Drilling optimization method fusing observation disturbance and dynamic decision

Abstract

The invention relates to the field of deep geological exploration, in particular to a drilling optimization method fusing observation disturbance and dynamic decision. The method comprises the following steps: firstly, based on a clustering result of a fuzzy C-means clustering method, determining a preliminary range for drilling parameter optimization; and then further determining a final operation parameter adjustment range and a drilling speed optimization target interval according to the data in the sliding window. And then, designing an improved whale optimization algorithm containing observation disturbance characteristics, and carrying out optimization adjustment on drilling parameters. By the adoption of the drilling parameter optimization method, the problems of non-convex optimization and unstable drilling state operation in the optimization process can be effectively solved, and a reliable analysis basis is provided for adjustment of the drilling process.

Description

Technical Field

[0001] This invention relates to the field of deep geological exploration, and in particular to a drilling optimization method that integrates observational disturbances and dynamic decision-making. Background Technology

[0002] In deep geological drilling activities, the optimization and adjustment of the drilling process plays an irreplaceable role, especially when encountering complex geological structures and significant differences in lithology. The optimization and adjustment of the drilling process is related to construction progress and economic costs. Scientific modeling and accurate prediction of drilling speed can not only rationally allocate drilling resources and reduce construction risks, but also extend equipment lifespan, thereby comprehensively improving the efficiency of resource development.

[0003] With continuous breakthroughs in deep drilling technology, the operating environment is becoming increasingly harsh and variable. Under conditions of high temperature, high pressure, and strong nonlinear stress, the optimization of the drilling process faces challenges such as non-convex optimization and instability in the drilling state. Therefore, it is necessary to develop suitable optimization methods for drilling processes in deep and complex geological conditions to achieve optimized adjustment of the drilling state. Summary of the Invention

[0004] To address the issues of non-convex optimization and unstable operation during deep geological drilling, a suitable and reliable drilling parameter optimization method is designed for complex deep geological environments. This method effectively improves drilling efficiency and provides effective optimization settings for subsequent drilling process control. This invention provides a drilling optimization method that integrates observed disturbances and dynamic decision-making, mainly including:

[0005] S1: Based on the initial drilling data, determine the category of the drilling state that needs to be optimized and adjusted. The adjustment range of the drilling parameters in the category is the initial adjustment range of the optimization adjustment.

[0006] S2: Based on the data in the sliding window, take the union of the adjustment range of the data in the sliding window and the initial adjustment range as the final range of the optimized adjustment, and determine the target range of drilling speed in the optimized adjustment based on the data in the window;

[0007] S3: Based on the final range of the optimized adjustment and the target range of the drilling rate, an improved white whale optimization algorithm with an elite fusion strategy, optimal solution of disturbance and observation of disturbance strategy is designed.

[0008] S4: The improved Beluga optimization algorithm is used to optimize and adjust the drilling process to obtain the best drilling parameter values ​​and the optimal drilling speed.

[0009] A computer device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method.

[0010] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above method.

[0011] A computer program product includes a computer program or instructions that, when executed by a processor, implement the steps of the above-described method.

[0012] The beneficial effects of the technical solution provided by this invention are as follows: the drilling parameter optimization adjustment range designed by this invention can effectively ensure the stable operation of drilling equipment; the determined target range for drilling optimization adjustment can reasonably design the upper and lower limits of the optimization target, ensuring the stable operation of the geological drilling process; based on the adjustment range and target range, the improved white whale optimization method designed by adopting elite fusion, optimal solution of disturbance and observation disturbance strategy can effectively solve the high-dimensional change and non-convex optimization problems faced by drilling process optimization adjustment, ensuring the stable operation of the drilling process while improving drilling efficiency, and providing good optimization settings for subsequent drilling process control. Attached Figure Description

[0013] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0014] Figure 1 This is a structural diagram of a drilling optimization method that integrates observational perturbation and dynamic decision-making in an embodiment of the present invention;

[0015] Figure 2 This is a simulation result diagram of the test function F10 in an embodiment of the present invention;

[0016] Figure 3 This is a simulation result diagram of the test function F20 in the embodiment of the present invention;

[0017] Figure 4 This is a comparison chart of the optimization results and the actual drilling speed in the embodiments of the present invention. Detailed Implementation

[0018] 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.

[0019] Example 1

[0020] Please refer to Figure 1 , Figure 1This is a structural diagram of a drilling optimization method integrating observational perturbation and dynamic decision-making in an embodiment of the present invention. This method first determines the initial range for drilling parameter optimization based on the clustering results of the fuzzy C-means clustering method. Then, based on the data in the sliding window, it further determines the final adjustment range of operating parameters and the target range of drilling speed during optimization. Next, an improved Beluga optimization algorithm incorporating observational perturbation features is designed to optimize and adjust the drilling parameters. The specific steps of this method are as follows:

[0021] S1: Based on the initial drilling data, determine the category of the drilling state that requires drilling optimization adjustment. The adjustment range of the drilling parameters in the category is the initial adjustment range for optimization adjustment. The specific implementation process of this step is as follows:

[0022] The initial drilling data is divided based on fuzzy C-means clustering, which determines the category of the drilling state that needs to be optimized. Based on the division results, the adjustment range of the operating parameters to be adjusted is selected as the initial adjustment range for optimization, as follows:

[0023]

[0024] In the formula, V wob V rpm V q and V mw These are drilling pressure, rotation speed, pump flow rate, and density, respectively. and These are the lower limits for drilling pressure, rotational speed, pump flow rate, and density, respectively. and These are the upper limits for drilling pressure, rotation speed, pump flow rate, and density, respectively.

[0025] In this embodiment, 1200 sets of initial drilling data from a geological drilling site were collected for optimization and adjustment of the drilling process. 1140 sets of data were selected for fuzzy C-means clustering, and the remaining 60 sets were used for further optimization and adjustment. Based on the aforementioned preliminary method for determining the optimization and adjustment range, the current operating state category was determined according to the clustering results, thus establishing the initial adjustment range for optimization and adjustment.

[0026] S2: Based on the data in the sliding window, take the union of the adjustment range of the data in the sliding window and the initial adjustment range as the final range of the optimized adjustment, and determine the target range of drilling speed in the optimized adjustment based on the data in the window; Figure 1 The optimization constraint in S2 refers to the final range of optimization adjustment and the target range of drilling speed; the specific implementation process of S2 is as follows:

[0027] S2.1: Select the adjustment range of drilling pressure, rotation speed, pump flow rate, and density in the sliding window, and take the union of these with the initial adjustment range as the final range for optimized adjustment during the drilling process, as follows:

[0028]

[0029] In the formula, and These are the lower limits for drilling pressure, rotation speed, pump rate, and density within the sliding window; and These are the upper limits of drilling pressure, rotation speed, pump rate, and density within the sliding window, respectively. and To optimize the adjustment, the final lower limit of drilling parameters, and This represents the final upper limit of the drilling parameters.

[0030] S2.2: Calculate the combined distance between the sliding window data and the initial drilling data. Based on the combined distance from largest to smallest, select the first 50% of the data in the sliding window to update the drilling rate model constructed using the support vector regression method, in order to ensure the accuracy of the drilling rate model. The combined distance is calculated as follows:

[0031]

[0032] In the formula, d i Let be the minimum combined distance from the i-th data sample in the sliding window to all samples in the initial drilling data. Let be the Mahalanobis distance from the i-th data sample in the sliding window to the j-th sample in the initial drilling data. Let be the Euclidean distance from the i-th data sample in the sliding window to the j-th sample in the initial drilling data.

[0033] S2.3: Select the drilling rate from the last 50% of the data in the sliding window and the drilling rate from the initial drilling data, and set the target range for the drilling rate in the drilling optimization adjustment. The specific target range is as follows:

[0034]

[0035] In the formula, and These are the minimum drilling speeds in the initial drilling data and the sliding window data, respectively. and These represent the maximum drilling speeds in the initial drilling data and the sliding window data, respectively, and 2·ROP. c It is twice the current drilling rate.

[0036] Further, 200 sets of drilling field data were collected as sliding window data. According to the method described above, the adjustment range in the sliding window was combined with the initial adjustment range to determine the final range of the optimized adjustment. Based on the comprehensive distance, 50% of the data was selected to update the drilling speed model and determine the optimization target interval.

[0037] S3: Based on the determined final range of adjustment and the target range of drilling speed, an improved white whale optimization algorithm with an elite fusion strategy, optimal solution perturbation, and observation perturbation strategy is designed. Compared with the standard white whale optimization algorithm, the specific implementation process of the improvement in S3 is as follows:

[0038] S3.1: The elite fusion strategy is as follows:

[0039]

[0040] In the formula, R1 and R2 are the positions of the i-th beluga whale during the t-th and t+1-th iterations, respectively, where R1 and R2 are random numbers between 0 and 1, and Elite(·) represents the elite pool. q represents the position of a randomly selected whale from n p whales. a and q b The calculation is as follows:

[0041] q a =0.5+2·(tt) max ) / (1-t max )

[0042] q b =0.6+2·(tt) max ) / (1-t max )

[0043] In the formula, t max This represents the maximum number of iterations.

[0044] S3.2: The optimal solution to the perturbation is as follows:

[0045]

[0046] In the formula, X * The current global optimal solution Let be the global optimal solution after perturbation, levy be the Lévy perturbation, and R3 be a random number between 0 and 1.

[0047] S3.3: If the perturbed global optimal solution is better than the current global optimal solution, accept the perturbed global optimal solution. Conversely, if the perturbed global optimal solution satisfies the following observation perturbation conditions, also accept the perturbed global optimal solution. The specific observation perturbation conditions are as follows:

[0048]

[0049] In the formula, exp is the natural logarithm, F(·) is the fitness function, and R4 is a random number between 0 and 1.

[0050] S4: An improved white whale optimization algorithm was used to optimize and adjust the drilling process. Before optimization, the performance of the improved white whale optimization algorithm was analyzed using test functions F10 and F20, respectively. The experimental results are as follows: Figure 2 and Figure 3 As shown, experimental results demonstrate that the improved beluga whale optimization method possesses strong global search capabilities, validating the effectiveness of the improvement. Further optimization and adjustment analysis was conducted on 60 sets of data, and the experimental results... Figure 4 As shown, the experimental results indicate that the average drilling speed after optimization of 60 sets of drilling data was 3.73 m / hr, while the average drilling speed before optimization was 3.11 m / hr. The comparison results show that the optimization can significantly improve the drilling speed, indicating the effectiveness of the optimization and adjustment in the drilling process.

[0051] Example 2

[0052] A computer device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method.

[0053] Example 3

[0054] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above method.

[0055] Example 4

[0056] A computer program product includes a computer program or instructions that, when executed by a processor, implement the steps of the above-described method.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A drilling optimization method integrating observational perturbation and dynamic decision-making, characterized in that, include: S1: Based on the initial drilling data, determine the category of the drilling state that needs to be optimized and adjusted. The adjustment range of the drilling parameters in the category is the initial adjustment range of the optimization adjustment. S2: Based on the data in the sliding window, take the union of the adjustment range of the data in the sliding window and the initial adjustment range as the final range of the optimized adjustment, and determine the target range of drilling speed in the optimized adjustment based on the data in the window; S3: Based on the final range of the optimized adjustment and the target range of the drilling rate, an improved white whale optimization algorithm with an elite fusion strategy, optimal solution of disturbance and observation of disturbance strategy is designed. S4: The improved Beluga optimization algorithm is used to optimize and adjust the drilling process to obtain the best drilling parameter values ​​and the optimal drilling speed.

2. The drilling optimization method integrating observational perturbation and dynamic decision-making as described in claim 1, characterized in that, The specific implementation process of S1 is as follows: The initial drilling data is divided based on fuzzy C-means clustering, which determines the category of the drilling state that needs to be optimized. Based on the division results, the adjustment range of the operating parameters to be adjusted is selected as the initial adjustment range for optimization, as follows: In the formula, V wob V rpm V q and V mw These are drilling pressure, rotation speed, pump flow rate, and density, respectively. and These are the lower limits for drilling pressure, rotational speed, pump flow rate, and density, respectively. and These are the upper limits for drilling pressure, rotation speed, pump flow rate, and density, respectively.

3. The drilling optimization method integrating observational perturbation and dynamic decision-making as described in claim 1, characterized in that, The specific implementation process of S2 is as follows: S2.1: Select the adjustment range of drilling pressure, rotation speed, pump flow rate, and density in the sliding window, and take the union of these with the initial adjustment range as the final range for optimized adjustment during the drilling process, as follows: In the formula, and These are the lower limits for drilling pressure, rotation speed, pump rate, and density within the sliding window; and These are the upper limits of drilling pressure, rotation speed, pump rate, and density within the sliding window, respectively. and To optimize the adjustment of the final lower limit of drilling parameters, and To optimize the final upper limit of drilling parameters during adjustment; S2.2: Calculate the combined distance between the sliding window data and the initial drilling data. Based on the combined distance from largest to smallest, select the first 50% of the data in the sliding window to update the drilling rate model constructed using the support vector regression method, in order to ensure the accuracy of the drilling rate model. The combined distance is calculated as follows: In the formula, d i Let be the minimum combined distance from the i-th data sample in the sliding window to all samples in the initial drilling data. Let be the Mahalanobis distance from the i-th data sample in the sliding window to the j-th sample in the initial drilling data. Let be the Euclidean distance from the i-th data sample in the sliding window to the j-th sample in the initial drilling data; S2.3: Select the drilling rate from the last 50% of the data in the sliding window and the drilling rate from the initial drilling data, and set the target range for the drilling rate in the drilling optimization adjustment. The specific target range is as follows: In the formula, and These are the minimum drilling speeds in the initial drilling data and the sliding window data, respectively. and These represent the maximum drilling speeds in the initial drilling data and the sliding window data, respectively, and 2·ROP. c It is twice the current drilling rate.

4. The drilling optimization method integrating observational perturbation and dynamic decision-making as described in claim 1, characterized in that, In S3, the elite fusion strategy is as follows: In the formula, R1 and R2 are the positions of the i-th beluga whale during the t-th and t+1-th iterations, respectively, where R1 and R2 are random numbers between 0 and 1, and Elite(·) represents the elite pool. Given n p whales, the position of a randomly selected whale is q. a and q b The calculation is as follows: q a =0.5+2·(t-t max ) / (1-t max ) q b =0.6+2·(t-t max ) / (1-t max ) In the formula, t max This represents the maximum number of iterations.

5. The drilling optimization method integrating observational perturbation and dynamic decision-making as described in claim 1, characterized in that, The optimal solution to the perturbation is as follows: In the formula, X * The current global optimal solution Let be the global optimal solution after perturbation, levy be the Lévy perturbation, and R3 be a random number between 0 and 1.

6. The drilling optimization method integrating observational perturbation and dynamic decision-making as described in claim 5, characterized in that, The specific observation perturbation strategy is as follows: If the perturbed global optimal solution is better than the current global optimal solution, then accept the perturbed global optimal solution; If the perturbed global optimal solution is inferior to the current global optimal solution, the perturbed global optimal solution will still be accepted as long as it satisfies the following observation perturbation condition: In the formula, exp is the natural logarithm, F(·) is the fitness function, and R4 is a random number between 0 and 1.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes a computer program to implement the drilling optimization method for fusing observational perturbations and dynamic decision-making as described in claims 1-6.

8. A computer-readable storage medium, characterized in that, The device contains a computer program that, when executed by a processor, implements the steps of the drilling optimization method for fusing observational perturbations and dynamic decision-making as described in claims 1-6.

9. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the steps of the drilling optimization method for fusing observational perturbations and dynamic decision-making as described in claims 1-6.

Images