Well leakage prediction method and device, electronic equipment and storage medium
By labeling and model training of historical well recording data, the target prediction model is determined, and the problem that well leakage prediction depends on manual experience in the existing technology is solved, and the accuracy and timeliness of well leakage prediction are achieved, thereby reducing the risk of well leakage.
Patent Information
- Application Number
- CN202310610361.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-26
- Publication Date
- 2024-11-26
AI Technical Summary
The existing well leakage prediction technology relies too much on manual experience, and the diagnostic results are not accurate, and it is impossible to predict well leakage before drilling, making it difficult to prevent the risk of well leakage.
By labeling historical well recording data, and training multiple prediction models based on historical well recording data and their labels, determining the target prediction model, inputting the current well recording data into the target prediction model, and obtaining the well leakage prediction results.
It improves the accuracy and timeliness of well leakage prediction, realizes well leakage warning before drilling, and reduces the risk of well leakage.
Smart Images

Figure CN119026719A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of oil and gas development, and in particular to a method, device, electronic equipment and storage medium for predicting well leakage. Background Art
[0002] Well leakage refers to a complex underground situation in which the drilling fluid completely penetrates into the bottom layer due to the pressure difference during drilling. After the well leakage occurs, it will lead to a waste of time, materials and manpower. Severe well leakage may even cause blowouts, well wall instability, reservoir damage and even wellbore abandonment. Early warning can minimize the risk of well leakage. Therefore, well leakage prediction is an important topic in the field of oil and gas development.
[0003] At present, the more mature well leakage prediction technology mainly uses field exploration conditions such as micro-overflow measurement, drilling fluid inlet and outlet flow measurement, and downhole micro-flow detection, combined with expert analysis, to diagnose well leakage. This method relies too much on manual experience, the diagnostic results are not accurate, and it is impossible to predict well leakage before drilling. Summary of the invention
[0004] The present invention provides a method, device, electronic equipment and storage medium for predicting well leakage, so as to improve the accuracy and timeliness of well leakage prediction.
[0005] In a first aspect, an embodiment of the present invention provides a method for predicting lost circulation, the method comprising:
[0006] Determining historical logging data and a label matching the historical logging data, wherein the label is a lost circulation or no lost circulation;
[0007] According to the historical logging data and the labels matching the historical logging data, respectively, at least two preset candidate prediction models are trained, and a target prediction model is determined from among the candidate prediction models obtained through the training;
[0008] The current logging data is input into the target prediction model to obtain the well leakage prediction result output by the target prediction model.
[0009] In a second aspect, an embodiment of the present invention further provides a device for predicting lost circulation, the device comprising:
[0010] A historical logging data determination module, used to determine historical logging data and a label matching the historical logging data, wherein the label is a lost circulation or no lost circulation;
[0011] A target prediction model determination module, used to train at least two preset candidate prediction models according to the historical logging data and the labels matched with the historical logging data, and determine a target prediction model from the candidate prediction models obtained through the training;
[0012] The well loss prediction result determination module is used to input the current logging data into the target prediction model to obtain the well loss prediction result output by the target prediction model.
[0013] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for predicting well leakage as described in any one of the embodiments of the present invention is implemented.
[0014] In a fourth aspect, an embodiment of the present invention further provides a storage medium storing computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to execute a method for predicting lost circulation as described in any one of the embodiments of the present invention.
[0015] The technical solution of the embodiment of the present invention labels the historical logging data, and trains multiple prediction models respectively according to the historical logging data and its labels, determines the target prediction model among the multiple prediction models obtained by training, inputs the current logging data into the target prediction model, and obtains the well leakage prediction result. The problem that the well leakage diagnosis method in the prior art relies too much on manual experience and the diagnosis result has low accuracy, and the problem that the well leakage prediction before drilling cannot be realized is solved, and the accuracy and timeliness of the well leakage prediction are improved.
[0016] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0018] Figure 1 is a flow chart of a method for predicting lost circulation provided by the first embodiment of the present invention;
[0019] Figure 2 is a flow chart of a method for predicting lost circulation provided by the second embodiment of the present invention;
[0020] Figure 3 It is a structural schematic diagram of a well leakage prediction device provided by the third embodiment of the present invention;
[0021] Figure 4 It is a structural schematic diagram of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION
[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0024] Embodiment 1
[0025] Figure 1 A flowchart of a method for predicting well leakage is provided for embodiment 1 of the present invention. This embodiment is applicable to the situation of predicting well leakage before drilling. The method can be executed by a well leakage prediction device. The well leakage prediction device can be implemented in the form of hardware and / or software. The well leakage prediction device can be configured in an electronic device.
[0026] like Figure 1 As shown, the method includes:
[0027] S110: Determine historical logging data and tags matching the historical logging data.
[0028] The label is leakage or no leakage. Specifically, the historical logging data can be the whole meter logging data of the drilled well, and the data of the well section where the leakage occurs is labeled as "leakage", and the other data is labeled as "no leakage". The historical logging data may include characteristics such as well depth, drilling time, drilling pressure, torque, hanging weight, pump pressure, displacement, inlet density, outlet density, inlet temperature, outlet temperature, total pool volume, difference, and lithology. Since the prediction model needs to be trained based on the historical logging data in the future, in order to improve the efficiency of model training and establish an effective data set, the historical logging data can be preprocessed.
[0029] Specifically, preprocessing may include but is not limited to normalization processing, correction processing for erroneous data, quantization processing, and normalization processing, etc. However, this embodiment does not limit the type, quantity, and specific method of preprocessing. Among them, normalization processing means that when the historical logging data from different sources have different expressions for the same feature, the expression of the feature is determined as a standardized expression. For example, if the historical logging data of drilled well A includes the feature "drilling fluid inlet temperature", the historical logging data of drilled well B includes the feature "inlet temperature", and the historical logging data of drilled well C includes the feature "temperature 1", the features "drilling fluid inlet temperature", "inlet temperature" and "temperature 1" are all used to represent the drilling fluid temperature at the inlet, if the standardized expression corresponding to the feature is "inlet temperature", then the feature name in the historical logging data of drilled wells A and drilled wells C is modified to "inlet temperature".
[0030] The error and missing data correction process refers to correcting the error data in the historical logging data and adding the missing data. Furthermore, it is possible to compare each historical logging data with a threshold value determined based on the empirical value of the same type of data to determine whether the historical logging data is error data, and it is possible to determine whether there is missing data by detecting whether the feature value corresponding to each feature in the whole meter logging data of each well that has been drilled is empty. When it is determined that a certain historical logging data is error data, the historical logging data can be adjusted to data that meets the threshold conditions of the same type of data. For example, the well depth corresponding to the drilled well A should be within the range of 3000-5000m. If the well depth of a certain historical logging data is 40000m, it can be known that the well depth data is error data and can be corrected to 4000m. When it is determined that a certain historical logging data is missing data, it can be supplemented according to the empirical value of the same type of data. This embodiment does not limit the detection of error data and missing data, and the method of correcting error data and missing data.
[0031] Quantization processing refers to quantifying text data in digital form. For example, for the feature of lithology, it can include fine sandstone, dolomite, gypsum, etc., and different lithologies can be represented by numbers, for example, fine sandstone is represented by number 0, dolomite is represented by number 1, gypsum is represented by number 2, etc. This embodiment does not limit the specific form of quantization processing.
[0032] Normalization processing refers to mapping each historical logging data into the same mapping range, such as [0,1] or [-1,1]. Normalizing the historical logging data can improve the performance of the data set and increase the convergence speed of the model during model training.
[0033] Furthermore, after preprocessing the historical logging data, if the number of samples of the historical logging data is small, data simulation can be performed based on the historical logging data, and the simulated logging data and the historical logging data can be used together as a data set for model training. This embodiment does not limit this.
[0034] S120: training at least two preset candidate prediction models respectively according to the historical logging data and the labels matched with the historical logging data, and determining a target prediction model from the candidate prediction models obtained through the training.
[0035] In this embodiment, after the historical logging data and its labels form a valid data set, multiple pre-set candidate prediction models are trained to obtain trained candidate prediction models. Among the trained candidate prediction models, the target prediction model is selected as the model for subsequent well leakage prediction.
[0036] The number of target prediction models may be one or more. Specifically, according to the actual needs of well leakage prediction, a target prediction model may be selected from each candidate prediction model for well leakage prediction, or multiple candidate prediction models may be selected as target prediction models to jointly perform well leakage prediction. As long as the prediction result of at least one target prediction model is that well leakage occurs, a well leakage warning is performed, and the risk of well leakage is eliminated by adjusting various drilling parameters.
[0037] Furthermore, the basis for selecting the target prediction model from among the candidate prediction models can be flexibly restricted according to the actual needs of well leakage prediction. For example, if the need for well leakage prediction is timeliness, a candidate prediction model with a short prediction time consumption is selected as the target prediction model; if the need for well leakage prediction is accuracy, a candidate prediction model with a high model accuracy is selected as the target prediction model; if there are multiple target prediction models, candidate prediction models with a short prediction time consumption and a high model accuracy can be selected respectively, and well leakage prediction is performed together according to multiple candidate prediction models.
[0038] Furthermore, the preset candidate prediction models include at least two of the following: a linear Gaussian function model, a quadratic Gaussian function model, a cubic Gaussian function model, a coarse Gaussian function model, a medium Gaussian function model, a fine Gaussian function model and an artificial neural network model.
[0039] Among them, the linear Gaussian function model, the quadratic Gaussian function model, the cubic Gaussian function model, the coarse Gaussian function model, the medium Gaussian function model and the fine Gaussian function model are all support vector machine models derived from kernel functions. Support Vector Machines (SVM) is a binary classification model. In this embodiment, six different kernel functions are used to map the input data of the model to a higher feature space, which can improve the accuracy of classification and thus improve the accuracy of well leakage prediction.
[0040] The artificial neural network (ANN) model is a widely parallel interconnected network composed of simple units with adaptability. The artificial neural network model is composed of multiple neuron models connected according to certain connection rules. During the training process of the artificial neural network model, the artificial neural network model can repeatedly update the connection weights and thresholds according to the learning rules, and finally achieve the minimum error between the predicted output and the actual output.
[0041] It should be noted that the above-mentioned model types are merely examples of candidate prediction model types, and other types of models may also be used as candidate prediction models, which is not limited in this embodiment.
[0042] Furthermore, the artificial neural network model is a four-layer neural network model, including an input layer, two hidden layers and an output layer.
[0043] The artificial neural network model in this embodiment adopts a four-layer neural network structure, including an input layer, two hidden layers and an output layer, so as to better process complex nonlinear data distribution.
[0044] The technical solution of this embodiment can meet the actual needs of well leakage prediction by training multiple candidate prediction models and selecting a target prediction model from each trained candidate prediction model to perform final well leakage prediction.
[0045] S130, inputting the current logging data into the target prediction model to obtain the well leakage prediction result output by the target prediction model.
[0046] The current logging data may be the whole meter logging data of the currently drilled well or the real-time data before drilling, which is not limited in this embodiment. The well leakage prediction result may include whether the well leakage occurs, the probability value of the well leakage occurrence, and when the prediction result is the well leakage occurrence, the characteristics related to the well leakage.
[0047] In this embodiment, after obtaining the lost circulation prediction result output by the target prediction model, the adjustment strategy of the drilling parameters can be determined in combination with the features related to lost circulation to avoid lost circulation.
[0048] Furthermore, this embodiment can update the historical logging data at preset time intervals, such as every month, or every time a preset number of drillings are completed, such as every time a drilling is completed, and repeat the operations of S110-S120 to redefine the target prediction model to maintain the accuracy of well leakage prediction.
[0049] It should be noted that, in this embodiment, the process of well leakage prediction is described by taking historical logging data as an example, and a more systematic well leakage prediction system can be built by combining different types of data (such as geological data, logging data, and formation data, etc.). Specifically, different target prediction models can be selected for different types of data, and the probability of well leakage can be analyzed by combining the prediction results of each well leakage.
[0050] The technical solution of the embodiment of the present invention labels the historical logging data, and trains multiple prediction models respectively according to the historical logging data and its labels, determines the target prediction model among the multiple prediction models obtained by training, inputs the current logging data into the target prediction model, and obtains the well leakage prediction result. The problem that the well leakage diagnosis method in the prior art relies too much on manual experience and the diagnosis result has low accuracy, and the problem that the well leakage prediction before drilling cannot be realized is solved, and the accuracy and timeliness of the well leakage prediction are improved.
[0051] Embodiment 2
[0052] Figure 2 A flowchart of a method for predicting lost circulation provided in the second embodiment of the present invention. Based on the above embodiments, the embodiment of the present invention further specifies the process of processing historical logging data, the process of model training, and the process of determining the target prediction model.
[0053] like Figure 2 As shown, the method includes:
[0054] S210: Acquire historical logging data, and determine tags matching the historical logging data.
[0055] The method of obtaining the historical logging data and determining the label for the historical logging data has been described in the above embodiment, and will not be repeated in this embodiment.
[0056] S220: If it is determined that the historical logging data includes text data, quantize the text data.
[0057] S230: normalize the historical logging data.
[0058] The above-mentioned embodiments have described the specific process of quantization and normalization of text data, which will not be repeated in this embodiment.
[0059] S240, performing dimensionality reduction processing on each feature according to the correlation between each feature and the lost circulation, and determining the feature after the dimensionality reduction processing and the feature value matching the feature after the dimensionality reduction processing.
[0060] The historical logging data includes at least two features and feature values matching the features.
[0061] In this embodiment, when the historical logging data includes multiple features, or some features have a low correlation with well leakage, if the model training is performed based on all the features of the historical logging data, it will not only be time-consuming and inefficient, but may also interfere with the accuracy of well leakage prediction. Therefore, this embodiment performs dimensionality reduction processing on each feature included in the historical logging data according to the correlation between the feature and well leakage, retains the features with a strong correlation with well leakage, and performs subsequent model training to improve the efficiency of model training and effectively remove the interference of potential redundant factors on the accurate prediction of well leakage.
[0062] Specifically, the principal component analysis (PCA) method can be used to reduce the dimension of each feature. The main idea of PCA is to map n-dimensional features to k-dimensional features. This k-dimensional feature is a new orthogonal feature, which is a k-dimensional feature reconstructed on the basis of the original n-dimensional feature. Taking the above historical logging data including well depth, drilling time, drilling pressure, torque, hanging weight, pump pressure, displacement, inlet density, outlet density, inlet temperature, outlet temperature, total pool volume, difference and lithology as an example, after feature dimensionality reduction, the new features are well depth, drilling time, drilling pressure and other features.
[0063] Table 1 provides a schematic diagram of the impact of feature dimensionality reduction on model performance. As shown in Table 1, after the features are reduced in dimensionality through PCA, the accuracy of each candidate prediction model is improved, the time consumption is shortened, and the computational efficiency of the prediction model is improved.
[0064]
[0065] S250: Divide the historical logging data and labels matching the historical logging data into a training set and a test set.
[0066] Specifically, the historical logging data and its labels may be randomly divided according to a ratio of 4:1 to obtain a training set and a test set. However, this embodiment does not limit the ratio of the training set to the test set.
[0067] S260: Train the preset candidate prediction model according to the training set, and test the trained candidate prediction model according to the test set.
[0068] In this embodiment, the training set is used to train the model, and the test set is used to test the model performance.
[0069] S270, determine whether the test result of the candidate prediction model meets the model training requirements, if so, execute S280, otherwise return to execute S260.
[0070] The model training requirement may be that the model accuracy is greater than or equal to a preset threshold, or that the number of model training times is greater than or equal to a preset number, which is not limited in this embodiment.
[0071] S280: Determine a target prediction model from among the candidate prediction models obtained through training according to the prediction accuracy and time consumption of each candidate prediction model.
[0072] In this embodiment, after training of multiple types of candidate prediction models is completed, a target prediction model is selected based on the prediction accuracy and time consumption of each candidate prediction model.
[0073] Specifically, weights can be determined for prediction accuracy and time consumption respectively, and the sum of the weights is 1. Since the smaller the time consumption, the better, for each candidate prediction model, the sum of the product of the prediction accuracy and the prediction accuracy weight and the product of the negative value corresponding to the time consumption and the time consumption weight is calculated to obtain the score value of the candidate prediction model. For each candidate prediction model, the order from large to small according to the score value is sorted, and the first preset number of candidate prediction models are selected as the target prediction model.
[0074] Taking Table 1 as an example, if one target prediction model is selected, the target prediction model is an artificial neural network model; if two target prediction models are selected, the target prediction models are an artificial neural network model and a fine Gaussian function model.
[0075] S290, inputting the current logging data into the target prediction model to obtain the well leakage prediction result output by the target prediction model.
[0076] The technical solution of this embodiment uses multiple types of candidate prediction models for training and comparison, and selects the most appropriate prediction model for lost circulation prediction, which can achieve real-time lost circulation prediction with good prediction effect and high timeliness.
[0077] The technical solution of the embodiment of the present invention is to label the historical logging data, and after quantization, normalization and feature dimension reduction processing, divide the historical logging data into training sets and test sets, respectively train seven prediction models including six kernel function derived support vector machine algorithm models and a neural network model, determine the target prediction model according to the model accuracy and time consumption in each prediction model obtained by training, input the current logging data into the target prediction model, and obtain the well leakage prediction result. The problem that the well leakage diagnosis method in the prior art relies too much on manual experience and the diagnosis result has low accuracy, and the problem that the well leakage prediction before drilling cannot be realized is solved. The accuracy of model training is improved through a series of processing of historical logging data. At the same time, the accuracy and timeliness of well leakage prediction are improved by training multiple models and selecting the optimal prediction model according to training accuracy and time consumption.
[0078] Embodiment 3
[0079] Figure 3 This is a schematic diagram of the structure of a device for predicting lost circulation provided by the third embodiment of the present invention. Figure 3 As shown, the device includes: a historical logging data determination module 310, a target prediction model determination module 320 and a well loss prediction result determination module 330. Among them:
[0080] A historical logging data determination module 310 is used to determine historical logging data and a label matching the historical logging data, wherein the label is a lost circulation or no lost circulation;
[0081] A target prediction model determination module 320 is used to train at least two preset candidate prediction models according to the historical logging data and the labels matched with the historical logging data, and determine a target prediction model from the candidate prediction models obtained through the training;
[0082] The lost circulation prediction result determination module 330 is used to input the current logging data into the target prediction model to obtain the lost circulation prediction result output by the target prediction model.
[0083] The technical solution of the embodiment of the present invention labels the historical logging data, and trains multiple prediction models respectively according to the historical logging data and its labels, determines the target prediction model among the multiple prediction models obtained by training, inputs the current logging data into the target prediction model, and obtains the well leakage prediction result. The problem that the well leakage diagnosis method in the prior art relies too much on manual experience and the diagnosis result has low accuracy, and the problem that the well leakage prediction before drilling cannot be realized is solved, and the accuracy and timeliness of the well leakage prediction are improved.
[0084] Based on the above embodiment, the historical logging data determination module 310 includes:
[0085] A historical logging data acquisition unit, used to acquire historical logging data and determine a tag matching the historical logging data;
[0086] A quantization processing unit, for quantizing the text data if it is determined that the historical logging data includes text data;
[0087] The normalization unit is used to normalize the historical logging data.
[0088] Based on the above embodiment, the historical logging data includes at least two features and feature values matching the features;
[0089] The device further comprises:
[0090] The dimension reduction processing module is used to perform dimension reduction processing on each feature according to the correlation between each feature and the well leakage, and determine the features after the dimension reduction processing and the feature values matching the features after the dimension reduction processing.
[0091] Based on the above embodiment, the target prediction model determination module 320 includes:
[0092] A historical logging data division unit, used for dividing the historical logging data and labels matching the historical logging data into a training set and a test set;
[0093] A model training unit, used to train a preset candidate prediction model according to a training set, and to test the trained candidate prediction model according to a test set;
[0094] The test result judgment unit is used to repeatedly perform the operation of training the preset candidate prediction model according to the training set until the test result of the candidate prediction model meets the model training requirements.
[0095] Based on the above embodiment, the preset candidate prediction models include at least two of the following: a linear Gaussian function model, a quadratic Gaussian function model, a cubic Gaussian function model, a coarse Gaussian function model, a medium Gaussian function model, a fine Gaussian function model and an artificial neural network model.
[0096] Based on the above embodiment, the artificial neural network model is a four-layer neural network model, including an input layer, two hidden layers and an output layer.
[0097] Based on the above embodiment, the target prediction model determination module 320 includes:
[0098] The target prediction model determination unit is used to determine the target prediction model from the candidate prediction models obtained through training according to the prediction accuracy and time consumption of each candidate prediction model.
[0099] The lost circulation prediction device provided in the embodiment of the present invention can execute the lost circulation prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0100] Embodiment 4
[0101] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.
[0102] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can 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 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.
[0103] A number of 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, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0104] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a method for predicting lost circulation.
[0105] In some embodiments, the method for predicting lost circulation may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for predicting lost circulation described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the method for predicting lost circulation in any other appropriate manner (e.g., by means of firmware).
[0106] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0107] Computer programs for implementing the methods of the present invention 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, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
[0108] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0109] To provide interaction with a user, the systems and techniques described herein may 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 the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the 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, voice input, or tactile input).
[0110] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.
[0111] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.
[0112] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.
[0113] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for predicting lost circulation, characterized in that: include: Determining historical logging data and a label matching the historical logging data, wherein the label is a lost circulation or no lost circulation; According to the historical logging data and the labels matching the historical logging data, respectively, at least two preset candidate prediction models are trained, and a target prediction model is determined from among the candidate prediction models obtained through the training; The current logging data is input into the target prediction model to obtain the well leakage prediction result output by the target prediction model.
2. The method according to claim 1, characterized in that Determining historical logging data and tags matching the historical logging data includes: Acquire historical logging data and determine tags matching the historical logging data; If it is determined that the historical logging data includes text data, quantizing the text data; Normalize the historical logging data.
3. The method according to claim 2, characterized in that The historical logging data includes at least two features and feature values matching the features; After normalizing the historical logging data, it also includes: According to the correlation between each feature and the lost circulation, each feature is subjected to dimensionality reduction processing, and the features after dimensionality reduction processing and the feature values matching the features after dimensionality reduction processing are determined.
4. The method according to claim 1, characterized in that: According to the historical logging data and the labels matching the historical logging data, at least two preset candidate prediction models are trained respectively, including: Dividing the historical logging data and labels matched with the historical logging data into a training set and a test set; The preset candidate prediction model is trained according to the training set, and the trained candidate prediction model is tested according to the test set; Repeat the operation of training the preset candidate prediction model according to the training set until the test results of the candidate prediction model meet the model training requirements.
5. The method according to claim 1, characterized in that The preset candidate prediction models include at least two of the following: a linear Gaussian function model, a quadratic Gaussian function model, a cubic Gaussian function model, a coarse Gaussian function model, a medium Gaussian function model, a fine Gaussian function model and an artificial neural network model.
6. The method according to claim 5, characterized in that The artificial neural network model is a four-layer neural network model, including an input layer, two hidden layers and an output layer.
7. The method according to claim 1, characterized in that Determine the target prediction model from the candidate prediction models obtained through training, including: According to the prediction accuracy and time consumption of each candidate prediction model, a target prediction model is determined from among the candidate prediction models obtained through training.
8. A device for predicting lost circulation, characterized in that: include: A historical logging data determination module, used to determine historical logging data and a label matching the historical logging data, wherein the label is a lost circulation or no lost circulation; A target prediction model determination module, used to train at least two preset candidate prediction models according to the historical logging data and the labels matched with the historical logging data, and determine a target prediction model from the candidate prediction models obtained through the training; The well loss prediction result determination module is used to input the current logging data into the target prediction model to obtain the well loss prediction result output by the target prediction model.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for predicting lost circulation as described in any one of claims 1-7 is implemented.
10. A storage medium storing computer executable instructions, characterized in that: When the computer executable instructions are executed by a computer processor, they are used to perform the method for predicting lost circulation as claimed in any one of claims 1 to 7.
