Plateau area well logging operation automation system

Abstract

The invention relates to the technical field of automatic well logging, in particular to a plateau area well logging operation automatic system which comprises a rock stratum characteristic analysis module, a data processing module, a data processing module, a data processing module, a data processing module and a data processing module, and the rock stratum characteristic analysis module is used for performing pixel point detection and area marking based on rock stratum texture and fault line image data of a plateau area and performing characteristic extraction and node correlation analysis on each node; and establishing a structured geological map. According to the method, details and accuracy of geological data are ensured through accurate analysis of rock stratum texture and fault line data in the plateau area, real-time feedback and parameter adjustment are performed by using the high-quality data, the drilling risk caused by rapid change of geological conditions is effectively reduced, and the drilling efficiency is improved. By dynamically adjusting the drilling parameters, the possibility of equipment failure and operation interruption is reduced, the continuity and safety of drilling operation are optimized, the establishment of a geological map is enhanced, and the data has higher prediction value and operation guiding significance by improving the relationship between data points, so that the accuracy of decision support and the overall efficiency of oilfield development are improved.

Description

Technical Field

[0001] The invention relates to the technical field of automated well logging, and in particular to an automated well logging operation system in plateau areas. Background Art

[0002] The field of automated logging technology involves the use of highly automated equipment and systems to perform logging tasks in drilling operations in underground oil and gas fields. Logging refers to the measurement of geological, geophysical, and geochemical characteristics using specialized tools and equipment during the drilling process. Automation technology makes the logging process more efficient and accurate, and can analyze underground structures in real time, providing key data to support oilfield development decisions. This includes automatic control of the operation of drilling equipment, real-time data collection, processing and interpretation, and instant adjustments based on data feedback. Automated logging systems typically integrate advanced software and hardware, such as artificial intelligence, machine learning, and remote operation technologies to improve operational safety and efficiency.

[0003] Among them, the automation system for logging operations in plateau areas involves the development and application of an automated logging system designed specifically for plateau areas. This system takes into account the unique environmental conditions in plateau areas, such as low oxygen, large temperature changes, and inconvenient transportation, and uses special equipment and highly adaptable control strategies. Its main purpose is to effectively perform logging operations under these complex and extreme conditions to ensure data accuracy and operational safety. Through automated control, the system can reduce personnel exposure to harsh environments, while improving the efficiency and accuracy of data collection, providing a scientific basis for oil and gas development.

[0004] Although existing automated logging technologies show high efficiency and accuracy in most cases, their application performance in special environments such as plateaus is not ideal. These technologies often lack the necessary adaptability, especially under conditions of low oxygen and drastic temperature fluctuations. The system may not be able to adjust to the optimal drilling parameters in a timely manner, resulting in inaccurate data collection and reduced operational efficiency. In addition, traditional systems often lack flexibility when processing highly variable geological data, making it difficult to effectively distinguish and filter noise or erroneous data, which in turn affects the accuracy of data interpretation. This lack of highly adaptable and optimized data processing capabilities is particularly evident in extreme conditions such as plateaus, often leading to major decision-making errors and increasing the risks and costs of oil and gas development. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an automated well logging operation system for plateau areas.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A well logging operation automation system in plateau areas comprises:

[0007] The rock formation characteristic analysis module performs pixel detection and region annotation based on the rock formation texture and fault line image data of the plateau area, and performs feature extraction and node association analysis on each node to establish a structured geological map;

[0008] The data purification module identifies and filters noise on the structured geological map, performs quality assessment on each data point and removes erroneous data, obtains a clean geological map, performs data enhancement processing on the basis of the clean geological map, adjusts the relationship strength between data points, and creates an enhanced geological map;

[0009] The parameter optimization module extracts rock hardness and wear rate from the enhanced geological map, performs parameter impact analysis and sensitivity test in combination with current drilling parameters, formulates a preliminary adjustment plan, and dynamically adjusts the learning rate and micro-batch data in the preliminary adjustment plan according to real-time drilling feedback to obtain optimized drilling parameters;

[0010] The efficiency evaluation module uses the optimized drilling parameters to perform drilling operations, collects drilling speed and drill bit wear data in real time, performs correlation analysis between speed and wear, generates drilling efficiency results, and evaluates the effectiveness of drilling parameters by comparing historical data and real-time data to obtain parameter efficiency evaluation.

[0011] As a further solution of the present invention, the steps of obtaining the structured geological map are specifically as follows:

[0012] Load the rock texture and fault line image data of the plateau area, and use the edge enhancement algorithm to process the image using the formula:

[0013]

[0014] Calculate edge strength to extract rock texture and fault lines and obtain characteristic boundary map;

[0015] Where E(x,y,λ,φ) represents the edge strength at position (x,y), λ and φ are parameters for adjusting the sensitivity of edge detection, and I represents the image intensity. Represent the partial derivatives of the image in the x and y directions respectively;

[0016] The feature boundary map is used to perform pixel detection and region annotation, and an image segmentation algorithm based on brightness and texture is used to refine the region division, using the formula:

[0017]

[0018] Perform weighted summation on the pixels in each area to obtain a regional map of geological characteristics;

[0019] Among them, S represents the score of the region, a i is the importance weight of the i-th pixel, P iis the attribute value of the i-th pixel, c i is the brightness reference value, η is the brightness adjustment coefficient;

[0020] Based on the geological feature area map, node association analysis is performed to construct a connection network between nodes using the formula:

[0021]

[0022] Evaluate the strength of associations between nodes to obtain a structured geological map;

[0023] Among them, R ij represents the association strength of node i to node j, w ij is the relationship weight between nodes i and j, and ∈ is the stability parameter.

[0024] As a further solution of the present invention, the steps of obtaining the clean geological map are specifically as follows:

[0025] The structured geological map is subjected to noise identification, and the image data is analyzed using the frequency domain filtering method, using the formula:

[0026]

[0027] The noise pattern was identified by Fourier transform and a Gaussian decay function was used to generate a preliminary filtered geological map;

[0028] Where N(x,y) represents the image intensity after noise attenuation at position (x,y). represents Fourier transform, I(x,y) is the original intensity of the image at position (x,y), α is the attenuation parameter, and x and y represent the coordinates of the pixel position;

[0029] The data point quality assessment was performed using the geological map after the preliminary filtering to exclude erroneous data using the formula:

[0030]

[0031] Calculate the consistency of each data point with its neighboring points, exclude abnormal data points by setting thresholds, and generate a clean geological map;

[0032] Where Q(i) represents the quality score of the i-th data point, v i 、v j They represent the attribute values ​​of data point i and neighboring point j respectively, N(i) is the set of neighboring points of point i, and ∈ is a small number to avoid the denominator being zero.

[0033] As a further solution of the present invention, the steps of acquiring the enhanced geological map are specifically as follows:

[0034] Starting from the clean geological map, the relationship between data points is strengthened using the formula:

[0035]

[0036] By adjusting the weights between data points, a data relationship reinforcement graph is generated;

[0037] Among them, S ij Indicates the strength of the relationship between data points i and j, v i 、v j is the attribute value of the corresponding data point, δ is the stability coefficient, avoiding the denominator to be zero, β is the spatial attenuation coefficient, d ij is the distance between points;

[0038] Using the data relationship enhancement diagram, spatial continuity analysis is performed using the formula:

[0039]

[0040] By referring to the effect of geometric distance on the strength of relationship, a spatial continuity enhancement graph is generated;

[0041] Where C(i) represents the spatial continuity score of data point i, and γ is the adjustment parameter of spatial attenuation;

[0042] Integrate the spatial continuity enhancement map and perform image rendering using the formula:

[0043] G=∑ i C(i)·log(1+ρv i )

[0044] Enhance the readability and detail of geological maps to obtain enhanced geological maps;

[0045] Where G represents the enhanced geological map, v i is the original attribute value of data point i, and ρ is the logarithmic gain parameter.

[0046] As a further solution of the present invention, the step of obtaining the optimized drilling parameters is specifically as follows:

[0047] From the enhanced geological map, the data of rock hardness and wear rate are extracted, and the rock formation characteristics are analyzed using image processing technology, using the formula:

[0048]

[0049] Generate rock hardness and wear data;

[0050] Among them, intensity(i) represents the light intensity value of the rock layer, texture(i) and edge(i) represent the texture and edge intensity information of the rock layer respectively, H(i) represents the hardness score of position i, and α, β, and γ are the weight coefficients of multiple attributes;

[0051] Based on the rock hardness and wear data, combined with the current drilling parameters, parameter impact analysis and sensitivity testing were performed using the formula:

[0052]

[0053] Determine key adjustment points and generate preliminary drilling adjustment plans;

[0054] Where D represents the drilling depth, P represents the drilling performance prediction, D and H represent the depth and hardness parameters respectively, and f k is the influence function of parameters D and H, γ k , τ k are the adjustment factors for impact and depth attenuation, respectively;

[0055] The learning rate and micro-batch data are dynamically adjusted according to the preliminary drilling adjustment plan and real-time drilling feedback, using the formula:

[0056]

[0057] After adjustment, the optimized drilling parameters are obtained;

[0058] Among them, λ is the learning rate, η is the fine-tuning parameter, and δ j represents the adjustment amount from drilling feedback, P j is the drilling performance index associated with j, and ΔP represents the optimization result of the drilling parameters.

[0059] As a further solution of the present invention, the steps for obtaining the drilling efficiency result are specifically as follows:

[0060] Start the drilling operation, use the optimized drilling parameters, monitor the drilling speed and drill bit wear data in real time, and use the formula:

[0061]

[0062] generating adjusted drilling speed data;

[0063] Among them, V(d) is the adjusted drilling speed data, v0 is the initial drilling speed, α and β adjust the influence of drilling depth and drill wear respectively, d represents the current drilling depth, and w is the drill wear index;

[0064] The adjusted drilling speed data was used to perform a speed-wear correlation analysis using the formula:

[0065]

[0066] Calculate the correlation between speed and wear, and generate correlation analysis results between drilling speed and wear data;

[0067] Among them, C is the correlation analysis result of drilling speed and wear data, V(i) is the drilling speed data, and W(i) is the corresponding drill bit wear rate;

[0068] According to the correlation analysis results of the drilling speed and wear data, the drilling efficiency is calculated using the formula:

[0069]

[0070] Get drilling efficiency results;

[0071] Among them, E is the drilling efficiency result.

[0072] As a further solution of the present invention, the step of obtaining the parameter efficiency evaluation is specifically as follows:

[0073] Based on the drilling efficiency results, the data of drilling speed and drill bit wear rate were collected from the historical database and real-time monitoring system, and the formula was used:

[0074]

[0075] Calculate and generate real-time drilling parameter data;

[0076] Among them, P real (t) is the real-time drilling parameter data, α represents the weight coefficient of speed, β represents the weight coefficient of wear, V(t) represents the square root of the real-time drilling speed, and W(t) represents the natural logarithm increment of the real-time drill wear rate;

[0077] The real-time drilling parameter data is compared with the stored historical drilling parameter data using the formula:

[0078]

[0079] Calculate the difference between historical data and real-time data to obtain parameter difference analysis results;

[0080] Among them, D is the result of parameter difference analysis, P hist (j) represents the historical drilling parameters, and m represents the number of data points;

[0081] According to the results of the parameter difference analysis, the formula is used:

[0082]

[0083] The average difference value was calculated to obtain parameter efficiency evaluation;

[0084] Among them, A is the parameter efficiency evaluation, and m represents the total number of data points involved in the calculation.

[0085] Compared with the prior art, the advantages and positive effects of the present invention are:

[0086] In the present invention, the geological data is ensured to be detailed and accurate through precise analysis of rock texture and fault line data in plateau areas, and these high-quality data are used for real-time feedback and parameter adjustment, effectively reducing the drilling risk caused by rapid changes in geological conditions, dynamically adjusting drilling parameters to reduce the possibility of equipment failure and operation interruption, optimizing the continuity and safety of drilling operations, and enhancing the creation of geological maps by improving the relationship between data points, making the data more predictive and operationally guiding, thereby improving the accuracy of decision support and the overall efficiency of oilfield development. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 is a system flow chart of the present invention;

[0088] Figure 2 A flow chart of the steps for obtaining a structured geological map of the present invention;

[0089] Figure 3 A flow chart of the steps for obtaining a clean geological map of the present invention;

[0090] Figure 4 A flow chart of the steps for obtaining an enhanced geological map of the present invention;

[0091] Figure 5 A flow chart of the steps for obtaining the optimized drilling parameters of the present invention;

[0092] Figure 6 A flowchart of the steps for obtaining the drilling efficiency results of the present invention;

[0093] Figure 7 Flow chart of the steps for obtaining parameter efficiency evaluation of the present invention. DETAILED DESCRIPTION

[0094] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0095] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0096] Embodiment 1

[0097] See also Figure 1 , a well logging operation automation system in plateau areas includes:

[0098] The rock formation characteristic analysis module performs pixel detection and region annotation based on the rock formation texture and fault line image data of the plateau area, and performs feature extraction and node association analysis on each node to establish a structured geological map;

[0099] The data purification module identifies and filters noise in structured geological maps, performs quality assessment and removes erroneous data for each data point, obtains clean geological maps, performs data enhancement processing on the basis of clean geological maps, adjusts the relationship strength between data points, and creates enhanced geological maps;

[0100] The parameter optimization module extracts rock hardness and wear rate from the enhanced geological map, combines the current drilling parameters, performs parameter impact analysis and sensitivity testing, formulates a preliminary adjustment plan, and dynamically adjusts the learning rate and micro-batch data in the preliminary adjustment plan based on real-time drilling feedback to obtain optimized drilling parameters;

[0101] The efficiency evaluation module uses optimized drilling parameters to perform drilling operations, collects drilling speed and drill wear data in real time, performs correlation analysis between speed and wear, generates drilling efficiency results, and evaluates the effectiveness of drilling parameters by comparing historical data with real-time data to obtain parameter efficiency evaluation.

[0102] The structured geological map includes geological node characteristics, connection relationship indicators, and terrain mapping data. The enhanced geological map specifically adjusts the relationship strength of data points. The optimized drilling parameters include hardness grade, wear rate range, and adjustment frequency. The parameter efficiency evaluation specifically includes efficiency improvement rate, parameter stability, and adjustment response time.

[0103] See also Figure 2 , the steps to obtain the structured geological map are as follows:

[0104] Load the rock texture and fault line image data of the plateau area, and use the edge enhancement algorithm to process the image using the formula:

[0105]

[0106] Calculate edge strength to extract rock texture and fault lines and obtain characteristic boundary map;

[0107] Where E(x,y,λ,φ) represents the edge strength at position (x,y), λ and φ are parameters for adjusting the sensitivity of edge detection, and I represents the image intensity. Represent the partial derivatives of the image in the x and y directions respectively;

[0108] The feature boundary map is used for pixel detection and region annotation, and the image segmentation algorithm based on brightness and texture is used to refine the region division. The formula is:

[0109]

[0110] Perform weighted summation on the pixels in each area to obtain a regional map of geological characteristics;

[0111] Among them, S represents the score of the region, a i is the importance weight of the i-th pixel, P i is the attribute value of the i-th pixel, c i is the brightness reference value, η is the brightness adjustment coefficient;

[0112] Based on the geological feature area map, node association analysis is performed to build a connection network between nodes using the formula:

[0113]

[0114] Evaluate the strength of associations between nodes to obtain a structured geological map;

[0115] Among them, R ij represents the association strength of node i to node j, w ij is the relationship weight between nodes i and j, and ∈ is the stability parameter.

[0116] Formula 1: Edge detection formula

[0117]

[0118] Parameter explanation and derivation:

[0119] I: Image intensity

[0120] The first-order derivative of the image in the x and y directions represents the gradient of the image in the horizontal and vertical directions.

[0121] The second derivative of the image in the x and y directions is used to detect the curvature of the edge.

[0122] λ, φ: adjustment parameters used to balance the influence of the first-order and second-order derivatives. They are usually adjusted according to the image characteristics in experiments. Here we set λ = 0.5 and φ = 1.5.

[0123] Example derivation:

[0124] Assume that at a certain point (x, y), the image intensity I = 100, and the first-order derivative in the x direction is First derivative in the y direction Second derivative in the x direction Second derivative in the y direction

[0125] Compute the contribution of the first-order derivative part:

[0126] Calculate the contribution of the second-order derivative: λ(5+3)=0.5×8=4

[0127] Final edge strength E(x,y)=4+27.041=31.041

[0128] Formula 2: Weighted summation formula

[0129]

[0130] Parameter explanation and derivation:

[0131] a i : The importance weight of the i-th pixel, assuming the weight is 1

[0132] P i : The attribute value of the i-th pixel, assumed to be 255 (the maximum brightness value of the image)

[0133] c i : Brightness reference value, assumed to be 100

[0134] η: Brightness adjustment coefficient, assumed to be 10

[0135] Example derivation:

[0136] Assume there is only one pixel n=1:

[0137] Calculate the ratio:

[0138] Weighted sum: S = 1 × (255 × 0.041956) = 10.69928

[0139] Formula 3: Node association strength formula

[0140]

[0141] Parameter explanation and derivation:

[0142] w ij : The relationship weight between node i and node j, assumed to be 0.8

[0143] ∈: stability parameter, assumed to be 0.01

[0144] Example derivation:

[0145] Assume that the total weight of node i and other nodes ∑ k≠i w ik =2.5:

[0146] Calculate the strength of association:

[0147] See also Figure 3 , the specific steps for obtaining a clean geological map are:

[0148] To identify the noise of structured geological maps, the frequency domain filtering method is used to analyze the image data, using the formula:

[0149]

[0150] The noise pattern was identified by Fourier transform and a Gaussian decay function was used to generate a preliminary filtered geological map;

[0151] Where N(x,y) represents the image intensity after noise attenuation at position (x,y). represents Fourier transform, I(x,y) is the original intensity of the image at position (x,y), α is the attenuation parameter, and x and y represent the coordinates of the pixel position;

[0152] The data point quality was evaluated using the geological map after preliminary filtering to exclude erroneous data using the formula:

[0153]

[0154] Calculate the consistency of each data point with its neighboring points, exclude abnormal data points by setting thresholds, and generate a clean geological map;

[0155] Where Q(i) represents the quality score of the i-th data point, v i 、v j They represent the attribute values ​​of data point i and neighboring point j respectively, N(i) is the set of neighboring points of point i, and ∈ is a small number to avoid the denominator being zero.

[0156] Formula 1: Noise Identification and Attenuation

[0157] formula:

[0158]

[0159] Parameter explanation and calculation process:

[0160] This is to perform Fourier transform on the image I (x, y). Fourier transform is used to convert the image from the time domain (spatial domain) to the frequency domain, mainly used to identify the frequency components in the image, that is, the noise and signal parts.

[0161] This represents the absolute value of the Fourier transform result and is used to obtain the amplitude at each point in the frequency domain, i.e. the noise intensity.

[0162] Taking a logarithmic transformation of the amplitude of the Fourier transform can help expand the dynamic range of low-intensity values ​​while compressing the range of high-intensity values, making the noise characteristics of the image more obvious.

[0163] α: Attenuation coefficient, which controls the attenuation rate of the Gaussian function. In actual use, this coefficient can be adjusted according to the distribution and density of the noise.

[0164] Gaussian attenuation function is used to reduce the impact of noise, where x and y represent the coordinates of the pixel position. This function decays faster as the distance from the center increases, effectively suppressing noise at the edge.

[0165] Calculation example:

[0166] Assume that there is a pixel point (x, y) = (10, 10), and the intensity I (10, 10) of the image at this point is 100. The modulus of the Fourier transform result is assumed to be 50, and the attenuation coefficient α is set to 0.001.

[0167]

[0168] N(10,10)=log(51)·e -0.2

[0169] N(10,10)=3.9318·0.8187=3.218

[0170] Here, the value of N(10, 10) indicates the strength of the noise attenuation process at position (10, 10). This value is low, indicating that the noise has been effectively attenuated.

[0171] Formula 2: Data point quality assessment and error data removal

[0172] formula:

[0173]

[0174] Parameter explanation and calculation process:

[0175] v i : The attribute value of the i-th data point.

[0176] v j : The attribute value of the neighboring point j of the i-th data point.

[0177] N(i): The set of neighboring points of point i.

[0178] ∈: A small amount used to avoid the denominator being zero, set to a small positive value such as 0.01.

[0179] Calculation example:

[0180] Assume that the value v at point i is i =80, the value v of its neighboring point j j =70 and 90.

[0181]

[0182] Q(i)=0.8749+0.8890=1.7639

[0183] Here, the value of Q(i) is high, indicating that point i has good consistency with its neighboring points and the data quality is high.

[0184] See also Figure 4 , the specific steps for obtaining the enhanced geological map are:

[0185] Starting from the clean geological map, the relationship between data points is strengthened using the formula:

[0186]

[0187] By adjusting the weights between data points, a data relationship reinforcement graph is generated;

[0188] Among them, S ij Indicates the strength of the relationship between data points i and j, v i 、v j is the attribute value of the corresponding data point, δ is the stability coefficient, avoiding the denominator to be zero, β is the spatial attenuation coefficient, d ij is the distance between points;

[0189] Use the data relationship enhancement diagram to perform spatial continuity analysis using the formula:

[0190]

[0191] By referring to the effect of geometric distance on the strength of relationship, a spatial continuity enhancement graph is generated;

[0192] Where C(i) represents the spatial continuity score of data point i, and γ is the adjustment parameter of spatial attenuation;

[0193] Integrate the spatial continuity enhancement map and perform image rendering using the formula:

[0194] G=∑ i C(i)·log(1+ρv i )

[0195] Enhance the readability and detail of geological maps to obtain enhanced geological maps;

[0196] Where G represents the enhanced geological map, v i is the original attribute value of data point i, and ρ is the logarithmic gain parameter.

[0197] Formula analysis for step 1

[0198] formula:

[0199]

[0200] Parameter explanation and examples:

[0201] v i 、v j : The attribute values ​​of data points i and j. For example, v i =20, v j =30.

[0202] δ: stability coefficient, used to avoid the denominator being zero, set to 1.

[0203] β: spatial attenuation coefficient, assumed to be 0.1.

[0204] d ij : The distance between points i and j, assumed to be 2.

[0205] Calculation process:

[0206] Calculate the numerator: v i ·v j =20×30=600

[0207] Calculate the denominator:

[0208] Apply attenuation factor: e -0.1×2 =e -0.2 ≈0.8187

[0209] Final result:

[0210] Result interpretation:

[0211] S ij=0.8172 represents the value of the relationship strength between data points i and j after attenuation processing. The closer it is to 1, the stronger the relationship.

[0212] Formula analysis for step 2

[0213] formula:

[0214]

[0215] Parameter explanation and examples:

[0216] S ij : Calculated from step 1, assuming that for all j, S ij =0.8172.

[0217] γ: The adjustment parameter of spatial attenuation, assumed to be 0.05.

[0218] d ij : Assume that for all j, d ij =2.

[0219] Calculation process:

[0220] For each j in the neighborhood of i, calculate

[0221] Assuming that i has 3 neighbors j, then C(i) = 3×0.8172×0.9317≈2.2832.

[0222] Result interpretation:

[0223] C(i)=2.2832 represents the spatial continuity score of point i after spatial attenuation adjustment. The higher the value, the better the spatial continuity.

[0224] Formula analysis for step 3

[0225] formula:

[0226]

[0227] Parameter explanation and examples:

[0228] C(i): obtained from step 2, assuming that for all i, C(i) = 2.2832.

[0229] v i : Attribute value, assumed to be 20.

[0230] ρ: logarithmic gain parameter, set to 0.1.

[0231] Calculation process:

[0232] Calculate log(1+0.1×20)=log(3)≈1.0986

[0233] Assuming there are 10 points i in total, then G = 10 × 2.2832 × 1.0986 ≈ 25.086.

[0234] Result interpretation:

[0235] G=25.086 represents the overall strength of the enhanced geological map. The higher this value is, the better the readability and detail of the image is.

[0236] See also Figure 5 , the specific steps for obtaining optimized drilling parameters are:

[0237] From the enhanced geological map, the rock hardness and wear rate data were extracted, and the rock formation characteristics were analyzed using image processing technology, using the formula:

[0238]

[0239] Generate rock hardness and wear data;

[0240] Among them, intensity(i) represents the light intensity value of the rock layer, texture(i) and edge(i) represent the texture and edge intensity information of the rock layer respectively, H(i) represents the hardness score of position i, and α, β, and γ are the weight coefficients of multiple attributes;

[0241] Based on the rock hardness and wear data, combined with the current drilling parameters, parameter impact analysis and sensitivity test were performed using the formula:

[0242]

[0243] Determine key adjustment points and generate preliminary drilling adjustment plans;

[0244] Where D represents the drilling depth, P represents the drilling performance prediction, D and H represent the depth and hardness parameters respectively, and f k is the influence function of parameters D and H, γ k , τ k are the adjustment factors for impact and depth attenuation, respectively;

[0245] Dynamically adjust the learning rate and micro-batch data according to the preliminary drilling adjustment plan and real-time drilling feedback, using the formula:

[0246]

[0247] After adjustment, the optimized drilling parameters are obtained;

[0248] Among them, λ is the learning rate, η is the fine-tuning parameter, and δ j represents the adjustment amount from drilling feedback, P jis the drilling performance index associated with j, and ΔP represents the optimization result of the drilling parameters.

[0249] Step 1: Rock hardness and wear data extraction

[0250] formula:

[0251]

[0252] Detailed explanation and derivation of formula parameters:

[0253] intensity(i): This is the intensity value of the rock image at pixel i. Assume that the intensity value of a pixel is 120.

[0254] texture(i): This represents the texture analysis result of pixel i in the image. Assume the texture value is 0.5.

[0255] edge(i): This represents the edge strength of pixel i in the image. Assume the edge strength is 0.3.

[0256] α, β, γ: These are weight coefficients used to balance the influence of various physical properties. Assume that the weight coefficients α = 0.8, β = 0.1, γ = 0.5.

[0257] Calculation process:

[0258] Calculate the squared contribution of the light intensity value:

[0259] intensitypart=α·intensity(i) 2 =0.8·120 2 =0.8·14400=11520

[0260] Calculate the product contribution of texture and edge:

[0261] textureandedgepart=β·texture(i)·edge(i)=0.1·0.5·0.3=0.015

[0262] Calculate the square root contribution of edge strength:

[0263]

[0264] Total hardness score:

[0265] H(i)=11520+0.015+0.27385≈11520.29

[0266] Explanation of the example results:

[0267] The final hardness score is approximately 11520.29, which is dominated by the contribution of the light intensity value, indicating that the rock has a higher hardness at this point and is suitable for predicting the resistance encountered in drilling.

[0268] Step 2: Parameter impact analysis and sensitivity testing

[0269] formula:

[0270]

[0271] Detailed explanation and derivation of formula parameters:

[0272] D: Drilling depth, assumed to be 300 meters.

[0273] H: Hardness score obtained from the step, i.e. 11520.29.

[0274] f k (D,H): influence function, assumed to be H / D.

[0275] γ k , τ k : Sensitivity coefficient and attenuation rate, assuming γ k =2,τ k =0.01.

[0276] Calculation process:

[0277] Calculate the influence function:

[0278] f k (D,H)=H / D=11520.29 / 300≈38.401

[0279] Calculate the effect of exponential decay:

[0280]

[0281] The resulting performance predictions are:

[0282]

[0283] Explanation of the example results:

[0284] The performance prediction value is 3.825, which reflects the drilling performance at a given depth and rock hardness, where a lower performance prediction value indicates more difficult drilling conditions.

[0285] Step 3: Dynamic Adjustment of Real-Time Drilling Feedback

[0286] formula:

[0287]

[0288] Detailed explanation and derivation of formula parameters:

[0289] λ, η: learning rate and fine-tuning parameters, assuming λ = 0.05, η = 0.1.

[0290] δ j : Adjustment amount, assumed to be 0.02.

[0291] P j : Performance index, using P = 3.825 from the previous step.

[0292] Calculation process:

[0293] Compute the time derivative contribution:

[0294] Calculate micro-batch adjustment contribution:

[0295] micro-batch adjustment=η·(δ j ·P j ) 1.5 =0.1·(0.02·3.825) 1.5 ≈0.1·(0.0765) 1.5 ≈0.1·0.066=0.0066

[0296] The total optimization parameter changes are:

[0297] ΔP=0.005+0.0066=0.0116

[0298] Explanation of the example results:

[0299] ΔP = 0.0116 indicates a small optimization of the drilling parameters under the current feedback mechanism, showing a slight improvement in drilling efficiency after fine-tuning.

[0300] See also Figure 6 , the specific steps for obtaining the drilling efficiency results are:

[0301] Start the drilling operation, use the optimized drilling parameters, and monitor the drilling speed and drill wear data in real time, using the formula:

[0302]

[0303] generating adjusted drilling speed data;

[0304] Among them, V(d) is the adjusted drilling speed data, v0 is the initial drilling speed, α and β adjust the influence of drilling depth and drill wear respectively, d represents the current drilling depth, and w is the drill wear index;

[0305] The speed-wear correlation analysis was performed using the adjusted drilling speed data using the formula:

[0306]

[0307] Calculate the correlation between speed and wear, and generate correlation analysis results between drilling speed and wear data;

[0308] Among them, C is the correlation analysis result of drilling speed and wear data, V(i) is the drilling speed data, and W(i) is the corresponding drill bit wear rate;

[0309] According to the correlation analysis results between drilling speed and wear data, the drilling efficiency is calculated using the formula:

[0310]

[0311] Get drilling efficiency results;

[0312] Among them, E is the drilling efficiency result.

[0313] Detailed explanation and derivation of the formula for step 1

[0314] formula:

[0315]

[0316] Parameter Description:

[0317] V(d): adjusted drilling speed, indicating the drilling speed at depth d.

[0318] v0: initial drilling speed, assumed to be 100m / s.

[0319] α: Depth attenuation coefficient, assumed to be 0.05 / m.

[0320] β: drill wear adjustment factor, assumed to be 0.03.

[0321] d: Drilling depth, assumed to be 100m.

[0322] w: drill wear index, assumed to be 0.5.

[0323] Calculation process:

[0324] Calculate α·d = 0.05 × 100 = 5

[0325] calculate

[0326] calculate

[0327] The exponential part of the comprehensive formula is -5 + 0.021 = -4.979

[0328] Calculate the drilling speed V(d) = 100·e -4.979 ≈0.007m / s

[0329] Result description:

[0330] The calculated result V(d)≈0.007m / s indicates that at a depth of 100 meters, after considering the influence of drill bit wear, the drilling speed slows down significantly and almost stops, showing extremely high attenuation.

[0331] Detailed explanation and derivation of the formula for step 2

[0332] formula:

[0333]

[0334] Parameter Description:

[0335] C: Correlation coefficient between speed and wear.

[0336] V(i): drilling speed data, assumed to be 0.007m / s.

[0337] W(i): drill wear rate data, assumed to be 50% or 0.5.

[0338] Calculation process:

[0339] Calculate log(1+W(i))=log(1.5)≈0.405

[0340] Calculate V(i)·log(1+W(i))=0.007×0.405≈0.002835

[0341] Calculate V(i) 2 =(0.007) 2 =0.000049

[0342] Calculate (log(1+W(i))) 2 =(0.405) 2 =0.164025

[0343] calculate

[0344] Result description:

[0345] The result C = 1 indicates that there is a perfect positive correlation between drilling speed and drill bit wear, that is, the slower the speed, the greater the wear, or vice versa, which in practical applications may indicate that drilling parameters need to be adjusted to optimize performance.

[0346] Detailed explanation and derivation of the formula for step 3

[0347] formula:

[0348]

[0349] Parameter Description:

[0350] E: Drilling efficiency.

[0351] C: The correlation coefficient calculated in step 2, known to be 1.

[0352] Calculation process:

[0353] Calculate |C| = 1

[0354] Use max(0.1,|C|)=1

[0355] calculate

[0356] Result description:

[0357] The result E=0 indicates that the current drilling efficiency is extremely low, which may be caused by the perfect positive correlation between drilling speed and drill bit wear.

[0358] See also Figure 7 , the specific steps for obtaining parameter efficiency evaluation are:

[0359] Based on the drilling efficiency results, the data of drilling speed and drill bit wear rate are collected from the historical database and real-time monitoring system, and the formula is used:

[0360]

[0361] Calculate and generate real-time drilling parameter data;

[0362] Among them, P real (t) is the real-time drilling parameter data, α represents the weight coefficient of speed, β represents the weight coefficient of wear, V(t) represents the square root of the real-time drilling speed, and W(t) represents the natural logarithm increment of the real-time drill wear rate;

[0363] The real-time drilling parameter data is compared with the stored historical drilling parameter data using the formula:

[0364]

[0365] Calculate the difference between historical data and real-time data to obtain parameter difference analysis results;

[0366] Among them, D is the result of parameter difference analysis, P hist (j) represents the historical drilling parameters, and m represents the number of data points;

[0367] According to the results of parameter difference analysis, the formula is used:

[0368]

[0369] The average difference value was calculated to obtain parameter efficiency evaluation;

[0370] Among them, A is the parameter efficiency evaluation, and m represents the total number of data points involved in the calculation.

[0371] Assume the following numbers:

[0372] The drilling speed V(t) is usually between 100 and 300 rpm (revolutions per minute) in real-time monitoring.

[0373] The drill wear rate W(t) is usually between 0.01% and 0.1%.

[0374] The weight coefficients α and β usually depend on the actual equipment and operating conditions, and here it is assumed that α=0.5 and β=0.3.

[0375] formula:

[0376]

[0377] Parameter details:

[0378] α=0.5: The weight coefficient of speed, which determines the influence of speed in the overall formula.

[0379] β = 0.3: Weight coefficient of wear, which determines the influence of wear rate in the overall formula.

[0380] V(t)=150rpm: real-time drilling speed.

[0381] W(t)=0.05% or 0.0005: real-time drill wear rate.

[0382] Calculation process and examples:

[0383] Impact of calculation speed:

[0384] Calculate the effect of wear rate: log(0.0005+1)≈0.0005 (using natural logarithm)

[0385] Combine these two parts: real (t)=0.5·12.25+0.3·0.0005≈6.12515

[0386] This result P real (t) is the weighted result of the calculated real-time drilling parameters, reflecting the combined effect of drilling speed and wear rate.

[0387] formula:

[0388]

[0389] Parameter details:

[0390] P hist (j) = 6: historical drilling parameters.

[0391] P real (j) = 6.12515: Real-time drilling parameters calculated above.

[0392] m=10: total number of data points.

[0393] Calculation process and examples:

[0394] For a single data point the difference is calculated:

[0395] Assuming that all points have similar differences, then D≈10×0.0195=0.195

[0396] This result D is the sum of the differences of all data points, reflecting the total amount of difference between historical data and real-time data.

[0397] formula:

[0398]

[0399] Parameter details:

[0400] D = 0.195: The sum of the differences calculated above.

[0401] m=10: total number of data points.

[0402] Calculation process and examples:

[0403] Calculate the mean difference:

[0404] This result E is an evaluation of the drilling parameter efficiency, showing the average degree of difference between historical and real-time data. This helps to evaluate the accuracy and reliability of the real-time monitoring system.

[0405] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A well logging operation automation system in plateau areas, characterized in that: The system comprises: The rock formation characteristic analysis module performs pixel detection and region annotation based on the rock formation texture and fault line image data of the plateau area, and performs feature extraction and node association analysis on each node to establish a structured geological map; The data purification module identifies and filters noise on the structured geological map, performs quality assessment on each data point and removes erroneous data, obtains a clean geological map, performs data enhancement processing on the basis of the clean geological map, adjusts the relationship strength between data points, and creates an enhanced geological map; The parameter optimization module extracts rock hardness and wear rate from the enhanced geological map, performs parameter impact analysis and sensitivity test in combination with current drilling parameters, formulates a preliminary adjustment plan, and dynamically adjusts the learning rate and micro-batch data in the preliminary adjustment plan according to real-time drilling feedback to obtain optimized drilling parameters; The efficiency evaluation module uses the optimized drilling parameters to perform drilling operations, collects drilling speed and drill bit wear data in real time, performs correlation analysis between speed and wear, generates drilling efficiency results, and evaluates the effectiveness of drilling parameters by comparing historical data and real-time data to obtain parameter efficiency evaluation.

2. The plateau logging operation automation system according to claim 1 is characterized in that: The steps for obtaining the structured geological map are specifically as follows: Load the rock texture and fault line image data of the plateau area, and use the edge enhancement algorithm to process the image using the formula: Calculate edge strength to extract rock texture and fault lines and obtain characteristic boundary map; Where E(x,y,λ,φ) represents the edge strength at position (x,y), λ and φ are parameters for adjusting the sensitivity of edge detection, and I represents the image intensity. Represent the partial derivatives of the image in the x and y directions respectively; The feature boundary map is used to perform pixel detection and region annotation, and an image segmentation algorithm based on brightness and texture is used to refine the region division, using the formula: Perform weighted summation on the pixels in each area to obtain a regional map of geological characteristics; Among them, S represents the score of the region, a i is the importance weight of the i-th pixel, P i is the attribute value of the i-th pixel, c i is the brightness reference value, η is the brightness adjustment coefficient; Based on the geological feature area map, node association analysis is performed to construct a connection network between nodes using the formula: Evaluate the strength of associations between nodes to obtain a structured geological map; Among them, R ij represents the association strength of node i to node j, w ij is the relationship weight between nodes i and j, and ∈ is the stability parameter.

3. The plateau logging operation automation system according to claim 1 is characterized in that: The steps for obtaining the clean geological map are specifically as follows: The structured geological map is subjected to noise identification, and the image data is analyzed using the frequency domain filtering method, using the formula: The noise pattern was identified by Fourier transform and a Gaussian decay function was used to generate a preliminary filtered geological map; Where N(x,y) represents the image intensity after noise attenuation at position (x,y). represents Fourier transform, I(x,y) is the original intensity of the image at position (x,y), α is the attenuation parameter, and x and y represent the coordinates of the pixel position; The data point quality assessment was performed using the geological map after the preliminary filtering to exclude erroneous data using the formula: Calculate the consistency of each data point with its neighboring points, exclude abnormal data points by setting thresholds, and generate a clean geological map; Where Q(i) represents the quality score of the i-th data point, v i 、v j They represent the attribute values ​​of data point i and neighboring point j respectively, N(i) is the set of neighboring points of point i, and ∈ is a small number to avoid the denominator being zero.

4. The plateau logging operation automation system according to claim 1 is characterized in that: The steps for obtaining the enhanced geological map are specifically as follows: Starting from the clean geological map, the relationship between data points is strengthened using the formula: By adjusting the weights between data points, a data relationship reinforcement graph is generated; Among them, S ij represents the strength of the relationship between data points i and j, v i 、v j is the attribute value of the corresponding data point, δ is the stability coefficient, avoiding the denominator to be zero, β is the spatial attenuation coefficient, d ij is the distance between points; Using the data relationship enhancement diagram, spatial continuity analysis is performed using the formula: By referring to the effect of geometric distance on the strength of relationship, a spatial continuity enhancement graph is generated; Where C(i) represents the spatial continuity score of data point i, and γ is the adjustment parameter of spatial attenuation; Integrate the spatial continuity enhancement map and perform image rendering using the formula: G=∑ i C(i)·log(1+ρv i ) Enhance the readability and detail of geological maps to obtain enhanced geological maps; Where G represents the enhanced geological map, v i is the original attribute value of data point i, and ρ is the logarithmic gain parameter.

5. The plateau logging operation automation system according to claim 1 is characterized in that: The steps for obtaining the optimized drilling parameters are specifically as follows: From the enhanced geological map, the data of rock hardness and wear rate are extracted, and the rock formation characteristics are analyzed using image processing technology, using the formula: Generate rock hardness and wear data; Among them, intensity(i) represents the light intensity value of the rock layer, texture(i) and edge(i) represent the texture and edge intensity information of the rock layer respectively, H(i) represents the hardness score of position i, and α, β, and γ are the weight coefficients of multiple attributes; Based on the rock hardness and wear data, combined with the current drilling parameters, parameter impact analysis and sensitivity testing were performed using the formula: Determine key adjustment points and generate preliminary drilling adjustment plans; Where D represents the drilling depth, P represents the drilling performance prediction, D and H represent the depth and hardness parameters respectively, and f k is the influence function of parameters D and H, γ k , τ k are the adjustment factors for impact and depth attenuation, respectively; The learning rate and micro-batch data are dynamically adjusted according to the preliminary drilling adjustment plan and real-time drilling feedback, using the formula: After adjustment, the optimized drilling parameters are obtained; Among them, λ is the learning rate, η is the fine-tuning parameter, and δ j represents the adjustment amount from drilling feedback, P j is the drilling performance index associated with j, and ΔP represents the optimization result of the drilling parameters.

6. The plateau logging operation automation system according to claim 1 is characterized in that: The steps for obtaining the drilling efficiency result are specifically as follows: Start the drilling operation, use the optimized drilling parameters, monitor the drilling speed and drill bit wear data in real time, and use the formula: generating adjusted drilling speed data; Among them, V(d) is the adjusted drilling speed data, v0 is the initial drilling speed, α and β adjust the influence of drilling depth and drill wear respectively, d represents the current drilling depth, and w is the drill wear index; The adjusted drilling speed data was used to perform a speed-wear correlation analysis using the formula: Calculate the correlation between speed and wear, and generate correlation analysis results between drilling speed and wear data; Among them, C is the correlation analysis result of drilling speed and wear data, V(i) is the drilling speed data, and W(i) is the corresponding drill bit wear rate; According to the correlation analysis results of the drilling speed and wear data, the drilling efficiency is calculated using the formula: Get drilling efficiency results; Among them, E is the drilling efficiency result.

7. The plateau logging operation automation system according to claim 1 is characterized in that: The steps for obtaining the parameter efficiency evaluation are specifically as follows: Based on the drilling efficiency results, the data of drilling speed and drill bit wear rate were collected from the historical database and real-time monitoring system, and the formula was used: Calculate and generate real-time drilling parameter data; Among them, P real (t) is the real-time drilling parameter data, α represents the weight coefficient of speed, β represents the weight coefficient of wear, V(t) represents the square root of the real-time drilling speed, and W(t) represents the natural logarithm increment of the real-time drill wear rate; The real-time drilling parameter data is compared with the stored historical drilling parameter data using the formula: Calculate the difference between historical data and real-time data to obtain parameter difference analysis results; Among them, D is the result of parameter difference analysis, P hist (j) represents the historical drilling parameters, and m represents the number of data points; According to the results of the parameter difference analysis, the formula is used: The average difference value was calculated to obtain parameter efficiency evaluation; Among them, A is the parameter efficiency evaluation, and m represents the total number of data points involved in the calculation.

Images