A rock drillability prediction method based on multi-source information fusion

Abstract

The application discloses a rock drillability prediction method based on multi-source information fusion, first, the drilling data and vibration data are fused to predict the rock drillability, aiming at the problem that the time scales of the drilling data and the vibration data are inconsistent, a segmented cubic Hermite interpolation method is adopted to unify the time scale, then, a mutual information method is adopted to analyze the data correlation, and suitable input variables are obtained, finally, an extreme learning machine method is adopted to establish a rock drillability prediction model to obtain the uniaxial compressive strength value. The drilling data and the vibration data are used to predict the rock drillability, the drilling efficiency is improved, the stability of the drilling process is ensured, the safety accidents are effectively reduced, and the method provides guidance for the research on intelligent control of the drilling process, and has practicability and applicability.

Description

Technical Field

[0001] This invention relates to the field of coal mine drilling engineering, and in particular to a method for predicting rock drillability based on multi-source information fusion. Background Technology

[0002] Energy resource supply is a key constraint on the sustainable development of a country's economy and an important component of national security.

[0003] my country currently faces numerous challenges in coal exploration and development, including complex geological conditions, low resource density, high exploration difficulty, and environmental protection concerns. Therefore, my country needs to accelerate the research and development of a new generation of transformative drilling technologies to meet the demands of complex coal resource exploration and development. Intelligent drilling technology in coal exploration and development can improve drilling success rates and reservoir encounter rates by controlling drill bit progress and direction in real time. Therefore, the development of intelligent drilling technology has significant strategic importance and broad application prospects. Among intelligent drilling technologies, lithology identification is a crucial one. Multi-source information fusion uses different geological data and information to determine the type, distribution, and physical characteristics of underground rocks, effectively identifying formation hardness and reducing drill bit wear and accidents.

[0004] In practical applications, the use of fused drilling and vibration data for rock drillability prediction has yielded certain results. Studies have shown that by real-time monitoring of vibration signals during drilling and combining this with changes in drilling parameters, the hardness, fracture degree, and heterogeneity of rock formations can be predicted in advance, thereby anticipating potential difficulties encountered during drilling operations. This method has been widely applied in the exploration of resources such as coal, oil, and natural gas, and has demonstrated particularly good results in drilling operations in coal mining areas. Summary of the Invention

[0005] This invention provides a rock drillability prediction method based on multi-source information fusion, which aims to solve the problem of rock drillability prediction during drilling.

[0006] To solve this technical problem, the technical solution adopted by the present invention is as follows:

[0007] A method for predicting rock drillability based on multi-source information fusion includes the following steps:

[0008] S1: Integrate drilling data and vibration data to predict rock drillability, and use piecewise cubic Hermite interpolation to unify the time scale of drilling data and vibration data;

[0009] S2: The mutual information method is used to perform correlation analysis on drilling data and vibration data, and the variables with the nearest correlation are used as inputs for the rock drillability prediction model.

[0010] S3: The rock drillability prediction model is established using the extreme learning machine method to obtain the uniaxial compressive strength value.

[0011] Optionally, in step S1, the step of unifying the time scale of drilling data and vibration data using piecewise cubic Hermite interpolation includes:

[0012] G(r) is the piecewise cubic Hermite interpolation polynomial of f(r) over n+1 nodes, and its specific expression is as follows:

[0013]

[0014] In the formula, r represents the current interpolation point, r k and r k+1 Let t represent the left and right endpoints of the k-th interval. k and t k+1 t represents the known values ​​of the function at both endpoints. k =f(r k ), t' k and t' k+1 s represents the first derivative (slope) of the function at both endpoints. k (r) and s k+1 (r) represents the interpolation basis function (multiplied by t) k and t k+1 (Previously, the influence of the control function value), S k (r) and S k+1 (r) is the interpolation basis function, G(r) represents the Hermite interpolation function, the estimated value at point x, and k is the number of the current interval, which is the kth segment in piecewise interpolation.

[0015] Optionally, in step S2, the data correlation analysis using the mutual information method includes:

[0016]

[0017] In the formula, ρ(x,y), ρ(x), and ρ(y) represent the joint probability of x and y, the marginal probability of x, and the marginal probability of y, respectively; X represents the variables such as feed rate, feed pressure, rotational speed, main pump pressure, auxiliary pump pressure, frequency amplitude, instantaneous frequency, mean, and kurtosis; Y represents the uniaxial compressive strength, where x is a specific value in X and y is a specific value in Y; the larger the value of I(X;Y), the stronger the correlation between x and y; I(X;Y) = 0 indicates no correlation, and I(X;Y) = I(Y;X).

[0018] Optionally, in step S3, the step of using the extreme learning machine method to establish a rock drillability prediction model to obtain the uniaxial compressive strength value includes:

[0019] The input to the rock drillability prediction model is the feed rate V. ROP mm / s, feed pressure F p MPa, rotational speed R, r / min, main pump pressure M p MPa, auxiliary pump pressure S p MPa, bandwidth amplitude A B m / s 2 Instantaneous passband B I m / s 2 Mean M e m / s 2 And kurtosis K, unitless, output is U cs The relationship between input and output, MPa, is expressed as follows:

[0020] U cs =f ucs (V ROP ,F p ,R,M p ,S p A B B I M e ,K);

[0021] The Extreme Learning Machine method is chosen to construct a rock drillability prediction model, denoted as:

[0022] f ucs =ELM(n,ω,b);

[0023] In the formula, n is the number of hidden layer neurons, ω is the weight connecting the input neurons and the hidden layer neurons, and b is the bias of the hidden layer neurons.

[0024] In an extreme learning machine, the output weight matrix β can be calculated according to the following formula;

[0025]

[0026] In the formula, N is the length of the selected dataset, L is the number of hidden layer nodes; T represents matrix transpose, C represents a constant, and I represents the identity matrix;

[0027] The output matrix H can be calculated using the following formula:

[0028]

[0029] In the formula, It is the activation function, specifically the Tanh activation function; ω and b are randomly generated.

[0030] Optionally, the random generation of ω and b includes:

[0031] A two-layer search strategy is adopted for selection. Inner layer search: First, a reasonable number of search times is set. Based on the characteristic that ω and b are randomly generated in the extreme learning machine, the root mean square error of the ten-fold cross is calculated, and ω and b with the minimum root mean square error are selected. Outer layer search: Based on the inner layer search, a reasonable range of hidden layer nodes and the number of search times are set. When the final root mean square error is minimized, the optimal number of hidden layer nodes n, ω, and b are obtained.

[0032] Optionally, in S1, the drilling data includes feed rate, feed pressure, rotational speed, main pump pressure, and auxiliary pump pressure.

[0033] Optionally, in S1, the vibration data includes the frequency amplitude, instantaneous frequency, mean, and kurtosis.

[0034] Optionally, in S2, the variables with close correlation include feed rate, feed pressure, rotational speed, main pump pressure, auxiliary pump pressure, frequency amplitude, instantaneous frequency, mean, and kurtosis.

[0035] The technical solution provided by this invention has the following beneficial effects:

[0036] First, drilling and vibration data are integrated to predict rock drillability. To address the inconsistency in time scales between the two data sets, piecewise cubic Hermite interpolation is used to unify the time scale. Then, mutual information analysis is employed to analyze data correlation. Finally, an extreme learning machine method is used to establish a rock drillability prediction model and obtain uniaxial compressive strength values. This invention utilizes drilling and vibration data to predict rock drillability, improving drilling efficiency while ensuring stability during the drilling process, effectively reducing safety accidents, and providing guidance for research on intelligent control of the drilling process. It is practical and applicable. Attached Figure Description

[0037] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0038] Figure 1 This invention provides the implementation steps of a rock drillability prediction method based on multi-source information fusion.

[0039] Figure 2 This is the segmented cubic Hermite interpolation result in the real-time example of this invention;

[0040] Figure 3 This is a diagram showing the predicted uniaxial compressive strength results by integrating drilling data and vibration data in an embodiment of the present invention.

[0041] Figure 4 This is an example of an error diagram for predicting uniaxial compressive strength by integrating drilling data and vibration data in an embodiment of the present invention.

[0042] Figure 5 In this embodiment of the invention, only drilling data is used to predict the uniaxial compressive strength diagram;

[0043] Figure 6 This is a prediction error diagram for predicting uniaxial compressive strength using only drilling data in an embodiment of the present invention. Detailed Implementation

[0044] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0045] refer to Figure 1 , Figure 1 The implementation steps of the rock drillability prediction method based on multi-source information fusion of the present invention specifically include the following steps:

[0046] S1: Integrate drilling data and vibration data to predict rock drillability, and use piecewise cubic Hermite interpolation to unify the time scale of drilling data and vibration data;

[0047] S2: The mutual information method is used to perform correlation analysis on drilling data and vibration data, and the variables with the nearest correlation are used as inputs for the rock drillability prediction model.

[0048] S3: The rock drillability prediction model is established using the extreme learning machine method to obtain the uniaxial compressive strength value.

[0049] In step S1, the step of using piecewise cubic Hermite interpolation to unify the time scale of drilling and vibration data includes the following interpolation principle: assuming that a known function f(r) satisfies f(r) at n+1 distinct nodes (i = 0, 1, ..., n) in the interpolation interval [p, q]. i )-f i and f'(r i )-f i (i = 0, 1, ..., n), if the function G(x) exists that satisfies the following condition:

[0050] (1) The polynomial degree of G(r) is 3 in each cell;

[0051] (2)G(r)∈C 1 [a,b];

[0052] (3)G(r i )=f(r i ),G'(r i )=f'(r i ), i = (0, 1, ..., n);

[0053] G(r) is then called the piecewise cubic Hermite interpolation polynomial of f(x) at n+1 nodes, as shown in the following equation.

[0054]

[0055] In the formula, r represents the current interpolation point, r k and r k+1 Let t represent the left and right endpoints of the k-th interval. k and t k+1 t represents the known values ​​of the function at both endpoints. k =f(r k ), t' k and t' k+1 s represents the first derivative (slope) of the function at both endpoints. k (r) and s k+1 (r) represents the interpolation basis function (multiplied by t) k and t k+1 (Previously, the influence of the control function value), S k (r) and S k+1 (r) Interpolation basis function (multiplied in t') k and t' k+1 (Previously, the influence of the control derivative) G(r) represents the Hermite interpolation function, the estimated value at point x, and k is the number of the current interval (in piecewise interpolation, the kth segment).

[0056] In step S2, the data correlation analysis using the mutual information method includes:

[0057] To reduce model redundancy and improve model accuracy, mutual information analysis is used for correlation analysis. The formula is expressed as follows:

[0058]

[0059] In the formula, ρ(x,y), ρ(x), and ρ(y) represent the joint probability of x and y, and the marginal probability of x and y, respectively; X represents the variables such as feed rate, feed pressure, rotational speed, main pump pressure, auxiliary pump pressure, frequency amplitude, instantaneous frequency, mean, and kurtosis; Y represents the uniaxial compressive strength, where x is a specific value in X and y is a specific value in Y; the larger the value of I(X;Y), the stronger the correlation between x and y; I(X;Y) = 0 indicates no correlation, and I(X;Y) = I(Y;X).

[0060] In step S3, the step of using the extreme learning machine method to establish a rock drillability prediction model to obtain the uniaxial compressive strength value includes:

[0061] (1) The input to the rock drillability prediction model is the feed rate V. ROP(mm / s), feed pressure F p (MPa), rotational speed R (r / min), main pump pressure M p (MPa), auxiliary pump pressure S p (MPa), bandwidth A B (m / s 2 Instantaneous passband B I (m / s 2 ), mean M e (m / s 2 The output is U, which is the kurtosis K (unitless). cs (MPa). The relationship between input and output is expressed as:

[0062] U cs =f ucs (V ROP ,F p ,R,M p ,S p A B B I M e ,K);

[0063] The Extreme Learning Machine (ELM) method was chosen to construct a rock drillability prediction model. This is because it is simple in structure, learns quickly, and has been widely applied in drilling processes. This can also be expressed as...

[0064] f ucs =ELM(n,ω,b);

[0065] In the formula, n is the number of hidden layer neurons, ω is the weight connecting the input neurons and the hidden layer neurons, and b is the bias of the hidden layer neurons.

[0066] In an extreme learning machine, the output weight matrix β can be calculated using the following formula.

[0067]

[0068] In the formula, N is the length of the selected dataset, and L is the number of hidden layer nodes. The output matrix H can be calculated according to the following formula.

[0069]

[0070] In the formula, It is the activation function; the activation function chosen in this paper is the Tanh activation function.

[0071] In the Extreme Learning Machine (ELM) algorithm, the number of hidden layer nodes needs careful design, while ω and b are randomly generated. To obtain the optimal number of hidden layer nodes and bias, we employ a two-layer search strategy. Inner layer search: First, a reasonable number of search iterations is set. Given that ω and b are randomly generated in ELM, the root mean square error (RMSE) of the ten-fold crossover is calculated, and ω and b are selected when the RMS error is minimized. Outer layer search: Based on the inner layer search, a reasonable range for the number of hidden layer nodes and the number of search iterations are set. The optimal number of hidden layer nodes n, ω, and b are obtained when the final RMS error is minimized.

[0072] Example 1:

[0073] In this embodiment, 18,852 sets of drilling data and 504 sets of vibration data from a drilling site were selected as the specific data. These included key operating variables such as feed rate, feed pressure, power head speed, main pump pressure, and auxiliary pump pressure, as well as vibration data such as frequency amplitude, instantaneous frequency, mean, and kurtosis. First, a 7-second time scale was used to unify the time scale of the drilling and vibration data. Second, a mutual information method was used to select appropriate input variables. Then, an extreme learning machine method was used to establish a uniaxial compressive strength prediction model.

[0074] The specific steps are as follows:

[0075] (1) The piecewise cubic Hermite interpolation method was used to unify the time scale of drilling and vibration data.

[0076] The characteristic data representing vibration include frequency amplitude, instantaneous frequency, mean, RMS value, and kurtosis. Their cubic Hermite interpolation polynomial results are as follows: Figure 2 As shown.

[0077] (2) Data correlation analysis using mutual information method

[0078] When data comes from different systems or sensors, their timestamps may differ. Standardizing the time scale ensures that the data are aligned on the same timeline, facilitating correlation analysis. ROP F p , R, M p and S p There is a strong correlation between uniaxial compressive strength and drilling variables. Drill bits primarily break formations through cutting, shearing, compression, or grinding, rotating and advancing under the influence of feed pressure and rotational speed. Feed rate is highly sensitive to changes in formation, indirectly reflecting the ease with which the drill bit penetrates the rock mass. Main pump pressure and auxiliary pump pressure control drilling operations and feed pressure regulation. From a data perspective, the mutual information between uniaxial compressive strength and drilling variables is shown in Table 1. ROP F p , R, M p and Sp The mutual information values ​​between the uniaxial compressive strength and the uniaxial compressive strength also indicate that they are strongly correlated.

[0079] Table 1 Mutual information between uniaxial compressive strength and drilling variables

[0080]

[0081] After standardizing the time scale, the mutual information method was used to measure the correlation between uniaxial compressive strength and vibration data. The results are shown in Table 2. (Passband amplitude (A)) B Instantaneous passband (B) I ), mean (M) e The mutual information values ​​of ) and kurtosis (K) are relatively large, and have a significant impact on uniaxial compressive strength; therefore, these four variables are selected as inputs. RMS value (R M The mutual information value of ) is small, so it is not used as input.

[0082] Table 2 Mutual information between uniaxial compressive strength and vibration variables

[0083]

[0084] In summary, the input to the rock drillability prediction model is the feed rate V. ROP (mm / s), feed pressure F p (MPa), rotational speed R (r / min), main pump pressure M p (MPa), auxiliary pump pressure S p (MPa), bandwidth A B (m / s 2 Instantaneous passband B I (m / s 2 ), mean M e (m / s 2 The output is U, which is the kurtosis K (unitless). cs (MPa).

[0085] (3) Using the extreme learning machine method, a rock drillability prediction model was established to obtain the uniaxial compressive strength value.

[0086] The predicted uniaxial compressive strength results are as follows Figure 3 As shown, the error results are as follows: Figure 4 As shown.

[0087] The comparative experiment results show that predictions were made using only drilling data, and the results are as follows: Figure 5 and Figure 6 As shown.

[0088] To more comprehensively evaluate the model, three performance metrics were used, which are expressed as follows:

[0089] (1) Root mean square error:

[0090]

[0091] (2) Mean absolute error:

[0092]

[0093] (3) Mean absolute percentage error:

[0094]

[0095] In the formula, This is the predicted value of uniaxial compressive strength. It is the actual value of uniaxial compressive strength.

[0096] Table 3 Comparison Results of Indicators from Different Data Sources

[0097]

[0098] Table 3 shows the comparison results of indicators using different data sources. As can be seen from the table, whether it's root mean square error, mean absolute error, or mean absolute percentage error, the results of fusing drilling and vibration data are better than those using only drilling data, indicating that combining multi-source data can effectively improve model performance. The combination of vibration and drilling data provides richer information, enabling the model to capture the complex characteristics of uniaxial compressive strength, thereby improving prediction accuracy.

[0099] The beneficial effects of this invention are as follows: First, it integrates drilling data and vibration data to predict rock drillability. Addressing the inconsistency in time scales between the two data sets, a piecewise cubic Hermite interpolation method is used to unify the time scale. Then, mutual information analysis is employed to analyze data correlation. Finally, an extreme learning machine method is used to establish a rock drillability prediction model and obtain uniaxial compressive strength values. This invention utilizes drilling and vibration data to predict rock drillability, improving drilling efficiency while ensuring the stability of the drilling process, effectively reducing safety accidents, and providing guidance for research on intelligent control of the drilling process. It is practical and applicable.

[0100] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0101] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In the unit claims listing several devices, several of these devices may be embodied by the same hardware item. The use of the terms first, second, and third, etc., does not indicate any order and can be interpreted as identifiers.

[0102] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the protection scope of the present invention.

Claims

1. A method for predicting rock drillability based on multi-source information fusion, characterized in that, Includes the following steps: S1: Integrate drilling data and vibration data to predict rock drillability, and use piecewise cubic Hermite interpolation to unify the time scale of drilling data and vibration data; In step S1, the step of using piecewise cubic Hermite interpolation to unify the time scale of drilling and vibration data includes: G( r )yes f ( r )exist n The piecewise cubic Hermite interpolation polynomial with +1 nodes is expressed as follows: ； In the formula, r Indicates the current interpolation point. and Indicates the first k The left and right endpoints of the segment interval, and This represents the known values ​​of the function at both endpoints. , and This represents the first derivative of the function at both endpoints. , , and Both represent interpolation basis functions, which respectively control the influence weights of endpoint function values ​​and derivative values ​​on the final interpolation result; This represents the Hermite interpolation function at the point... r The estimated value; k It is the index of the current interval, which in the piecewise interpolation function represents the th interval. k The interval of the segment; S2: Correlation analysis of drilling data and vibration data is performed using the mutual information method to obtain variables with adjacent correlations, which are then used as inputs for the rock drillability prediction model. Step S2, the data correlation analysis using the mutual information method, includes: ； In the formula, , and They represent x and y joint probability, x marginal probability and y The marginal probability; X These variables represent feed rate, feed pressure, rotational speed, main pump pressure, auxiliary pump pressure, bandwidth amplitude, instantaneous bandwidth, mean bandwidth, and kurtosis. Y Represents uniaxial compressive strength. x yes X The specific value in, y yes Y The specific value in; The larger the value, the more it indicates X and Y The stronger the correlation between them; Indicates that it is irrelevant. ; S3: Establish a rock drillability prediction model using the extreme learning machine method to obtain the uniaxial compressive strength value; Step S3, establishing a rock drillability prediction model using the extreme learning machine method to obtain the uniaxial compressive strength value, includes: The input to the rock drillability prediction model is the feed rate. mm / s, feed pressure MPa, rotational speed R r / min, main pump pressure MPa, auxiliary pump pressure MPa, bandwidth amplitude m / s², instantaneous bandwidth m / s², mean m / s² and kurtosis K Unitless, output is The relationship between input and output, MPa, is expressed as follows: ； The Extreme Learning Machine method is chosen to construct a rock drillability prediction model, denoted as: ； In the formula, n It is the number of neurons in the hidden layer. ω These are the weights connecting the input neurons and the hidden layer neurons. b It is the bias of neurons in the hidden layer; Output weight matrix in extreme learning machine β The following formula can be used for calculation: ； In the formula, N It is the length of the selected dataset. L It represents the number of hidden layer nodes; T represents matrix transpose, C represents a constant, and I represents the identity matrix; Output matrix H The following formula can be used for calculation: ； In the formula, It's an activation function, specifically the Tanh activation function; randomly generated. ω and b .

2. The rock drillability prediction method based on multi-source information fusion according to claim 1, characterized in that, The aforementioned random generation ω and b include: A two-layer search strategy is employed for selection. The inner layer search involves first setting a reasonable number of search iterations, based on the Extreme Learning Machine... ω and b The random generation characteristic is used to calculate the root mean square error of the ten-fold cross, and the option with the smallest root mean square error is selected. ω and b Outer layer search: Based on the inner layer search, set a reasonable range for the number of hidden layer nodes and the number of searches. The optimal number of hidden layer nodes is obtained when the root mean square error is minimized. n , ω and b .

3. The rock drillability prediction method based on multi-source information fusion according to claim 1 or 2, characterized in that, In S1, the drilling data includes feed rate, feed pressure, rotational speed, main pump pressure, and auxiliary pump pressure.

4. The rock drillability prediction method based on multi-source information fusion according to claim 1 or 2, characterized in that, In S1, the vibration data includes the frequency amplitude, instantaneous frequency, mean, and kurtosis.

5. The rock drillability prediction method based on multi-source information fusion according to claim 1 or 2, characterized in that, In S2, the variables with close correlation include feed rate, feed pressure, rotational speed, main pump pressure, auxiliary pump pressure, frequency amplitude, instantaneous frequency, mean, and kurtosis.

Images