Drill bit optimization and structure optimization method fusing big data classification and golden section method

Abstract

The invention discloses a drill bit optimization and structure optimization method fusing big data classification and a golden section method, and the method comprises the steps: 1, collecting the data of a drill bit application adjacent well region, and constructing a drill bit database; 2, according to the borehole size of the target well, drill bit data corresponding to the same borehole size and the same stratum of an adjacent well are extracted, the drill bit data comprise the mechanical drilling speed and the drilling footage, normalization processing is carried out, and primary screening classification is carried out through a golden section method; step 3, on the basis of primary classification, performing normalization processing on wear classification of the drill bit, and performing clustering analysis by using a neural network method to complete secondary screening classification; step 4, performing optimization solution by fusing the drilling cycle and the cost constraint condition, and further providing suggestions for drill bit structure improvement and performance optimization; the defects in the prior art are overcome, and therefore the requirements for refined, scientific and reasonable application of the regional drill bit are met.

Description

Technical Field

[0001] This invention belongs to the field of drill bit design technology, and in particular relates to a method for drill bit selection and structural optimization that integrates big data classification and the golden ratio. Background Technology

[0002] In the drilling engineering system of energy industries such as oil and gas exploration and development, and solid mineral resource exploration, the drill bit, as the core rock-breaking tool, directly determines the efficiency, safety, and economic benefits of drilling operations due to its cutting performance, wear resistance, and operational stability. A suitable drill bit can not only improve the rate of penetration (ROP) and extend its life, but also avoid complex downhole conditions and reduce well construction costs. Its performance is affected by various factors such as drill string assembly, formation lithology, drilling fluid properties, and drilling technology. In practical applications, the compatibility between the drill bit and the formation can be comprehensively evaluated through multiple dimensions such as ROP, cumulative footage, drill string vibration, wear degree, and rock removal efficiency.

[0003] Current drill bit selection methods include field data statistical methods, technical benefit index methods, reference drill bit selection methods, equal probability drill bit selection methods, Cartesian coordinate selection methods, and rock mechanics parameter methods. However, these methods generally face multiple application bottlenecks: first, the evaluation methods are relatively simple, making it difficult to form a comprehensive and systematic analysis; second, there are dual barriers of technology and cost in obtaining logging data; third, the complexity of formation lithology leads to increased uncertainty; fourth, the real-time and accurate measurement of drilling parameters is difficult; and fifth, a continuous iterative plan has not been formed after evaluation. More importantly, most selection methods do not set core objectives such as the shortest drilling cycle and the optimal drilling cost as constraints, making it difficult to achieve refined, scientific, and rational drill bit selection.

[0004] Therefore, there is an urgent need for a fast, efficient, and comprehensive technical method for drill bit selection and structural optimization. Summary of the Invention

[0005] The technical problem this invention aims to solve is to provide a method for drill bit selection and structural optimization that integrates big data classification and the golden ratio. This method overcomes the shortcomings of existing technologies by setting boundary conditions that minimize drilling cycles and optimize costs, combined with iterative design of drill bit structure and materials, thereby meeting the needs for refined, scientific, and rational application of drill bits in specific regions.

[0006] The technical solution of this invention is:

[0007] A method for drill bit selection and structure optimization that integrates big data classification and the golden ratio, the method comprising:

[0008] Step 1: Collect data from adjacent well areas used by the drill bit to build a drill bit database;

[0009] Step 2: Based on the target wellbore size, extract the drill bit data corresponding to the same wellbore size and formation of adjacent wells, including mechanical drilling rate and footage, perform normalization processing, and use the golden section method for initial screening and classification.

[0010] Step 3: Based on the initial classification, normalize the wear classification of the drill bit and use neural network cluster analysis to complete the secondary screening and classification.

[0011] Step 4: Combine drilling cycle and cost constraints to perform optimization solution, and then propose suggestions for drill bit structure improvement and performance optimization.

[0012] Step 1 involves collecting data from adjacent wells using the drill bit, including the well number, wellhead coordinates, formation lithology, drilling fluid density, logging data, drill bit model, drill bit size, drill bit type, drill bit manufacturer, mechanical drilling rate, pure drilling time, stroke drilling rate, wear characteristic classification, and drillability extreme values ​​of the target well and adjacent wells.

[0013] The methods for determining adjacent wells include: based on the principle of stratigraphic similarity, using the target wellhead coordinates as boundary conditions to determine reference adjacent wells; in development zones and rolling development zones, adjacent wells with a straight-line distance of less than 10 km are selected; in exploration blocks, the well spacing is increased based on the actual drilling situation, and the spacing is set independently according to the number of wells drilled in the area. The well spacing calculation formula is as follows:

[0014]

[0015] The distance between the target well and the wellhead of the adjacent well. The vertical coordinate of the adjacent wellhead. The x-coordinate of the adjacent wellhead. The vertical coordinate of the target wellhead. The x-coordinate of the target wellhead.

[0016] Normalization methods include:

[0017] The formula for normalizing the mechanical drilling rate is:

[0018] (i=1,2,3……n)

[0019] To normalize the mechanical drilling rate of the sample, This is the drilling speed of the machine. The minimum mechanical drilling rate for the sample set. The maximum mechanical drilling rate for the sample set;

[0020] The normalization formula for advance is:

[0021] (i=1,2,3……n)

[0022] To normalize the sample advance, For sample advance, The minimum advance size for the sample set. Let i be the maximum footage of the sample set, i be the sample number (dimensionless), and n be the sample number. The average mechanical drilling rate is:

[0023] (i=1,2,3……n)

[0024] This represents the average mechanical drilling speed. For pure diamond time;

[0025] Normalization formula for average mechanical drilling rate:

[0026] (i=1,2,3……n)

[0027] Normalize the average mechanical drilling rate;

[0028] Average footage:

[0029] (i=1,2,3……n)

[0030] This is the average advance length.

[0031] Normalization formula for average footage:

[0032] (i=1,2,3……n)

[0033] To normalize the average footage,

[0034] Golden section function for mechanical drilling rate and footage indicators:

[0035] (i=1,2,3……n)

[0036] Golden Ratio Factor:

[0037] (i=1,2,3……n)

[0038] -Golden section factor;

[0039] Golden ratio threshold: when When the value is greater than 1, the sample value is above the golden section line, belonging to the dominant set. When the value is less than 1, the sample value is below the golden section line and belongs to the inferior set.

[0040] Methods for normalizing drill bit wear classification include:

[0041] Thresholds were set for the characteristics of inner tooth I, outer tooth O, and gauge diameter G in the drill bit wear classification, and then normalization was performed. The specific threshold setting method is as follows: According to the IADC drill bit wear classification, inner tooth I and outer tooth O are divided into 9 levels: 0, 1, 2...8, where 0 represents no wear and 9 represents complete wear. Gauge diameter G is divided into 5 levels: 0, 1 / 16 in, 2 / 16 in, 3 / 16 in, and 4 / 16 in, where 0 represents no wear and 4 / 16 represents diameter wear of 4 / 16 in. The classification assignment formula is as follows:

[0042] Wear grading assignment for internal gear I:

[0043] (i=1,2,3……9)

[0044] Characterizing the wear of the internal gear I; The wear grading threshold for internal gear I;

[0045] Wear grading assignment for external gear teeth:

[0046] (i=1,2,3……9)

[0047] Characterizing the wear amount of the outer tooth O;

[0048] The wear grading threshold for the outer tooth O;

[0049] Wear grading assignment for gauge G:

[0050] (i=1,2,3……5)

[0051] Characterized by the wear amount of gauge diameter G;

[0052] The wear grading threshold is the gauge wear threshold.

[0053] The normalization formula for wear classification is:

[0054] Normalized formula for wear classification of internal gear I:

[0055] (i=1,2,3……n)

[0056] This represents the minimum wear value of the outer tooth rack in the sample set. This represents the maximum wear value of the outer tooth rack in the sample set;

[0057] Normalized formula for wear classification of external gear teeth:

[0058] (i=1,2,3……n)

[0059] This represents the minimum wear value of the tooth rack within the sample set.

[0060] This represents the maximum wear value of the tooth rack within the sample set.

[0061] Normalized formula for wear classification of gauge G:

[0062] (i=1,2,3……n)

[0063] This represents the minimum wear value of the gauge diameter in the sample set. This represents the maximum wear value of the gauge diameter in the sample set.

[0064] Wear weight normalization: ,in

[0065] x, y, and z are weighting factors, typically x∈(0.25~0.35), y∈(0.30~0.40), and z∈(0.35~0.45).

[0066] The method of using neural network cluster analysis to complete secondary screening and classification includes: using neural network optimization to cluster and classify drill bit wear characteristics, including drill bit diameter reduction (PB), self-sharpening wear (SS), and scouring wear (WO), and proposing drill bit improvement suggestions based on different formations, wellbore trajectories, drilling fluid systems, and drilling process conditions and requirements.

[0067] The optimization solution that integrates drilling cycle and cost constraints includes:

[0068] Drilling cycle formula:

[0069] (i=1,2,3……5)

[0070] (j=1,2,3……5)

[0071] -Optimal drilling cycle, d;

[0072] - Single drill bit advance, meters;

[0073] - Pure diamond aging, dimensionless;

[0074] -Minimum cost, 10,000 yuan;

[0075] -Daily fee (drilling rig daily fee, tool daily fee), 10,000 yuan;

[0076] - Tool and rice cost, 10,000 yuan.

[0077] The method further includes:

[0078] Step 5: Based on the structural optimization suggestions, make personalized improvements to the target drill bit, conduct numerical simulation using 3D modeling software, and determine the final 2D drawings and 3D model based on the simulation results of temperature field and flow field.

[0079] Step 6: Produce drill bits, track and evaluate their field use, and carry out continuous iterative upgrades.

[0080] The beneficial effects of this invention are:

[0081] The drill bit optimization method proposed in this invention breaks through the limitations of traditional technology. It can accurately construct and solve the optimization model based on the user's core requirements for drilling cycle or cost control, effectively avoiding the decision-making dilemma caused by multiple solutions in conventional screening methods.

[0082] The drill bit parameter normalization, golden ratio calculation, and sample data extraction techniques of this invention significantly improve the efficiency and accuracy of data processing, completely eliminate interference caused by differences in data dimensions and numerical fluctuations, and ensure the reliability of analysis results.

[0083] The drill bit wear classification and quantitative method and neural network optimization method of this invention can deeply integrate formation characteristics, wellbore trajectory design, drilling fluid system configuration and process technology requirements to form a highly targeted drill bit iterative optimization scheme, which can significantly shorten the regional drill bit development cycle.

[0084] This invention achieves deep cross-integration of multiple disciplines such as geological analysis, drilling engineering, fluid mechanics, wellbore trajectory control, mechanical design and simulation, and builds an integrated collaborative innovation platform of "geology-drilling-machinery". It overcomes the technical problems of unclear structural design path and low material iteration efficiency under the dual constraints of drilling cycle and cost in traditional drill bit optimization mode, and meets the current industry's urgent needs for refined selection, scientific design and rational application of drill bits. Attached Figure Description

[0085] Figure 1 This is a schematic diagram of the process of the present invention;

[0086] Figure 2 This is a schematic diagram of a single screening and classification process in a specific embodiment of the present invention;

[0087] Figure 3 This is a schematic diagram of secondary screening and classification in a specific embodiment of the present invention;

[0088] Figure 4 This is a schematic diagram illustrating the proposed improvements to the neural network method in a specific embodiment of the present invention.

[0089] Figure 5 This is a schematic diagram of the optimization solution process in a specific embodiment of the present invention;

[0090] Figure 6 This is a schematic diagram of the drill bit improvement process in a specific embodiment of the present invention. Detailed Implementation

[0091] This invention relates to the field of drill bit design technology, proposing a drill bit selection and structural optimization method that integrates big data classification and the golden section method. The method achieves precise drill bit optimization through the following steps: First, the system collects data from adjacent well areas where the drill bit is applied, constructing a drill bit database. Second, the data is classified based on the wellbore size of the target well. Data is extracted from the drill bit database according to the target well's wellbore size, and invalid data is removed. Formation lithology is screened, and drill bit data corresponding to the same wellbore size and formation in adjacent wells is extracted, including mechanical drilling rate and footage. The screening results are then normalized, and the golden section method is used for a first screening and classification. Next, based on the first classification, the wear classification of advantageous drill bits is normalized, and neural network clustering analysis is used to complete a second screening and classification. Subsequently, constraints such as drilling cycle and cost, which are of key concern to users, are integrated to perform optimization solutions, thereby proposing drill bit structural improvement and performance optimization suggestions. Finally, the drill bit manufacturer conducts numerical simulations based on the optimization suggestions to determine the final improvement scheme and conducts drill bit trial production and engineering verification.

[0092] This invention specifically includes:

[0093] Step 1: Collect regional data, including but not limited to well number, wellhead coordinates, formation lithology, drilling fluid density, logging data, drill bit model, drill bit size, drill bit type, drill bit manufacturer, mechanical drilling rate, pure drilling time, stroke drilling rate, wear characteristic classification, and drillability extreme values ​​of the target well and adjacent wells, and establish a drill bit database.

[0094] The database data comes from data automatically stored at the drilling site logging end, drilling design, geological design, logging interpretation data, and drill bit structure data.

[0095] The determination of adjacent wells is based on the principle of stratigraphic similarity (strata position, lithology, burial depth, thickness, drillability extremes, etc.). Reference adjacent wells are determined using the target wellhead coordinates as boundary conditions. In development zones and rolling development zones, adjacent wells with a straight-line distance of less than 10 km, or even less than 5 km, can be selected. In exploration blocks, the well spacing is increased based on the actual drilling situation. The actual spacing can be set according to the number of wells drilled in the area. The formula for calculating the well spacing is:

[0096] ;

[0097] - Distance between the target well and adjacent wells, in meters;

[0098] - The ordinate of the adjacent wellhead, in meters;

[0099] - The x-coordinate of the adjacent wellhead, in meters;

[0100] - Target wellhead ordinate, in meters;

[0101] - Target wellhead x-coordinate, m.

[0102] Step 2: Extract data from the drill bit database based on the target wellbore size and clean up invalid data; screen the formation lithology and extract drill bit data corresponding to the same wellbore size and formation in adjacent wells, including mechanical drilling rate and footage, and then normalize the screening results and use the golden section method for a second screening and classification.

[0103] The normalization formulas for drilling rate and footage are as follows:

[0104] Normalization formula for mechanical drilling rate:

[0105] (i=1,2,3……n);

[0106] - Sample mechanical drilling speed is normalized and dimensionless;

[0107] -Sample mechanical drilling rate, m / h;

[0108] -Minimum mechanical drilling rate in the sample set, m / h;

[0109] - Maximum mechanical drilling rate in the sample set, m / h.

[0110] normalized advance formula:

[0111] (i=1,2,3……n);

[0112] - Sample advance normalization, dimensionless;

[0113] - Sample advance length, m;

[0114] -Minimum advance measurement of the sample set, m;

[0115] -Maximum advance size of the sample set, in meters;

[0116] i - Sample number, dimensionless;

[0117] n - number of samples, dimensionless.

[0118] Average mechanical drilling rate:

[0119] (i=1,2,3……n);

[0120] - Average mechanical drilling rate, m / h;

[0121] - Pure diamond time, h;

[0122] Normalization formula for average mechanical drilling rate:

[0123] (i=1,2,3……n);

[0124] - The average mechanical drilling rate is normalized and dimensionless.

[0125] Average footage:

[0126] (i=1,2,3……n);

[0127] - Average advance length, in meters.

[0128] Normalization formula for average footage:

[0129] (i=1,2,3……n);

[0130] - The average advance rate is normalized and dimensionless.

[0131] Golden ratio values ​​for mechanical drilling rate and footage:

[0132] (i=1,2,3……n);

[0133] Golden Ratio Factor:

[0134] (i=1,2,3……n);

[0135] - The golden ratio factor is dimensionless.

[0136] Golden ratio threshold: when When the value is greater than 1, the sample value is above the golden section line, belonging to the dominant set. When the value is less than 1, the sample value is below the golden section line and belongs to the inferior set.

[0137] Step 3, as follows Figure 2 As shown, in the first classification result, the main characteristic descriptions of drill bit wear classification, such as inner tooth I, outer tooth O, and diameter G, are numerically transformed and normalized; the wear characteristics and locations in the drill bit wear classification are clustered using a neural network method for secondary screening and classification, such as... Figure 3 As shown. The specific method is as follows:

[0138] Thresholds were set for key characteristics in drill bit wear classification, including inner tooth I, outer tooth O, and gauge diameter G, and then normalization was performed. The specific threshold setting method is as follows: According to the IADC drill bit wear classification, inner tooth I and outer tooth O are divided into 9 levels: 0, 1, 2...8, where 0 represents no wear and 9 represents complete wear. Gauge diameter G is divided into 5 levels: 0, 1 / 16 in, 2 / 16 in, 3 / 16 in, and 4 / 16 in, where 0 represents no wear and 4 / 16 represents diameter wear of 4 / 16 in. The classification assignment formula is as follows:

[0139] Wear grading assignment for internal gear I:

[0140] (i=1,2,3……9);

[0141] - Wear measurement of internal gear I, dimensionless;

[0142] - Wear grading threshold for internal gear teeth I, dimensionless.

[0143] Wear grading assignment for external gear teeth:

[0144] (i=1,2,3……9);

[0145] - Wear measurement of external gear teeth, dimensionless;

[0146] - Wear grading threshold for external gear teeth, dimensionless.

[0147] Wear grading assignment for gauge G:

[0148] (i=1,2,3……5);

[0149] - Wear measurement characterization of gauge G, in;

[0150] -Diameter wear grading threshold, dimensionless.

[0151] The normalization formula for wear classification is:

[0152] Normalized formula for wear classification of internal gear I:

[0153] (i=1,2,3……n);

[0154] -Minimum wear value of external tooth rack in the sample set, dimensionless;

[0155] - Maximum wear value of external tooth rack in the sample set, dimensionless.

[0156] Normalized formula for wear classification of external gear teeth:

[0157] (i=1,2,3……n);

[0158] -Minimum wear value of toothed rack within the sample set, dimensionless;

[0159] - The maximum wear value of tooth racks within the sample set, dimensionless.

[0160] Normalized formula for wear classification of gauge G:

[0161] (i=1,2,3……n);

[0162] -Minimum wear value of the caliber in the sample set;

[0163] -Maximum wear value of the sample set caliber, in.

[0164] Wear weight normalization: ,in ;

[0165] x, y, and z are weighting factors, typically x∈(0.25~0.35), y∈(0.30~0.40), and z∈(0.35~0.45).

[0166] The neural network optimization method is used to cluster and classify drill bit wear characteristics (drill bit diameter reduction PB, self-sharpening wear SS, erosion WO, etc.), and can propose drill bit improvement suggestions based on different formations, wellbore trajectories, drilling fluid systems, drilling technology and other conditions and requirements.

[0167] Step 4: After secondary classification, integrate user needs and use the shortest cycle or best cost as constraints to find the optimal solution, and clarify the improvement suggestions for the structure and performance of the selected target drill bit.

[0168] Drill bit structure optimization involves using 3D modeling software to improve the model based on the original design drawings. According to the actual working conditions and conditions such as well depth, drilling fluid performance, drilling parameters, and stress, simulations of temperature field, flow field, and stress field are carried out to evaluate the impact resistance and wear resistance of drill bit teeth, temperature field distribution, flow field distribution, and stress field distribution. If there are defects in the simulation results, the design and simulation are repeated after verification and adjustment.

[0169] Based on the results of structural optimization, secondary classification, and cost cycle assessment, the final optimization scheme is determined, and numerical simulation of the drill bit is carried out.

[0170] Drilling cycle formula:

[0171] (i=1,2,3……5);

[0172] (j=1,2,3……5);

[0173] -Optimal drilling cycle, d;

[0174] - Single drill bit advance, meters;

[0175] - Pure diamond aging, dimensionless;

[0176] -Minimum cost, 10,000 yuan;

[0177] -Daily fee (drilling rig daily fee, tool daily fee), 10,000 yuan;

[0178] - Tool and rice cost, 10,000 yuan.

[0179] Step 5: Based on the structural optimization suggestions, the drill bit company makes personalized improvements to the target drill bit, uses 3D modeling software to conduct numerical simulations, and determines the final 2D drawings and 3D model based on the simulation results of temperature field, flow field, etc.

[0180] Step 6: Produce drill bits, track and evaluate their field use, and carry out continuous iterative upgrades.

Claims

1. A method of bit optimization and structure optimization by fusing big data classification and golden section method, characterized in that: The method comprises: Step 1, collecting adjacent well area data of drill bit application, and constructing a drill bit database; Step 2, according to the borehole size of the target well, extracting the drill bit data corresponding to the same borehole size and the same formation of the adjacent well, including the rate of penetration and footage, performing normalization processing, and using the golden section method for primary screening and classification; Step 3, on the basis of the primary classification, performing normalization processing on the wear grading of the drill bit, and using a neural network method for cluster analysis to complete secondary screening and classification; Step 4, performing optimal solution under the constraint conditions of drilling cycle and cost, and then proposing suggestions for drill bit structure improvement and performance optimization.

2. The method of claim 1, wherein the method is a method of bit selection and optimization that combines big data classification and the golden section method. The collection of adjacent well area data of drill bit application in step 1 includes well number, wellhead coordinates, formation lithology, drilling fluid density, logging data, drill bit model, drill bit size, drill bit category, drill bit manufacturer, rate of penetration, net drilling time, travel drilling speed, wear characteristic grading, and drillability extreme value of the target well and adjacent wells.

3. The method of claim 1, wherein the method is characterized by: The method for determining adjacent wells comprises: determining reference adjacent wells based on the formation similarity principle and taking the wellhead coordinates of the target well as the boundary condition; selecting adjacent wells with a straight-line distance of less than 10 km in the development area and the rolling development area; increasing the inter-well spacing of adjacent wells in the exploration block according to the actual drilled exploration wells; and setting the inter-well spacing according to the number of regional drilling wells, with the inter-well spacing calculation formula being: ； is the distance from the wellhead of the target well to the wellhead of the neighboring well, is the longitudinal coordinate of the wellhead of the neighboring well, is the lateral coordinate of the wellhead of the neighboring well, is the longitudinal coordinate of the wellhead of the target well, is the lateral coordinate of the wellhead of the target well.

4. The method of claim 1, wherein the method is a method of bit selection and optimization that combines big data classification and the golden section method. The normalization processing method comprises: The rate of penetration normalization formula is: (i = 1, 2, 3...n); normalizing the mechanical specific drilling rate for the sample, mechanical specific drilling rate for the sample, mechanical specific drilling rate minimum for the sample set, mechanical specific drilling rate maximum for the sample set; The footage normalization formula is: (i = 1, 2, 3...n); for sample footage normalization, for sample footage, for sample set footage minimum, for sample set footage maximum, i is sample number; n is sample number. The average value of the rate of penetration is: (i = 1, 2, 3...n); is the average value of the mechanical drilling speed, is the net drilling time; The average value of the rate of penetration normalization formula is: (i = 1, 2, 3...n); RPM is normalized by the average value of the mechanical penetration rate; The average value of the footage is: (i = 1, 2, 3...n); for average footage, The average value of the footage normalization formula is: (i = 1, 2, 3...n); To normalize the average of footage, The golden section value of the rate of penetration and footage indicators is: (i = 1, 2, 3...n); - a golden split value; The golden section point factor is: (i = 1, 2, 3...n); - the golden section point factor.

5. The drill bit optimization and structure optimization method fusing big data classification and the golden section method according to claim 4, characterized in that: Golden section threshold: when >1, the sample value is above the golden section line, belonging to the dominant set, when <1, the sample value is below the golden section line, belonging to the inferior set.

6. The method of claim 1, wherein the method is a method of bit selection and optimization that combines big data classification and the golden section method. The method for performing normalization processing on the wear grading of the drill bit comprises: Threshold values are set for the inner row teeth I, outer row teeth O and gauge G characteristics in the drill bit wear grading, and then normalization processing is performed, and the specific threshold setting method is as follows: according to the IADC drill bit wear grading, the inner row teeth I and the outer row teeth O are divided into 0, 1, 2, …, 8 in total, 0 represents no wear, and 9 represents complete wear; the gauge G is divided into 0, 1 / 16 in, 2 / 16 in, 3 / 16 in, 4 / 16 in in total, 0 represents no wear, and 4 / 16 represents a diameter wear of 4 / 16 in; and the grading assignment formula is as follows: The inner row teeth I wear grading assignment is: (i=1,2,3……9)； is an inner row tooth I wear amount characteristic; is an inner row tooth I wear classification threshold value; The outer row teeth O wear grading assignment is: (i=1,2,3……9)； to characterize the wear of the outer row teeth O; tooth O wear grading threshold; The gauge G wear grading assignment is: (i=1,2,3……5)； To characterize the wear of the G gauge; to regulate the wear grading threshold; The wear grading normalization processing formula is: The inner row teeth I wear grading normalization formula is: (i = 1, 2, 3...n); min_outlet_wear_sample_set is the minimum wear value for the outlet teeth for the sample set; max_outlet_wear_sample_set is the maximum wear value for the outlet teeth for the sample set; The outer row teeth O wear grading normalization formula is: (i = 1, 2, 3...n); Min for sample set for gear tooth wear; Max for sample set of gear tooth wear; The gauge G wear grading normalization formula is: (i = 1, 2, 3...n); minimizing wear for the sample set; maximizing wear for the sample set; wear weight normalization: wherein ; x, y, z are weight factors, usually x∈(0.25~0.35), y∈(0.30~0.40), and z∈(0.35~0.45).

7. The method of claim 1, wherein the method is a method of bit selection and optimization that combines big data classification and the golden section method. The method for using a neural network method for cluster analysis to complete secondary screening and classification comprises: using a neural network optimization method to cluster and classify the drill bit wear characteristics, including drill bit diameter reduction PB, self-sharpening wear SS and scouring WO; and according to different formations, borehole trajectories, drilling fluid systems and drilling process conditions and requirements, proposing suggestions for drill bit improvement.

8. The method of claim 1, wherein the method is a method of bit selection and optimization that combines big data classification and the golden section method. The optimal solution is obtained by fusing drilling cycle and cost constraints, including: Drilling cycle formula: (i=1,2,3……5)； (j=1,2,3……5)； for the optimum drilling cycle, for the single bit footage, for the net drilling age, for the lowest cost, for the rig day cost, for the tool cost.

9. The method of claim 1, wherein the method is a method of bit selection and optimization that combines big data classification and the golden section method. The method further includes: Step 5. According to the structure optimization suggestion, individual improvement is carried out on the basis of the target drill bit, numerical simulation is carried out by using three-dimensional modeling software, and the final two-dimensional drawing and three-dimensional model are determined according to the simulation results of temperature field and flow field; Step 6. Trial production of drill bit, track and evaluate the on-site use, and continuously iterate and upgrade.

Images