Shale gas well shaft flow pressure gradient intelligent calculation method

Abstract

The invention relates to the field of oil and gas resource development methods, and discloses a shale gas well shaft flow pressure gradient intelligent calculation method, which comprises the following steps: searching historical production data of a single well most similar to the current state of a target single well, performing analysis and calculation based on the historical production data to obtain the current shaft flow pressure gradient of the target single well, and calculating the current shaft flow pressure gradient of the target single well. The future production condition of the target single well is predicted according to the follow-up production data of the most similar single well, and optimization guidance is provided for the production strategy of the target single well; an existing shaft liquid accumulation judging method cannot meet the requirements of quickly, efficiently and accurately recognizing the flow state of a shaft and judging whether liquid accumulation exists in a gas well or not in shale gas well production.

Description

Technical Field

[0001] This invention relates to the field of oil and gas resource development methods, specifically to an intelligent calculation method for the wellbore pressure gradient of shale gas wells. Background Technology

[0002] Shale gas reservoirs are characterized by low porosity and low permeability, and gas well production relies on artificial fracture networks. During production, formation water enters the wellbore along with the gas. If the gas's liquid-carrying capacity is insufficient, liquid will accumulate in the wellbore, forming accumulated fluid. Accumulated fluid in the wellbore has adverse effects on shale gas well production. On one hand, accumulated fluid increases the fluid density in the wellbore, raising the bottomhole back pressure and reducing the production pressure differential, thus decreasing the gas production rate and output. As the amount of accumulated fluid increases, the bottomhole back pressure continues to rise, rapidly reducing the well's production capacity. In severe cases, it can even lead to water flooding and production shutdown, rendering the well unproductive. On the other hand, accumulated fluid corrodes wellbore equipment and tubing, shortening equipment lifespan and increasing maintenance and replacement costs. Furthermore, it may affect the effectiveness of subsequent enhancement measures, reducing their success rate and overall production impact.

[0003] To effectively control fluid accumulation in the wellbore, existing technologies for assessing fluid accumulation include: The wellbore flow pressure gradient change discrimination method involves monitoring the flow pressure at different depths in the wellbore, calculating the pressure gradient, and judging whether fluid has accumulated in the wellbore based on its changes. However, in practice, factors such as the accuracy of measuring instruments, the measurement location, and the non-uniformity of the fluid in the wellbore can lead to errors in the measurement results, affecting the accuracy of the pressure gradient calculation. Moreover, pressure gradient changes may be affected by multiple factors such as production conditions and formation pressure fluctuations, making it difficult to accurately distinguish whether they are caused by fluid accumulation, resulting in a low accuracy rate.

[0004] Wellhead-Casing Pressure Difference Judgment Method: Under normal circumstances, the wellhead-casing pressure difference is relatively stable with gas well production. When fluid accumulates in the wellbore, the pressure difference changes, generally manifested as a decrease in tubing pressure, an increase in casing pressure, and a decrease in the pressure difference. However, after fluid accumulation in the wellbore, changes in the pressure difference are easily affected by factors such as the sealing performance of wellhead equipment, surface pipeline resistance, and changes in gas composition, leading to pressure difference distortion and failing to accurately reflect the fluid accumulation situation in the wellbore. Furthermore, changes in the pressure difference exhibit a lag; by the time an anomaly is detected, a large amount of fluid may have already accumulated in the wellbore, missing the optimal time for intervention.

[0005] Methods for determining gas and water production changes: When fluid accumulates in the wellbore, gas production decreases, while water production may increase or remain stable; changes in gas and water production can be used as a basis for judgment. However, factors such as changes in formation gas supply capacity, adjustments to production systems, and measurement errors can interfere with changes in gas and water production. For example, a decrease in formation gas supply capacity over time can be difficult to distinguish from a decrease in gas volume caused by fluid accumulation in the wellbore. Furthermore, errors exist in gas and water production measurements, such as flow meter accuracy and environmental factors, further reducing the accuracy of the judgment.

[0006] The manual calculation method for wellbore flowing pressure gradient can accurately determine whether there is fluid accumulation in the wellbore. However, traditional manual calculation relies on gas production engineering models and surface parameters, which are difficult to obtain accurately in practice and change with the production process, making it difficult to update the calculations in a timely and accurate manner. Furthermore, manual calculation is greatly affected by human factors; different personnel use different calculation methods and assumptions, leading to large errors and low accuracy.

[0007] The wireline drilling test method involves lowering a testing instrument into the wellbore to directly measure fluid properties and fluid level, providing accurate results. However, it has drawbacks such as high labor intensity, high operational risks, and high costs. Wireline drilling requires specialized equipment and personnel, necessitates production shutdowns, and disrupts normal gas well production. Furthermore, it carries safety risks, such as stuck instruments and blowouts, which could damage the well and personnel. In addition, the high operating costs increase gas production costs, making it unsustainable for wells with poor economic returns.

[0008] Existing methods for determining wellbore fluid accumulation are all inadequate and cannot meet the needs of shale gas well production for rapid, efficient, and accurate identification of wellbore flow patterns and determination of whether a gas well has fluid accumulation. Summary of the Invention

[0009] The present invention aims to provide an intelligent calculation method for the flow pressure gradient in shale gas wells, as existing methods for judging wellbore liquid accumulation cannot meet the needs of shale gas well production for rapid, efficient, and accurate identification of wellbore flow patterns and determination of whether a gas well has accumulated liquid.

[0010] To achieve the above objectives, the present invention adopts the following technical solution: a smart calculation method for the flowing pressure gradient of a shale gas wellbore, comprising the following steps: S1. Acquire and preprocess relevant production data of the target single well to form target window data; S2. Obtain the production data of the candidate wells. The candidate wells are the other wells besides the target well. Construct a historical dataset. S3. For the historical data of the single well to be selected, use a sliding window to generate multiple sets of candidate window data; S4. Calculate the distance between each group of candidate window data for each candidate well and the target window data for the target well; S5. Determine the candidate window data that is most similar to the target well for each candidate well based on the distance value; S6. Compare the minimum average distance of each candidate well and output the well and its production period data that are most similar to the current state of the target well. S7. Calculate the current wellbore flowing pressure gradient of the target well by analyzing the historical data of the most similar single well, and predict the future production of the target single well based on the subsequent production data of the similar single well, so as to provide optimization guidance for the production strategy of the target single well.

[0011] The principle and advantages of this solution are as follows: This application adopts a data-driven approach, collecting and analyzing historical production data of gas wells to provide scientific decision support for calculating the wellbore pressure gradient. This method avoids over-reliance on human experience or simple rules, reduces errors caused by human factors, and can fully utilize information from actual production data to improve the accuracy and reliability of calculations. It effectively solves the problems of existing technologies where manual calculations rely on parameters that are difficult to obtain precisely and are greatly affected by human factors, resulting in large errors and low accuracy in judgment.

[0012] This application introduces a similarity matching mechanism. By calculating the distance between each candidate window data set of each candidate well and the target window data of the target well, the most similar candidate window data for each candidate well is determined. This process then identifies the gas well with the highest similarity to the target gas well among all producing wells. This method can draw on the production experience of other similar gas wells, providing a more accurate production strategy for the target gas well. It overcomes the problem of low accuracy in existing technologies such as wellbore flow pressure gradient change discrimination method, wellhead oil-casing pressure difference discrimination method, and gas-water production change discrimination method, which are affected by various factors.

[0013] By analyzing the historical production data of the target gas well, this proposed solution can formulate personalized production strategies based on the actual conditions of the gas well. Different gas wells are significantly affected by various factors during production, such as formation conditions and extraction history. This solution fully considers the differences among gas wells, enabling precise analysis and decision-making tailored to the characteristics of the target gas well, improving production efficiency and extraction effectiveness, and solving the problem that existing technologies struggle to accurately distinguish the impact of different factors on gas well production.

[0014] This application utilizes big data technology to mine and analyze gas well production data, identifying the wells with the smallest deviations and their production status at specific times. By generating multiple sets of candidate window data through a sliding window and performing comprehensive distance calculations and similarity matching, it is possible to deeply uncover potential patterns and correlations within the data. The application of this technology can improve the depth and breadth of production data analysis, providing more accurate decision support for production, helping to promptly detect problems such as wellbore fluid accumulation, avoiding missing the optimal treatment opportunity, and solving the problem of lag in existing wellhead oil-casing pressure differential discrimination methods.

[0015] Preferably, as an improvement, in S1, the historical production data of the target single well for the past seven days is obtained as the target window data, and the production data includes tubing inner diameter, casing pressure, oil pressure, gas production and water production.

[0016] The beneficial effects of this improvement are: selecting data from the past seven days as the target window allows for timely reflection of the current production status of the target well. Shale gas well production changes continuously over time, and recent data more accurately reflects the current fluid state and production characteristics within the wellbore. This provides more realistic baseline data for subsequent similarity matching and flow pressure gradient calculations, helping to improve the accuracy and timeliness of judgments. The data includes production data such as tubing inner diameter, casing pressure, oil pressure, gas production, and water production, comprehensively covering key parameters in the gas well production process. Tubing inner diameter affects the flow characteristics of fluid within the wellbore; casing pressure and oil pressure reflect the pressure distribution within the wellbore; and gas production and water production are important indicators for determining whether fluid has accumulated in the wellbore. Acquiring this data provides rich information for subsequent analysis and calculations, ensuring accurate judgment of the wellbore flow regime.

[0017] Preferably, as an improvement, in S3, a sliding window of fixed size of 7 days is used to process the historical data of the selected single well. The sliding window starts from the production date and moves forward by one day each time to generate multiple sets of candidate window data for 7 consecutive days. Each set of candidate window data includes various types of production data of the corresponding candidate single well, and each type of production data includes 7 data arranged in chronological order within the sliding window range.

[0018] The beneficial effects of this improvement are as follows: Using a fixed-size 7-day sliding window ensures a consistent time span for each candidate window, facilitating fair and accurate comparison with the target window data. The 7-day timeframe includes sufficient production data points to reflect the production trends of the gas wells without becoming overly complex, increasing computational difficulty and analytical errors. The sliding window starts from the production date and moves forward one day at a time, generating multiple sets of consecutive 7-day candidate window data. This method comprehensively covers the historical production process of the candidate wells, ensuring no potentially similar production stages are overlooked. By generating multiple sets of candidate window data, the likelihood of finding production states similar to the target well is increased, improving the accuracy and reliability of similarity matching.

[0019] Preferably, as an improvement, in S4, for each set of candidate window data for each candidate well, the corresponding weighted absolute distance is calculated sequentially with the target window data of the target well; each candidate well will obtain multiple sets of distance values; The method for calculating the weighted absolute value distance includes first calculating the absolute value difference between the target window data of the target well and the corresponding production data of the candidate window data of the single well, then multiplying these absolute value differences by the weight corresponding to the production data, and finally summing the weighted absolute value differences of all production data to obtain a distance value between the time series of the candidate window data and the target window data; for a set of candidate window data and target window data for 7 consecutive days, 7 distance values ​​will be obtained.

[0020] The beneficial effects of this improvement are: by employing a weighted absolute value distance calculation method, the importance of different production data to the judgment of wellbore flow and fluid accumulation is fully considered. Different production data play different roles in reflecting wellbore conditions; for example, gas production and water production may be more critical for judging fluid accumulation, while the influence of tubing inner diameter is relatively small. By setting corresponding weights for different production data, the difference between each group of candidate window data and target window data can be more reasonably measured, improving the accuracy and scientific nature of the distance calculation.

[0021] For a set of candidate window data and target window data spanning seven consecutive days, seven distance values ​​are obtained. These seven distance values ​​reflect the degree of difference between the two sets of data at different time points, providing more detailed information for subsequent calculation of the average and determination of the most similar window data. By analyzing the distance values ​​at multiple time points, the similarity between the two sets of data can be more comprehensively assessed, avoiding misjudgments caused by data anomalies at individual time points.

[0022] Preferably, as an improvement, in S5, for each set of distance values ​​of each candidate well, the average value is calculated, the average values ​​of different sets of distance values ​​of the same candidate well are compared, the number of calculations corresponding to the minimum average value is found, and the candidate window data that is most similar to the target well is determined based on the number of calculations.

[0023] The beneficial effect of this improvement is that calculating the average of each group of distance values ​​allows for a comprehensive consideration of distance differences across multiple time points, resulting in a more intuitive and representative similarity index. A smaller average value indicates a smaller difference between the candidate window data and the target window data, and thus a higher similarity. By comparing the average values ​​of different groups of distance values, the candidate window data most similar to the target well can be identified quickly and accurately.

[0024] By determining the most similar candidate window data based on the number of calculations, the specific location and time range of the similar window data within the historical data of the candidate wells were clarified. This provides an accurate foundation for subsequent analysis of the historical data of similar wells to calculate the current wellbore flowing pressure gradient of the target well, and for predicting the future production of the target well based on the subsequent production data of similar wells. This helps to provide more effective optimization guidance for the production strategy of the target well. Attached Figure Description

[0025] Figure 1 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0026] The following detailed description illustrates the specific implementation method: Example The basics are as follows: Figure 1 As shown, a smart calculation method for the wellbore pressure gradient of a shale gas well includes the following steps: S1. Identify the target well and obtain its historical production data for the past seven days as the target window data. The production data includes tubing inner diameter (d), casing pressure (c), oil pressure (t), gas production (g), and water production (w). The target window data is then... ,in, , ...; Each category of production data in the target window data includes 7 data points arranged in chronological order.

[0027] Missing data can be supplemented using interpolation or mean imputation, and outlier data can be cleaned using thresholding, statistical methods, or data smoothing.

[0028] S2. Obtain production data of the candidate wells from the start of production from the historical database. The candidate wells are the other wells besides the target well. Construct a comprehensive historical dataset for subsequent similarity comparison with the window data of the target well.

[0029] S3. For the historical data of each single well obtained in S2, a sliding window of fixed size of 7 days is used for processing. The sliding window starts from the production date and moves forward by one day each time to generate multiple sets of candidate window data for 7 consecutive days.

[0030] Each candidate window data set includes various production data for the corresponding candidate well, and each type of production data within the candidate window data set includes seven data sets arranged in chronological order within the sliding window range.

[0031] S4. For each set of candidate window data for each candidate well, calculate the corresponding weighted absolute distance with the target window data of the target well in turn.

[0032] The distance between a candidate well and a target well is calculated. There are multiple sets of candidate window data for the candidate well. These multiple sets of candidate window data are arranged in sequence according to the sliding direction of the sliding window. The distance between each candidate window data and the target window data is calculated in sequence according to the sliding direction of the sliding window. Each candidate window data and the target window data form a set of distance values. Finally, multiple sets of distance values ​​for the target well corresponding to each candidate well are obtained.

[0033] For example, in the first calculation, the weighted absolute distance is calculated between the first day's production data of the first group of candidate window data for the candidate well and the first day's production data of the target window data for the target well. The weighted absolute distance is calculated between the second day's production data of the first group of candidate window data for the candidate well and the second day's production data of the target window data for the target well. This process continues until the weighted absolute distance is calculated between the seventh day's production data of the first group of candidate window data for the candidate well and the seventh day's production data of the target window data, resulting in a set of distance values, which includes 7 distance values.

[0034] Starting from the sliding window and moving one position to the right, the weighted absolute distance between the second set of candidate window data for the candidate well and the target window data is calculated. This calculation process is repeated until all candidate window data for the candidate well have been calculated with the target window data of the target well. In this way, multiple sets of distance values ​​will be obtained for each candidate well.

[0035] The methods for calculating weighted absolute distance include: First, calculate the absolute difference between the target window data of the target well and the corresponding production data of the candidate window data of the shortlisted wells. Then, multiply these absolute differences by the weight corresponding to the production data. Finally, sum the weighted absolute differences of all production data to obtain a distance value between the time series of the candidate window data and the target window data. For a set of candidate window data and target window data for 7 consecutive days, 7 such distance values ​​will be obtained. These 7 distance values ​​reflect the degree of difference between the two sets of data at the corresponding sliding time points.

[0036] The calculation formula is as follows: Where △ represents the distance value, , , , and The weights for tubing inner diameter, casing pressure, oil pressure gas production, and water production are respectively. , , , , These are the tubing inner diameter, casing pressure, oil pressure gas production, and water production of the candidate window data, respectively. , , , , These are the tubing inner diameter, casing pressure, oil pressure gas production, and water production of the target window data, respectively.

[0037] Methods for determining the weights of production data include: The expert experience method involves inviting experts in shale gas well production to assign weights to each parameter based on their understanding of the importance of various production data in the calculation of wellbore flowing pressure gradient. For example, experts believe that gas production has a greater impact on the wellbore flowing pressure gradient, so they may assign a higher weight to gas production, such as 0.4; while water production has a relatively smaller impact, with a weight of 0.2; the weights of tubing inner diameter, casing pressure, and oil pressure can be assigned as 0.15, 0.15, and 0.1, respectively, depending on the actual situation.

[0038] Data analysis methods can employ correlation analysis and principal component analysis. Correlation analysis calculates the correlation coefficients between various production data and the wellbore flowing pressure gradient in the historical data of the target well. A higher correlation coefficient indicates a closer relationship between the parameter and the wellbore flowing pressure gradient, and should therefore be assigned a higher weight. For example, if calculations show a correlation coefficient of 0.8 for gas production and 0.5 for water production, 0.3 for tubing diameter, 0.4 for casing pressure, and 0.6 for oil pressure, then the weights can be normalized based on these correlation coefficients to obtain the weights for each parameter. Principal component analysis performs principal component analysis on the historical production data of the target well, transforming multiple correlated parameters into a few uncorrelated principal components. The weight of each parameter is determined based on the contribution rate of the principal components. Principal components with higher contribution rates should contain original parameters with relatively larger weights. For example, if the first principal component has a contribution rate of 50% and is mainly composed of gas production and oil pressure, then the weights of gas production and oil pressure can be appropriately increased.

[0039] Model optimization uses weights as parameters of the model. By continuously adjusting the weights, the error between the wellbore pressure gradient calculated by the model and the actual measured value is minimized. Optimization algorithms (such as genetic algorithms and particle swarm optimization) can be used to search for the optimal weight combination. For example, with the goal of minimizing the mean square error between the wellbore pressure predicted by the model and the actual wellbore pressure, a genetic algorithm can be used to iteratively optimize the weights, ultimately obtaining a set of optimal weight values.

[0040] S5. For each set of distance values ​​for each candidate well, calculate its average value. This average value reflects the overall similarity between the candidate window data and the target window data of the target well. Compare the average values ​​of different sets of distance values ​​for the same candidate well, find the number of calculations corresponding to the minimum average value, and determine the candidate window data that is most similar to the target well based on the number of calculations. This most similar candidate window data corresponds to the production period in which the current production status of the candidate well and the target well is most similar. Thus, the most similar candidate window data for each candidate well is determined.

[0041] S6. Compare the minimum average distance of each candidate well. The candidate well with the smallest average distance and its most similar candidate window data are the wells and their production periods that are most similar to the current state of the target well. Output the well number of the most similar well and the window start period of the most similar candidate window data.

[0042] S7. Analyze the production data of the most similar single well during the most similar production period, calculate the current wellbore flowing pressure gradient of the target single well, and then predict the future production of the target single well based on the production data of the similar single well after the most similar production period, so as to provide optimization guidance for the production strategy of the target single well.

[0043] Methods for calculating the current wellbore flowing pressure gradient of a target single well include: Based on the physical principles of shale gas wellbore flow, a gas-liquid two-phase flow model is selected as the theoretical basis for calculating the wellbore pressure gradient. The Beggs-Brill model is chosen here, as it considers the flow characteristics of the gas and liquid phases in the wellbore and is applicable to wellbores with different inclination angles.

[0044] The parameters of the Beggs-Brill model are fitted using wellbore flowing pressure data and corresponding well depths from the most similar well at the most similar period. The Beggs-Brill model involves multiple parameters, such as the slip velocity of the gas-liquid two-phase system and the liquid holdup. These parameters are fitted using a nonlinear regression method, with the objective of minimizing the error between the model's calculated results and actual wellbore flowing pressure data. For example, the objective function is constructed using the least squares method, and the optimal values ​​of the parameters are solved using an iterative algorithm.

[0045] The fitted model is applied to data from other periods or different well sections of the most similar single well to verify the model's accuracy and reliability. The relative error between the wellbore flowing pressure predicted by the model and the actual measured wellbore flowing pressure is calculated. If the relative error is within an acceptable range (e.g., less than 10%), the model is considered valid; otherwise, the model parameters need to be readjusted or other influencing factors need to be considered.

[0046] The standardized tubing inner diameter, casing pressure, oil pressure, gas production, and water production data of the target well are input into the modified Beggs-Brill model. The wellbore of the target well is divided into several segments based on the well depth. The length of each segment can be determined according to actual conditions, for example, 100 meters per segment. Using the wellhead oil pressure as the initial condition, the bottomhole pressure of each segment is iteratively calculated to obtain the wellbore flowing pressure distribution.

[0047] When measured data is available, compare the calculated results with the measured data, analyze the errors, and identify the causes. Conduct sensitivity analysis to determine key parameters. Optimize the model based on the analysis results, such as adjusting parameters or introducing new factors to improve the model.

[0048] Compared to traditional manual calculations, this method for calculating wellbore flowing pressure gradients relies heavily on gas production engineering models and surface parameters. These parameters are often difficult to obtain accurately in practice and are greatly influenced by human factors, leading to significant errors in the calculation results. This method corrects the model parameters by collecting actual data from the most similar wells at the most similar times, making the model more closely reflect reality and reducing errors caused by inaccurate parameters, resulting in more accurate calculations. By using actual data to correct and validate the model, this method improves the model's ability to predict wellbore flowing pressure gradients, achieving higher accuracy and providing a more reliable basis for the production optimization and management of shale gas wells.

[0049] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for intelligent calculation of the flowing pressure gradient in a shale gas wellbore, characterized in that, Includes the following steps: S1. Acquire and preprocess relevant production data of the target single well to form target window data; S2. Obtain the production data of the candidate wells. The candidate wells are the other wells besides the target well. Construct a historical dataset. S3. For the historical data of the single well to be selected, use a sliding window to generate multiple sets of candidate window data; S4. Calculate the distance between each group of candidate window data for each candidate well and the target window data for the target well; S5. Determine the candidate window data that is most similar to the target well for each candidate well based on the distance value; S6. Compare the minimum average distance of each candidate well and output the well and its production period data that are most similar to the current state of the target well. S7. Calculate the current wellbore flowing pressure gradient of the target well by analyzing the historical data of the most similar single well, and predict the future production of the target single well based on the subsequent production data of the similar single well, so as to provide optimization guidance for the production strategy of the target single well.

2. The intelligent calculation method for the flowing pressure gradient of a shale gas wellbore according to claim 1, characterized in that: In S1, the historical production data of the target single well for the past seven days is obtained as the target window data. The production data includes tubing inner diameter, casing pressure, oil pressure, gas production and water production.

3. The intelligent calculation method for the flowing pressure gradient of a shale gas wellbore according to claim 2, characterized in that: In S3, a sliding window of fixed size of 7 days is used to process the historical data of the candidate well. The sliding window starts from the production date and moves forward by one day each time to generate multiple sets of candidate window data for 7 consecutive days. Each set of candidate window data includes various types of production data for the corresponding candidate well, and each type of production data includes 7 data points arranged in chronological order within the sliding window range.

4. The intelligent calculation method for the flowing pressure gradient of a shale gas wellbore according to claim 3, characterized in that: In S4, for each set of candidate window data for each candidate well, the corresponding weighted absolute distance is calculated sequentially with the target window data of the target well; each candidate well will obtain multiple sets of distance values; The method for calculating the weighted absolute value distance includes first calculating the absolute value difference between the target window data of the target well and the corresponding production data of the candidate window data of the single well, then multiplying these absolute value differences by the weight corresponding to the production data, and finally summing the weighted absolute value differences of all production data to obtain a distance value between the time series of the candidate window data and the target window data; for a set of candidate window data and target window data for 7 consecutive days, 7 distance values ​​will be obtained.

5. The intelligent calculation method for the flowing pressure gradient of a shale gas wellbore according to claim 4, characterized in that: In S5, for each set of distance values ​​for each candidate well, calculate its average value, compare the average values ​​of different sets of distance values ​​for the same candidate well, find the number of calculations corresponding to the minimum average value, and determine the candidate window data that is most similar to the target well based on the number of calculations.

Images