CN117932446A - Drilling overflow identification and prediction method, device, equipment and medium - Google Patents
Drilling overflow identification and prediction method, device, equipment and medium Download PDFInfo
- Publication number
- CN117932446A CN117932446A CN202311550703.7A CN202311550703A CN117932446A CN 117932446 A CN117932446 A CN 117932446A CN 202311550703 A CN202311550703 A CN 202311550703A CN 117932446 A CN117932446 A CN 117932446A
- Authority
- CN
- China
- Prior art keywords
- data
- drilling
- overflow
- real
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005553 drilling Methods 0.000 title claims abstract description 353
- 238000000034 method Methods 0.000 title claims abstract description 137
- 238000012549 training Methods 0.000 claims abstract description 80
- 230000008569 process Effects 0.000 claims abstract description 59
- 239000012530 fluid Substances 0.000 claims description 43
- 230000002706 hydrostatic effect Effects 0.000 claims description 29
- 238000012360 testing method Methods 0.000 claims description 27
- 238000004422 calculation algorithm Methods 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 16
- 238000012706 support-vector machine Methods 0.000 claims description 16
- 238000005457 optimization Methods 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 13
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000002790 cross-validation Methods 0.000 claims description 10
- 239000012535 impurity Substances 0.000 claims description 10
- 238000004140 cleaning Methods 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 125000004122 cyclic group Chemical group 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 abstract description 11
- 238000010586 diagram Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 238000010801 machine learning Methods 0.000 description 6
- 230000015572 biosynthetic process Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000009530 blood pressure measurement Methods 0.000 description 3
- 238000012217 deletion Methods 0.000 description 3
- 230000037430 deletion Effects 0.000 description 3
- 238000011010 flushing procedure Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000001105 regulatory effect Effects 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 229930195733 hydrocarbon Natural products 0.000 description 2
- 150000002430 hydrocarbons Chemical class 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 239000003921 oil Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000003209 petroleum derivative Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000005086 pumping Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 239000002002 slurry Substances 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application discloses a method, a device, equipment and a medium for identifying and predicting well drilling overflow. The method comprises the following steps: determining candidate data according to real drilling data acquired in the real drilling process of the historical drilling; intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data; model training is carried out based on training data to obtain a drilling overflow identification model so as to identify the overflow condition of drilling based on the drilling overflow identification model. According to the technical scheme provided by the embodiment of the application, the model is trained to generate the drilling overflow identification model according to the training data obtained in the mode of real drilling data and the sliding window, and the real-time monitoring and early warning of the overflow condition in the drilling process can be realized through the drilling overflow identification model.
Description
Technical Field
The application relates to the field of petroleum and natural gas exploration, in particular to a drilling overflow identification and prediction method, device, equipment and medium.
Background
In the drilling process, when the density of the drilling fluid is smaller than the density of the formation fluid and the pressure of the formation fluid cannot be balanced, the pressure of a shaft is smaller than the pressure of the formation, oil, gas and water in the formation are pressed into the shaft to cause overflow, the overflow treatment time is short, and if the overflow treatment time is not controlled in time, blowout events can occur to cause great safety, economy and environmental hazard.
In the prior art, the overflow condition is determined by adopting a field observation method, and the method has higher feasibility, but has certain hysteresis, depends on experience, responsibility and consciousness of monitoring staff, does not effectively utilize the prior drilling parameters, has larger limitation, and can not provide early identification and prediction of overflow.
Disclosure of Invention
The application provides a drilling overflow identification and prediction method, device, equipment and medium, which are used for training a model according to training data obtained in a real drilling data and sliding window mode to generate a drilling overflow identification model, and can realize real-time monitoring and early warning of overflow conditions in the drilling process through the drilling overflow identification model.
According to an aspect of the present application, there is provided a method of identifying and predicting well overflow, the method comprising:
determining candidate data according to real drilling data acquired in the real drilling process of the historical drilling;
Intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data;
Model training is carried out based on the training data to obtain a drilling overflow identification model so as to identify the overflow condition of drilling based on the drilling overflow identification model.
According to another aspect of the present application there is provided a drilling overflow identification and prediction apparatus, the apparatus comprising:
The candidate data determining module is used for determining candidate data according to the real drilling data collected in the real drilling process of the historical drilling;
The training data acquisition module is used for intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data;
And the recognition model generation module is used for carrying out model training based on the training data to obtain a drilling overflow recognition model so as to recognize the overflow condition of the drilling based on the drilling overflow recognition model.
According to another aspect of the present application, there is provided an electronic device including:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the well overflow identification method of any of the embodiments of the present application.
According to another aspect of the application, there is provided a computer readable storage medium having stored thereon computer instructions for causing a processor to perform the well overflow identification method of any of the embodiments of the application when executed.
According to the technical scheme, candidate data are determined according to the real drilling data collected in the real drilling process of the historical drilling; intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data; model training is carried out based on training data to obtain a drilling overflow identification model so as to identify the overflow condition of drilling based on the drilling overflow identification model. According to the technical scheme provided by the embodiment of the application, the model is trained to generate the drilling overflow identification model according to the training data obtained in the mode of real drilling data and the sliding window, and the real-time monitoring and early warning of the overflow condition in the drilling process can be realized through the drilling overflow identification model.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for identifying and predicting well overflow according to a first embodiment of the application;
FIG. 2 is a schematic diagram of a processing operation of test data and training data according to a first embodiment of the present application;
FIG. 3 is a flow chart for training and evaluating a drilling overflow identification model according to a first embodiment of the application;
FIG. 4 is a flow chart of a method for identifying and predicting well overflow according to a second embodiment of the application;
Fig. 5 is a schematic flow chart of real-time monitoring and early warning and early prediction of overflow based on a drilling overflow identification model according to a second embodiment of the application;
FIG. 6 is a flowchart of a specific implementation of a method for identifying and predicting overflow of a well according to a third embodiment of the application;
Fig. 7 is a partial real-drill data diagram after impurity data deletion according to the third embodiment of the present application;
FIG. 8 is a thermodynamic diagram of the correlation between one feature provided in accordance with a third embodiment of the present application;
FIG. 9 is a schematic structural view of a drilling overflow identification and prediction device according to a fourth embodiment of the present application;
fig. 10 is a schematic structural diagram of an electronic device for implementing the drilling overflow identification method provided in the fifth embodiment of the application.
Detailed Description
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," "third," "fourth," "actual," "preset," and the like in the description and the claims of the present application and in the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of a method for identifying and predicting overflow of a well according to an embodiment of the present application, which is applicable to identifying overflow during a well drilling process. The method may be performed by a well overflow identification and prediction device, which may be implemented in hardware and/or software, which may be configured in an electronic device. As shown in fig. 1, the method includes.
S110, determining candidate data according to the real drilling data acquired in the real drilling process of the historical drilling.
The real drilling data comprise real drilling engineering data, pressure measurement while drilling data and pressure control drilling completion data, the pressure control drilling completion data comprise a plurality of characteristic parameters related to pressure change in the drilling process, such as measured back pressure, outlet flow, outlet density, circulating back pressure and the like, and the real drilling data can be measured through various sensors.
According to the embodiment of the application, the candidate data can be determined according to the real drilling data collected in the history drilling real drilling process. Specifically, real drilling data acquired in the historical drilling process can be acquired, the real drilling data is processed, and the processed real drilling data is used as candidate data.
It should be noted that, the history drilling process corresponding to the obtained real drilling data includes overflow working conditions, and the data amounts of the normal working conditions and the overflow working conditions in the real drilling data should not differ too much, when the ratio of the data amounts of the overflow working conditions to the normal working conditions is lower than a certain ratio, for example, 1:100, the data amount of the overflow working condition sample needs to be expanded in an undersampling or oversampling manner.
S120, intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data.
Wherein the sliding window is a time-series analysis method for extracting features and target variables from time-series data, which divides the time-series data into successive, overlapping windows, and extracts features and target variables from each window.
It will be appreciated that the real drilling data collected during the history drilling real drilling process is continuous and contains time series, and at the same time, the real drilling data in the same time period does not change suddenly under the fine time series division. Therefore, the real drilling data collected in the history drilling real drilling process is suitable for continuous research by adopting a sliding window method, and the candidate data obtained by processing the real drilling data is also suitable for continuous research by adopting the sliding window method, so that the relation between the candidate data in adjacent and a period of time is accurately grasped.
According to the embodiment of the application, the candidate data of the whole time dimension can be traversed in a sliding window mode based on the preset time sliding window and the preset step length, the candidate data of the time dimension with the size of the sliding window is intercepted from the current position each time in the process of traversing the candidate data of the whole time dimension, and the intercepted candidate data form a feature matrix of one sample, so that training data are obtained. The preset time sliding window and the preset step length can be set according to actual conditions, and the embodiment of the application is not limited to the preset time sliding window and the preset step length.
And S130, performing model training based on training data to obtain a drilling overflow identification model so as to identify the overflow condition of the drilling based on the drilling overflow identification model.
In the embodiment of the application, the model training can be performed through the training data obtained in the step S120, so as to obtain the drilling overflow identification model for identifying the overflow condition of the drilling.
Specifically, model training is performed based on training data to obtain a drilling overflow identification model, which comprises the following steps: inputting training data into a support vector machine model, optimizing preset super parameters by using a Bayesian optimization algorithm and a K-fold cross validation method, and determining an optimal parameter combination; and giving the optimal parameter combination to a support vector machine model to obtain the drilling overflow identification model.
The support vector machine model is a powerful machine learning algorithm, which is widely used for classification and regression problems, and separates data samples of different categories by constructing a decision boundary. The Bayesian optimization algorithm is an optimization method and can be used for automatically adjusting the super-parameters of the machine learning algorithm so as to improve the performance of the machine learning algorithm. The K-fold cross validation method is a common technology in machine learning and statistical modeling, and can improve the generalization capability and stability of the model.
In the embodiment of the application, a support vector machine model can be defined first, and a Bayesian optimization algorithm and a K-fold cross validation method are used for optimizing preset super parameters before training data is input into the support vector machine model, so that an optimal parameter combination is determined, and the support vector machine model is endowed, so that the model has better applicability. By adopting the support vector machine model optimized based on the Bayesian optimization algorithm and the K-fold cross validation method, the overflow event in the drilling process can be predicted more accurately, and the prediction accuracy is improved.
In addition, in order to verify the validity and the accuracy of the identification effect of the drilling overflow identification model, the embodiment of the application also tests the drilling overflow identification model.
Specifically, after model training is performed based on training data to obtain a drilling overflow identification model, the method further comprises: inputting the test data into a drilling overflow identification model to test the drilling overflow identification model, and determining an evaluation report of the drilling overflow identification model; wherein the evaluation report comprises at least one of the following: the method comprises the steps of normal working condition identification accuracy rate, overflow working condition identification accuracy rate, recall rate, F1 fraction, test data quantity corresponding to different identification results, overall accuracy rate, macro average value and weighted average value.
Wherein the test data and the training data may originate from the same data set. It should be noted that when the test data and the training data originate from the same data set, the data in the data set needs to be divided, for example, the first 70% of the data in the data set is used as the training data, and the last 30% of the data in the data set is used as the test data. Meanwhile, the processing operations of the test data and the training data are also different, and an exemplary schematic diagram of the processing operations of the test data and the training data is shown in fig. 2, where the processing operations of the training data include adding a target column, a target code, a null value deletion/substitution, and time-series data processing, and the target column and the target code are added for training the model. The test data does not need to add a target column and target codes, and only comprises deletion/replacement of a blank value and time sequence data processing so as to facilitate the input of a model.
The recall rate is the proportion of the correct recognition of the model as the positive example in all actual positive examples, the F1 score is the harmonic mean of the accuracy rate and the recall rate, and the macro average is one of the commonly used evaluation indexes in the multi-classification problem and is used for calculating the average value of the evaluation indexes of each class.
In the embodiment of the application, the test data are input into the drilling overflow identification model to test the drilling overflow identification model, so that the evaluation report of a plurality of indexes including the normal working condition identification accuracy rate, the overflow working condition identification accuracy rate, the recall rate, the F1 fraction, the number of the test data corresponding to different identification results, the overall accuracy rate, the macro average value, the weighted average value and the like can be obtained, and the performance of the drilling overflow identification model is verified.
Illustratively, fig. 3 shows a training and evaluating flowchart of a drilling overflow identification model, and as shown in fig. 3, the training process of the drilling overflow identification model includes support vector machine model construction, preset super-parameter optimization, training data input and generating a drilling overflow identification model, and the testing process of the drilling overflow identification model includes testing data input and obtaining a model evaluation report.
According to the technical scheme, candidate data are determined according to the real drilling data collected in the real drilling process of the historical drilling; intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data; model training is carried out based on training data to obtain a drilling overflow identification model so as to identify the overflow condition of drilling based on the drilling overflow identification model. According to the technical scheme provided by the embodiment of the application, the model is trained to generate the drilling overflow identification model according to the training data obtained in the mode of real drilling data and the sliding window, and the real-time monitoring and early warning of the overflow condition in the drilling process can be realized through the drilling overflow identification model.
Example two
Fig. 4 is a flowchart of a method for identifying and predicting overflow of a well according to a second embodiment of the present application, where the method is optimized based on the foregoing embodiment, and a scheme not described in detail in the embodiment of the present application is shown in the foregoing embodiment. As shown in fig. 4, the method in the embodiment of the present application specifically includes the following steps:
S210, determining candidate data according to real drilling data acquired in the real drilling process of the historical drilling.
Specifically, determining candidate data according to real drilling data collected in a history drilling real drilling process comprises: according to the real drilling data collected in the history drilling real drilling process, calculating hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density; and identifying a corresponding overflow judgment result according to real drilling data acquired in the real drilling process of the historical drilling, representing the overflow judgment result by a preset numerical value, and combining the overflow judgment result with the real drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the drilling fluid equivalent circulating density to obtain candidate data.
According to the embodiment of the application, according to the actual drilling data acquired in the actual drilling process of the historical drilling, the hydrostatic column pressure, the circulating pressure consumption data and the equivalent circulating density of drilling fluid are calculated through a physical calculation model mainly according to the pressure control drilling completion data obtained in the pressure control drilling process, so that the effects of enriching and improving the data quality input by the model are achieved, and the series fusion of the physical model and the machine learning model on the data is realized.
According to the real drilling data collected in the history drilling real drilling process, calculating hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density, comprising:
Determining hydrostatic column pressure according to the drilling fluid density and the drilling fluid depth;
correcting the Van Ning friction coefficient according to the cyclic pressure loss in the drilling data of the drilled or adjacent well;
Calculating cyclic pressure consumption data based on the corrected van der waals friction coefficient and the real drilling data;
and calculating the corrected drilling fluid equivalent circulating density according to the circulating pressure consumption data and the drilling fluid equivalent circulating density model.
The calculation formulas of hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density are respectively as follows:
Ph＝0.00981ρH；
Wherein, P h is hydrostatic column pressure, P f is circulation pressure consumption data, ECD is drilling fluid equivalent circulation density, ρ is drilling fluid density, H is vertical well depth, f' is Van Ning friction coefficient corrected according to measured drilling data, v is average flow rate of drilling fluid, D is drill pipe inner diameter, P is bottom hole pressure, and P bp is wellhead back pressure.
In the embodiment of the application, the corresponding overflow judgment result can be identified according to the real drilling data acquired in the real drilling process, and the overflow judgment result is represented by using a preset value, for example, the value of the overflow judgment result is 0 or 1, wherein 0 indicates that no overflow occurs, and 1 indicates that overflow occurs. After determining the value of the preset value, the preset value representing the overflow judgment result can be combined with the real drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the drilling fluid equivalent circulating density, and specifically, the corresponding preset value can be added to the last column of the real drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the drilling fluid equivalent circulating density as the target feature of model training.
In the embodiment of the application, the situation that data distortion and vacancies possibly exist in the data is considered, and data processing is also needed.
Specifically, after the overflow judgment result is represented by a preset numerical value and combined with the real drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the drilling fluid equivalent circulating density, the method further comprises the following steps: carrying out data cleaning on the combined data, deleting impurity data in a time dimension, and carrying out data cleaning according to the correlation coefficient of the combined data; wherein the impurity data comprises at least one of: the combined data exceeds a preset range, the data is missing, and the combined data is a preset value representing data abnormality; and performing type conversion on the combined data according to a preset mode, and performing normalization or standardization processing on the combined data to obtain candidate data.
In the embodiment of the application, the data cleaning comprises the steps of deleting impurity data with the combined data exceeding a preset range and missing data and being a preset value representing data abnormality, and extracting the characteristics of the combined data by adopting pearson correlation coefficients.
And before the overflow judgment result is represented by a preset value and combined with the real drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the drilling fluid equivalent circulating density to obtain the candidate data, the method further comprises the steps of:
Traversing the real drilling data, and determining the correlation coefficients of the real drilling data currently traversed, other real drilling data and overflow results;
and taking and removing the real drilling data with the correlation coefficient lower than a preset correlation coefficient threshold value.
In the embodiment of the application, the pearson correlation coefficient is mainly used for measuring the linear relation strength and direction between each feature and the target feature of the target column, namely the target feature of the last column, and the calculation formula is as follows:
Wherein, r: pearson correlation coefficient; x and y: observations of two consecutive variables; Is the mean value of x,/> Is the mean value of y.
The value range is-1 to 1, wherein when the value is 1, the positive correlation between the two features is completely shown, when the value is-1, the negative correlation between the two features is completely shown, and when the value is 0, the linear relation between the two features is not shown. The preset correlation coefficient threshold value can be determined according to actual conditions, if the correlation coefficient of the two real drilling data is lower than the preset correlation coefficient threshold value, the correlation of the two real drilling data is smaller, or if the correlation coefficient of the real drilling data and the overflow result is lower than the preset correlation coefficient threshold value, the correlation of the real drilling data and the overflow result is smaller, and the real drilling data is removed.
After the data is cleaned, the combined data is subjected to type conversion according to a preset mode, namely, the combined data is subjected to time sequence data type processing. The real drilling data are acquired at fixed time intervals, the acquired time stamp time type time data are not standard and are difficult to directly use, therefore, the time stamp time type time data are required to be converted into a date time datetime object to form standard time data in the format of "% Y/% M/% d% H:% M:% S", and then the time data in the format of "% Y/% M/% d% H:% M:% S" are converted into Unix time stamps to form numerical time sequence data. The processing method for converting the time sequence type is added on the basis of processing the non-time sequence data, so that universality of the data is improved.
And finally, carrying out normalization or standardization treatment on the combined data to obtain candidate data. Since the normalization process requires the data to approximately follow a normal distribution, the combined data is preferably normalized. Specifically, a minimum-maximum normalization MinMaxScaler method is adopted to map the combined data to between 0 and 1 according to a linear proportion, so that candidate data are obtained.
S220, intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data.
S230, training a model based on the training data to obtain a drilling overflow identification model so as to identify the overflow condition of the drilling based on the drilling overflow identification model.
S240, determining hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density as actual data according to real drilling data acquired in the real drilling process of the target drilling.
In the embodiment of the application, the real drilling data acquired in the real drilling process of the target drilling can be acquired in real time, the hydrostatic column pressure, the circulating pressure consumption data and the equivalent circulating density of the drilling fluid are calculated according to the formula in the step S210, and the data are used as actual data.
S250, intercepting actual data of the time dimension based on a preset time sliding window and a preset step length to obtain input data.
According to the embodiment of the application, the real data of the whole time dimension can be traversed in a sliding window mode based on the preset time sliding window and the preset step length, the real data of the time dimension of the sliding window is intercepted from the current position each time in the process of traversing the real data of the whole time dimension, and the intercepted real data forms a feature matrix of one sample, so that input data is obtained. The preset time sliding window and the preset step length can be set according to actual conditions, and the embodiment of the application is not limited to the preset time sliding window and the preset step length.
S260, inputting the input data into a drilling overflow identification model, and determining a target drilling overflow identification result.
In the embodiment of the application, the obtained input data is input into the drilling overflow identification model, so that the target drilling overflow identification result can be determined, and the real-time monitoring and early warning of the overflow condition in the drilling process are realized.
Optionally, after inputting the input data into the drilling overflow identification model, comprising: the drilling overflow identification model outputs and displays the target drilling overflow identification result and corresponding input data, and facilitates on-site engineers to regulate and control drilling parameters according to the input data while overflow early warning is carried out. For example, the drilling fluid density is timely regulated and controlled based on the equivalent circulating density of the drilling fluid, and complex working conditions such as overflow and the like are avoided in a well section with a narrow safety density window.
S270, based on a preset time sliding window, intercepting input data closest to the current moment and a target drilling overflow identification result as target data.
In the embodiment of the application, whether overflow occurs in the target drilling process is monitored and early-warned in real time according to the data acquired in real time in the target drilling process, and input data closest to the current moment and a target drilling overflow identification result can be intercepted as target data based on a preset time sliding window to realize the prediction of the overflow.
S280, predicting the overflow condition of the target drilling in the period after the current moment according to the target data based on a preset prediction algorithm.
In the embodiment of the application, the overflow condition of the target drilling in the period after the current moment can be predicted according to the target data based on a preset prediction algorithm, such as predict () method.
Fig. 5 is a schematic flow chart of real-time monitoring, early warning and early prediction of overflow based on a drilling overflow recognition model, and as shown in fig. 5, firstly, determining a target drilling overflow recognition result by inputting input data into the drilling overflow recognition model, outputting and displaying the input data and the target drilling overflow recognition result to realize visualization of the monitoring result, then, intercepting the input data closest to the current moment and the target drilling overflow recognition result, and predicting the overflow condition of the target drilling in a period after the current moment according to a preset prediction algorithm.
It should be noted that, the embodiment of the present application may also implement multi-step prediction of overflow conditions, where the time interval between each step is the acquisition time interval of real drilling data, specifically, how many steps of prediction are to be performed is defined first, the first step of overflow condition prediction is performed by using target data, after the first step of overflow condition prediction result is obtained, the previous step of overflow condition prediction result is used as input data of the subsequent step of overflow condition prediction, multi-step prediction is performed in a circulating manner, and after the circulation is completed, the multi-step prediction result is output.
The embodiment of the application provides a drilling overflow identification and prediction method, which is used for determining candidate data according to real drilling data acquired in a historical drilling real drilling process; intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data; model training is carried out based on training data to obtain a drilling overflow identification model so as to identify the overflow condition of drilling based on the drilling overflow identification model; according to the real drilling data collected in the real drilling process of the target drilling, determining hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density as actual data; intercepting actual data of time dimension based on a preset time sliding window and a preset step length to obtain input data; inputting the input data into a drilling overflow identification model, and determining a target drilling overflow identification result; based on a preset time sliding window, intercepting input data closest to the current moment and a target drilling overflow identification result as target data; and predicting the overflow condition of the target drilling in a period after the current moment according to the target data based on a preset prediction algorithm. According to the technical scheme provided by the embodiment of the application, the model is trained to generate the drilling overflow identification model according to the training data obtained in the mode of real drilling data and the sliding window, and the real-time monitoring and early warning of the overflow condition in the drilling process can be realized through the drilling overflow identification model. Meanwhile, based on a prediction algorithm, the prediction of overflow risk of a period after the current moment can be realized by intercepting input data closest to the current moment and a target drilling overflow identification result.
Example III
Fig. 6 is a flowchart for specifically implementing a method for identifying and predicting overflow of drilling according to a third embodiment of the present application, and on the basis of the foregoing embodiment, a preferred embodiment is provided, which is specifically as follows:
s310, acquiring real drilling data acquired in the real drilling process of the target drilling.
The real drilling data comprise real drilling engineering data, pressure measurement while drilling data and pressure control drilling completion data, and relate to DateTime, time, well depth (m), vertical depth (m), drill bit vertical depth (m), drilling time (min/m), drilling pressure (KN), hanging weight (KN), rotating speed (rpm), torque (KN.m), square in (m), hook position (m), hook speed (m/S), vertical pressure log (MPa), casing pressure (MPa), pump flushing 1 (spm), pump flushing 2 (spm), pump flushing 3 (spm), total pool volume (m 3), delay time (min), slurry overflow (m 3), inlet flow (L/S), outlet flow log (%), inlet density log (g/cc), outlet density log (g/cc), inlet temperature (DEG C), outlet temperature (DEG C), total hydrocarbons, H2S (ppm), C1 (%), C2 (%), PWD depth (m), D depth (m), PWD pressure (MPa), D Pressure (PWD), annular volume (PWD), volume (PWD) measurement of the annulus volume (PWD), PWD (PWD), and annulus volume (PWD) measurement of the annulus volume (64 DEG) and the annulus volume (PWD The system comprises 60 characteristic parameters such as an operation mode, vertical pressure (MPa), measured back pressure (MPa), outlet flow (L/s), outlet density (g/cc), back pressure pump flow (L/s), circulating back pressure (MPa), additional back pressure (MPa), inlet flow (L/s), fixed point depth (m), fixed point vertical depth (m), fixed point pressure (MPa), wellhead regulating pressure (MPa), fixed point pressure consumption (MPa), fixed point ECD (g/cm 3), circulating pressure consumption (MPa), hydrostatic pressure (MPa), ECD (g/cc) and the like.
After the real drilling data acquired in the real drilling process of the target drilling are acquired, the Vanning friction resistance coefficient can be corrected through the data, and the hydrostatic column pressure, the circulating pressure consumption data and the equivalent circulating density of the drilling fluid are calculated in real time.
S320, preprocessing the real drilling data.
And firstly, performing target feature coding on the real drilling data. And adding overflow target characteristics in the last column of the real drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the drilling fluid equivalent circulating density data which are acquired in the real drilling process of the target drilling, and determining overflow judgment results according to the real drilling engineering data, the pressure measurement while drilling data and the pressure control drilling completion data which are acquired in the real drilling process to encode the overflow target characteristics, wherein the overflow working condition corresponds to 1, and the normal drilling standard is 0.
And secondly, deleting the impurity data in the time dimension. Specifically, the characteristics of distortion data (constant-999.25) and data column blank and constant 0 are deleted, and in the embodiment of the application, the following 10 characteristic parameters are removed altogether: the method comprises the steps of feeding (m), casing pressure (MPa), pumping 3 (spm), H2S (ppm), PWD depth (m), PWD drill string pressure (MPa), PWD annular space temperature (DEG C), PWD measurement ECD (g/cc), injection volume (m 3) and return depth (m), wherein 50 characteristic parameters are remained, and an overflow target characteristic is added for 51 characteristics. Illustratively, fig. 7 shows a partial real-drill data map after removing impurity data.
Thirdly, screening artificial features according to the features directly related to overflow and indirectly related to overflow. And removing inaccurate, overflow-irrelevant and repeated characteristics in the working condition records, and removing the following 19 characteristic parameters: time, vertical depth (m), drilling Time (min/m), hook position (m), hook speed (m/s), vertical pressure log (MPa), delay Time (min), inlet flow log (L/s), outlet flow log (%), PWD vertical depth (m), PWD well deviation (°), PWD azimuth (°), operation mode, fixed point depth (m), fixed point vertical depth (m), fixed point pressure (MPa), wellhead regulating pressure (MPa), fixed point pressure consumption (MPa), fixed point ECD (g/cm 3), the remaining 31 feature parameters, plus 32 features of overflow target feature.
And step four, cleaning the data according to the correlation coefficient. Specifically, the pearson correlation coefficient is adopted to continuously perform feature extraction on the data after the artificial feature selection, the correlation coefficient between each feature is calculated, a feature matrix is generated, and a correlation thermodynamic diagram is established. Illustratively, FIG. 8 shows a thermodynamic diagram of the correlation between features. According to the size of the correlation coefficient of each feature in the feature matrix and the target feature overflow, removing the PWD annular pressure (MPa) and back pressure pump flow by combining the actual engineering requirements on the basis of the correlation coefficient of the bit pressure, and obtaining the following 30 feature parameters: dateTime, well depth (m), bit sag (m), weight on bit (KN), weight on weight (KN), rotational speed (rpm), torque (KN.m), pump 1 (spm), pump 2 (spm), total sump volume (m 3), mud overflow (m 3), inlet density log (g/cc), outlet density log (g/cc), inlet temperature (DEG C), outlet temperature (DEG C), total hydrocarbons (%), C1 (%), C2 (%), stand-alone pressure (MPa), measured back pressure (MPa), outlet flow (L/s), outlet density (g/cc), circulating back pressure (MPa), additional back pressure (MPa), inlet flow (L/s), wellhead conditioning pressure (MPa), ECD (g/cc), circulating pressure (MPa), hydrostatic pressure (MPa), overflow.
And fifthly, performing type conversion on the cleaned data according to a preset mode, namely performing time sequence data type processing on the cleaned data.
And sixthly, carrying out normalization processing on the data subjected to the time sequence data type processing.
S330, dividing the preprocessed real drilling data into a training set and a testing set, wherein the proportion is 6:4.
S340, intercepting the training set data of the time dimension based on a preset time sliding window and a preset step length to obtain training data.
S350, optimizing preset super parameters by using a Bayesian optimization algorithm and a five-fold cross validation method, and determining an optimal parameter combination.
The first step, defining search_space as a dictionary of the super-parameter search space of the Bayesian optimization algorithm, wherein each key in the dictionary represents a super-parameter name, and the corresponding value comprises a range of possible values of the super-parameter and a sampling mode. Wherein the search range of the super parameter C is 0.1 to 100.0; the search range of the super parameter gamma is 0.001 to 1.0, and the sampling mode of Gaussian Process is adopted. The superparameter kernel is randomly sampled from four kernel functions [ ' linear ', ' rbf ', ' poly ', ' sigmoid ].
Second, a BayesSearchCV object opt is created using the defined search space. Some parameters need to be specified when creating an object: estimator: the model to be optimized, i.e. the classifier model SVC in the SVM; search_space: super parameter search space, namely defined super parameter search space; n_iter: the iteration times, namely how many groups of super parameter combinations the algorithm tries, are set for 25 times; cv: cross-verifying the number of folds, here five folds cross-verifying; to speed up model training and optimization, n_jobs= -1 is additionally added, and all available CPU cores are used for parallel computation.
And thirdly, executing bayes _search.fit () method, transmitting training data generated through the sliding window, and performing super-parameter search. And the training set of the data is equally divided into five folds through five-fold cross-validation, the hyper-parameter combination obtained by each iteration is transmitted into an SVC model before being not optimized, training and testing are carried out on the cross-validation set to evaluate the performance of the model, and then an evaluation index (the average value of accuracy calculated in the cross-validation) mean_test_score is taken and stored in a list. After 25 iterations, 25 sets of optimal superparameter combinations are formed. And reserving four significant digits after decimal points of all the super parameters, sorting 25 groups of super parameter combinations from large to small by means of means_test_score, obtaining optimized super parameter combination sorting, and determining optimal parameter combinations.
S360: and selecting an optimal parameter combination, and transmitting the optimal parameter combination into a Support Vector Machine (SVM) model for training to obtain an SVM fusion model based on sliding window and Bayesian optimization algorithm and five-fold cross verification super-parameter optimization, namely a drilling overflow recognition model.
S370: testing the drilling overflow identification model using the test set data and determining an evaluation report of the drilling overflow identification model.
S380: and outputting and displaying the input data and the target drilling overflow identification result so as to realize the visualization of the monitoring result.
S390: intercepting input data closest to the current moment and a target drilling overflow identification result as target data, and predicting the overflow condition of the target drilling in a period after the current moment according to the target data based on a preset prediction algorithm.
The embodiment of the application has the same beneficial effects as the embodiment.
Example IV
Fig. 9 is a schematic structural diagram of a drilling overflow identifying and predicting device according to a fourth embodiment of the present application, where the device may execute the drilling overflow identifying method according to any embodiment of the present application, and the device has functional modules and beneficial effects corresponding to the executing method. As shown in fig. 9, the apparatus includes:
A candidate data determining module 410, configured to determine candidate data according to real drilling data collected during a history drilling real drilling process;
The training data obtaining module 420 is configured to intercept candidate data of a time dimension based on a preset time sliding window and a preset step length, so as to obtain training data;
The recognition model generation module 430 is configured to perform model training based on the training data to obtain a drilling overflow recognition model, so as to recognize the overflow situation of the drilling based on the drilling overflow recognition model.
Optionally, the candidate data determination module 410 includes:
The characteristic parameter calculation unit is used for calculating hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density according to real drilling data acquired in the real drilling process of the historical drilling;
And the candidate data determining unit is used for identifying a corresponding overflow judgment result according to the real drilling data acquired in the real drilling process of the historical drilling, representing the overflow judgment result by a preset value and combining the overflow judgment result with the real drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the equivalent circulating density of the drilling fluid to obtain the candidate data.
Optionally, the candidate data determining module 410 is specifically configured to:
Determining hydrostatic column pressure according to the drilling fluid density and the drilling fluid depth;
correcting the Van Ning friction coefficient according to the cyclic pressure loss in the drilling data of the drilled or adjacent well;
Calculating cyclic pressure consumption data based on the corrected van der waals friction coefficient and the real drilling data;
and calculating the corrected drilling fluid equivalent circulating density according to the circulating pressure consumption data and the drilling fluid equivalent circulating density model.
Optionally, the apparatus further comprises:
The traversing module is used for traversing the real drilling data and determining the correlation coefficients of the real drilling data which is traversed currently, other real drilling data and overflow results;
And the removing module is used for removing the real drilling data with the correlation coefficient lower than the preset correlation coefficient threshold value.
Optionally, the candidate data determining unit includes:
The data cleaning subunit is used for carrying out data cleaning on the combined data, deleting impurity data in a time dimension and carrying out data cleaning according to the correlation coefficient of the combined data; wherein the impurity data comprises at least one of: the combined data exceeds a preset range, the data is missing, and the combined data is a preset value representing data abnormality;
and the data conversion subunit is used for carrying out type conversion on the combined data according to a preset mode, and carrying out normalization or standardization processing on the combined data to obtain the candidate data.
Optionally, the recognition model generation module 430 includes:
the optimal parameter combination determining unit is used for inputting the training data into a support vector machine model, optimizing preset super parameters by using a Bayesian optimization algorithm and a K-fold cross validation method, and determining an optimal parameter combination;
And the identification model generating unit is used for endowing the optimal parameter combination to the support vector machine model to obtain the drilling overflow identification model.
Optionally, the apparatus further comprises:
The recognition model testing module is used for inputting testing data into the drilling overflow recognition model to test the drilling overflow recognition model and determining an evaluation report of the drilling overflow recognition model; wherein the evaluation report comprises at least one of the following: the method comprises the steps of normal working condition identification accuracy rate, overflow working condition identification accuracy rate, recall rate, F1 fraction, test data quantity corresponding to different identification results, overall accuracy rate, macro average value and weighted average value.
Optionally, the apparatus further comprises:
the actual data determining module is used for determining hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density as actual data according to real drilling data acquired in the real drilling process of the target drilling well;
The input data determining module is used for intercepting actual data of the time dimension based on a preset time sliding window and a preset step length to obtain input data;
And the recognition result determining module is used for inputting the input data into the drilling overflow recognition model to determine a target drilling overflow recognition result.
Optionally, the apparatus further comprises:
The target data determining module is used for intercepting input data closest to the current moment and a target drilling overflow identification result as target data based on a preset time sliding window;
And the overflow condition prediction module is used for predicting the overflow condition of the target drilling in a period after the current moment according to the target data based on a preset prediction algorithm.
The drilling overflow identifying and predicting device provided by the embodiment of the application can execute the drilling overflow identifying and predicting method provided by any embodiment of the application, and has the corresponding functional modules and beneficial effects of the executing method.
Example five
Fig. 10 shows a schematic diagram of the structure of an electronic device 10 that may be used to implement an embodiment of the application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the applications described and/or claimed herein.
As shown in fig. 10, the electronic device 10 includes at least one processor 11, and a memory, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, in which the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12 and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.
Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the well overflow identification method.
In some embodiments, the well overflow identification method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the well overflow identification method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the well overflow identification method in any other suitable manner (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable article of manufacture such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present application, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be executed in parallel, sequentially, or in a different order, so long as the information desired by the technical solution of the present application can be achieved, and the present application is not limited herein.
The above embodiments do not limit the scope of the present application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of the present application.
Claims (12)
1. A method of identifying and predicting well overflow, the method comprising:
determining candidate data according to real drilling data acquired in the real drilling process of the historical drilling;
Intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data;
Model training is carried out based on the training data to obtain a drilling overflow identification model so as to identify the overflow condition of drilling based on the drilling overflow identification model.
2. The method of claim 1, wherein determining candidate data based on real drilling data collected during a historical real drilling process comprises:
According to the real drilling data collected in the history drilling real drilling process, calculating hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density;
According to the actual drilling data collected in the actual drilling process of the historical drilling, candidate data are determined, corresponding overflow judgment results are identified, the overflow judgment results are represented by preset values and are combined with the actual drilling data, the hydrostatic column pressure, the circulating pressure consumption data and the drilling fluid equivalent circulating density, and the candidate data are obtained.
3. The method of claim 2, wherein calculating hydrostatic column pressure, cycle pressure consumption data, drilling fluid equivalent cycle density from real drilling data collected during a historical drilling real drilling process comprises:
Determining hydrostatic column pressure according to the drilling fluid density and the drilling fluid depth;
correcting the Van Ning friction coefficient according to the cyclic pressure loss in the drilling data of the drilled or adjacent well;
Calculating cyclic pressure consumption data based on the corrected van der waals friction coefficient and the real drilling data;
and calculating the corrected drilling fluid equivalent circulating density according to the circulating pressure consumption data and the drilling fluid equivalent circulating density model.
4. The method of claim 2, wherein the overflow determination is represented by a predetermined value and combined with the real drilling data, hydrostatic column pressure, circulating pressure consumption data, drilling fluid equivalent circulating density, and prior to obtaining the candidate data, the method further comprises:
Traversing the real drilling data, and determining the correlation coefficients of the real drilling data currently traversed, other real drilling data and overflow results;
and taking and removing the real drilling data with the correlation coefficient lower than a preset correlation coefficient threshold value.
5. The method of claim 2, wherein after representing the overflow determination result with a preset value and combining with the real drilling data, hydrostatic column pressure, cycle pressure consumption data, drilling fluid equivalent cycle density, the method further comprises:
Carrying out data cleaning on the combined data, deleting impurity data in a time dimension, and carrying out data cleaning according to the correlation coefficient of the combined data; wherein the impurity data comprises at least one of: the combined data exceeds a preset range, the data is missing, and the combined data is a preset value representing data abnormality;
and performing type conversion on the combined data according to a preset mode, and performing normalization or standardization processing on the combined data to obtain the candidate data.
6. The method of claim 1, wherein model training based on the training data results in a drilling overflow identification model, comprising:
Inputting the training data into a support vector machine model, optimizing preset super parameters by using a Bayesian optimization algorithm and a K-fold cross validation method, and determining an optimal parameter combination;
And giving the optimal parameter combination to the support vector machine model to obtain the drilling overflow identification model.
7. The method of any of claims 1-6, wherein after model training based on the training data to obtain a drilling overflow identification model, the method further comprises:
inputting test data into the drilling overflow identification model to test the drilling overflow identification model, and determining an evaluation report of the drilling overflow identification model; wherein the evaluation report comprises at least one of the following: the method comprises the steps of normal working condition identification accuracy rate, overflow working condition identification accuracy rate, recall rate, F1 fraction, test data quantity corresponding to different identification results, overall accuracy rate, macro average value and weighted average value.
8. The method of claim 1, wherein after model training based on the training data to obtain a drilling overflow identification model, the method further comprises:
According to the real drilling data collected in the real drilling process of the target drilling, determining hydrostatic column pressure, circulating pressure consumption data and drilling fluid equivalent circulating density as actual data;
intercepting actual data of time dimension based on a preset time sliding window and a preset step length to obtain input data;
And inputting the input data into the drilling overflow identification model, and determining a target drilling overflow identification result.
9. The method of claim 8, wherein the method further comprises:
Based on a preset time sliding window, intercepting input data closest to the current moment and a target drilling overflow identification result as target data;
and predicting the overflow condition of the target drilling in a period after the current moment according to the target data based on a preset prediction algorithm.
10. A drilling overflow identification and prediction device, the device comprising:
The candidate data determining module is used for determining candidate data according to the real drilling data collected in the real drilling process of the historical drilling;
The training data acquisition module is used for intercepting candidate data of the time dimension based on a preset time sliding window and a preset step length to obtain training data;
And the recognition model generation module is used for carrying out model training based on the training data to obtain a drilling overflow recognition model so as to recognize the overflow condition of the drilling based on the drilling overflow recognition model.
11. A drilling overflow identification and prediction apparatus, the apparatus comprising:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the drilling overflow identification and prediction method of any of claims 1-9.
12. A computer readable storage medium storing computer instructions for causing a processor to perform the well overflow identification and prediction method of any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311550703.7A CN117932446A (en) | 2023-11-20 | 2023-11-20 | Drilling overflow identification and prediction method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311550703.7A CN117932446A (en) | 2023-11-20 | 2023-11-20 | Drilling overflow identification and prediction method, device, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117932446A true CN117932446A (en) | 2024-04-26 |
Family
ID=90756200
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311550703.7A Pending CN117932446A (en) | 2023-11-20 | 2023-11-20 | Drilling overflow identification and prediction method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117932446A (en) |
-
2023
- 2023-11-20 CN CN202311550703.7A patent/CN117932446A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110148285B (en) | Intelligent oil well parameter early warning system based on big data technology and early warning method thereof | |
| CN112529341B (en) | Drilling well leakage probability prediction method based on naive Bayesian algorithm | |
| Leng et al. | A hybrid data mining method for tunnel engineering based on real-time monitoring data from tunnel boring machines | |
| Zheng et al. | Sucker rod pumping diagnosis using valve working position and parameter optimal continuous hidden Markov model | |
| CN111311059A (en) | Knowledge graph-based water mill room fault diagnosis method | |
| CN110795888A (en) | Petroleum drilling risk prediction method | |
| CN112529053A (en) | Short-term prediction method and system for time sequence data in server | |
| CN116629620A (en) | Risk level determining method and device, electronic equipment and storage medium | |
| Yang et al. | A Fault Identification Method for Electric Submersible Pumps Based on DAE‐SVM | |
| Liang et al. | Overflow warning and remote monitoring technology based on improved random forest | |
| Geekiyanage et al. | Feature selection for kick detection with machine learning using laboratory data | |
| CN117108269A (en) | Method, device, equipment and medium for dynamically adjusting parameters of oil pumping well | |
| CN117932446A (en) | Drilling overflow identification and prediction method, device, equipment and medium | |
| CN116629686A (en) | Method and device for evaluating enterprise energy consumption data | |
| US20230111036A1 (en) | Integrated drilling dysfunction prediction | |
| CN111461461B (en) | Hydraulic engineering abnormity detection method and system | |
| Shaohu et al. | Prediction of drilling plug operation parameters based on incremental learning and CNN-LSTM | |
| CN113392576A (en) | Staying balloon main cable tension state assessment early warning method | |
| CN113537706A (en) | Oil field production increasing measure optimization method based on intelligent integration | |
| Fernández Berrocal et al. | Evaluation and Interpretation on Data Driven ROP Models From Engineering Perspectives | |
| Feder | Machine-learning approach determines spatial variation in shale decline curves | |
| CN116302848B (en) | Detection method and device for bias of evaluation value, electronic equipment and medium | |
| CN117077060A (en) | Shale gas well lost circulation early warning method and device, electronic equipment and storage medium | |
| Zhang et al. | Intelligent kick detection using a parameter adaptive neural network | |
| Wang et al. | A Seq2Seq-Based Method in the Prediction of Oil Production with Long Short-Term Memory Network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |