A well kick risk assessment method based on pre-stack inversion
Patent Information
- Application Number
- CN202510488987.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-18
- Publication Date
- 2026-10-23
AI Technical Summary
这种实时监测方式虽然能够在井漏发生的瞬间及时察觉，但无法在钻前对井漏风险进行预测，存在明显的局限性
[0031]本发明综合考虑了裂缝、多孔隙和地层破裂等因素，可以提前预测井漏，为钻井工程提供预警，降低成本和风险。相比现有技术，本发明综合考虑多种因素，提高了预测准确性；同时，利用地震数据进行预测，可以实现提前预警，为钻井工程提供更多时间和空间进行预防和处理。
Smart Images
Figure CN122962636A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of drilling (exploration) engineering technology and relates to a well leakage risk assessment method based on pre-stack inversion. Background Technology
[0002] Throughout the development of drilling engineering, well leakage has always been a pressing technical challenge. Once well leakage occurs, it can trigger a series of serious consequences. From an economic perspective, significant drilling fluid loss leads to a substantial increase in drilling costs due to the need for continuous replenishment of expensive drilling fluid. From a time perspective, well leakage prolongs the drilling cycle, delays production progress, and significantly reduces production efficiency. Therefore, accurate prediction of well leakage in advance is of paramount importance for avoiding these unnecessary losses and ensuring the smooth progress of drilling projects.
[0003] Well leakage is not caused by a single factor, but rather by the combined effect of multiple factors. First, when the drill bit penetrates a cavity or porous formation, the formation's unique structure allows drilling fluid to easily flow into it, leading to well leakage. Second, encountering fracture zones or faults can also provide seepage channels for drilling fluid, increasing the risk of well leakage. Third, if the mud circulation pressure exceeds the formation fracturing pressure, the formation will be fractured, resulting in well leakage. Additionally, rapid drilling or excessive casing installation can also cause well leakage. Of these factors, the first three are closely related to the formation's inherent characteristics and can be predicted to some extent using scientific methods. However, factors related to the engineering process, such as rapid drilling or excessive casing installation, are difficult to predict in advance due to their inherent randomness and uncontrollability.
[0004] Currently, during drilling and logging, real-time monitoring of well leakage can be achieved by monitoring changes in parameters such as formation pressure, drilling fluid outlet flow rate, and total mud pit volume. While this real-time monitoring method can detect leakage instantly upon its occurrence, it cannot predict leakage risk before drilling, presenting a significant limitation. Seismic methods, however, have demonstrated certain advantages in predicting well leakage, as they can provide basic predictions of leakage caused by the aforementioned three factors.
[0005] Among numerous earthquake prediction methods, conventional methods largely focus on studying fracture development because most well leakage is intricately linked to formation fractures. For example, researchers have developed a relatively effective technique for predicting severe well leakage earthquakes using coherence analysis and continuous spectrum analysis. Another method uses curvature to reflect fractures, suggesting that significant changes in formation curvature indicate a higher degree of fracture development, making well leakage more likely during drilling. However, reality shows that a high degree of fracture development does not necessarily lead to well leakage, and a low degree of fracture development does not mean that well leakage will not occur. This clearly demonstrates that while fractures are an important factor influencing well leakage, they are not the only factor; other factors such as porous formations and formation fracture pressure also need to be considered.
[0006] Therefore, there is an urgent need for a well leakage risk assessment method that breaks through the limitations of traditional prediction methods, can accurately predict well leakage risk before drilling operations begin, and provide timely and effective early warning information for drilling projects. Summary of the Invention
[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a well leakage risk assessment method based on pre-stack inversion. This method can not only help drilling teams to formulate response strategies in advance and reduce cost losses during the drilling process, but also minimize the safety risks caused by well leakage.
[0008] To achieve the above objectives, the present invention employs the following technical solutions:
[0009] A well leakage risk assessment method based on pre-stack inversion includes the following steps:
[0010] S1. Perform frequency division processing on the post-stack seismic data to separate the intermediate-frequency post-stack seismic data and the high-frequency post-stack seismic data;
[0011] S2. Perform crack identification processing based on ant tracking technology on the mid-frequency post-stack seismic data and the high-frequency post-stack seismic data respectively to obtain crack intensity volume 1 representing large and medium-level cracks and crack intensity volume 2 representing small-level cracks.
[0012] S3. Based on pre-stack gather data, the Young's modulus data volume E of rock mechanics parameters is obtained by pre-stack simultaneous inversion.
[0013] S4. Process crack strength body 1 and crack strength body 2, and calculate the relative crack strength Rx and the apparent crack length Rs;
[0014] S5. Calculate and assess well leakage risk based on the comprehensive index model L=Rx×Rs×E.
[0015] As a preferred approach, in step S4, based on the numerical range of the comprehensive index L, the well leakage risk level is divided into four levels: low, low-medium, medium-high, and high, for risk warning before drilling.
[0016] Further preferred, the classification standard for the well leakage risk level is as follows:
[0017] Low risk: Among the sample points within the comprehensive index L range, the well leakage rate is less than 20%;
[0018] Low to medium risk: Among the sample points within the comprehensive index L range, the well leakage rate is 20%-50%;
[0019] Medium to high risk: Among the sample points within the comprehensive index L range, the well leakage rate is 50%-80%;
[0020] High risk: Among the sample points within the comprehensive index L range, the well leakage rate is greater than 80%.
[0021] As a preferred approach, in step S2, the ant tracking technology employs an improved ant colony optimization algorithm, which iteratively optimizes the crack tracking path by simulating the path selection behavior of ants.
[0022] Further preferred, the parameters of the ant colony optimization algorithm include pheromone evaporation coefficient ρ = 0.5, heuristic factor β = 2, and ant number N = 100.
[0023] As a preferred embodiment, in step S3, the pre-stack simultaneous inversion is a Bayesian-based elastic impedance synchronous inversion, jointly inverting the P-wave velocity, S-wave velocity, and density. The inversion objective function is:
[0024] min[||d obs -G(m)|| 2 +λ||mm prior || 2 ]
[0025] Where, d obs For the observed data, G is the forward modeling operator, and λ is the regularization coefficient.
[0026] As a preferred method, in step S4, the relative crack strength Rx is calculated as follows: It is calculated using a normalized weighted method based on the ant tracking results of mid-frequency and high-frequency post-stack seismic data.
[0027] R x =α·max(S1)+(1-α)·max(S2)
[0028] Wherein, S1 and S2 are the local maximum values of fracture strength body 1 and fracture strength body 2, respectively, and the weighting coefficient α is adjusted according to the actual geological conditions, with a default value of 0.6.
[0029] As a preferred method, in step S4, the crack apparent length Rs is calculated as follows: based on the crack network generated by ant tracking, the main axis extension length of each crack is statistically analyzed, and noise interference is eliminated by Gaussian filtering. Finally, the weighted average of the lengths of all cracks in the region is taken.
[0030] The present invention has the following advantages:
[0031] This invention comprehensively considers factors such as fractures, porosity, and formation fracturing, enabling early prediction of well leakage and providing early warning for drilling operations, thereby reducing costs and risks. Compared to existing technologies, this invention comprehensively considers multiple factors, improving prediction accuracy; furthermore, by utilizing seismic data for prediction, it can achieve early warning, providing drilling operations with more time and space for prevention and intervention. Attached Figure Description
[0032] Figure 1 This is a flowchart of the method of the present invention;
[0033] Figure 2 The images show the post-stack crack prediction profile and the pre-stack inverted Young's modulus profile in the embodiments. Detailed Implementation
[0034] The present invention will now be described in detail with reference to the accompanying drawings.
[0035] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0036] Example:
[0037] A well leakage risk assessment method based on pre-stack inversion includes the following steps:
[0038] S1. Perform frequency division processing on the post-stack seismic data to separate the intermediate-frequency post-stack seismic data and the high-frequency post-stack seismic data;
[0039] S2. Perform crack identification processing based on ant tracking technology on the mid-frequency post-stack seismic data and the high-frequency post-stack seismic data respectively to obtain crack intensity volume 1 representing large and medium-level cracks and crack intensity volume 2 representing small-level cracks.
[0040] S3. Based on pre-stack gather data, the Young's modulus data volume E of rock mechanics parameters is obtained by pre-stack simultaneous inversion.
[0041] S4. Process crack strength body 1 and crack strength body 2, and calculate the relative crack strength Rx and the apparent crack length Rs;
[0042] S5. Calculate and assess well leakage risk based on the comprehensive index model L=Rx×Rs×E.
[0043] In step S4, based on the numerical range of the comprehensive index L, the well leakage risk level is divided into four levels: low, low-medium, medium-high, and high, for pre-drilling risk warning. The classification standard for well leakage risk level is as follows:
[0044] Low risk: Among the sample points within the comprehensive index L range, the well leakage rate is less than 20%;
[0045] Low to medium risk: Among the sample points within the comprehensive index L range, the well leakage rate is 20%-50%;
[0046] Medium to high risk: Among the sample points within the comprehensive index L range, the well leakage rate is 50%-80%;
[0047] High risk: Among the sample points within the comprehensive index L range, the well leakage rate is greater than 80%.
[0048] In step S2, the ant tracking technology uses an improved ant colony optimization algorithm to iteratively optimize the crack tracking path by simulating the path selection behavior of ants.
[0049] The parameters of the ant colony optimization algorithm include the pheromone evaporation coefficient ρ = 0.5, the heuristic factor β = 2, and the number of ants N = 100.
[0050] In step S3, the pre-stack simultaneous inversion is a Bayesian-based elastic impedance synchronous inversion, jointly inverting P-wave velocity, S-wave velocity, and density. The inversion objective function is:
[0051] min[||d obs -G(m)|| 2 +λ||mm prior || 2 ]
[0052] Where, d obs For the observed data, G is the forward modeling operator, and λ is the regularization coefficient.
[0053] In step S4, the relative crack strength Rx is calculated as follows: It is calculated using a normalized weighted method based on the ant tracking results of mid-frequency and high-frequency post-stack seismic data.
[0054] R x =α·max(S1)+(1-α)·max(S2)
[0055] Wherein, S1 and S2 are the local maximum values of fracture strength body 1 and fracture strength body 2, respectively, and the weighting coefficient α is adjusted according to the actual geological conditions, with a default value of 0.6.
[0056] In step S4, the crack apparent length Rs is calculated as follows: based on the crack network generated by ant tracking, the main axis extension length of each crack is statistically analyzed, and noise interference is eliminated by Gaussian filtering. Finally, the weighted average of the lengths of all cracks in the region is taken.
[0057] Multi-region validation: In study area B (carbonate rock formation), this method was applied to 128 fracture samples from 12 wells. The well leakage early warning accuracy rate reached 82% (105 / 128), which is better than the traditional fracture strength model (accuracy rate 65%).
[0058] Improved Comparison: Compared with the traditional single-factor model (crack strength only), the average accuracy of the missing data prediction of this model in study areas A and B is improved by about 18% (see Table 1).
[0059] Table 1 Comparison of prediction accuracy of different models
[0060]
[0061] Based on the comprehensive index model for well leakage risk assessment, a well leakage risk assessment model and grading evaluation criteria were established for study area E. Statistical analysis of drilled wells in the study area was conducted to determine whether well leakage occurred in areas with fracture development. Three parameters—relative fracture intensity, apparent fracture length, and Young's modulus—were statistically analyzed, and a comprehensive well leakage index was calculated for each point. Based on this, the range of the comprehensive well leakage index was graded according to well leakage risk, taking into account the occurrence of well leakage.
[0062] The results of fracture prediction in Well A within the study area were obtained using a well leakage early warning model (see attached). Figure 2 Five fractures indicated medium to high risk, of which one had leakage, three had engineering anomalies such as pressure buildup and top drive, and one had no engineering anomalies; three fractures indicated low risk, of which one had leakage and two had no engineering anomalies; the success rate of quality control early warning based on the comprehensive index model for well leakage risk assessment was 75% (6 / 8), as detailed in Table 2.
[0063] Table 2. Well Leakage Early Warning Model Quality Control Table (Well A)
[0064]
[0065] This invention is not limited to the specific embodiments described above. The invention extends to any new feature or combination disclosed in this specification, as well as any new method or process step or combination disclosed herein.
Claims
1. A well leakage risk assessment method based on pre-stack inversion, characterized in that, Includes the following steps: S1. Perform frequency division processing on the post-stack seismic data to separate the intermediate-frequency post-stack seismic data and the high-frequency post-stack seismic data; S2. Perform crack identification processing based on ant tracking technology on the mid-frequency post-stack seismic data and the high-frequency post-stack seismic data respectively to obtain crack intensity volume 1 representing large and medium-level cracks and crack intensity volume 2 representing small-level cracks. S3. Based on pre-stack gather data, the Young's modulus data volume E of rock mechanics parameters is obtained by pre-stack simultaneous inversion. S4. Process crack strength body 1 and crack strength body 2, and calculate the relative crack strength Rx and the apparent crack length Rs; S5. Calculate and assess well leakage risk based on the comprehensive index model L=Rx×Rs×E.
2. The well leakage risk assessment method based on pre-stack inversion according to claim 1, characterized in that: In step S4, based on the numerical range of the comprehensive index L, the well leakage risk level is divided into four levels: low, low-medium, medium-high, and high, for risk warning before drilling.
3. The well leakage risk assessment method based on pre-stack inversion according to claim 2, characterized in that: The classification criteria for well leakage risk levels are as follows: Low risk: Among the sample points within the comprehensive index L range, the well leakage rate is less than 20%; Low to medium risk: Among the sample points within the comprehensive index L range, the well leakage rate is 20%-50%; Medium to high risk: Among the sample points within the comprehensive index L range, the well leakage rate is 50%-80%; High risk: Among the sample points within the comprehensive index L range, the well leakage rate is greater than 80%.
4. The well leakage risk assessment method based on pre-stack inversion according to claim 1, characterized in that: In step S2, the ant tracking technology uses an improved ant colony optimization algorithm to iteratively optimize the crack tracking path by simulating the path selection behavior of ants.
5. The well leakage risk assessment method based on pre-stack inversion according to claim 4, characterized in that: The parameters of the ant colony optimization algorithm include the pheromone evaporation coefficient ρ = 0.5, the heuristic factor β = 2, and the number of ants N = 100.
6. The well leakage risk assessment method based on pre-stack inversion according to claim 1, characterized in that: In step S3, the pre-stack simultaneous inversion is a Bayesian-based elastic impedance synchronous inversion, jointly inverting P-wave velocity, S-wave velocity, and density. The inversion objective function is: min[||d obs -G(m)|| 2 +λ||mm prior || 2 ] Where, d obs For the observed data, G is the forward modeling operator, and λ is the regularization coefficient.
7. The well leakage risk assessment method based on pre-stack inversion according to claim 1, characterized in that: In step S4, the relative crack strength Rx is calculated as follows: It is calculated using a normalized weighted method based on the ant tracking results of mid-frequency and high-frequency post-stack seismic data. R x =α·max(S1)+(1-α)·max(S2) Wherein, S1 and S2 are the local maximum values of fracture strength body 1 and fracture strength body 2, respectively, and the weighting coefficient α is adjusted according to the actual geological conditions.
8. The well leakage risk assessment method based on pre-stack inversion according to claim 1, characterized in that: In step S4, the crack apparent length Rs is calculated as follows: based on the crack network generated by ant tracking, the main axis extension length of each crack is statistically analyzed, and noise interference is eliminated by Gaussian filtering. Finally, the weighted average of the lengths of all cracks in the region is taken.