While-drilling lithology identification method based on improved Harris eagle algorithm and support vector machine

Abstract

The invention discloses a while-drilling lithology identification method based on an improved Harris eagle algorithm and a support vector machine, and belongs to the technical field of lithology identification. The while-drilling lithology identification method based on the improved Harris eagle algorithm and the support vector machine comprises the following steps: collecting while-drilling data; according to the while-drilling data, optimizing and training a support vector machine model by using a multi-strategy improved Harris eagle algorithm to obtain a CDHHO-SVM model; inputting the acquired while-drilling data into the CDHHO-SVM model to output a lithology identification result; drilling parameters are obtained in real time through a drilling test, rock mechanics parameters are inversed, the lithology and the state of the surrounding rock of the roadway can be obtained, a traditional rock mechanics parameter detection method is effectively improved, fitting of a lithology identification model of a support vector machine optimized based on the improved Harris eagle algorithm to a nonlinear relation is more excellent, and the method is more suitable for large-scale popularization and application. The lithology identification and prediction accuracy is greatly improved, and the method has good expression ability and generalization ability.

Description

Technical Field

[0001] The invention belongs to the technical field of lithology identification, and in particular relates to a lithology identification while drilling method based on an improved Harris Eagle algorithm and a support vector machine. Background Art

[0002] Coal mine geological exploration is the premise for ensuring safe and efficient mining of coal mines. At present, most traditional detection methods are borehole peek method, drilling coring method, etc., but these methods are time-consuming and costly. The lithology identification while drilling technology is to invert the stratum structure parameters according to the working parameters of the drill rig during the drilling process, which can effectively obtain the characteristic parameters of the coal mine strata. Therefore, the lithology identification while drilling method greatly reduces the construction cost, supplements the traditional stratum lithology identification method, and is of great significance to the safe, fast and intelligent mining of coal mines.

[0003] Support vector machines have good performance in processing nonlinear data. A support vector machine model is established to learn existing drilling data, establish the response relationship between drilling data and formation parameters, and realize the lithology identification of unknown drilling formations. However, a single support vector machine model has the problem of insufficient prediction accuracy. Summary of the invention

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a method for lithology identification while drilling based on an improved Harris Eagle algorithm and a support vector machine, thereby solving the problems in the prior art.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] The method for identifying lithology while drilling based on the improved Harris Eagle algorithm and support vector machine includes the following steps:

[0007] Collecting data while drilling;

[0008] According to the drilling data, a multi-strategy improved Harris Eagle algorithm is used to optimize and train a support vector machine model to obtain a CDHHO-SVM model;

[0009] The collected while-drilling data are input into the CDHHO-SVM model to output the lithology identification results.

[0010] Furthermore, the drilling data includes: water inlet pressure, drilling displacement, thrust pressure, impact pressure, drilling speed and rotation pressure.

[0011] Furthermore, the strategies used to improve the Harris Hawk algorithm include: borrowing reverse learning strategy, mutualism strategy and differential evolution strategy.

[0012] Furthermore, a reverse learning strategy is introduced into the Harris Hawk algorithm to increase the quality of the initial solution of the population, and a mutualistic symbiosis strategy is introduced in the exploration phase to enhance the probability of the Harris Hawk algorithm finding the best solution. Through the joint perturbation of the reverse learning strategy and differential evolution, the probability of the Harris algorithm jumping out of the local optimal solution is increased.

[0013] Further, the steps of using the reverse learning strategy, the mutualism strategy and the differential evolution strategy to improve the Harris Hawk algorithm include:

[0014] Step 1: Initialize the basic parameters and initial population of the Harris Hawk algorithm, determine the dimension D of the population, the number N of the population, and generate a reverse population within a given range through a reverse learning strategy;

[0015] Step 2: Calculate the fitness value and update the energy factor E and the escape distance J of the prey in the Harris Hawk algorithm;

[0016] Step 3: Generate a new population X based on the energy factor E and the prey's escape distance J;

[0017] Step 4: If t is an odd number, then for population X t Execute the differential mutation strategy and sort the individuals according to their fitness values ​​to obtain If t is an even number, then for the population X t Execute the reverse learning strategy and update right The individuals in the population are sorted from small to large according to their fitness values, and the first N individuals are selected to form a population.

[0018] Step 5: Determine whether the algorithm meets the convergence conditions. If not, return to step 2; otherwise, output the optimal solution; terminate the loop.

[0019] Furthermore, the steps of optimizing and training the support vector machine model by using the multi-strategy improved Harris Eagle algorithm include:

[0020] S21, dividing the drilling data into a training set and a test set in a ratio of 8:2;

[0021] S22, setting the value range of the penalty factor C and the kernel function g in the support vector machine model, as well as the specific parameters of the multi-strategy improved Harris Eagle algorithm;

[0022] S23, calculating the fitness value of the CDHHO-SVM model;

[0023] S24, updating the population position according to the reverse learning, mutualism and differential mutation strategies, calculating the fitness value of the current position, and comparing it with the previous fitness value to select the better one;

[0024] S25, select the maximum number of iterations as the end indicator, and use the optimal value output by the multi-strategy improved Harris Eagle algorithm as the penalty factor C and kernel function g of the support vector machine model.

[0025] The lithology identification system while drilling based on the improved Harris Eagle algorithm and support vector machine includes:

[0026] Data acquisition module: collects drilling data;

[0027] Model training and optimization module: based on the drilling data, the multi-strategy improved Harris Eagle algorithm is used to optimize and train the support vector machine model to obtain the CDHHO-SVM model;

[0028] And, the identification result output module: inputs the collected while-drilling data into the CDHHO-SVM model to output the lithology identification result.

[0029] A computer storage medium stores a readable program, which can execute the above-mentioned lithology identification method while drilling based on the improved Harris Eagle algorithm and support vector machine when the program is running.

[0030] An electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus;

[0031] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the above-mentioned method for identifying lithology while drilling based on the improved Harris Eagle algorithm and support vector machine.

[0032] A computer program product includes computer instructions, wherein the computer instructions instruct a computing device to execute operations corresponding to the above-mentioned method for identifying lithology while drilling based on an improved Harris Eagle algorithm and a support vector machine.

[0033] Beneficial effects of the present invention:

[0034] 1. Through the drilling test, the drilling parameters are obtained in real time, and the rock mechanics parameters are inverted to obtain the lithology and state of the tunnel surrounding rock, which effectively improves the traditional rock mechanics parameter detection method. Compared with the traditional method, it greatly reduces manpower and material resources and can effectively increase the tunnel surrounding rock excavation efficiency.

[0035] 2. Aiming at the defects of Harris Hawk algorithm that it is easy to fall into local optimal solution and lacks global exploration ability, a multi-strategy fusion Harris Hawk algorithm (CDHHO) is proposed by drawing on the reverse learning strategy, mutualistic symbiosis strategy and differential evolution strategy, which has better performance than the initial Harris Hawk algorithm.

[0036] 3. The lithology identification model based on the support vector machine optimized by the improved Harris Eagle algorithm has a better fit for nonlinear relationships, greatly increases the accuracy of lithology identification prediction, and has good expression and generalization capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 It is a flow chart of the lithology identification method of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] like Figure 1 As shown, the method for lithology identification while drilling based on the improved Harris Eagle algorithm and support vector machine includes the following steps:

[0041] S1, collecting data while drilling;

[0042] Among them, the drilling data include: water inlet pressure, drilling displacement, thrust pressure, impact pressure, drilling speed and rotation pressure;

[0043] The above-mentioned drilling data were obtained through: rock drilling test collection by laboratory rotary drilling system and coal mine geological report; specifically:

[0044] The indoor rotary drilling system includes a mechanical drilling system, a hydraulic control system, and an electrical control system. The intelligent sensing system for drilling parameters collects, plots, and stores the drilling parameters such as water inlet pressure, drilling displacement, thrust pressure, impact pressure, drilling speed, and rotation pressure during rotary drilling by installing various sensors at various parts of the mechanical drilling system. The size of the rock specimens for the drilling test is 300mm×300mm×300mm.

[0045] When the drill bit contacts the rock, the thrust pressure will increase. When the thrust pressure is greater than the preset impact start pressure, it means that the rock drill starts drilling and data collection starts automatically. When the drill bit penetrates the test piece and the thrust pressure reaches below the preset impact start pressure range, the rock drill automatically stops impacting, stops data collection, returns to its original position, and then performs the next drilling and collection work.

[0046] S2, based on the drilling data collected in S1, the multi-strategy improved Harris Hawk algorithm (CDHHO) is used to optimize and train the support vector machine (SVM) model to obtain the CDHHO-SVM model;

[0047] Harris Hawk Algorithm

[0048] Harris Hawk Algorithm (HHO) is a bionic optimization algorithm inspired by the behavior of Harris Hawks in nature that prey on prey. Its advantages are flexible algorithm structure, excellent performance and high-quality results. The HHO algorithm consists of three stages: exploration stage, transition stage and development stage. The transition between stages is controlled by the energy factor E.

[0049] In the exploration phase, Harris's hawks randomly inhabit some locations according to the locations of other members and prey, and use two strategies to find prey. The mathematical model is as follows:

[0050]

[0051] Where X rand (t) represents an individual randomly selected from the current population, X rabbit (t) represents the location of the prey, X m (t) represents the average position of the current population, random parameters q, r i ∈ rand om(0,1),i=1,2,3,4,[LB,UB] represents the upper and lower bounds of the population, and N is the number of populations.

[0052] The search to development stage is controlled by the energy factor E, the formula is as follows:

[0053] E=2E0(1-t / T) (3)

[0054] Where E is the escape energy of the prey, t represents the current iteration, T represents the maximum number of iterations, and E0 represents a random number between (-1, 1).

[0055] In the exploitation phase, prey often tries to escape, and Harris's hawk will chase the prey and try to catch it. Harris's hawk has four strategies to surround prey in the exploitation phase, and their mathematical models are as follows:

[0056] ① Soft encirclement: When |E|≥0.5 and r≥0.5, the prey has sufficient energy to escape.

[0057] X(t+1)=(X rabbit (t)-X(t))-E|JX rabbit (t)-X(t)| (4)

[0058] Among them, J is the escape distance of the prey, J=2(1-r5), and r5 is a random number between (0,1).

[0059] ②Hard Encirclement: When |E|<0.5 and r≥0.5, the prey has low energy and the Harris Hawk will attack the prey.

[0060] X(t+1)=X rabbit (t)-E|X rabbit (t)-X(t)| (5)

[0061] ③ Rapid dive soft encirclement: When |E|≥0.5 and r<0.5, the prey has enough energy to escape, so the Harris Hawk first performs soft encirclement and then dives to prey. This process uses Levy walk to simulate the prey escape model.

[0062]

[0063] Y=X rabbit (t)-E|JX rabbit (t)-X(t)| (7)

[0064] Z=Y+S×LF(D) (8)

[0065] Where D is the solution dimension and S is the D-dimensional random flight vector.

[0066] ④ Rapid dive and hard encirclement: When |E|<0.5, r<0.5, the prey does not have enough energy to escape. The Harris Hawk first performs hard encirclement and then quickly dives to attack the prey.

[0067]

[0068] Y=X rabbit (t)-E|JX rabbit (t)-X m (t)| (10)

[0069] Multi-strategy improved Harris Hawk algorithm;

[0070] Aiming at the defects of Harris Hawk algorithm that it is easy to fall into local optimal solution and has insufficient global exploration ability, a multi-strategy fusion Harris Hawk algorithm (CDHHO) is proposed by referring to the reverse learning strategy, mutualism strategy and differential evolution strategy. The reverse learning strategy is introduced into the basic Harris Hawk algorithm to increase the quality of the initial solution of the population, and the mutualism strategy is introduced in the exploration stage to further enhance the probability of algorithm optimization. Finally, the reverse learning strategy and differential evolution are used to jointly perturb to increase the probability of the algorithm jumping out of the local optimal solution. The specific contents include:

[0071] 1) Reverse learning strategy

[0072] Compared with other algorithms, the efficiency of the reverse learning strategy to obtain new solutions takes time. For example, genetic algorithms obtain new solutions through gene mutation, which requires several generations or more to introduce new directions. In recent years, the reverse learning strategy (OBL) has been effectively applied to various swarm intelligence algorithms. When solving problems, it is considered that there may be better solutions on the opposite side of the invalid solution. Introducing opposing solutions into the population can improve the quality of the population more than introducing two independent random solutions.

[0073] Assume that there is a number x on [l,u], then the opposite point of x is defined as Extend the definition of opposite points to n-dimensional space, let p = (x1, x2, ... x n ), is a point in n-dimensional space, where x i ∈[l,u],i=1,2,…,n, then its opposite point is p'=(x'1,x'2,…,x' n ). where x' i = l i +u i -x i .

[0074] Assume x is a random number belonging to [l,u], is its reverse solution, f(x) is the objective function, g(·) is the appropriate evaluation function, and f(x) and If g(f(x)) is greater than Then keep the value of x, otherwise take

[0075] 2) Mutually beneficial symbiosis strategy

[0076] The symbiotic organisms search algorithm (SOS) is a heuristic search algorithm based on the phenomenon of symbiosis in organisms. The algorithm is derived from the simulation of the co-existence and reproduction strategies of organisms in natural ecosystems through mutual influence. The most common symbiotic relationships between organisms in nature are divided into three types: mutualism, partial mutualism, and parasitism. In the mutualism stage, individual X i and individual X j Maintaining the interactive relationship, and then learning from the best individual, the new solution generated and The calculation formulas are:

[0077]

[0078] γ=round(1+rand(0,1)) (13)

[0079] Where: i, j∈{1,2,…,n} and j≠i; X best represents the current optimal individual in the ecosystem; X mean is the mutualism vector, representing the characteristic vector of the relationship between organisms; rand(0,1) represents a random number between [0, 1]; the value of γ is 1 or 2. In the mutualism stage, individual X i and individual X j By comparing the corresponding objective function values, it is determined whether the new solution is retained. Mutualism is introduced into the Harris Hawk algorithm, and the formula is updated as follows in the exploration phase:

[0080]

[0081] Where, X rand (t) is the randomly selected Harris Hawk position; r1 is a random number between 0 and 1; γ is the interest factor.

[0082] 3) Differential Evolution Strategy (DE)

[0083] The basic idea of ​​the DE algorithm is to use the differential mutation and crossover recombination of the current population individuals to obtain the intermediate population, and then select individuals from the parent population and the intermediate population based on the greedy strategy to obtain the new generation population. The algorithm process is as follows:

[0084] 3.1) Mutation Operation

[0085] The DE algorithm achieves individual mutation through differential mutation, that is, through several different individuals in the population, their vectors are differentially scaled and then synthesized with the vectors of the individuals to be mutated to generate an intermediate population P V (t+1), where the i-th mutated intermediate individual is as follows:

[0086] Vi =(t+1)=(V i,1 (t+1),(V i,2 (t+1),…,(V i,j (t+1)) (15)

[0087] This paper adopts the DE / rand / 1 mutation strategy to obtain the mutant individuals as follows:

[0088] V i =(t+1)=X r1 (t)+F·(X r2 (t)-X r3 (t)) (16)

[0089] Where, i≠r1≠r2≠r3; F is the scaling factor; the advantage of using this differential strategy is that it can maximize the range of X, making it more likely to jump out of the local optimal solution.

[0090] 3.2) Crossover Operation

[0091] For the tth generation population individual X i (t) and its variant intermediate individual V i =(t+1) to perform crossover operations between individuals and obtain the experimental individual U i =(t+1)=(U i,1 (t+1),(U i,2 (t+1),…,(U i,j (t+1)), where U i,j (t+1) is:

[0092]

[0093] 3.3) Select an operation

[0094] The DE algorithm uses a greedy strategy to select the next generation population P from the parent generation and the experimental individuals. X (t+1) individual X i =(t+1), as shown below:

[0095]

[0096] In summary, the steps to improve the Harris Eagle algorithm by using the reverse learning strategy, mutualism strategy and differential evolution strategy include:

[0097] Step 1: Initialize the basic parameters and initial population of the Harris Hawk algorithm, determine the dimension D of the population, the number N of the population, and generate a reverse population within a given range through a reverse learning strategy;

[0098] Step 2: Calculate the fitness value and update the energy factor E and the escape distance J of the prey in the Harris Hawk algorithm;

[0099] Step 3: Generate a new population X based on the energy factor E and the prey's escape distance J, calculated using formulas (3) and (4);

[0100] Step 4: If t is an odd number, then for population X t Execute the differential mutation strategy and sort the individuals according to their fitness values ​​to obtain If t is an even number, then for the population X t Execute the reverse learning strategy and update right The individuals in the population are sorted from small to large according to their fitness values, and the first N individuals are selected to form a population.

[0101] Step 5: Determine whether the algorithm meets the convergence conditions. If not, return to step 2; otherwise, output the optimal solution; terminate the loop.

[0102] Support Vector Machine (SVM) Model

[0103] Support Vector Machine (SVM) is a machine learning method that has been widely used to analyze and recognize patterns. The classification of data is achieved by using a training set to divide the data into two categories to obtain the Optimum Separate Hyperplane (OSH);

[0104] The linear support vector machine problem can be transformed into the following problem:

[0105]

[0106] sty i (w·x i +b)≥1-ξ i ,i=1,2,…,N (20)

[0107] ξ i ≥0,i=1,2,…,N (21)

[0108] Among them, w is the hyperplane normal vector, b is the classification threshold, ξ i ≥0 is the slack variable of input, C is the penalty factor; the size of C indicates the size of the penalty for misclassification; the Lagrange multiplier method is used to obtain the optimal decision function:

[0109] f(x)=sign[y i a i (x·x i +b)] (22)

[0110] The nonlinear problem is converted into a linear problem, and then converted into a high-dimensional space through nonlinear transformation to solve the classification problem. Finally, the optimal decision function becomes:

[0111]

[0112] Among them, k(x·x i ) is the kernel function.

[0113] SVM is suitable for solving problems such as small samples, nonlinearity, high latitude and local minima

[17] . In the SVM model using the radial basis function (RBF) as the kernel function, the penalty factor C and the kernel function g = k (x·x i ) jointly affect the performance of SVM.

[0114] In the invention, a multi-strategy improved Harris Eagle algorithm is used to optimize and train a support vector machine model, that is, to optimize the penalty factor C and the kernel function g in the support vector machine model. The specific steps include:

[0115] S21, the drilling data collected in S1 are divided into a training set and a test set in a ratio of 8:2, and the input variables are the above six variables (i.e., water inlet pressure, drilling displacement, thrust pressure, impact pressure, drilling speed, and rotation pressure);

[0116] S22, setting the value range of the penalty factor C and the kernel function g in the SVM, as well as the specific parameters of CDHHO, including the dimension D of the population and the number N of the population;

[0117] S23, calculating the fitness value of the CDHHO-SVM model;

[0118] S24, updating the population position according to the reverse learning, mutualism and differential mutation strategies, which can be updated by equations (8)-(13), calculating the fitness value of the current position, and comparing it with the previous fitness value to select a better one;

[0119] S25, select the maximum number of iterations as the end indicator, and use the optimal value output by CDHHO as the penalty factor C and kernel function g in the SVM model.

[0120] S3, input the collected while-drilling data into the CDHHO-SVM model to output the lithology identification results.

[0121] The method of the present invention may be implemented in hardware, firmware, or as software or computer code that may be stored in a recording medium (such as a CDROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code that is originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded over a network and will be stored in a local recording medium, so that the method described herein may be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, processor, or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the method shown herein.

[0122] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A method for identifying lithology while drilling based on improved Harris Eagle algorithm and support vector machine, characterized in that: The following steps are involved: Collecting data while drilling; According to the drilling data, a multi-strategy improved Harris Eagle algorithm is used to optimize and train a support vector machine model to obtain a CDHHO-SVM model; The collected while-drilling data are input into the CDHHO-SVM model to output the lithology identification results.

2. The method for lithology identification while drilling based on improved Harris Eagle algorithm and support vector machine according to claim 1, characterized in that: The drilling data include: water inlet pressure, drilling displacement, thrust pressure, impact pressure, drilling speed and rotation pressure.

3. The method for lithology identification while drilling based on improved Harris Eagle algorithm and support vector machine according to claim 1, characterized in that: The strategies used to improve the Harris Hawk algorithm include: borrowing reverse learning strategy, mutualism strategy and differential evolution strategy.

4. The method for identifying lithology while drilling based on improved Harris Eagle algorithm and support vector machine according to claim 3, characterized in that: The reverse learning strategy is introduced into the Harris Hawk algorithm to increase the quality of the initial solution of the population. The mutualistic symbiosis strategy is introduced in the exploration phase to enhance the probability of the Harris Hawk algorithm to find the best solution. The reverse learning strategy and differential evolution are used to jointly perturb the algorithm to increase the probability of the Harris algorithm to jump out of the local optimal solution.

5. The method for identifying lithology while drilling based on improved Harris Eagle algorithm and support vector machine according to claim 3 or 4, characterized in that: The steps to improve the Harris Hawk algorithm using reverse learning strategy, mutualism strategy and differential evolution strategy include: Step 1: Initialize the basic parameters and initial population of the Harris Hawk algorithm, determine the dimension D of the population, the number N of the population, and generate a reverse population within a given range through a reverse learning strategy; Step 2: Calculate the fitness value and update the energy factor E and the escape distance J of the prey in the Harris Hawk algorithm; Step 3: Generate a new population X based on the energy factor E and the prey's escape distance J; Step 4: If t is an odd number, then for population X t Execute the differential mutation strategy and sort the individuals according to their fitness values ​​to obtain If t is an even number, then for the population X t Execute the reverse learning strategy and update right The individuals in the population are sorted from small to large according to their fitness values, and the first N individuals are selected to form a population. Step 5: Determine whether the algorithm meets the convergence conditions. If not, return to step 2; otherwise, output the optimal solution; terminate the loop.

6. The method for lithology identification while drilling based on improved Harris Eagle algorithm and support vector machine according to claim 4, characterized in that: The steps of using multi-strategy improved Harris Eagle algorithm to optimize and train the support vector machine model include: S21, dividing the drilling data into a training set and a test set in a ratio of 8:2; S22, setting the value range of the penalty factor C and the kernel function g in the support vector machine model, as well as the specific parameters of the multi-strategy improved Harris Eagle algorithm; S23, calculating the fitness value of the CDHHO-SVM model; S24, updating the population position according to the reverse learning, mutualism and differential mutation strategies, calculating the fitness value of the current position, and comparing it with the previous fitness value to select the better one; S25, select the maximum number of iterations as the end indicator, and use the optimal value output by the multi-strategy improved Harris Eagle algorithm as the penalty factor C and kernel function g of the support vector machine model.

7. The lithology identification system while drilling based on the improved Harris Eagle algorithm and support vector machine is characterized by: include: Data acquisition module: collects drilling data; Model training and optimization module: based on the drilling data, the multi-strategy improved Harris Eagle algorithm is used to optimize and train the support vector machine model to obtain the CDHHO-SVM model; And, the identification result output module: inputs the collected while-drilling data into the CDHHO-SVM model to output the lithology identification result.

8. A computer storage medium storing a readable program, characterized in that: When the program is running, the method for lithology identification while drilling based on the improved Harris Eagle algorithm and support vector machine as described in any one of claims 1 to 6 can be executed.

9. An electronic device, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the method for drilling lithology identification based on the improved Harris Eagle algorithm and support vector machine as described in any one of claims 1 to 6.

10. A computer program product, comprising computer instructions, wherein the computer instructions instruct a computing device to perform operations corresponding to the method for lithology identification while drilling based on an improved Harris Eagle algorithm and a support vector machine as claimed in any one of claims 1 to 6.

Images