Drilling data intelligent analysis system and method

Abstract

The present invention relates to the field of drilling data analysis, and in particular to a drilling data intelligent analysis system and method. The drilling data intelligent analysis system comprises: a drilling data acquisition module, a working parameter adjustment module and a fault prediction module. The present invention uses advanced data analysis technology to monitor and analyze drilling environment data and drilling working parameters in real time, and provides the optimal drill speed and pressure settings for drilling operations, thereby dynamically adjusting drilling parameters to adapt to the ever-changing underground environment; this intelligent adjustment mechanism not only improves the efficiency of drilling operations, but also significantly reduces equipment wear and failure risks; this real-time, data-driven decision support system makes drilling operations more accurate and efficient, while ensuring the safety of operations.

Description

Technical Field

[0001] The present invention relates to the field of drilling data analysis, and in particular to a drilling data intelligent analysis system and method. Background Art

[0002] In the fields of oil, natural gas, geological exploration, etc., drilling operations are a key step in obtaining underground resources; with the development of technology, drilling operations have transformed from traditional manual operations to highly automated processes; however, the complexity and uncertainty of the environment during the drilling process, such as formation hardness, formation pressure, temperature, etc., pose challenges to drilling efficiency and equipment safety; traditional drilling methods often rely on the operator's experience and intuition to adjust drilling parameters. This method is not only inefficient, but also difficult to adapt to the changing underground environment, resulting in low drilling efficiency, severe equipment wear, and may even cause safety accidents.

[0003] In order to improve the efficiency and safety of drilling operations, in recent years, researchers have begun to explore the use of intelligent analysis methods to optimize the drilling process. These methods usually involve real-time monitoring and analysis of drilling environment data and drilling working parameters to dynamically adjust drilling parameters to adapt to formation changes; however, existing intelligent analysis methods still face some challenges in practical applications; traditional drilling data acquisition methods often only focus on a limited number of parameters, such as drill bit speed and pressure, while ignoring key environmental factors such as formation hardness, formation pressure, temperature, etc., resulting in limited accuracy and applicability of analysis results; at the same time, existing drilling parameter adjustment methods are usually based on static models and rules, lack the ability to respond quickly to real-time data, and cannot adapt to formation changes in time, resulting in drilling efficiency and equipment safety not being effectively guaranteed; on the other hand, drilling equipment fault prediction usually relies on limited historical data and simple statistical methods, lacks in-depth understanding and prediction capabilities of equipment status changes, resulting in low accuracy of fault prediction; in response to the above problems, the present invention proposes a drilling data intelligent analysis system and method, which dynamically adjusts drilling parameters to improve drilling efficiency and equipment safety by real-time monitoring and analysis of drilling environment data and drilling working parameters, and realizes early warning and prevention of drilling equipment failures through advanced fault prediction models. Summary of the invention

[0004] The present invention uses advanced data analysis technology to monitor and analyze drilling environment data and drilling working parameters in real time, and provides the optimal drill speed and pressure settings for drilling operations, thereby dynamically adjusting drilling parameters to adapt to the ever-changing underground environment; this intelligent adjustment mechanism not only improves the efficiency of drilling operations, but also significantly reduces equipment wear and failure risks; this real-time, data-driven decision support system makes drilling operations more precise and efficient while ensuring the safety of operations.

[0005] Intelligent analysis method of drilling data, including:

[0006] At the current monitoring time point, drilling environment data and drilling working parameters are obtained, the drilling environment data including formation hardness, formation pressure, temperature, and friction coefficient; the drilling working parameters including drill bit speed and drill bit pressure; based on the currently obtained drilling environment data, it is determined whether the drilling working parameters need to be adjusted; if so, the obtained drilling environment data is used as the input of the working parameter acquisition model, the optimal working parameters are output, and the obtained optimal working parameters are applied to adjust the drilling working parameters; if not, no operation is performed;

[0007] Acquire drilling equipment data at the current monitoring time point, the drilling equipment data including equipment temperature, vibration frequency and operating pressure;

[0008] A frequency adjustment cycle is set. At the beginning of any frequency adjustment cycle, the drilling environment data obtained in the previous frequency adjustment cycle is used to calculate the drilling environment characteristic value. A frequency adjustment cycle includes several fault prediction time points, and the fault prediction frequency is adjusted according to the drilling environment characteristic value.

[0009] At the current fault prediction time point, the most recently acquired N drilling equipment data are used as a fault prediction data set, the fault prediction data set is used as the input of the fault prediction model, and the fault prediction result is output; based on the fault prediction result, corresponding response measures are executed.

[0010] Preferably, it is determined whether the drilling working parameters need to be adjusted, and the specific operations are as follows:

[0011] The currently acquired drilling environment data is compared with the drilling environment data when the drilling working parameters were last adjusted to obtain the formation hardness difference, formation pressure difference, temperature difference and friction coefficient difference; the formation hardness change threshold, formation pressure change threshold, temperature change threshold and friction coefficient change threshold are set. If at least one of the formation hardness difference, formation pressure difference, temperature difference and friction coefficient difference is greater than the corresponding change threshold, the judgment result is that the drilling working parameters need to be adjusted; otherwise, the judgment result is that the drilling working parameters do not need to be adjusted.

[0012] Preferably, the drilling environment characteristic value is calculated, and the specific operations are as follows:

[0013] Set the frequency adjustment cycle. At the beginning of any frequency adjustment cycle, select the formation hardness data from the drilling environment data obtained in the previous frequency adjustment cycle. , friction coefficient data and maximum formation pressure data F, i=1, 2, ..., n; n represents the number of selected formation hardness data and friction coefficient data;

[0014] For formation hardness data , calculate the average formation hardness , set the formation hardness range to to , when the average formation hardness lie in to When between Calculate the first eigenvalue ; When the average formation hardness Less than or equal to When the first eigenvalue is 0; when the average formation hardness Greater than or equal to When the first eigenvalue is 1;

[0015] According to the formation hardness data And the average formation hardness obtained , using the formula Calculate the second eigenvalue ;

[0016] For friction coefficient data , calculate the average friction coefficient , set the friction coefficient range to to , when the average friction coefficient lie in to When between Calculate the third eigenvalue ; When the average friction coefficient Less than or equal to When the third eigenvalue is 0; when the average friction coefficient Greater than or equal to When the third eigenvalue is 1;

[0017] For the maximum formation pressure F, set the formation pressure range to to , when the maximum formation pressure F is located at to When between Calculate the fourth eigenvalue ; When the maximum formation pressure F is less than or equal to When the fourth eigenvalue is 0; when the maximum formation pressure F is greater than or equal to When the fourth eigenvalue is 1;

[0018] The final formula Calculate and obtain the drilling environment characteristic values.

[0019] Preferably, the fault prediction frequency is adjusted by the drilling environment characteristic value, and the specific operation is as follows:

[0020] The adjustable range of the fault prediction frequency Q is set to to , using the formula The fault prediction frequency Q of the current frequency adjustment period is calculated and obtained.

[0021] Preferably, the working parameter acquisition model is established based on a random forest model. The specific operations for training the working parameter acquisition model include:

[0022] Historical drilling data is obtained, and according to all drilling operation data in the historical drilling data and the degree of drill bit wear and drilling work efficiency after the completion of the drilling operation, the drilling operation data with a degree of drilling wear less than a preset wear threshold and a drilling work efficiency greater than a preset efficiency threshold is used as a target training sample; each target training sample contains a number of drilling environment data and corresponding drilling work parameters; all target training samples are divided into a training set and a validation set, the drilling environment data is used as the input of the working parameter acquisition model, the drilling work parameters are used as the output of the working parameter acquisition model, and the training set is input into the parameter-initialized fault risk assessment model for training; the working parameter acquisition model is verified using the validation set to obtain a first verification result; it is determined whether the first verification result meets the preset first training condition, and if so, the trained working parameter acquisition model is output; if not, the training set is used to continue training the working parameter acquisition model.

[0023] Preferably, the fault prediction model is established based on the LSTM model, including N fault prediction LSTM units, 1 fully connected layer and 1 output layer, wherein the fault prediction LSTM unit is used to extract the temporal features in the fault prediction data set; the fully connected layer is used to extract features from the outputs of all fault prediction LSTM units; and the output layer is used to output the fault prediction results.

[0024] Preferably, for the training of the fault prediction model, the specific operations are as follows:

[0025] The N continuous drilling equipment data in the historical drilling data, which are sorted by time under normal conditions, are taken as the normal data set and marked as "normal", and the N continuous drilling equipment data, which are sorted by time before the failure occurs, are taken as the fault data set and marked as "fault"; a number of equal normal data sets and fault data sets are selected in the historical drilling data as the fault training sample set; the N drilling equipment data in the fault training sample are taken as the input data of the fault prediction model, and the fault prediction results corresponding to the drilling equipment data in the fault training sample are taken as the output data of the fault prediction model; the fault training sample set is divided into a fault prediction training set and a fault prediction verification set, the fault prediction training set is sent to the fault prediction model with parameter initialization for training, and then the fault prediction verification set is sent to the fault prediction model to obtain the verification result, and it is judged whether the verification result meets the training condition. If the training condition is met, the trained fault prediction model is output; otherwise, the fault prediction model is continuously trained through the fault prediction training set.

[0026] Preferably, the drilling data intelligent analysis system comprises:

[0027] The drilling data acquisition module is used to acquire drilling environment data and drilling working parameters. The drilling environment data includes formation hardness, formation pressure, temperature, and friction coefficient; the drilling working parameters include drill bit speed and drill bit pressure;

[0028] The working parameter adjustment module is used to determine whether the drilling working parameters need to be adjusted according to the acquired drilling environment data. If so, the acquired drilling environment data is used as the input of the working parameter acquisition model, the optimal working parameters are output, and the acquired optimal working parameters are used to adjust the drilling working parameters; if not, no operation is performed;

[0029] The fault prediction module includes a drilling equipment data acquisition unit, a fault prediction frequency adjustment unit and a fault prediction unit; the drilling equipment data acquisition unit is used to acquire drilling equipment data, and the drilling equipment data includes equipment temperature, vibration frequency and operating pressure; the fault prediction frequency adjustment unit is used to use the drilling environment data acquired in the previous frequency adjustment cycle at the beginning of any frequency adjustment cycle, calculate the drilling environment characteristic value, and adjust the fault prediction frequency by the drilling environment characteristic value; the fault prediction unit is used to use the most recently acquired N drilling equipment data as a fault prediction data set at the current fault prediction time point, use the fault prediction data set as the input of the fault prediction model, and output the fault prediction result; based on the fault prediction result, execute corresponding response measures.

[0030] The present invention has the following advantages:

[0031] 1. The present invention uses advanced data analysis technology to monitor and analyze drilling environment data and drilling working parameters in real time, and provides the optimal drill speed and pressure settings for drilling operations, thereby dynamically adjusting drilling parameters to adapt to the ever-changing underground environment; this intelligent adjustment mechanism not only improves the efficiency of drilling operations, but also significantly reduces equipment wear and failure risks; this real-time, data-driven decision support system makes drilling operations more accurate and efficient, while ensuring the safety of operations.

[0032] 2. The present invention utilizes drilling equipment data and is based on in-depth study and analysis of historical drilling data to predict potential equipment failures in advance, thereby taking preventive measures to avoid failures or mitigate their impacts. At the same time, by calculating the characteristic values ​​of the drilling environment, the fault prediction frequency is dynamically adjusted in each frequency adjustment cycle. In this way, the present invention not only improves the reliability and life of the equipment, but also reduces the downtime and maintenance costs caused by failures, providing continuous protection for drilling operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a schematic diagram of the structure of the drilling data intelligent analysis system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0035] Embodiment 1, a drilling data intelligent analysis method, comprising:

[0036] Drilling environment data and drilling working parameters are obtained at the current monitoring time point. Drilling environment data include formation hardness, formation pressure, temperature, and friction coefficient. Formation hardness refers to the hardness of formation rock, which affects drill bit wear and drilling speed. Formation pressure refers to the fluid pressure in the formation. Excessive pressure may cause safety accidents such as blowouts. Temperature refers to the temperature in the drilling area. Temperature changes may affect the performance of drilling equipment and the stability of drilling materials. Friction coefficient refers to the degree of friction between the drill bit and the formation, which affects drill bit wear and drilling efficiency. Drilling working parameters include drill bit speed and drill bit pressure. Drill bit speed refers to the temperature in the drilling area. Temperature changes may affect the performance of drilling equipment and the stability of drilling materials. Friction coefficient refers to the degree of friction between the drill bit and the formation, which affects drill bit wear and drilling efficiency. The speed of the drill bit rotation affects the drilling speed and the wear of the drill bit. The drill bit pressure refers to the force applied to the drill bit, which affects the drilling speed and the degree of formation damage. Based on the currently acquired drilling environment data, it is determined whether the drilling working parameters need to be adjusted. If so, the acquired drilling environment data is used as the input of the working parameter acquisition model, the optimal working parameters are output, and the drilling working parameters are adjusted using the acquired optimal working parameters. If not, no operation is performed. Such intelligent adjustment not only improves the efficiency and safety of drilling operations, but also reduces the dependence on the operator's experience, and realizes the optimization and automation of the drilling process.

[0037] Acquire drilling equipment data at the current monitoring time point. Drilling equipment data includes equipment temperature, vibration frequency and operating pressure. The purpose of this step is to collect the key operating parameters of the current drilling equipment. These parameters can reflect the health status and performance level of the equipment, which are essential for preventing equipment failures and maintaining the normal operation of the equipment. Equipment temperature refers to the temperature of the drilling equipment during operation. Overheating of the equipment may indicate a failure or performance problem, which needs to be dealt with in time to avoid damage. Vibration frequency refers to the vibration of the drilling equipment during operation. Abnormal vibration frequency may indicate that the equipment is unbalanced or has other mechanical problems. Operating pressure refers to the pressure that the equipment is subjected to during operation. Too high or too low pressure may affect equipment performance and even cause equipment damage. By analyzing equipment data, potential failures can be predicted, so that preventive measures can be taken to reduce unexpected downtime and maintenance costs.

[0038] Set the frequency adjustment cycle, which is the time interval for the system to regularly evaluate and adjust the fault prediction frequency. By setting such a cycle, the system can regularly update its predicted frequency to adapt to changes in the drilling environment. At the beginning of any frequency adjustment cycle, the drilling environment data obtained in the previous frequency adjustment cycle is used to calculate the drilling environment characteristic value, which can reflect the current state of the drilling environment. A frequency adjustment cycle contains several fault prediction time points, and the fault prediction frequency is adjusted by the drilling environment characteristic value. If the environmental characteristic value indicates that the current drilling conditions are more complex or the risk is higher, the system may increase the frequency of fault prediction to monitor the equipment status more closely.

[0039] At the current fault prediction time point, the most recently acquired N drilling equipment data are used as a fault prediction data set, and the fault prediction data set is used as the input of the fault prediction model to output the fault prediction result. Based on the fault prediction result, corresponding response measures are executed to timely discover and handle equipment problems, which can prevent safety accidents caused by equipment failure and ensure the safety of operators and equipment.

[0040] Determine whether the drilling parameters need to be adjusted. The specific operations are as follows:

[0041] The currently acquired drilling environment data is compared with the drilling environment data when the drilling working parameters were last adjusted to obtain the formation hardness difference, formation pressure difference, temperature difference and friction coefficient difference; the formation hardness change threshold, formation pressure change threshold, temperature change threshold and friction coefficient change threshold are set. If at least one of the formation hardness difference, formation pressure difference, temperature difference and friction coefficient difference is greater than the corresponding change threshold, the judgment result is that the drilling working parameters need to be adjusted; otherwise, the judgment result is that the drilling working parameters do not need to be adjusted; this judgment logic ensures that the drilling parameters are adjusted only when the environmental conditions change significantly, so as to adapt to the new drilling conditions, maintain drilling efficiency and safety, and avoid too frequent adjustments to the drilling working parameters.

[0042] Calculate the drilling environment characteristic value. The specific operations are as follows:

[0043] Set the frequency adjustment cycle. At the beginning of any frequency adjustment cycle, select the formation hardness data from the drilling environment data obtained in the previous frequency adjustment cycle. , friction coefficient data and maximum formation pressure data F, i=1, 2, ..., n; n represents the number of selected formation hardness data and friction coefficient data;

[0044] For formation hardness data , calculate the average formation hardness , set the formation hardness range to to , when the average formation hardness lie in to When between Calculate the first eigenvalue ; When the average formation hardness Less than or equal to When the first eigenvalue is 0; when the average formation hardness Greater than or equal to When the first eigenvalue is 1;

[0045] According to the formation hardness data And the average formation hardness obtained , using the formula Calculate the second eigenvalue ;

[0046] For friction coefficient data , calculate the average friction coefficient , set the friction coefficient range to to , when the average friction coefficient lie in to When between Calculate the third eigenvalue ; When the average friction coefficient Less than or equal to When the third eigenvalue is 0; when the average friction coefficient Greater than or equal to When the third eigenvalue is 1;

[0047] For the maximum formation pressure F, set the formation pressure range to to , when the maximum formation pressure F is located at to When between Calculate the fourth eigenvalue ; When the maximum formation pressure F is less than or equal to When the fourth eigenvalue is 0; when the maximum formation pressure F is greater than or equal to When the fourth eigenvalue is 1;

[0048] The final formula Calculate and obtain the drilling environment characteristic values.

[0049] The fault prediction frequency is adjusted by the drilling environment characteristic value. The specific operations are as follows:

[0050] The adjustable range of the fault prediction frequency Q is set to to , using the formula The fault prediction frequency Q of the current frequency adjustment period is calculated and obtained.

[0051] The working parameter acquisition model is based on the random forest model. The random forest model is an integrated learning method that improves the accuracy and robustness of predictions by building multiple decision trees and aggregating their results. During the training process, the random forest extracts multiple subsets from the original data set, and each tree is trained on its subset. Each decision point is derived from a randomly selected feature subset, which can increase the diversity of the model and reduce the risk of overfitting. Finally, the random forest combines the prediction results of each tree through majority voting or averaging to obtain a more reliable output. The specific operations for the training of the working parameter acquisition model include:

[0052] Historical drilling data is obtained, and according to all drilling operation data in the historical drilling data and the degree of drill bit wear and drilling work efficiency after the completion of the drilling operation, the drilling operation data with a degree of drilling wear less than a preset wear threshold and a drilling work efficiency greater than a preset efficiency threshold is used as a target training sample; each target training sample contains a number of drilling environment data and corresponding drilling work parameters; all target training samples are divided into a training set and a validation set, the drilling environment data is used as the input of the working parameter acquisition model, the drilling work parameters are used as the output of the working parameter acquisition model, and the training set is input into the parameter-initialized fault risk assessment model for training; the working parameter acquisition model is verified using the validation set to obtain a first verification result; it is determined whether the first verification result meets the preset first training condition, and if so, the trained working parameter acquisition model is output; if not, the training set is used to continue training the working parameter acquisition model.

[0053] The fault prediction model is established based on the LSTM model. The LSTM model is a special recurrent neural network designed to solve the gradient vanishing or gradient exploding problems encountered by traditional RNNs when processing long sequence data. By introducing a gating mechanism, it can selectively retain or forget information, thereby effectively capturing long-term dependencies in time series. The LSTM model performs well in tasks that require consideration of long-term dependent information, such as natural language processing, speech recognition, and time series prediction. It can learn the temporal dynamic features in the data and generate accurate prediction results. The fault prediction model includes N fault prediction LSTM units, 1 fully connected layer, and 1 output layer. The fault prediction LSTM unit is used to extract the temporal features in the fault prediction data set. The fully connected layer is used to extract features from the outputs of all fault prediction LSTM units. The output layer is used to output the fault prediction results.

[0054] The specific operations for training the fault prediction model are as follows:

[0055] The N continuous drilling equipment data in the historical drilling data, which are sorted by time under normal conditions, are taken as the normal data set and marked as "normal", and the N continuous drilling equipment data, which are sorted by time before the failure occurs, are taken as the fault data set and marked as "fault"; a number of equal normal data sets and fault data sets are selected in the historical drilling data as the fault training sample set; the N drilling equipment data in the fault training sample are taken as the input data of the fault prediction model, and the fault prediction results corresponding to the drilling equipment data in the fault training sample are taken as the output data of the fault prediction model; the fault training sample set is divided into a fault prediction training set and a fault prediction verification set, the fault prediction training set is sent to the fault prediction model with parameter initialization for training, and then the fault prediction verification set is sent to the fault prediction model to obtain the verification result, and it is judged whether the verification result meets the training condition. If the training condition is met, the trained fault prediction model is output; otherwise, the fault prediction model is continuously trained through the fault prediction training set.

[0056] Embodiment 2, drilling data intelligent analysis system, such as Figure 1 As shown, including:

[0057] The drilling data acquisition module is used to acquire drilling environment data and drilling working parameters. The drilling environment data includes formation hardness, formation pressure, temperature, and friction coefficient; the drilling working parameters include drill bit speed and drill bit pressure;

[0058] The working parameter adjustment module is used to determine whether the drilling working parameters need to be adjusted according to the acquired drilling environment data. If so, the acquired drilling environment data is used as the input of the working parameter acquisition model, the optimal working parameters are output, and the acquired optimal working parameters are used to adjust the drilling working parameters; if not, no operation is performed;

[0059] The fault prediction module includes a drilling equipment data acquisition unit, a fault prediction frequency adjustment unit and a fault prediction unit; the drilling equipment data acquisition unit is used to acquire drilling equipment data, and the drilling equipment data includes equipment temperature, vibration frequency and operating pressure; the fault prediction frequency adjustment unit is used to use the drilling environment data acquired in the previous frequency adjustment cycle at the beginning of any frequency adjustment cycle, calculate the drilling environment characteristic value, and adjust the fault prediction frequency by the drilling environment characteristic value; the fault prediction unit is used to use the most recently acquired N drilling equipment data as a fault prediction data set at the current fault prediction time point, use the fault prediction data set as the input of the fault prediction model, and output the fault prediction result; based on the fault prediction result, execute corresponding response measures.

[0060] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A drilling data intelligent analysis method, characterized in that: include: Acquire drilling environment data and drilling working parameters at the current monitoring time point, the drilling environment data including formation hardness, formation pressure, temperature, and friction coefficient; The drilling working parameters include the drill bit speed and the drill bit pressure; based on the currently acquired drilling environment data, it is determined whether the drilling working parameters need to be adjusted. If so, the acquired drilling environment data is used as the input of the working parameter acquisition model, the optimal working parameters are output, and the acquired optimal working parameters are applied to adjust the drilling working parameters; if not, no operation is performed; Acquire drilling equipment data at the current monitoring time point, the drilling equipment data including equipment temperature, vibration frequency and operating pressure; A frequency adjustment cycle is set. At the beginning of any frequency adjustment cycle, the drilling environment data obtained in the previous frequency adjustment cycle is used to calculate the drilling environment characteristic value. A frequency adjustment cycle includes several fault prediction time points, and the fault prediction frequency is adjusted according to the drilling environment characteristic value. At the current fault prediction time point, the most recently acquired N drilling equipment data are used as a fault prediction data set, the fault prediction data set is used as the input of the fault prediction model, and the fault prediction result is output; based on the fault prediction result, corresponding response measures are executed; The specific operation of calculating the drilling environment characteristic value is as follows: Set the frequency adjustment cycle. At the beginning of any frequency adjustment cycle, select the formation hardness data from the drilling environment data obtained in the previous frequency adjustment cycle. , friction coefficient data and maximum formation pressure data F, i=1, 2, ..., n; n represents the number of selected formation hardness data and friction coefficient data; For formation hardness data , calculate the average formation hardness , set the formation hardness range to to , when the average formation hardness lie in to When between Calculate the first eigenvalue ; When the average formation hardness Less than or equal to When the first eigenvalue is 0; when the average formation hardness Greater than or equal to When the first eigenvalue is 1; According to the formation hardness data And the average formation hardness obtained , using the formula Calculate the second eigenvalue ; For friction coefficient data , calculate the average friction coefficient , set the friction coefficient range to to , when the average friction coefficient lie in to When between Calculate the third eigenvalue ; When the average friction coefficient Less than or equal to When the third eigenvalue is 0; when the average friction coefficient Greater than or equal to When the third eigenvalue is 1; For the maximum formation pressure F, set the formation pressure range to to , when the maximum formation pressure F is located at to When between Calculate the fourth eigenvalue ; When the maximum formation pressure F is less than or equal to When the fourth eigenvalue is 0; when the maximum formation pressure F is greater than or equal to When the fourth eigenvalue is 1; The final formula Calculate and obtain drilling environment characteristic values; The fault prediction frequency is adjusted by the drilling environment characteristic value. The specific operations are as follows: The adjustable range of the fault prediction frequency Q is set to to , using the formula The fault prediction frequency Q of the current frequency adjustment period is calculated and obtained.

2. The drilling data intelligent analysis method according to claim 1, characterized in that: Determine whether the drilling parameters need to be adjusted. The specific operations are as follows: Compare the currently acquired drilling environment data with the drilling environment data when the drilling working parameters were last adjusted to obtain formation hardness difference, formation pressure difference, temperature difference and friction coefficient difference; Set the formation hardness change threshold, formation pressure change threshold, temperature change threshold and friction coefficient change threshold. If at least one of the formation hardness difference, formation pressure difference, temperature difference and friction coefficient difference is greater than the corresponding change threshold, the judgment result is that the drilling working parameters need to be adjusted; otherwise, the judgment result is that the drilling working parameters do not need to be adjusted.

3. The drilling data intelligent analysis method according to claim 2, characterized in that: The working parameter acquisition model is established based on the random forest model. The specific operations for training the working parameter acquisition model include: Historical drilling data is obtained, and according to all drilling operation data in the historical drilling data and the degree of drill bit wear and drilling work efficiency after the completion of the drilling operation, the drilling operation data with a degree of drilling wear less than a preset wear threshold and a drilling work efficiency greater than a preset efficiency threshold is used as a target training sample; each target training sample contains a number of drilling environment data and corresponding drilling work parameters; all target training samples are divided into a training set and a validation set, the drilling environment data is used as the input of the working parameter acquisition model, the drilling work parameters are used as the output of the working parameter acquisition model, and the training set is input into the parameter-initialized fault risk assessment model for training; the working parameter acquisition model is verified using the validation set to obtain a first verification result; it is determined whether the first verification result meets the preset first training condition, and if so, the trained working parameter acquisition model is output; if not, the training set is used to continue training the working parameter acquisition model.

4. The drilling data intelligent analysis method according to claim 3, characterized in that: The fault prediction model is established based on the LSTM model, including N fault prediction LSTM units, 1 fully connected layer and 1 output layer. The fault prediction LSTM unit is used to extract the temporal features in the fault prediction data set. The fully connected layer is used to extract features from the outputs of all fault prediction LSTM units; the output layer is used to output the fault prediction results.

5. The drilling data intelligent analysis method according to claim 4, characterized in that: The specific operations for training the fault prediction model are as follows: The N consecutive drilling equipment data in the historical drilling data that are in a normal state and sorted by time are taken as the normal data set and marked as "normal", and the N consecutive drilling equipment data that are sorted by time before the failure occurs are taken as the fault data set and marked as "fault"; Selecting a number of normal data sets and fault data sets of equal size from historical drilling data as fault training sample sets; Using N drilling equipment data in the fault training sample as input data of the fault prediction model, and using the fault prediction results corresponding to the drilling equipment data in the fault training sample as output data of the fault prediction model; The fault training sample set is divided into a fault prediction training set and a fault prediction verification set. The fault prediction training set is sent to the fault prediction model with parameter initialization for training. Then the fault prediction verification set is sent to the fault prediction model to obtain the verification result. It is judged whether the verification result meets the training conditions. If the training conditions are met, the trained fault prediction model is output; otherwise, the fault prediction model is continuously trained through the fault prediction training set.

6. Drilling data intelligent analysis system, characterized in that: The system is applied to the drilling data intelligent analysis method according to any one of claims 1 to 5, comprising: The drilling data acquisition module is used to acquire drilling environment data and drilling working parameters. The drilling environment data includes formation hardness, formation pressure, temperature, and friction coefficient; the drilling working parameters include drill bit speed and drill bit pressure; The working parameter adjustment module is used to determine whether the drilling working parameters need to be adjusted according to the acquired drilling environment data. If so, the acquired drilling environment data is used as the input of the working parameter acquisition model, the optimal working parameters are output, and the acquired optimal working parameters are used to adjust the drilling working parameters; if not, no operation is performed; The fault prediction module includes a drilling equipment data acquisition unit, a fault prediction frequency adjustment unit and a fault prediction unit; the drilling equipment data acquisition unit is used to acquire drilling equipment data, and the drilling equipment data includes equipment temperature, vibration frequency and operating pressure; the fault prediction frequency adjustment unit is used to use the drilling environment data acquired in the previous frequency adjustment cycle at the beginning of any frequency adjustment cycle, calculate the drilling environment characteristic value, and adjust the fault prediction frequency by the drilling environment characteristic value; the fault prediction unit is used to use the most recently acquired N drilling equipment data as a fault prediction data set at the current fault prediction time point, use the fault prediction data set as the input of the fault prediction model, and output the fault prediction result; based on the fault prediction result, execute corresponding response measures.

Images