modification, equivalent replacement and improvement within the principles and spirit of the application are intended to be included in the protection scope of the application.

[0111] Not described in detail are prior art.

Claims

1. A method of calculating a probability of a distribution of well deviation, characterized by Comprising the following specific steps: Step one establishes a matrix of drill shock source data for horizontal wells within the study block M S ; Step two recombines the drill source data matrix M S into five recombined augmented horizontal coordinate matrices of equal size M X , recombined augmented vertical coordinate matrices M Y , recombined augmented drill spacing matrices M L , recombined augmented point depth matrices M D and recombined augmented bed matrices M H ; Step three extracts the well-to-seismic anomaly degree matrix M m ; Step 2 builds the score anchoring relationship matrix M a determining the anchor score; Step five calculates the bias degree distribution probability matrix M I .

2. The method of claim 1, wherein: The step one drilling source data matrix M S There are 5 columns in total, the first column is composed of the horizontal coordinate values of each point of the drilled trajectory of the completed horizontal well, the second column is composed of the vertical coordinate values of each point of the drilled trajectory of the completed horizontal well, the third column is composed of the length values of the horizontal section of each point of the drilled trajectory of the completed horizontal well, the fourth column is composed of the depth values of each point of the drilled trajectory of the completed horizontal well, and the fifth column is composed of the depth values of the corresponding depth domain structure interpretation horizon of each point of the drilled trajectory of the completed horizontal well, and each row of the drilling source data matrix is in one-to-one correspondence.

3. The method of claim 1, wherein: The second step is to calculate the point distance vector V D1 ; determine the augmented recombination matrix by calculation M X , M Y , M L , M D , M H the number of rows n ; Compute accumulated interval distance vector V D2 ; to drill source data matrix M S Interpolate to get augmented vector V X , V Y , V L , V D , V H ; recombine augmented vector V X , V Y , V L , V D , V H , get augmented recombined horizontal coordinate matrix M X , augmented recombined vertical coordinate matrix M Y , augmented recombined drill distance matrix M L , augmented recombined point depth matrix M D , and augmented recombined horizon matrix M H .

4. The method of claim 3, wherein: Point spacing vector V D1 is a column vector, the number of elements is equal to the number of columns of the drill seismic source data matrix M S the number of rows is equal, and the value of the first element is 0, the calculation method of the value of the i-th element is to take the drill seismic source data matrix M S the first and second columns as a new matrix, denoted as matrix M 1 , the matrix M 1 transpose, get the transpose matrix, on the transpose matrix, subtract the elements of the i-th column from the elements of the i-1-th column, get the column vector, and then take the vector module operation, the value is the vector V D1 i-th element value of the vector, i is an integer not equal to 1; Determine the expanded recombination matrix M X , M Y , M L , M D , M H Number of lines n The method is if the point spacing vector V D1 One-eighth of the minimum value of the elements in the array is not less than 100, and the number of rows is... n The value is 100; If the dot spacing vector V D1 If one-eighth of the minimum value of an element is less than 100, divide the minimum value by 16 and round it to the nearest integer. Use the result as the starting term value to construct an arithmetic sequence with a common difference of 2 and a ending term value not exceeding 100. S 1 If the first term of the sequence is less than 1, then set the starting point to 1; S 1 Each item is used as the class interval of the statistical frequency distribution. The group point value corresponding to the maximum frequency in each frequency distribution is found and combined into a new sequence. A 1 ;Will A 1 After removing terms with a median value of 0, the sequence is obtained. A 1 The arithmetic mean and median are calculated, and the minimum of the mean and median is taken and rounded to the nearest integer. The result is the number of rows. n The possible values ​​of ; accumulated distance vector V D2 the first element of the accumulated distance vector is equal to the point distance vector V D1 the first element; accumulated distance vector V D2 the value of the i-th element of the accumulated distance vector is equal to the sum of the point distance vector V D1 the i-th element of the accumulated distance vector is equal to the sum of all elements of the accumulated distance vector V D1 preceding the i-th element of the accumulated distance vector, i being an integer different from 1 extended vector V X , V Y , V L , V D , V H The calculation method of the extended vector is that firstly, linear interpolation is performed on the accumulated interval vector V D2 , the interpolation step is not limited, and an interpolated accumulated interval vector V D3 is obtained; the first, second, third, fourth and fifth columns of the drill seismic source data matrix M S are respectively interpolated by the same interpolation method according to the proportion of the vector V D2 and the vector V D3 , and five column vectors V X , V Y , V L , V D , V H are obtained. Recombining augmented vectors V X , V Y , V L , V D , V H The rules are as follows: define the initial sampling range as [1,N] , N The value of the number of rows of the augmented matrix M X ; define the sampling step, whose value is equal to N divided by 2 and rounded to the nearest integer; for V X loop sampling: each time the loop is, the vector V X If the element sequence number is within the sampling range, the elements are taken in order and combined into a column, and incorporated into the matrix M X , and the sampling range is updated by adding the step distance; this is done until the last element of the vector V X is sampled, and the loop ends; In the same way, the vector V Y recombined into a matrix M Y , the vector V L recombined into a matrix M L , the vector V D recombined into a matrix M D , the vector V H recombined into a matrix V H .

5. The method of claim 1, wherein: The third step is to calculate the depth variation rate vector V G , azimuth variation rate vector V O ; calculate the matching degree vector V V , transpose the vector V V ; transpose the vector V V , map the numerical range to 0-1 to obtain the mapped vector V V ; merge V G , V O , the mapped vector V V to obtain the well-seismic deviation degree matrix M m .

6. The method of claim 5, wherein: Depth change rate vector V G Azimuth change rate vector V O The calculation method is to take out the first row of matrix M X as row vector V 1 , take out the last row of matrix M X as row vector V 2 , subtract row vector V 2 from row vector V 1 , and get row vector V 3 ; similarly, take out the first row of matrix M Y as row vector V 4 , take out the last row of matrix M Y as row vector V 5 , subtract row vector V 5 from row vector V 4 , and get row vector V 6 ; similarly, take out the first row of matrix M L as row vector V 7 , take out the last row of matrix M L as row vector V 8 , subtract row vector V 8 from row vector V 7 , and get row vector V 9 ; similarly, take out the first row of matrix M H as row vector V 10 , take out the last row of matrix M H as row vector V 11 , subtract row vector V 11 from row vector V 10 , and get row vector V 12 ; take V 3 With V 6 After squaring all elements, then corresponding addition and square root, get row vector V 13 ; vector V 12 Divide each element by vector V 9 Corresponding elements, get depth rate of change vector V G ; vector V 12 Divide each element by vector V 13 Corresponding elements, get azimuth rate of change vector V O ; vector V G , vector V O Transpose, get column vector V G And column vector V O ; Matching degree vector V V The calculation method of the i-th element value of the vector is to take out the matrix M D Each column, and i is an integer not less than 1, and the matrix M D The i-th column is taken out, and i is any value from the 1st column to the last column, denoted as the column vector V d_i ; The column vector V d_i Each element is subtracted by the minimum value in the vector and then added by 1, and the column vector obtained is denoted as the vector V d2_i ; Take out the matrix M H The corresponding i-th column is denoted as the column vector V h_i , and V h_i is subtracted from V d2_i , and the minimum value is found, and the vector V h_i Each element is subtracted by the minimum value, and the column vector obtained is denoted as V h2_i ; Take V h_i , subtract V d2_i , and take the absolute value, and the column vector obtained is denoted as V hd_i ; Add all the elements of the vector V hd_i , multiply the result by 2, and subtract the value of the first element and the value of the tail element of the vector V hd_i , and the result is the i-th element value of the vector V V ; Then, transpose the vector V V to obtain the row vector V V ; Matching degree vector V V The method of operation for mapping a numerical range into 0 to 1 is to divide the row vector V V by the maximum value of the row vector V V to get a new row vector, and replace each element of the original row vector V V with the element of the new row vector. merge V G , V O , V V for well seismic deviation matrix M m when, first, row vector V V is transposed into column vector, then matrix M m the first column is column vector V G , the second column is column vector V O , the third column is transposed column vector V V , and there are 3 columns in total.

7. The method of claim 1, wherein: The step four is calculating vector V a1 ; calculating vector V a2 ; merging V a1 , V a2 is a score anchoring relation matrix M a ; determining an anchoring score according to the score anchoring relation matrix M a .

8. The method of claim 7, wherein: vector V a1 from the well-to-seismic misfit matrix M m after sorting the 3rd column in ascending order; vector V a2 The calculation method is to use vectors V a1 Data standardization, the resulting data is denoted as V 14 ;Will V 14 Subtract each element V 14 The minimum value in the vector is denoted as . V 15 ,Will V 15 Perform an exponentiation operation with the natural constant e on each element, and add 1 to obtain the vector. V 16 ,Pick V 16 Multiplying the reciprocal of each element by 200 yields the vector. V a2 ; score anchoring relationship matrix M a by V a1 , V a2 obtained by combining, V a1 i.e. the first column of the matrix M a , the second column of the matrix V a2 , the third column of the matrix M a ; The method for determining the anchor score is based on the anchor score matrix M a The second column specifies the acceptable threshold score, whose matrix M a The second column is the score value of the deviation score, which ranges from 0 to 100, and 100 points represent complete agreement without deviation; the matrix corresponding to the row where the threshold score is located M a The first column element value is the anchor score.

9. The method of claim 1, wherein: Step five is the matrix M I of the size determination; the construction interval number sequence S d ; the calculation of the frequency distribution matrix M f ; the determination of the cycle number, cycle scale number; the calculation of the cycle sampling total number matrix M n ; the update of the well-seismic deviation degree matrix M m ; the operation of the one-time calculation matrix M p ; the calculation of the deviation degree distribution probability matrix M I .

10. The method of claim 9, wherein: determining matrix M I The size of the method is to specify the number of partition intervals by the researchers, which is an integer greater than 1, and the matrix M I The number of rows and columns of the matrix is equal to the number of partition intervals. Constructing interval number series S d That is, the method of arithmetic sequence is that the initial term is 0, the terminal term is 1, and the common difference is obtained by dividing 1 by the number of partitions. Frequency distribution matrix M f With matrix M I The size is consistent, the calculation method of the value of the element of the ith row and the jth column is to take out the deviation degree matrix M m The first column is recorded as V 17 The element of which is not greater than the interval number sequence S d The ith+1 term is set to 1, and the rest is set to 0; take out the second column of matrix M m Recorded as V 18 The element of which is not greater than the interval number sequence S d The jth+1 term is set to 1, and the rest is set to 0; let V 17 And V 18 Add, count the number of values greater than 1, and divide the result by the number of rows of matrix M m , that is, matrix M f The value of the element of the ith row and the jth column, the value range of i and j of the ith row and the jth column is 1 to k, k is the number of rows of matrix M I ; The number of cycles, cycle size is specified by the researcher, cycle size is valued between 0 and 100%; cyclically sampled total number matrix M n with matrix M f of the same size, calculated by subtracting 1 from the matrix M f of the same size, calculated by subtracting 1 from the matrix M m of the same size, calculated by subtracting 1 from the matrix M n ; Updating well-to-seismic misfit matrix M m The operation method is to take out the 3rd column of the matrix M m Compare the numerical value with the anchor score, and set the element smaller than the anchor score to 1 and the rest to 0 to obtain a new vector, which replaces the original misfit matrix M m The 3rd column is updated. One-time calculation matrix M p With matrix M I The same size, the value of the i-th row and j-th column element is to take out the updated deviation degree matrix M m The first column is recorded as V 19 , whose elements are not greater than the interval number sequence S d The i+1th item is set to 1, and the rest is set to 0. The value range of i and j in the i-th row and j-th column of the matrix is 1 to k, and k is the number of rows of the matrix M I Take out the second column of the updated deviation degree matrix M m Recorded as V 20 , whose elements are not greater than the interval number sequence S d The j+1th item is set to 1, and the rest is set to 0; Let V 19 And V 20 Add, set the elements less than 2 to 0, and the rest to 1, to get the vector V 21 ; Take out the third column of the updated deviation degree matrix M m Recorded as vector V 22 ; The row where the element value of vector V 21 is 1, the corresponding V 22 elements are selected to form a new vector V 23 ; Randomly select elements in vector V 23 , the number of times of selection is equal to the total number of matrix M n The value of the i-th row and j-th column element; The selected elements form a new vector V 24 ; Count the number of elements with a value of 1 in vector V 24 , and divide by the cycle size number, and then divide by the number of rows of matrix M m , and the value is the value of the i-th row and j-th column element of matrix M p ; Deviation degree distribution probability matrix M I The calculation method is to repeatedly operate the matrix M p a number of times equal to the number of cycles. Then calculate all the matrices once. M p By summing them up, we get a new matrix; Divide each element of the new matrix by the number of cycles, and subtract 1 from each element, which is the bias distribution probability matrix M I .

Images