A method for detecting water breakthrough in a formation

Abstract

The present application provides a kind of formation water detection method, comprising: obtaining the drilling data corresponding to oil-based drilling and water-based drilling respectively, corresponding to obtain oil-based drilling data and water-based drilling data;Oil-based formation water detection model is trained using the oil-based drilling data, and the trained oil-based formation water detection model is obtained;Water-based formation water detection model is trained using the water-based drilling data, and the trained water-based formation water detection model is obtained;The oil-based drilling data and water-based drilling data to be detected are respectively input into the corresponding model to predict the formation water yield of oil-based drilling and water-based drilling respectively.The present application improves the prediction accuracy of the formation water yield of oil-based drilling and water-based drilling.

Description

Technical Field

[0001] This invention relates to the field of drilling technology, and more particularly to a method for detecting formation water. Background Technology

[0002] Formation water inrush is an unavoidable occurrence during drilling, and it can lead to the introduction of large amounts of sodium into the drilling fluid. + Ca 2+ Mg 2+ Plasma can cause serious contamination of drilling fluids; formation water inflow can also lead to decreased drilling fluid density, increased filtration loss, and damage to the filter cake, resulting in formation collapse and wellbore instability. Once the density is increased, the lubricity and rheological properties of the drilling fluid deteriorate, increasing the bottom hole pressure differential and causing stuck pipe accidents. In addition, failure to deal with overflow water in time can lead to serious accidents such as well kick or blowout. Therefore, accurately judging the formation water inflow situation is an important factor in ensuring the safe conduct of the drilling process.

[0003] Currently, drilling technologies are categorized into oil-based drilling, water-based drilling, and pneumatic drilling. Pneumatic drilling has many mature methods for determining formation water production. However, determining formation water production under oil-based and water-based conditions is relatively simple, relying on a rough estimate based on changes in drilling fluid outflow. If the change in drilling fluid after it exits the well is directly assessed, the presence of drilling cuttings and other solids will affect the formation water production assessment. Conversely, if solids are removed before assessment, some drilling fluid will be lost during the removal process, leading to inaccurate formation water production determination.

[0004] Therefore, in the case of oil-based drilling and water-based drilling, the formation water production cannot be accurately determined. Summary of the Invention

[0005] This invention provides a method for detecting formation water production, which addresses the shortcomings of existing technologies where the formation water production cannot be accurately determined in the case of oil-based and water-based drilling.

[0006] A method for detecting formation water discharge includes:

[0007] Drilling data for oil-based and water-based wells were obtained separately, resulting in corresponding oil-based drilling data and water-based drilling data.

[0008] The oil-based drilling data is used to train an oil-based formation water detection model to obtain a trained oil-based formation water detection model; the water-based drilling data is used to train a water-based formation water detection model to obtain a trained water-based formation water detection model.

[0009] The oil-based drilling data and water-based drilling data to be tested are input into the corresponding models to predict the formation water production of oil-based drilling and water-based drilling, respectively.

[0010] Furthermore, in the formation water detection method described above, the oil-based drilling data includes:

[0011] Drilling fluid inflow rate, drilling fluid inflow oil-water ratio, drilling fluid outflow oil-water ratio.

[0012] Furthermore, in the formation water detection method described above, the oilfield formation water detection model is trained using the following model:

[0013]

[0014] Furthermore, in the formation water detection method described above, the water-based drilling data includes:

[0015] Drilling fluid inflow rate, K + Well concentration, K + Well discharge concentration, Cl - Well concentration, Cl - Well output concentration, formation water K + Concentration, formation water cl - concentration.

[0016] Furthermore, in the formation water emission detection method described above, the water-based formation water emission detection model is trained using the following model:

[0017]

[0018] Furthermore, in the formation water detection method described above, both the oil-based formation water detection model and the water-based formation water detection model are support vector machine-based formation water detection models, and both models include the following parameters:

[0019] The penalty coefficient C, the kernel function K, and the kernel parameters.

[0020] The formation water production detection method provided by this invention uses oil-based drilling data and water-based drilling data respectively, and trains oil-based formation water production detection models and water-based formation water production detection models using the oil-based drilling data and water-based drilling data respectively. Then, the trained oil-based formation water production detection models and water-based formation water production detection models are used to predict the formation water production of oil-based drilling and water-based drilling respectively. This method improves the prediction accuracy of formation water production of oil-based drilling and water-based drilling. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 The formation water detection method flow provided by the present invention Figure 1 ;

[0023] Figure 2 The formation water detection method flow provided by the present invention Figure 2 . Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] Figure 1 The formation water detection method flow provided by the present invention Figure 1 , Figure 2 The formation water detection method flow provided by the present invention Figure 2 ;like Figure 1 , Figure 2 As shown, the method includes the following steps:

[0026] S1: Obtain drilling data for oil-based drilling and water-based drilling respectively, and obtain corresponding oil-based drilling data and water-based drilling data.

[0027] Specifically, oil-based drilling data includes: drilling fluid inlet flow rate, drilling fluid inlet oil-water ratio, and drilling fluid outlet oil-water ratio. Among these, the drilling fluid inlet flow rate can be accurately measured, and the oil phase volume of the drilling fluid does not change during drilling, but the water phase volume increases with formation water production. Therefore, the drilling fluid oil-water ratio directly reflects the formation water production status. This invention uses the above-mentioned oil-based drilling data to determine formation water production, ensuring reliable data sources, high data accuracy, and strong timeliness. Therefore, this invention, by combining drilling fluid inlet flow rate, drilling fluid inlet oil-water ratio, and drilling fluid outlet oil-water ratio, can accurately determine formation water production.

[0028] The water-based drilling data includes: drilling fluid inflow rate, K... + Well concentration, K + Real-time concentration at well outlet, Cl -Well concentration, Cl - Real-time concentration at well outlet, formation water cl + Concentration, formation water cl - concentration.

[0029] Because the drilling fluid inflow rate can be accurately measured, and K in the drilling fluid + cl - The concentrations are relatively stable, and only when formation water enters the drilling fluid will their concentrations change. Therefore, it is important to measure the concentration changes of these two ions after they enter and exit the well, combined with the drilling fluid inflow rate and the stable K+ concentration in the formation water. + cl - Concentration can accurately determine the formation water yield. Moreover, the method of using the above-mentioned water-based drilling data to determine the formation water yield in this invention has reliable data sources, high data accuracy, and strong timeliness.

[0030] S2: Use the oil-based drilling data to train an oil-based formation water detection model to obtain a trained oil-based formation water detection model; use the water-based drilling data to train a water-based formation water detection model to obtain a trained water-based formation water detection model.

[0031] Specifically, this invention establishes an intelligent formation water production judgment model based on support vector machine, trains the model using historical drilling fluid data, and imports real-time acquired drilling fluid data into the trained model for real-time prediction of formation water.

[0032] The following model is used to establish the formation water detection model for the oil base:

[0033]

[0034] The following model was used to establish the water effluent detection model for the aforementioned water base:

[0035]

[0036] S3: Input the oil-based drilling data and water-based drilling data to be tested into the corresponding models to predict the formation water production of oil-based drilling and water-based drilling respectively.

[0037] Furthermore, both the oil-based formation water detection model and the water-based formation water detection model are formation water detection models based on support vector machines. Both models include the following parameters: penalty coefficient C, kernel function K, and kernel parameters.

[0038] This invention uses the support vector machine algorithm to linearly and accurately predict formation water yield by dividing intervals.

[0039] The method of this invention establishes a water production detection model by collecting drilling fluid data in real time. The model predicts the real-time formation water production, providing important guidance for on-site construction, effectively avoiding complex drilling accidents, and improving the efficiency of drilling operations.

[0040] This invention integrates formation water ion concentration data, real-time drilling fluid injection data, real-time oil-water ratio data of oil-based drilling fluid before and after injection, and real-time K-value data of water-based drilling fluid before and after injection. + cl - The ion concentration data is reliable and timely. A support vector machine model was used to determine formation water production. A quantitative calculation model for formation water production based on oil-based oil-water ratio and water-based ion concentration parameters was constructed, which improved the detection accuracy of formation water production.

[0041] This invention can provide real-time detection results during on-site construction, offering guidance to on-site engineers, avoiding complex drilling accidents, improving drilling efficiency, and saving costs.

[0042] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting formation water discharge, characterized in that, include: Drilling data for oil-based and water-based wells were obtained separately, resulting in corresponding oil-based drilling data and water-based drilling data. Using the oil-based drilling data, a formation water production detection model for oil-based wells is trained to obtain a formation water production model for oil-based wells; using the water-based drilling data, a formation water production detection model for water-based wells is trained to obtain a formation water production model for water-based wells. The oil-based drilling data and water-based drilling data to be tested are input into the corresponding models to predict the formation water production of oil-based drilling and water-based drilling respectively. The oil-based drilling data includes: Drilling fluid inlet flow rate, drilling fluid inlet oil-water ratio, drilling fluid outlet oil-water ratio; The formation water production model for the oil-based drilling is as follows: ; The water-based drilling data includes: Drilling fluid inflow rate Well concentration Real-time concentration at well outlet Well concentration Real-time concentration at well outlet and formation water Concentration, formation water concentration; The formation water production model for the water-based drilling is as follows: 。 2. The formation water detection method according to claim 1, characterized in that, Both the formation water production model for oil-based drilling and the formation water production model for water-based drilling are formation water production detection models based on support vector machines. Both models include the following parameters: The penalty coefficient C, the kernel function K, and the kernel parameters.

Images