CN113052195A - Logging curve abnormal value marking method and device, computer equipment and storage medium - Google Patents
Logging curve abnormal value marking method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN113052195A CN113052195A CN201911380142.4A CN201911380142A CN113052195A CN 113052195 A CN113052195 A CN 113052195A CN 201911380142 A CN201911380142 A CN 201911380142A CN 113052195 A CN113052195 A CN 113052195A
- Authority
- CN
- China
- Prior art keywords
- machine learning
- trained machine
- logging curve
- learning model
- abnormal value
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 125
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000010801 machine learning Methods 0.000 claims abstract description 181
- 238000002372 labelling Methods 0.000 claims abstract description 70
- 238000005457 optimization Methods 0.000 claims abstract description 13
- 238000012360 testing method Methods 0.000 claims description 42
- 238000012937 correction Methods 0.000 claims description 37
- 238000012549 training Methods 0.000 claims description 31
- 238000012163 sequencing technique Methods 0.000 claims description 16
- 238000004891 communication Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 11
- 238000004590 computer program Methods 0.000 description 12
- 238000012545 processing Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 238000010276 construction Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002285 radioactive effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
The application relates to a method, a device, computer equipment and a storage medium for marking abnormal values of a logging curve, wherein the method comprises the following steps: and reading a preset trained machine learning model set, selecting the optimal trained machine learning model under the current optimization standard from the set, and inputting the logging curve to be labeled into the selected trained machine learning model to obtain a logging curve abnormal value labeling result. In the whole process, the optimal trained machine learning model under the current optimization standard is selected, the abnormal value of the logging curve to be labeled is labeled through the selected trained machine learning model, manual abnormal value labeling is not needed, and the abnormal value labeling of the logging curve can be efficiently and accurately achieved.
Description
Technical Field
The application relates to the technical field of automation, in particular to a logging curve abnormal value labeling method, a logging curve abnormal value labeling device, a logging curve abnormal value labeling computer device and a storage medium.
Background
Well logging is a branch of geophysical exploration and is a generic term for geophysical exploration methods used in boreholes. According to the physical properties of the rock, the method can be classified into electric logging, radioactive logging, magnetic logging, acoustic logging, thermal logging, gravity logging and the like. The parameter condition of the stratum can be obtained through the logging curve, so that exploration and development of petroleum are facilitated. However, the logging curve is influenced by various factors such as environment during the measurement process, and some abnormal values are generated. These values are not a true reflection of the formation properties and if these outliers are not processed in the early curve preprocessing stage, they can severely affect the results of the final log interpretation. Therefore, the identification of log outliers is particularly important.
The traditional abnormal value labeling mode of the logging curve is mainly characterized in that a computer runs some logging interpretation software, the logging curve is presented in a visual mode, experienced interpreters check the curves one by one from top to bottom to find out abnormal values in the curves, the personnel operate on the computer to input the found abnormal values into the computer, and the computer integrates the input data and the original logging curve to obtain the logging curve with the abnormal values finally.
Although the above method can realize the abnormal value labeling of the logging curve, because the logging depth is very large, a large amount of data needs to be processed, a large amount of time is consumed for processing and identifying the part of data by manpower, and the cognition and experience of different people in the processing process are deviated, so that the abnormal value labeling of the part of data is inaccurate.
Disclosure of Invention
In view of the above, there is a need to provide a method, an apparatus, a computer device and a storage medium for labeling abnormal values of a well logging curve with high efficiency and accuracy.
A method of logging curve outlier labeling, the method comprising:
acquiring a logging curve to be marked and an optimization standard;
reading a preset trained machine learning model set, wherein the trained machine learning model set comprises a plurality of trained machine learning models, and the trained machine learning models are obtained by training sample well logging curves carrying abnormal value labels;
selecting the optimal trained machine learning model under the tuning standard from the trained machine learning model set;
and inputting the logging curve to be labeled into the selected trained machine learning model to obtain a logging curve abnormal value labeling result.
In one embodiment, before reading the preset set of trained machine learning models, the method further includes:
acquiring a sample logging curve carrying abnormal value labels;
randomly dividing the sample data into training sample data and test sample data;
training a plurality of initial machine learning models respectively according to the training sample data to obtain a plurality of trained machine learning models;
respectively testing the plurality of trained machine learning models according to the test sample data to obtain model test results;
and obtaining the optimal trained machine learning models under different tuning standards according to the model test results, and constructing a preset trained machine learning model set.
In one embodiment, the acquiring a sample log with an outlier label includes:
acquiring an original logging curve;
carrying out depth correction on the original logging curve to obtain an original logging curve after the depth correction;
and obtaining a sample logging curve carrying abnormal value labels by a third party for carrying out abnormal value labeling on the original logging curve after the depth correction.
In one embodiment, the depth correcting the original well log to obtain a depth-corrected original well log includes:
randomly selecting any one of the original logging curves as a reference curve;
comparing other curves in the original logging curve with the reference curve to obtain depth errors of the other curves in the original logging curve and the reference curve;
and performing depth correction on other curves in the original logging curve according to the depth error to obtain the original logging curve after depth correction.
In one embodiment, the obtaining, according to the model test result, the optimal trained machine learning models under different tuning standards, and the constructing a preset set of trained machine learning models includes:
according to the model test result, respectively carrying out model quality sequencing on the trained machine learning model under different tuning standards to respectively obtain a plurality of sequencing sequences and record corresponding tuning standards;
and selecting the optimal trained machine learning model in the plurality of sequencing sequences to be matched with the corresponding tuning standard, and constructing a preset trained machine learning model set.
In one embodiment, the inputting the logging curve to be labeled into the selected trained machine learning model to obtain a logging curve abnormal value labeling result includes:
inputting the logging curve to be marked into the selected trained machine learning model;
acquiring output data of the selected trained machine learning model to obtain an abnormal value and a depth corresponding to the abnormal value;
and according to the depth corresponding to the abnormal value, the abnormal value is marked to the logging curve to be marked, and a logging curve abnormal value marking result is obtained.
In one embodiment, the inputting the well log to be labeled to the selected trained machine learning model includes:
carrying out depth correction on the well logging curve to be marked;
and inputting the logging curve to be marked after the depth correction into the selected trained machine learning model.
A well log outlier labeling apparatus, the apparatus comprising:
the data acquisition module is used for acquiring a logging curve to be marked and an optimization standard;
the reading module is used for reading a preset trained machine learning model set, the trained machine learning model set comprises a plurality of trained machine learning models, and the trained machine learning models are obtained by training sample logging curves carrying abnormal value labels;
the model selection module is used for selecting the optimal trained machine learning model under the tuning standard from the trained machine learning model set;
and the abnormal value marking module is used for inputting the logging curve to be marked to the selected trained machine learning model to obtain a logging curve abnormal value marking result.
In addition, the application also provides an electronic device, which comprises at least one processor, at least one memory and a bus, wherein the memory and the bus are connected with the processor; the processor and the memory complete mutual communication through the bus; the processor is used for calling the program instructions in the memory to execute the logging curve abnormal value labeling method.
In addition, the application also provides a storage medium which comprises a stored program, wherein when the program runs, the equipment on which the storage medium is controlled to execute is used for realizing the logging curve abnormal value labeling method.
According to the logging curve abnormal value labeling method, the logging curve abnormal value labeling device, the computer equipment and the storage medium, a preset trained machine learning model set is read, a trained machine learning model which is optimal under the current tuning standard is selected from the set, the logging curve to be labeled is input to the selected trained machine learning model, and a logging curve abnormal value labeling result is obtained. In the whole process, the optimal trained machine learning model under the current optimization standard is selected, the abnormal value of the logging curve to be labeled is labeled through the selected trained machine learning model, manual abnormal value labeling is not needed, and the abnormal value labeling of the logging curve can be efficiently and accurately achieved.
Drawings
FIG. 1 is a diagram of an environment in which the method for labeling abnormal values of a well log is applied in one embodiment;
FIG. 2 is a schematic flow chart illustrating a method for labeling abnormal values of a well log in one embodiment;
FIG. 3 is a schematic flow chart illustrating a method for labeling abnormal values of a well log in another embodiment;
FIG. 4 is a schematic view of a log depth correction;
FIG. 5 is a block diagram of an apparatus for labeling abnormal values of a well log according to an embodiment;
fig. 6 is an internal structural diagram of the apparatus in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The method for labeling the abnormal value of the logging curve can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The server 104 stores a trained machine learning model set in advance, the terminal 102 sends the well logging curve to be labeled and the tuning standard under the current application scene to the server 104, the server 104 obtains the well logging curve to be labeled and the tuning standard, reads the preset trained machine learning model set, selects the best trained machine learning model under the tuning standard from the trained machine learning model set, inputs the well logging curve to be labeled to the selected trained machine learning model to obtain a well logging curve abnormal value labeling result, and the server 104 can feed back the well logging curve abnormal value labeling result to the terminal 102. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, a method for labeling abnormal values of a well log is provided, which is illustrated by applying the method to the server 104 in fig. 1, and includes the following steps:
s200: and obtaining a logging curve to be marked and an adjusting and optimizing standard.
The logging curve to be annotated refers to an object for annotating an abnormal value of the logging curve, and specifically, the logging curve to be annotated can be submitted to a server on a terminal by a user, and the server directly acquires the logging curve to be annotated uploaded by the terminal. The tuning standard can be understood as the evaluation scale of the model, and in the subsequent operation, different trained machine learning models can be selected for logging curve abnormal value labeling under different evaluation scales. Specifically, the tuning criteria include mean _ absolute _ error, mean _ squared _ error, mean _ absolute _ error, and r2_ score (goodness of fit), and the 4 tuning criteria have their corresponding mathematical scaling formulas. In practical application, a professional operator operates at a terminal, a logging curve to be labeled is uploaded to a server through the terminal, and in addition, a user selects a corresponding optimization standard at the terminal according to the service requirement (the purpose or the use of the abnormal value labeling of the logging curve and the like) and uploads the selected optimization standard to the server. For example, in the current application scenario, the porosity needs to be calculated, the corresponding tuning standard is r2_ score, and the professional operator selects the tuning standard, r2_ score, to upload to the server. Unnecessary, in practical application, the tuning standard can be directly displayed on the terminal for a professional operator to click and select; and the terminal can only display the service demand option, and automatically search the tuning index (evaluation scale) corresponding to the service demand option and upload the tuning index to the server.
S400: reading a preset trained machine learning model set, wherein the trained machine learning model set comprises a plurality of trained machine learning models, and the trained machine learning models are obtained by training sample well logging curves carrying abnormal value labels.
The preset trained machine learning model set is a set of preset trained machine learning models and comprises a plurality of trained machine learning models, the plurality of trained machine learning models can be obtained by training different machine learning models through sample data, tuning standards can be considered in the trained machine learning model set, and the model set can only contain the optimal trained machine learning models under different tuning standards. The machine learning model may specifically include a model based on 10 regression algorithms such as HuberRegressor, adaboost regressor, Lasso, Ridge, SGDRegressor, LinearSVR, SVR, decisiontresereregressor, bagggingregressor, GradientBoostRegressor, randomforteregraessor, and the like.
S600: and selecting the optimal trained machine learning model under the tuning standard from the trained machine learning model set.
The trained machine learning model set carries a plurality of trained machine learning models, and tuning standards are associated with the trained machine learning models, so that the trained machine learning model which is optimal under the current tuning standard can be selected from the trained machine learning models. Generally, a tuning standard-optimal trained machine learning model correspondence table is stored in the trained machine learning model set, and based on the tuning standard obtained in step S200 and the correspondence table, an optimal trained machine learning model under the tuning standard can be obtained.
S800: and inputting the logging curve to be labeled into the selected trained machine learning model to obtain the logging curve abnormal value labeling result.
And inputting the labeled logging curve into the selected trained machine learning model, so as to obtain the logging curve abnormal value labeling result corresponding to the logging curve labeling task. Specifically, the abnormal value and the depth corresponding to the abnormal value can be directly obtained from the output data of the selected trained machine learning model, and the abnormal value is labeled to the logging curve to be labeled according to the depth corresponding to the abnormal value, so that the labeling result of the abnormal value of the logging curve is obtained.
The logging curve abnormal value labeling method comprises the steps of reading a preset trained machine learning model set, selecting the optimal trained machine learning model under the current tuning standard from the set, and inputting a logging curve to be labeled to the selected trained machine learning model to obtain a logging curve abnormal value labeling result. In the whole process, the optimal trained machine learning model under the current optimization standard is selected, the abnormal value of the logging curve to be labeled is labeled through the selected trained machine learning model, manual abnormal value labeling is not needed, and the abnormal value labeling of the logging curve can be efficiently and accurately achieved.
As shown in fig. 3, in one embodiment, before step S400, the method further includes:
s310: and acquiring a sample logging curve carrying abnormal value labels.
The sample logging curve refers to a logging curve with completed abnormal value marking, and may be a logging curve with marked abnormal values recently obtained from a historical record, or a logging curve with marked abnormal values and uploaded manually by a professional technician in real time.
S320: and randomly dividing the sample data into training sample data and test sample data.
The obtained sample well logging curve is randomly divided into training sample data and test sample data, generally, the data size of the training sample data is larger than that of the test sample data, for example, the sample well logging curve can be divided into 10 parts on average, 8 parts of the sample well logging curve are used as the training sample data, and 2 parts of the sample well logging curve are used as the test sample data; for another example, 7 of the samples can be used as training sample data, and 3 of the samples can be used as test sample numbers.
S330: and training the plurality of initial machine learning models respectively according to the training sample data to obtain a plurality of trained machine learning models.
Training a plurality of initial machine learning models by taking training sample data as training data to obtain a plurality of trained machine learning models, wherein the initial machine learning models comprise initial models based on 10 regression algorithms such as Huber regressor, AdaBoostregressor, Lasso, Ridge, SGDRegresor, LinearSVR, SVR, Desision TreeRegessor, BaggingRegessor, GradientBoostRegessor and RandomForest Regessor, and the training sample data is adopted to respectively train the 10 initial machine learning models to finally obtain a plurality of (10) trained machine learning models.
S340: and respectively testing the trained machine learning models according to the test sample data to obtain model test results.
S350: and according to the model test result, obtaining the optimal trained machine learning model under different tuning standards, and constructing a preset trained machine learning model set.
Selecting the test sample data to test a plurality of trained machine learning models, calculating the error between the predicted value of each trained machine learning model and the true value in the test sample data, performing reference calculation on the error values by adopting different tuning standards, selecting the optimal trained machine learning model under the current tuning standard, finally obtaining the optimal trained machine learning model under the different tuning standards, and constructing a preset trained machine learning model set. Taking the tuning standard of mean _ absolute _ error as an example, calculating the errors obtained by processing each group of test sample data by a plurality of trained machine learning models, calculating the average absolute value errors of the errors, and selecting the corresponding trained machine learning model with the minimum average absolute value error as the optimal trained machine learning model with the tuning standard of mean _ absolute _ error.
In one embodiment, obtaining a sample log carrying an outlier label comprises:
acquiring an original logging curve; carrying out depth correction on the original logging curve to obtain the original logging curve after the depth correction; and obtaining a sample logging curve carrying the abnormal value mark by performing the abnormal value mark on the original logging curve after the depth correction by a third party.
Because the original logging curve is influenced by the self weight and tension change of the instrument in the field acquisition process, different instruments have difference on depth display of the same stratum during data acquisition, and the original logging curve needs to be calibrated so as to ensure that all curves have corresponding response at the same depth. In practical application, the server may send the original logging curve after depth correction to a terminal of a professional, the professional performs abnormal value labeling on the terminal, the terminal uploads a sample logging curve with the abnormal value labeling to the server, and the server obtains the sample logging curve with the abnormal value labeling.
In one embodiment, depth correcting the original well log to obtain a depth corrected original well log comprises:
randomly selecting any one of the original logging curves as a reference curve; comparing other curves in the original logging curve with the reference curve to obtain the depth errors of the other curves in the original logging curve and the reference curve; and (4) correcting depths of other curves in the original logging curve according to the depth error to obtain the original logging curve after depth correction.
The process of logging curve depth correction can specifically refer to the example shown in fig. 4, taking fig. 4 as an example, first selecting a curve as a reference (natural gamma GR curve in fig. 4); the other curves are compared with the reference curve, and if there is an error in depth, the depth of the erroneous curve is added or subtracted by a corresponding number to match the curves in depth.
In one embodiment, obtaining the optimal trained machine learning models under different tuning standards according to the model test result, and constructing a preset trained machine learning model set includes:
according to the model test result, respectively carrying out model quality sequencing on the trained machine learning model under different tuning standards to respectively obtain a plurality of sequencing sequences and record corresponding tuning standards; and selecting the optimal trained machine learning model in the plurality of sequencing sequences to be matched with the corresponding tuning standard, and constructing a preset trained machine learning model set.
According to the model test result, different tuning standards are adopted to perform model goodness and badness sequencing on the trained machine learning models, namely different evaluation scales are adopted to evaluate the trained machine learning models to obtain a trained machine learning model goodness and badness sequencing sequence corresponding to the different tuning standards, the sequencing sequence can be arranged from goodness to badness under the current tuning standard, the optimal trained machine learning model is selected, the optimal trained machine learning model is matched with the corresponding tuning standard, and a preset trained machine learning model set is constructed. In other words, the trained machine learning models that are optimal under a plurality of different tuning criteria are stored in the preset set of trained machine learning models.
In one embodiment, inputting the well log to be labeled to the selected trained machine learning model comprises: carrying out depth correction on the well logging curve to be marked; and inputting the logging curve to be marked after the depth correction into the selected trained machine learning model.
For the logging curve to be marked, the depth error may exist, so that the logging curve to be marked needs to be subjected to depth correction in the same manner, and then the logging curve to be marked after the depth correction is input into the selected trained machine learning model, so that the accuracy of marking the abnormal value of the logging curve is improved.
It should be understood that although the various steps in the flow charts of fig. 2-3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-3 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.
As shown in fig. 5, an apparatus for labeling abnormal value of well log includes:
the data acquisition module 200 is used for acquiring a logging curve to be marked and an optimization standard;
the reading module 400 is configured to read a preset trained machine learning model set, where the trained machine learning model set includes a plurality of trained machine learning models, and the trained machine learning models are obtained by training sample logging curves carrying abnormal value labels;
the model selection module 600 is configured to select a trained machine learning model that is optimal under the tuning standard from the trained machine learning model set;
and the abnormal value labeling module 800 is used for inputting the logging curve to be labeled into the selected trained machine learning model to obtain a logging curve abnormal value labeling result.
The logging curve abnormal value labeling device reads a preset trained machine learning model set, selects the optimal trained machine learning model under the current tuning standard from the set, and inputs the logging curve to be labeled to the selected trained machine learning model to obtain the logging curve abnormal value labeling result. In the whole process, the optimal trained machine learning model under the current optimization standard is selected, the abnormal value of the logging curve to be labeled is labeled through the selected trained machine learning model, manual abnormal value labeling is not needed, and the abnormal value labeling of the logging curve can be efficiently and accurately achieved.
In one embodiment, the logging curve abnormal value labeling device comprises a model construction module, a model identification module and a model identification module, wherein the model construction module is used for acquiring a sample logging curve carrying abnormal value labeling; randomly dividing sample data into training sample data and test sample data; training the plurality of initial machine learning models respectively according to training sample data to obtain a plurality of trained machine learning models; respectively testing the trained machine learning models according to the test sample data to obtain model test results; and according to the model test result, obtaining the optimal trained machine learning model under different tuning standards, and constructing a preset trained machine learning model set.
In one embodiment, the model building module is further configured to obtain a raw log; carrying out depth correction on the original logging curve to obtain the original logging curve after the depth correction; and obtaining a sample logging curve carrying the abnormal value mark by performing the abnormal value mark on the original logging curve after the depth correction by a third party.
In one embodiment, the model construction module is further configured to randomly select any one of the original well logging curves as a reference curve; comparing other curves in the original logging curve with the reference curve to obtain the depth errors of the other curves in the original logging curve and the reference curve; and (4) correcting depths of other curves in the original logging curve according to the depth error to obtain the original logging curve after depth correction.
In one embodiment, the model construction module is further configured to perform model goodness sorting on the trained machine learning model under different tuning standards according to the model test result, respectively obtain a plurality of sorting sequences, and record corresponding tuning standards; and selecting the optimal trained machine learning model in the plurality of sequencing sequences to be matched with the corresponding tuning standard, and constructing a preset trained machine learning model set.
In one embodiment, the outlier labeling module 800 is further configured to input the well log to be labeled to the selected trained machine learning model; acquiring output data of the selected trained machine learning model to obtain an abnormal value and a depth corresponding to the abnormal value; and according to the depth corresponding to the abnormal value, the abnormal value is marked to the logging curve to be marked, and the marking result of the abnormal value of the logging curve is obtained.
In one embodiment, the outlier labeling module 800 performs depth correction on the log to be labeled; and inputting the logging curve to be marked after the depth correction into the selected trained machine learning model.
For the specific definition of the logging curve abnormal value labeling device, reference may be made to the above definition of the logging curve abnormal value labeling method, and details are not described herein again. All or part of the modules in the logging curve abnormal value labeling device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
The logging curve abnormal value labeling device comprises a processor and a memory, wherein the data acquisition module 200, the reading module 400, the model selection module 600 and the abnormal value labeling module 800 are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions. The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. One or more than one kernel can be set, and logging curve abnormal value marking is realized by adjusting kernel parameters.
The embodiment of the application provides a storage medium, wherein a program is stored on the storage medium, and when the program is executed by a processor, the abnormal value labeling method of the logging curve is realized.
The embodiment of the application provides a processor which is used for running a program, wherein the program runs to execute the logging curve abnormal value labeling method.
As shown in fig. 6, the embodiment of the present application provides an apparatus 70, which includes at least one processor 701, at least one memory 702 connected to the processor 701, and a bus 703; the processor 701 and the memory 702 complete mutual communication through a bus 703; the processor 701 is configured to call program instructions in the memory 702 to perform the above-mentioned logging curve outlier labeling method. The device herein may be a server, a PC, a PAD, a mobile phone, etc.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device:
acquiring a logging curve to be marked and an optimization standard;
reading a preset trained machine learning model set, wherein the trained machine learning model set comprises a plurality of trained machine learning models, and the trained machine learning models are obtained by training sample well logging curves carrying abnormal value labels;
selecting an optimal trained machine learning model under the tuning standard from the trained machine learning model set;
and inputting the logging curve to be labeled into the selected trained machine learning model to obtain the logging curve abnormal value labeling result.
In an embodiment, the computer program product, when being executed on a data processing device, is adapted to carry out a procedure for initializing the following method steps:
acquiring a sample logging curve carrying abnormal value labels; randomly dividing sample data into training sample data and test sample data; training the plurality of initial machine learning models respectively according to training sample data to obtain a plurality of trained machine learning models; respectively testing the trained machine learning models according to the test sample data to obtain model test results; and according to the model test result, obtaining the optimal trained machine learning model under different tuning standards, and constructing a preset trained machine learning model set.
In an embodiment, the computer program product, when being executed on a data processing device, is adapted to carry out a procedure for initializing the following method steps:
acquiring an original logging curve; carrying out depth correction on the original logging curve to obtain the original logging curve after the depth correction; and obtaining a sample logging curve carrying the abnormal value mark by performing the abnormal value mark on the original logging curve after the depth correction by a third party.
In an embodiment, the computer program product, when being executed on a data processing device, is adapted to carry out a procedure for initializing the following method steps:
randomly selecting any one of the original logging curves as a reference curve; comparing other curves in the original logging curve with the reference curve to obtain the depth errors of the other curves in the original logging curve and the reference curve; and (4) correcting depths of other curves in the original logging curve according to the depth error to obtain the original logging curve after depth correction.
In an embodiment, the computer program product, when being executed on a data processing device, is adapted to carry out a procedure for initializing the following method steps:
according to the model test result, respectively carrying out model quality sequencing on the trained machine learning model under different tuning standards to respectively obtain a plurality of sequencing sequences and record corresponding tuning standards; and selecting the optimal trained machine learning model in the plurality of sequencing sequences to be matched with the corresponding tuning standard, and constructing a preset trained machine learning model set.
In an embodiment, the computer program product, when being executed on a data processing device, is adapted to carry out a procedure for initializing the following method steps:
inputting the logging curve to be marked into the selected trained machine learning model; acquiring output data of the selected trained machine learning model to obtain an abnormal value and a depth corresponding to the abnormal value; and according to the depth corresponding to the abnormal value, the abnormal value is marked to the logging curve to be marked, and the marking result of the abnormal value of the logging curve is obtained.
In an embodiment, the computer program product, when being executed on a data processing device, is adapted to carry out a procedure for initializing the following method steps:
carrying out depth correction on the well logging curve to be marked; and inputting the logging curve to be marked after the depth correction into the selected trained machine learning model.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a device includes one or more processors (CPUs), memory, and a bus. The device may also include input/output interfaces, network interfaces, and the like.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method of logging curve outlier labeling, the method comprising:
acquiring a logging curve to be marked and an optimization standard;
reading a preset trained machine learning model set, wherein the trained machine learning model set comprises a plurality of trained machine learning models, and the trained machine learning models are obtained by training sample well logging curves carrying abnormal value labels;
selecting the optimal trained machine learning model under the tuning standard from the trained machine learning model set;
and inputting the logging curve to be labeled into the selected trained machine learning model to obtain a logging curve abnormal value labeling result.
2. The method of claim 1, wherein prior to reading the preset set of trained machine learning models, further comprising:
acquiring a sample logging curve carrying abnormal value labels;
randomly dividing the sample data into training sample data and test sample data;
training a plurality of initial machine learning models respectively according to the training sample data to obtain a plurality of trained machine learning models;
respectively testing the plurality of trained machine learning models according to the test sample data to obtain model test results;
and obtaining the optimal trained machine learning models under different tuning standards according to the model test results, and constructing a preset trained machine learning model set.
3. The method of claim 2, wherein obtaining a sample log carrying an outlier label comprises:
acquiring an original logging curve;
carrying out depth correction on the original logging curve to obtain an original logging curve after the depth correction;
and obtaining a sample logging curve carrying abnormal value labels by a third party for carrying out abnormal value labeling on the original logging curve after the depth correction.
4. The method of claim 3, wherein depth correcting the raw well log to obtain a depth corrected raw well log comprises:
randomly selecting any one of the original logging curves as a reference curve;
comparing other curves in the original logging curve with the reference curve to obtain depth errors of the other curves in the original logging curve and the reference curve;
and performing depth correction on other curves in the original logging curve according to the depth error to obtain the original logging curve after depth correction.
5. The method according to claim 2, wherein the obtaining of the optimal trained machine learning models under different tuning standards according to the model test results and the constructing of the preset set of trained machine learning models comprises:
according to the model test result, respectively carrying out model quality sequencing on the trained machine learning model under different tuning standards to respectively obtain a plurality of sequencing sequences and record corresponding tuning standards;
and selecting the optimal trained machine learning model in the plurality of sequencing sequences to be matched with the corresponding tuning standard, and constructing a preset trained machine learning model set.
6. The method of claim 1, wherein inputting the well log to be labeled into the selected trained machine learning model to obtain a well log outlier labeling result comprises:
inputting the logging curve to be marked into the selected trained machine learning model;
acquiring output data of the selected trained machine learning model to obtain an abnormal value and a depth corresponding to the abnormal value;
and according to the depth corresponding to the abnormal value, the abnormal value is marked to the logging curve to be marked, and a logging curve abnormal value marking result is obtained.
7. The method of claim 1, wherein inputting the well log to be labeled to the selected trained machine learning model comprises:
carrying out depth correction on the well logging curve to be marked;
and inputting the logging curve to be marked after the depth correction into the selected trained machine learning model.
8. An apparatus for labeling outliers in a well log, said apparatus comprising:
the data acquisition module is used for acquiring a logging curve to be marked and an optimization standard;
the reading module is used for reading a preset trained machine learning model set, the trained machine learning model set comprises a plurality of trained machine learning models, and the trained machine learning models are obtained by training sample logging curves carrying abnormal value labels;
the model selection module is used for selecting the optimal trained machine learning model under the tuning standard from the trained machine learning model set;
and the abnormal value marking module is used for inputting the logging curve to be marked to the selected trained machine learning model to obtain a logging curve abnormal value marking result.
9. An electronic device comprising at least one processor, and at least one memory, bus connected to the processor; the processor and the memory complete mutual communication through the bus; the processor is used for calling the program instructions in the memory to execute the logging curve abnormal value labeling method according to any one of claims 1 to 7.
10. A storage medium comprising a stored program, wherein the program when executed controls an apparatus on the storage medium to implement the method for labeling abnormal values of a well logging curve of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911380142.4A CN113052195A (en) | 2019-12-27 | 2019-12-27 | Logging curve abnormal value marking method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911380142.4A CN113052195A (en) | 2019-12-27 | 2019-12-27 | Logging curve abnormal value marking method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113052195A true CN113052195A (en) | 2021-06-29 |
Family
ID=76506651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911380142.4A Pending CN113052195A (en) | 2019-12-27 | 2019-12-27 | Logging curve abnormal value marking method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113052195A (en) |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2636400A1 (en) * | 2007-12-10 | 2009-06-10 | Tellabs Vienna, Inc. | Identifying communication errors caused by the exchange of unsupported messages among devices in a network |
| CN104278990A (en) * | 2013-07-02 | 2015-01-14 | 中国石油天然气集团公司 | Logging data quality recovery method and device |
| US20180247220A1 (en) * | 2017-02-28 | 2018-08-30 | International Business Machines Corporation | Detecting data anomalies |
| CN108615230A (en) * | 2018-03-16 | 2018-10-02 | 北京邮电大学 | A kind of hub surface method for detecting abnormality and system |
| CN109242041A (en) * | 2018-09-28 | 2019-01-18 | 南方电网科学研究院有限责任公司 | A kind of electric energy meter abnormal deviation data examination method, device, equipment and storage medium |
| CN109299156A (en) * | 2018-08-21 | 2019-02-01 | 平安科技（深圳）有限公司 | Electronic device, the electric power data predicting abnormality method based on XGBoost and storage medium |
| US20190072942A1 (en) * | 2017-09-01 | 2019-03-07 | Doosan Heavy Industries & Construction Co., Ltd. | Method of generating plant normal state prediction data and apparatus using the same |
| CN109840312A (en) * | 2019-01-22 | 2019-06-04 | 新奥数能科技有限公司 | A kind of rejecting outliers method and apparatus of boiler load factor-efficiency curve |
| CN109902832A (en) * | 2018-11-28 | 2019-06-18 | 华为技术有限公司 | Training method, predicting abnormality method and the relevant apparatus of machine learning model |
| CN109919184A (en) * | 2019-01-28 | 2019-06-21 | 中国石油大学(北京) | A kind of more well complex lithology intelligent identification Methods and system based on log data |
| CN109978172A (en) * | 2017-12-27 | 2019-07-05 | 中国移动通信集团设计院有限公司 | A kind of resource pool usage forecast method and device based on extreme learning machine |
| CN110324316A (en) * | 2019-05-31 | 2019-10-11 | 河南恩湃高科集团有限公司 | A kind of industry control anomaly detection method based on a variety of machine learning algorithms |
| CN110334317A (en) * | 2019-07-11 | 2019-10-15 | 东北石油大学 | A kind of picture type borehole log data extraction method |
| CN110363231A (en) * | 2019-06-27 | 2019-10-22 | 平安科技（深圳）有限公司 | Abnormality recognition method, device and storage medium based on semi-supervised deep learning |
| CN110378386A (en) * | 2019-06-20 | 2019-10-25 | 平安科技（深圳）有限公司 | Based on unmarked abnormality recognition method, device and the storage medium for having supervision |
| CN110379463A (en) * | 2019-06-05 | 2019-10-25 | 山东大学 | Marine algae genetic analysis and concentration prediction method and system based on machine learning |
| CN110555514A (en) * | 2019-08-20 | 2019-12-10 | 北京迈格威科技有限公司 | Neural network model searching method, image identification method and device |
-
2019
- 2019-12-27 CN CN201911380142.4A patent/CN113052195A/en active Pending
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2636400A1 (en) * | 2007-12-10 | 2009-06-10 | Tellabs Vienna, Inc. | Identifying communication errors caused by the exchange of unsupported messages among devices in a network |
| CN104278990A (en) * | 2013-07-02 | 2015-01-14 | 中国石油天然气集团公司 | Logging data quality recovery method and device |
| US20180247220A1 (en) * | 2017-02-28 | 2018-08-30 | International Business Machines Corporation | Detecting data anomalies |
| US20190072942A1 (en) * | 2017-09-01 | 2019-03-07 | Doosan Heavy Industries & Construction Co., Ltd. | Method of generating plant normal state prediction data and apparatus using the same |
| CN109978172A (en) * | 2017-12-27 | 2019-07-05 | 中国移动通信集团设计院有限公司 | A kind of resource pool usage forecast method and device based on extreme learning machine |
| CN108615230A (en) * | 2018-03-16 | 2018-10-02 | 北京邮电大学 | A kind of hub surface method for detecting abnormality and system |
| CN109299156A (en) * | 2018-08-21 | 2019-02-01 | 平安科技（深圳）有限公司 | Electronic device, the electric power data predicting abnormality method based on XGBoost and storage medium |
| CN109242041A (en) * | 2018-09-28 | 2019-01-18 | 南方电网科学研究院有限责任公司 | A kind of electric energy meter abnormal deviation data examination method, device, equipment and storage medium |
| CN109902832A (en) * | 2018-11-28 | 2019-06-18 | 华为技术有限公司 | Training method, predicting abnormality method and the relevant apparatus of machine learning model |
| CN109840312A (en) * | 2019-01-22 | 2019-06-04 | 新奥数能科技有限公司 | A kind of rejecting outliers method and apparatus of boiler load factor-efficiency curve |
| CN109919184A (en) * | 2019-01-28 | 2019-06-21 | 中国石油大学(北京) | A kind of more well complex lithology intelligent identification Methods and system based on log data |
| CN110324316A (en) * | 2019-05-31 | 2019-10-11 | 河南恩湃高科集团有限公司 | A kind of industry control anomaly detection method based on a variety of machine learning algorithms |
| CN110379463A (en) * | 2019-06-05 | 2019-10-25 | 山东大学 | Marine algae genetic analysis and concentration prediction method and system based on machine learning |
| CN110378386A (en) * | 2019-06-20 | 2019-10-25 | 平安科技（深圳）有限公司 | Based on unmarked abnormality recognition method, device and the storage medium for having supervision |
| CN110363231A (en) * | 2019-06-27 | 2019-10-22 | 平安科技（深圳）有限公司 | Abnormality recognition method, device and storage medium based on semi-supervised deep learning |
| CN110334317A (en) * | 2019-07-11 | 2019-10-15 | 东北石油大学 | A kind of picture type borehole log data extraction method |
| CN110555514A (en) * | 2019-08-20 | 2019-12-10 | 北京迈格威科技有限公司 | Neural network model searching method, image identification method and device |
Non-Patent Citations (1)
| Title |
|---|
| 汤晓军 等: "基于Tikhonov 正则化特征光谱选择与最优网络参数选择的轻烷烃气体分析", 《光谱学与光谱分析》, vol. 31, no. 6, pages 1674 - 1677 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108538296A (en) | Speech recognition test method and test terminal | |
| US10783442B1 (en) | Demand forecasting via direct quantile loss optimization | |
| WO2015122575A1 (en) | Method for predicting personality trait and device therefor | |
| CN110750694A (en) | Data annotation implementation method and device, electronic equipment and storage medium | |
| CN110009614A (en) | Method and apparatus for output information | |
| CN107545301B (en) | Page display method and device | |
| CN111611797B (en) | Method, device and equipment for marking prediction data based on Albert model | |
| CN111898904B (en) | Data processing method and device | |
| CN108334484B (en) | Data entry method and device | |
| CN108921323A (en) | Method and apparatus for generating information | |
| CN110929764A (en) | Picture auditing method and device, electronic equipment and storage medium | |
| CN112241805A (en) | Defect prediction using historical inspection data | |
| CN112632179A (en) | Model construction method and device, storage medium and equipment | |
| CN108255654A (en) | Automatic testing method and device | |
| CN111966580A (en) | Automatic testing method, device, equipment and storage medium based on artificial intelligence | |
| US10769732B2 (en) | Expertise determination based on shared social media content | |
| CN114428677A (en) | Task processing method, processing device, electronic equipment and storage medium | |
| CN111340015A (en) | Positioning method and device | |
| CN113868498A (en) | Data storage method, electronic device, device and readable storage medium | |
| CN111402220B (en) | Method and device for acquiring information | |
| CN113052195A (en) | Logging curve abnormal value marking method and device, computer equipment and storage medium | |
| KR102406603B1 (en) | Razor blade automatic order device | |
| CN113645357B (en) | Call quality inspection method, device, computer equipment and computer readable storage medium | |
| US20220391727A1 (en) | Analysis apparatus, control method, and program | |
| CN114266239A (en) | Data set generation method and device |
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 |