High temperature and high pressure well shut-in control method and device

Abstract

The application discloses a high-temperature and high-pressure well shut-in control method and device, and the control method constructs and trains an LSTM model through the following steps: S11, constructing a data set, wherein each piece of data in the data set comprises shut-in stand pressure P s , shut-in casing pressure P a , wellbore bottom pressure P w , wellbore temperature T and shut-in time m; S12, calculating the activation values of input gates, forgetting gates and output gates, updating the cell state C t of the LSTM model and calculating output h t ; S13, calculating the change rate of the wellbore bottom pressure P w , updating the number n L of layers and the number N h of neurons of the LSTM model and updating the cell state; S14, correcting h t based on the wellbore temperature and the wellbore bottom pressure; and S15, setting a learning rate theta for learning. The application constructs the LSTM model, and is used for calculating the shut-in time m according to the shut-in stand pressure P s , the shut-in casing pressure P a , the wellbore bottom pressure P w and the wellbore temperature T when overflow occurs in a high-temperature and high-pressure well.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of petroleum engineering, and particularly relates to a high-temperature and high-pressure well shut-in control method and device. BACKGROUND

[0002] Nowadays, oil and gas development in China gradually marches into deep water and deep formation, and with the progress of technology and the growth of market demand, more and more high-temperature and high-pressure wells begin to be developed. High-temperature and high-pressure wells are usually located in deep water areas or complex geological areas, and the temperature and pressure level of the wellbore is significantly higher than that of conventional wells. The development of high-temperature and high-pressure wells improves the efficiency of oil and gas resource exploitation, but also brings new challenges.

[0003] During the drilling of high-temperature and high-pressure wells, overflow is prone to occur due to abnormal temperature and pressure conditions and complex geological conditions. After overflow occurs, the formation pressure usually needs to be obtained by first shutting in, and then subsequent circulation of exhaust gas, well killing and other well control measures are carried out according to the formation pressure. Therefore, whether the formation pressure can be accurately obtained is particularly important for handling overflow and other downhole abnormal conditions.

[0004] Conventionally, after overflow occurs in a conventional well, based on the subsequent flow effect after shut-in and the ascending migration speed of gas, the well is shut in for 10-15 minutes (10-15 minutes is the time point at which the bottom hole pressure reaches the formation pressure), then a small displacement pump is started, the float valve in the drill string is opened, and the shut-in standing pressure or shut-in casing pressure is read, and then the formation pressure is calculated by adding the static column pressure to the shut-in standing pressure or shut-in casing pressure.

[0005] However, during the shut-in process of high-temperature and high-pressure wells, the invasion rate of formation gas into the wellbore gradually decreases, the invasion time is prolonged, and the migration speed of gas in the wellbore changes due to the influence of the formation wellbore temperature and pressure, and the thermal expansion effect of drilling fluid and gas in the wellbore is more obvious than that of conventional wells, and many factors cause the dynamic changes of wellhead pressure and downhole pressure to be often nonlinear.

[0006] Therefore, 10-15 minutes of shut-in is not the time point at which the bottom hole pressure reaches the formation pressure, and the shut-in standing pressure or shut-in casing pressure read at this time cannot be used to calculate the formation pressure. In summary, the time point (i.e. shut-in time) for shut-in pressure calculation when overflow occurs in a high-temperature and high-pressure well is difficult to determine due to the coupling effect of many factors, resulting in difficulties in shut-in formation pressure calculation.

[0007] In view of this, the present application is proposed. SUMMARY

[0008] The present application aims to solve the problems in the prior art and provides a high-temperature and high-pressure well shut-in control method, which constructs an LSTM model for calculating the formation pressure when overflow occurs in a high-temperature and high-pressure well according to the shut-in standing pressure P s , the shut-in casing pressure P a, wellbore bottom pressure P w , wellbore temperature T calculation shut-in time m.

[0009] In order to achieve the above purpose, the application adopts the following technical solutions:

[0010] A high temperature and high pressure well shut-in control method, comprising the following steps:

[0011] S1: constructing and constructing an LSTM model, the S1 comprising the following steps:

[0012] S11: constructing a data set, each data in the data set comprising shut-in stand pressure P s , shut-in casing pressure P a , wellbore bottom pressure P w , wellbore temperature T, shut-in time m;

[0013] S12: calculating the activation value of the input gate, the forgetting gate and the output gate, updating the cell state C t of the LSTM model, calculating the output h t ;

[0014] S13: calculating the change rate of the wellbore bottom pressure P w , updating the number of layers n L and the number of neurons N h of the LSTM model, updating the cell state;

[0015] S14: correcting h t based on the wellbore temperature and the wellbore bottom pressure;

[0016] S15: setting the learning rate θ for learning.

[0017] Further, in S11, the shut-in stand pressure P s , the shut-in casing pressure P a , the wellbore bottom pressure P w and the wellbore temperature T are the inputs of the LSTM model, and the shut-in time m is the output of the LSTM model.

[0018] Further, in S12, the activation values of the input gate, the forgetting gate and the output gate are calculated by the following formula:

[0019] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0020] f t =σ(W f ·[h t-1 ,x t ]+b f)

[0021] o t = σ(W o ·[h t-1 ,x t ]+b o )

[0022] where i t is the activation value of the input, f t is the activation value of the forget gate, o t is the activation value of the output gate, W i , W f , and W o are the weight matrices of the input gate, the forget gate, and the output gate, respectively, b i , b f , and b o are the bias vectors of the input gate, the forget gate, and the output gate, respectively, σ is a sigmoid activation function, x t is the input at the current time step, and h t-1 is the output at the previous time step.

[0023] Further, in S12, the cell state C t of the LSTM model is updated by the following formula, and the output h t is calculated:

[0024]

[0025] h t = o t *tanh(C t )

[0026] where C t is the cell state at the current time step, C t-1 is the cell state at the previous time step; is the candidate state.

[0027] Further, the number of layers n L and the number of neurons N h of the LSTM model are updated by the following steps in S13:

[0028] A131: Calculate the change amount ΔL of the loss value at the current time step by the following formula:

[0029] ΔL = |L t -L t-1 |

[0030] where L t is the loss value calculated by the loss function at the current time step, and L t-1 is the loss value calculated by the loss function at the previous time step.

[0031] A132: Update the number of layers n of the LSTM model using the following formula. L and the number of neurons N h :

[0032] n L =n L +ΔL·λ1

[0033] N h =N h +ΔL·λ2

[0034] Where λ1 and λ2 are hyperparameters, determined based on the bottom pressure P of the wellbore. w The rate of change is dynamically adjusted.

[0035] Furthermore, S3 includes the following steps to update the cell state of the LSTM model:

[0036] B131: The dynamic gating weights of the LSTM model are calculated using the following formula to control whether the cell state at the current time step is updated with weighted weights based on the wellbore bottom pressure data at the current time step:

[0037] G t =σ(W g r t +b g )

[0038] Among them, W g Let b be the weight matrix of the gating mechanism. g σ is the bias term of the gating mechanism, and σ is the sigmoid activation function;

[0039] B132: Update the cell state of the LSTM model using the following formula.

[0040] C′ t =G t ·C t +(1-G t )·C t-1 .

[0041] Furthermore, step S14 includes the following steps:

[0042] S141: Calculate the wellbore temperature influence factor T using the following formula. f This is used to represent the effect of wellbore temperature T on downhole fluid properties:

[0043] T f =αT+β

[0044] Where α and β are constants.

[0045] S142: calculate the corrected wellbore bottom pressure P by the following formula adj :

[0046]

[0047] wherein gamma is a compressibility influence factor;

[0048] S143: correct h by the following formula t :

[0049] h' t = h t · P adj .

[0050] Further, in S15, the learning rate theta is dynamically set by the following formula:

[0051] m t = beta1m t-1 + (1-beta1)g t

[0052]

[0053] wherein m t is the first moment, v t is the second moment, g t is the current gradient, beta1 and beta2 are the decay rates of the first moment and the second moment, and o is a small constant to prevent the denominator from being 0.

[0054] Further, the method further comprises the following steps:

[0055] S2: when overflow occurs, shut in the well, and obtain the shut-in standing pressure P s , the shut-in casing pressure P a , the wellbore bottom pressure P w , and the wellbore temperature T after 10-15 minutes after shut-in;

[0056] S3: calculate the shut-in time m based on the LSTM model constructed in S1, and control the small pump amount of the mud pump to open the float valve in the drill string after the shut-in time is reached;

[0057] S4: obtain the shut-in standing pressure P s or the shut-in casing pressure P a , and calculate the formation pressure by adding the hydrostatic pressure to the shut-in standing pressure P s or the shut-in casing pressure P a .

[0058] In order to achieve the above purpose, the application also adopts the following technical solutions:

[0059] A well shut-in control device for high temperature and high pressure wells, comprising:

[0060] a memory for storing the computer program; and

[0061] a processor for implementing the steps of the control method according to any one of the present application when executing the computer program.

[0062] Compared with the prior art, the present application has the following beneficial effects:

[0063] An LSTM model is constructed, which is used to calculate the shut-in time m according to the stand pressure P s , the casing pressure P a , the wellbore bottom pressure Pw and the wellbore temperature T when overflow occurs in a high-temperature and high-pressure well; the LSTM model can intelligently predict time series data, and when the LSTM model is constructed, the number of layers n w , the number of neurons N L and the cell state of the LSTM model are updated according to the change rate of the wellbore bottom pressure P h , and the correction h t is based on the wellbore temperature and the wellbore bottom pressure, so that the LSTM constructed in the present application has high accuracy when calculating the shut-in time m; the method of the present application can greatly improve the accuracy and efficiency of calculating the formation pressure, and provides a strong guarantee for the implementation of subsequent well control measures. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 is a flowchart for constructing the LSTM model;

[0065] Figure 2 is a structure diagram of the LSTM model;

[0066] Figure 3 is a flowchart of the shut-in control method for a high-temperature and high-pressure well. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments.

[0068] Embodiment one:

[0069] A shut-in control method for a high-temperature and high-pressure well, as shown in the figure, comprises the following steps: Figures 1-2

[0070] S1: constructing and training an LSTM model, wherein the S1 comprises the following steps:

[0071] S11: constructing a data set, wherein each piece of data in the data set comprises the stand pressure P s , the casing pressure P a , the wellbore bottom pressure P w ​, wellbore temperature T, shut-in time m;

[0072] S12: Calculate the activation values of the input gate, the forget gate and the output gate, update the cell state C of the LSTM model t , calculate the output h t ;

[0073] S13: Calculate the rate of change of the wellbore bottom pressure P w , update the number of layers n L and the number of neurons N h of the LSTM model, update the cell state;

[0074] S14: Correct h t based on wellbore temperature and wellbore bottom pressure;

[0075] S15: Set the learning rate theta for learning.

[0076] The high-temperature and high-pressure well shut-in control method of the embodiment constructs an LSTM model for calculating the shut-in time m when overflow occurs in the high-temperature and high-pressure well according to the shut-in standing pressure P s , the shut-in casing pressure P a , the wellbore bottom pressure P w and the wellbore temperature T. The LSTM model can intelligently predict time series data, and in the construction of the LSTM model, the number of layers n w , the number of neurons N L and the cell state of the LSTM model are updated according to the rate of change of the wellbore bottom pressure P h , and h t is corrected based on the wellbore temperature and the wellbore bottom pressure. Therefore, the LSTM constructed in the embodiment has high accuracy when calculating the shut-in time m. The method based on the embodiment can greatly improve the accuracy and efficiency of calculating the formation pressure, and provides a strong guarantee for the implementation of subsequent well control measures.

[0077] The high-temperature and high-pressure well shut-in control method of the embodiment first constructs a data set according to the known data of the current high-temperature and high-pressure well, and then trains and tests based on the data set to construct an LSTM model. When constructing data, data can be obtained according to the empirical method. For example, when overflow occurs, the shut-in time is set according to experience to shut in and calculate the formation pressure. If the calculated formation pressure is too large, the well cannot be successfully shut in, and gas invasion will occur. If the calculated formation pressure is too small, the well cannot be successfully shut in, and loss will occur. Based on this, the accuracy of the shut-in time can be deduced, and the accurate shut-in time is finally obtained to construct the data set.

[0078] In an optional embodiment, in S11, the shut-in standing pressure P s , the shut-in casing pressure P a , the wellbore bottom pressure P wand wellbore temperature T as the input of the LSTM model, and the shut-in time m as the output of the LSTM model.

[0079] In an optional embodiment, in S12, the activation values of the input gate, the forget gate and the output gate are calculated by the following formulas:

[0080] i t = σ(W i ·[h t-1 ,x t ]+b i )

[0081] f t = σ(W f ·[h t-1 ,x t ]+b f )

[0082] o t = σ(W o ·[h t-1 ,x t ]+b o )

[0083] where i t is the activation value of the input gate, used to control the influence of the current input, f t is the activation value of the forget gate, used to control the degree of forgetting of the previous state, o t is the activation value of the output gate, used to control the output of the current cell state, W i , W f , and W o are the weight matrices of the input gate, the forget gate and the output gate respectively, b i , b f , and b o are the bias vectors of the input gate, the forget gate and the output gate respectively, σ is the sigmoid activation function, x t is the input of the current time step, and h t-1 is the output of the previous time step.

[0084] In an optional embodiment, in S12, the cell state C t of the LSTM model is updated, and the output h t is calculated by the following formulas:

[0085]

[0086] h t = o t *tanh(C t )

[0087] where C tis the cell state of the current time step, used to store long-term memory, C t is 0, C t-1 is the cell state of the previous time step; is the candidate state.

[0088] In an optional embodiment, the rate of change r w of the wellbore bottom pressure P t at the current time step is calculated at S13 by the following formula:

[0089]

[0090] In an optional embodiment, the S13 updates the number of layers n L and the number of neurons N h of the LSTM model by the following steps:

[0091] A131: Calculate the change amount ΔL of the loss value of the current time step by the following formula:

[0092] ΔL = |L t - L t-1 |

[0093] wherein L t is the loss value calculated by the loss function at the current time step, and L t-1 is the loss value calculated by the loss function at the previous time step;

[0094] A132: Update the number of layers n L and the number of neurons N h of the LSTM model by the following formula:

[0095] n L = n L + ΔL·λ1

[0096] N h = N h + ΔL·λ2

[0097] wherein λ1 and λ2 are hyperparameters, which are dynamically adjusted according to the rate of change of the wellbore bottom pressure P w .

[0098] In order to adapt to the complex changes in the shut-in pressure process of high temperature and high pressure wells, the network structure of the LSTM model is dynamically adjusted in this optional embodiment: specifically, at the initial stage of shut-in, the number of layers and neurons of the LSTM model is increased to enhance the short-term memory ability, because the wellbore bottom pressure changes rapidly at this time; in the later stage of shut-in, the number of layers and neurons of the LSTM model is gradually reduced to reduce the calculation complexity and avoid overfitting, because the wellbore bottom pressure tends to be stable at this time.

[0099] In practice, λ1 and λ2 can be adjusted using the following formula:

[0100] λ t =λ t-1 (1+α*r t )

[0101] The value of α ranges from 0.1 to 0.5.

[0102] In an optional embodiment, S3 includes the step of updating the cell state of the LSTM model as follows:

[0103] B131: The dynamic gating weights of the LSTM model are calculated using the following formula to control whether the cell state at the current time step is updated with weighted weights based on the wellbore bottom pressure data at the current time step:

[0104] G t =σ(W g r t +b g )

[0105] Among them, W g This is the weight matrix for the gating mechanism, used to adjust the impact of the rate of pressure change on the network state, b g σ is the bias term of the gating mechanism, and σ is the sigmoid activation function;

[0106] B132: Update the cell state of the LSTM model using the following formula.

[0107] C′ t =G t ·C t +(1-G t )·C t-1 .

[0108] In this optional embodiment, the weights of the forget gate, input gate, and output gate of the cells in the LSTM model are adjusted by calculating dynamic gating values ​​based on the rate of change of pressure at the bottom of the wellbore. In the early stage of shutting in a high-temperature and high-pressure well, when the rate of change of pressure at the bottom of the wellbore is large, the LSTM model can more flexibly adjust the activation of these gates in order to better capture rapidly changing features.

[0109] In an optional embodiment, S14 includes the following steps:

[0110] S141: Calculate the wellbore temperature influence factor T using the following formula. f This is used to represent the effect of wellbore temperature T on downhole fluid properties:

[0111] T f =αT+β

[0112] wherein a and b are constants used to describe the magnitude and offset of the effect of the wellbore temperature T on the downhole fluid; the wellbore temperature change of the high-temperature and high-pressure well has an important influence on the fluid density, viscosity and other characteristics, and further affects the change of the wellbore bottom pressure 0, so the wellbore temperature influence factor T is introduced in this optional embodiment to represent the influence of the wellbore temperature on the wellbore bottom pressure;

[0113] S142: The corrected wellbore bottom pressure P is calculated by the following formula: adj :

[0114]

[0115] wherein g is a compressibility influence factor used to adjust the degree of influence of the wellbore temperature on the wellbore bottom pressure, and the value range is 1x10 -3 ~ 3x10 -3 ; under high-temperature and high-pressure conditions, the compressibility of the fluid becomes significant, which affects the change of the wellbore bottom pressure, so the compressibility influence factor is introduced in this optional embodiment to simulate this effect;

[0116] S143: h is corrected by the following formula: t :

[0117] h′ t = h t · P adj .

[0118] In the shut-in process of the high-temperature and high-pressure well, the changes of the wellbore temperature and the wellbore bottom pressure have a mutual influence on the downhole fluid, so in this optional embodiment the influences of the wellbore temperature and the wellbore bottom pressure are combined to correct ht.

[0119] In other words, in this optional embodiment ht is corrected by the wellbore temperature-wellbore bottom pressure double scale, and the goal is to consider the double influence of the wellbore temperature and the pressure on the behavior of the fluid under the coupling relationship between the wellbore temperature and the wellbore bottom pressure under high-temperature and high-pressure conditions, and further improve the accuracy of the LSTM model.

[0120] In addition, in this optional embodiment the output is dynamically adjusted on the basis of considering the interaction between the wellbore temperature and the wellbore bottom pressure, so that the model is more in line with the actual situation of the high-temperature and high-pressure well, and further improves the accuracy of the shut-in pressure calculation.

[0121] In one optional embodiment, in S15, the learning rate q is dynamically set by the following formula:

[0122] mt = β1m t-1 + (1 - β1)g t

[0123]

[0124] wherein m t is the first moment, the first moment is an exponentially weighted average of past gradients, indicating the direction of the gradient, v t is the second moment, the second moment is an exponentially weighted average of the square of each gradient, reflecting the "change in the magnitude of the gradient", g t is the current gradient, the current gradient represents the gradient of the loss function calculated by the current model parameters, β1and β2are the decay rates of the first moment and the second moment, usually in the range of 0-1, and ò is a small constant to prevent the denominator from being 0.

[0125] In addition, it needs to be explained that gradient descent can be used for iterative training, and the training ends when the maximum number of iterations is reached or the error condition is met. This process can be completed using existing technology, which will not be expanded here.

[0126] In an optional embodiment, as Figure 3 shown, it further comprises the following steps:

[0127] S2: When overflow occurs, shut in the well, and obtain the shut-in standing pressure P s , the shut-in casing pressure P a , the bottom hole pressure P w , and the wellbore temperature T within 10-15 minutes after shut-in;

[0128] S3: Calculate the shut-in time m based on the LSTM model constructed in S1, and control the small pump of the mud pump to open the float valve in the drill string after the shut-in time is reached;

[0129] S4: Obtain the shut-in standing pressure P s or the shut-in casing pressure P a , and calculate the formation pressure by adding the hydrostatic pressure to the shut-in standing pressure P s or the shut-in casing pressure P a .

[0130] To verify the effectiveness of the high temperature and high pressure well shut-in control method of the embodiment, the data calculated by the embodiment is compared with the simulated data, and the basic parameters are shown in Table 1.

[0131] Table 1- Basic parameters

[0132] Parameter name / Unit Value Well depth / m 3000 Wellhead temperature / °C 30 Drilling fluid density / g / cm 3 ]] 1.35 Formation porosity / % 18 Formation pressure / MPa 50

[0133] Input the shut-in standing pressure P s , the shut-in casing pressure Pa , wellbore bottom pressure P w , wellbore temperature T data, time interval is 1s, the well shut-in time m calculated by the embodiment is 32.3min, the wellhead casing pressure P a of this moment is 10.6MPa, the formation pressure is 50MPa, the actual data of the simulation calculation is 33min, the wellhead casing pressure of this moment is 10.46MPa, the formation pressure is 50.4MPa, the well shut-in time error is 2.4%; the prediction time of the method of the embodiment is less than 100s, which can quickly respond.

[0134] Embodiment two:

[0135] A high temperature and high pressure well shut-in control device, comprising:

[0136] a memory for storing a computer program; and,

[0137] a processor for executing the computer program to implement the steps of the control method provided by any one of the embodiments.

[0138] The above is only the preferred specific implementation of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art can make equivalent replacement or change according to the technical scheme and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A method of high temperature, high pressure well shut-in control, c h a r a c t e r i s e d in that Comprising the following steps: S1: constructing and training an LSTM model, the S1 comprising the following steps: S11: Construct a data set, each piece of data in the data set including shut-in standing pressure P s , shut-in casing pressure P a , wellbore bottom hole pressure P w , wellbore temperature T, shut-in time m; S12: Calculate the activation values of the input gate, the forget gate and the output gate, update the cell state C of the LSTM model t , calculate the output h t ; S13: calculate the rate of change of the wellbore bottom pressure P w , update the number of layers n of the LSTM model L and the number of neurons N h , update the cell state; S14: Correct h based on wellbore temperature and wellbore bottomhole pressure t ; S15: setting a learning rate θ for learning; The S13 updates the number n of layers of the LSTM model by the following steps L and the number N of neurons h : A131: calculating the change amount ΔL of the current time step loss value by the following formula: ； wherein L t is the loss value calculated by the loss function at the current time step, L t-1 is the loss value calculated by the loss function at the previous time step; A132: Update the number of layers n of the LSTM model by the following formula L and the number of neurons N h : ； wherein λ1, λ2 are hyperparameters, dynamically adjusted according to the rate of change of the bottom hole pressure P w of the wellbore The S13 comprises the following steps of updating the cell state of the LSTM model: B131: calculating the dynamic gating weight of the LSTM model by the following formula, for controlling whether the cell state of the current time step is updated by the wellbore bottom pressure data of the current time step: ； where W g is the weight matrix of the gating mechanism, b g is the bias term of the gating mechanism, r t is the rate of change of the wellbore bottomhole pressure P w , and σ is the sigmoid activation function. B132: updating the cell state of the LSTM model by the following formula: 。 2. The method of claim 1, wherein, In S11, the shut-in stand pressure P s , the shut-in casing pressure P a , the wellbore bottom pressure P w and the wellbore temperature T are inputs of the LSTM model, and the shut-in time m is the output of the LSTM model.

3. The method of claim 1, wherein, In S12, the activation values of the input gate, the forget gate and the output gate are calculated by the following formula: ； where i t is the activation value of the input gate, f t is the activation value of the forget gate, o t is the activation value of the output gate, W i , W f , W o are the weight matrices of the input gate, the forget gate, and the output gate, respectively, b i , b f , b o are the bias vectors of the input gate, the forget gate, and the output gate, respectively, σ is the sigmoid activation function, x t is the input at the current time step, h t-1 is the output at the previous time step.

4. The method of claim 3, wherein, In S12, the cell state C of the LSTM model is updated by the following formula t , the output h t is calculated: ； ； where C t is the cell state at the current time step, C t-1 is the cell state at the previous time step; is the candidate state.

5. The method of claim 1, wherein, The S14 comprises the following steps: S141: Calculate a wellbore temperature impact factor T by the following equation f , for representing the impact of wellbore temperature T on the characteristics of downhole fluid: ； Wherein, α and β are constants; S142: Calculate the corrected wellbore bottomhole pressure P by the following equation adj : ； Wherein, γ is a compressibility influence factor; S143: Correct h by the following equation t : 。 6. The method of claim 1, wherein, In S15, the learning rate θ is dynamically set by the following formula: ； where m t is the first moment, v t is the second moment, g t is the current gradient, β1, β2 are the decay rates of the first moment and the second moment, respectively, is a small constant to prevent the denominator from being zero.

7. The method of claim 1, wherein, Further comprising the following steps: S2: shut in after overflow occurs, and obtain the shut-in standing pressure P after 10-15 minutes of shut-in s , shut-in casing pressure P a , wellbore bottom hole pressure P w , wellbore temperature T; S3: calculating the shut-in time m based on the LSTM model constructed in S1, and opening the top drill string inner float valve after the shut-in time arrives to control the small pump volume of the mud pump to open the drill string; S4: Obtain shut-in standing pressure P s or shut-in casing pressure P a , calculate formation pressure by shut-in standing pressure P s or shut-in casing pressure P a plus hydrostatic pressure.

8. A high temperature high pressure well shut-in control device, characterized by Comprising: A memory for storing a computer program; And, A processor for executing the computer program to implement the steps of the control method according to any one of claims 1 to 7.

Images