A decision tree drilling condition identification method and device based on time series characteristics

Abstract

The present invention provides a method and device for identifying drilling conditions based on a decision tree of time series characteristics, including: constructing a first sample data set based on historical drilling data corresponding to characteristic parameters; calculating the time series mapping relationship between the characteristic parameters and the drilling conditions; calculating the time series characteristics of each sample data in the first sample data set based on the time series mapping relationship, and converting the first sample data set into a second sample data set based on the time series characteristics; and performing data mining on the second sample data set using a decision tree algorithm to obtain a decision tree for identifying drilling conditions. Based on data dimensionality reduction based on characteristic parameters, this method calculates the time series variation trend between characteristic parameters and conditions, thereby obtaining a decision tree for identifying conditions by mining sample data including time series characteristics. The decision tree algorithm has the advantages of high efficiency and accuracy.

Description

Technical Field

[0001] The invention relates to the field of oil drilling data processing, and in particular to a decision tree drilling condition identification method and device based on time series characteristics. Background Art

[0002] Drilling is an important means of oil and gas exploration and development. Oil drilling operations require a series of complex technologies. At the same time, as the drilling depth increases, geological changes will become more complex, and emergencies may occur at any time, seriously affecting the safety of workers and the efficiency of drilling. Therefore, analyzing the real-time data generated during the drilling process can take measures in advance for the upcoming dangers, provide guarantees for fast and safe drilling, and further reduce drilling costs and improve efficiency.

[0003] With the development of artificial intelligence technology, drilling condition identification methods are developing in the direction of intelligence. The existing intelligent drilling condition identification methods mainly include gray correlation analysis, Bayesian discriminant (Bayes), support vector machine (SVM), artificial neural network (ANN), decision tree, etc. These methods are mainly used to detect abnormal conditions. Compared with deep learning models, which require a large number of labeled samples, have a long training time and a slow running speed, decision trees have the advantages of being easy to understand and explain, easy to extract rules, and running fast. Its fast running speed is very suitable for projects such as drilling projects with a very large amount of monitoring data. When constructing a decision tree model related to drilling projects, the existing technology reduces the dimension of the original drilling data (selecting some parameters for analysis and identification), but the temporal nature of the sample data set is usually ignored in the analysis process. However, during the drilling process, the drilling data and drilling conditions are dynamically changing. Therefore, if the temporal characteristics of the data are ignored and the decision tree model is directly used to mine the original data samples, the accuracy of the final condition identification results will be affected.

[0004] For example: Chinese patent application number 2021102050556 discloses a drilling loss mechanism prediction and auxiliary decision-making method based on a fuzzy decision tree. It performs drilling loss analysis by selecting drilling parameters, but it does not involve analysis of the time series characteristics of drilling data. Summary of the invention

[0005] The purpose of the present invention is to overcome the defect in the prior art that the temporal sequence of the sample data set is usually ignored when constructing a decision tree model related to drilling engineering, which will affect the accuracy of the final working condition identification result. A decision tree drilling condition identification method and device based on temporal characteristics are provided, which realizes real-time and accurate judgment of working condition data through a decision tree algorithm based on temporal characteristics, and has the advantages of high efficiency and accuracy.

[0006] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0007] A decision tree drilling condition identification method based on time series characteristics, comprising:

[0008] Constructing a first sample data set based on historical drilling data corresponding to the characteristic parameters;

[0009] Calculating a time series mapping relationship between the characteristic parameter and the drilling condition; calculating a time series feature of each sample data in the first sample data set based on the time series mapping relationship, and converting the first sample data set into a second sample data set based on the time series feature;

[0010] A decision tree algorithm is used to perform data mining on the second sample data set to obtain a drilling condition identification decision tree.

[0011] According to a specific implementation, in the above-mentioned decision tree drilling condition identification method based on time series features, the first sample data set is established based on the historical drilling data corresponding to the characteristic parameters, including:

[0012] Original historical drilling data is obtained, data cleaning is performed on the original historical drilling data, invalid parameters other than the characteristic parameters in the original historical drilling data are filtered out, and the first sample data set is established based on the filtered historical drilling data.

[0013] According to a specific implementation, in the above-mentioned decision tree drilling condition identification method based on time series features, the calculating of the time series mapping relationship between the feature parameter and the drilling condition includes:

[0014] Select at least two time periods, and draw multiple time series variation diagrams of the characteristic parameter and the drilling condition in different time periods respectively;

[0015] A time-series mapping relationship between the characteristic parameter and the drilling condition is obtained based on the multiple time-series change diagrams.

[0016] According to a specific implementation, in the above-mentioned decision tree drilling condition identification method based on time series characteristics, the characteristic parameters include: standard well depth, drill bit position, hook load, top drive speed, vertical pressure, pump stroke;

[0017] The drilling conditions include: pulling out the drill, drilling down the drill, drilling, sliding drilling, connecting the drill bit, reaming, and circulation.

[0018] According to a specific embodiment, in the above-mentioned decision tree drilling condition identification method based on time series features, the time series features include: the difference with the standard well depth of the previous sample data, the difference with the hook load parameter of the sample data at a first preset time interval, the difference between the current standard well depth and the drill bit position, and the difference with the drill bit position parameter of the sample data at a second preset time interval.

[0019] According to a specific implementation, in the above-mentioned decision tree drilling condition identification method based on time series features, converting the first sample data set into a second sample data set based on the time series features includes:

[0020] The time series feature is added to each sample data of the first sample data set, and the characteristic parameters of standard well depth, hook load and drill bit position in each sample data are deleted.

[0021] According to a specific implementation, in the above-mentioned decision tree drilling condition identification method based on time series characteristics, the first preset time is 3 minutes and the second preset time is 8 minutes.

[0022] According to a specific implementation, in the above-mentioned decision tree drilling condition identification method based on time series characteristics, the decision tree algorithm is an ID3 algorithm.

[0023] Another aspect of the present invention provides a device for identifying drilling conditions based on a decision tree of timing features, comprising a processor, a network interface and a memory, wherein the processor, the network interface and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the above-mentioned method for identifying drilling conditions based on timing features by using a decision tree.

[0024] Compared with the prior art, the invention has the following beneficial effects:

[0025] The method provided by the embodiment of the present invention establishes a first sample data set through historical drilling data corresponding to characteristic parameters; at the same time, calculates the time series mapping relationship between the characteristic parameters and the drilling conditions; determines the time series characteristics of the sample based on the time series mapping relationship, thereby converting the first sample data set into a second sample data set; uses a decision tree algorithm to perform data mining on the second sample data set with time series characteristics, and obtains a drilling condition identification decision tree that can be used to identify the drilling conditions; on the basis of realizing data dimensionality reduction based on characteristic parameters, this method calculates the time series change trend of characteristic parameters and conditions, thereby using sample data including time series characteristics to mine to obtain a condition identification decision tree, and the decision tree algorithm has the advantages of high efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1This is a flow chart of a method for identifying drilling conditions based on a decision tree of time series features according to an embodiment of the present invention;

[0027] Figure 2 A time sequence mapping diagram of characteristic parameters and drilling conditions according to an embodiment of the present invention;

[0028] Figure 3 A time sequence mapping diagram of characteristic parameters and drilling conditions according to an embodiment of the present invention;

[0029] Figure 4 A time sequence mapping diagram of characteristic parameters and drilling conditions according to an embodiment of the present invention;

[0030] Figure 5 A time sequence mapping diagram of characteristic parameters and sliding drilling conditions according to an embodiment of the present invention;

[0031] Figure 6 A time sequence mapping diagram of characteristic parameters and drilling tool connection working conditions according to an embodiment of the present invention;

[0032] Figure 7 A time sequence mapping diagram of characteristic parameters and eye-ream working conditions according to an embodiment of the present invention;

[0033] Figure 8 A timing mapping diagram of characteristic parameters and cycle conditions according to an embodiment of the present invention;

[0034] Fig. 9 A first schematic diagram of a decision tree drilling condition identification model according to an embodiment of the present invention;

[0035] Fig.10 A second schematic diagram of the decision tree drilling condition identification model according to an embodiment of the present invention;

[0036] Fig.11 It is a schematic diagram of the structure of a decision tree drilling condition identification device based on time series features according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The present invention is further described in detail below in conjunction with test examples and specific implementation methods. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0038] Example 1

[0039] Figure 1 A decision tree drilling condition identification method based on time series features according to an exemplary embodiment of the present invention is shown, comprising:

[0040] Constructing a first sample data set based on historical drilling data corresponding to the characteristic parameters;

[0041] Calculating a time series mapping relationship between the characteristic parameter and the drilling condition; calculating a time series feature of each sample data in the first sample data set based on the time series mapping relationship, and converting the first sample data set into a second sample data set based on the time series feature;

[0042] A decision tree algorithm is used to perform data mining on the second sample data set to obtain a drilling condition identification decision tree.

[0043] This embodiment reduces the complexity of operation by establishing data samples based on characteristic parameters that are highly correlated with the working conditions. Each working condition is judged using characteristic parameters, and the judgment basis is jointly determined based on the time series mapping characteristics of the data and the attribute information between each working condition and the parameters, thereby improving the accuracy of the judgment. Compared with deep learning, the decision tree algorithm can realize real-time and accurate judgment of each input data, and has the advantages of high efficiency and accuracy.

[0044] Example 2

[0045] In one possible implementation, the method further includes: before constructing the first sample data set based on the historical drilling data corresponding to the characteristic parameters, performing a correlation analysis between the drilling engineering parameters and the drilling conditions, and selecting multiple characteristic parameters that are most relevant to the conditions from more than 70 drilling engineering parameters.

[0046] In a possible implementation, the above-mentioned selecting multiple characteristic parameters most relevant to the working condition from more than 70 drilling engineering parameters includes: analyzing the working condition, and selecting multiple characteristic parameters related to the working condition from the drilling engineering according to the definition of the working condition itself, wherein the working condition definition and its related parameters are shown in Table 1.

[0047] Table 1 Related parameters and their functions

[0048]

[0049] Based on this, multiple characteristic parameters most relevant to the working conditions can be selected from more than 70 drilling engineering parameters according to the above table, and 8 characteristic parameters can be further optimized from the multiple characteristic parameters: standard well depth (BZJS), drill bit position (ZTWZ), hook load (DGFH), top drive speed (DQZS), standing pressure (LY), and pump stroke 123 (BC1, 2, 3).

[0050] In this embodiment, the characteristic parameters with the highest correlation are found in combination with the definition of the working condition itself, so as to reduce the dimension of the data and simplify the input, effectively reduce the amount of data processing, and perform efficient and accurate identification of the working condition based on accurate characteristic parameters.

[0051] In a possible implementation, the establishing a first sample data set based on the characteristic parameters and the historical drilling data includes:

[0052] Original historical drilling data is obtained, data cleaning is performed on the historical drilling data, invalid parameters other than the characteristic parameters in the historical drilling data are filtered out, and the first sample data set is established based on the historical drilling data corresponding to the filtered characteristic parameters.

[0053] In this embodiment, after parameter dimension reduction is performed, invalid parameters other than the characteristic parameters in the historical drilling data are filtered out through data cleaning, so that a clear sample data set can be established.

[0054] Taking the real drilling data of 16 wells in the shale gas block of the Sichuan Basin as an example, according to the results collected by the comprehensive logging instrument, a total of 74 parameter values ​​were collected, with a collection frequency of 12 times per minute. On average, one well has 3 million rows of data, and the total data volume is more than 100G. The collected raw data can be subjected to the above-mentioned dimensionality reduction and data cleaning to obtain the first sample data, wherein Table 2 shows part of the first sample data cleaned by the embodiment of the present invention.

[0055] Table 2 Data after cleaning

[0056] BYZGR ZGar DGF QUR LY BC1 BC2 BC3 2138.9 2138.9 1366.27 72.01 2 0 0 0 2138.91 2138.91 1366 72.01 2 0 0 0 2138.92 2138.92 1365.07 72.01 2 0 0 0 2138.93 2138.93 1364.67 72.01 2 0 0 0 2138.94 2138.94 1364.67 72.01 2 0 0 0 2138.95 2138.95 1364.67 72.01 2 0 0 0 2138.97 2138.97 1366.27 72.01 2 0 0 0 2138.98 2138.98 1368.17 72.01 2 0 0 0 2138.99 2138.99 1368.17 72.09 2 0 0 0 2139.02 2139.02 1365.87 71.92 2 0 0 0 2139.03 2139.03 1363.66 72.01 2 0 0 0 2139.04 2139.04 1361.66 72.01 2 0 0 0 2139.07 2139.07 1357.05 72.01 2 0 0 0 2139.08 2139.08 1355.15 72.01 2 0 0 0 2139.1 2139.1 1354.65 72.01 2 0 0 0 2139.11 2139.11 1353.14 72.01 2 0 0 0 2139.13 2139.13 1352.24 72.01 2 0 0 0

[0057] Furthermore, a decision tree represents a mapping between object attributes and object values. Each node in the tree represents the judgment condition of the object attribute, its branches represent the objects that meet the node conditions, and the leaf nodes of the tree represent the prediction results to which the objects belong. Therefore, drawing a decision tree requires finding the relationship between object attributes and object values, that is, the relationship between working conditions and parameters.

[0058] In a possible implementation, in order to accurately summarize the temporal mapping relationship between various working conditions and engineering parameters, 316,842 pictures were drawn for each field in the 16 wells with the time period of year, month, week, and day, and the mapping relationship between the working conditions and the trend of engineering parameters was obtained as follows: Figures 2 to 8 As shown. Figures 2 to 8 The timing mapping relationship shown in Table 3 is summarized to determine the working condition.

[0059] Table 3 Working condition judgment method

[0060]

[0061] In this embodiment, by drawing a time series mapping diagram between characteristic parameters and working conditions for multiple time periods, the relationship between working conditions and parameters is accurately found, and an empirical formula is formed, thereby providing a basis for the accurate drawing of a decision tree.

[0062] Furthermore, drilling data is a kind of data about time series, and the decision tree itself cannot judge this kind of time series relationship. In order to retain the time series relationship between the data, it is necessary to integrate the time series information into a single data feature through feature engineering. Therefore, in a further embodiment of the present invention, according to the working condition judgment method described in Table 3, judging the working condition requires knowing the relationship between the standard well depth, hook load, and drill bit position of the current data and the previous time series data, and adding time series features (i.e., four fields representing time series information) to the cleaned data, as shown in Table 4, while deleting the three fields of standard well depth, hook load, and drill bit position to avoid data redundancy.

[0063] Table 4 Timing Information Field

[0064]

[0065] Finally, the second sample data shown in Table 5 is obtained, wherein the time sequence information of a single piece of data is added to each sample data.

[0066] Table 5 Final training data

[0067]

[0068]

[0069] Furthermore, the decision tree algorithm is an important data mining method with a simple working process, fast classification speed, easy generation of classification rules, and high accuracy of the mined classification rules. The decision tree algorithm treats the sample attribute as a node, builds a tree structure based on the value of the sample attribute, finds the changing rules of the problem from the sample data set, and mines valuable information for decision-making. There are many decision tree algorithms currently, such as ID3, C4.5, CART, etc.

[0070] In a possible implementation, the most classic ID3 algorithm is used as the decision tree algorithm of the embodiment of the present application, wherein the ID3 algorithm adopts a top-down recursive, divide-and-conquer learning principle to continuously recursively establish an ideal decision tree.

[0071] The set of data samples is U, which contains u samples, and the category attribute value is n. Then the information entropy is defined as formula (1).

[0072]

[0073] In the formula, p(u i ) represents sample u i Belongs to category C i probability.

[0074] U contains v subsets {U1,U2,…,U v}, the attribute A takes the value v j When v j Under the premise of appearance i The conditional probability of occurrence is calculated as formula (2).

[0075]

[0076] Conditional entropy represents the entropy of the subset divided by attribute A, and its specific definition is as shown in formula (3).

[0077]

[0078] Information increase can be expressed as formula (4).

[0079] Gain(A)=I(U)-I(U|V) (4)

[0080] The smaller the value of I(U|V), the larger the value of Gain(A), which means that the attribute A provides more information for classification. ID3, C4.5, and CART algorithms use the maximum information gain, the maximum information gain ratio, and the minimum Gini index as the optimal features, respectively, starting from the root node and recursively generating a decision tree.

[0081] (3) Constructing a working condition classification decision tree

[0082] The simplest ID3 algorithm is used to continuously select parameters with the largest information gain as nodes to create branches. The parameters here include time series parameters, so that each piece of data contains time series information, so that the current working condition category can be accurately determined with only one piece of data, which effectively improves the recognition efficiency.

[0083] Steps to create a decision tree:

[0084] Step 1: Treat all features as nodes.

[0085] Step 2: Traverse each segmentation method of the current feature and find the best segmentation point through information gain.

[0086] Step 3: Use the second step to traverse all features, select the best feature, and the best way to divide the feature, and get the final child nodes N1, N2...N m .

[0087] Step 4: For child nodes N1, N2…N m Continue with steps 2-3 respectively until each final child node is sufficiently "pure".

[0088] Finally, based on the second sample data shown in Table 5, after a first preset number of iterations based on the ID3 algorithm, the following can be obtained: Fig. 9After a preset number of iterations, the decision tree shown in Fig.10 The decision tree for drilling condition identification is shown.

[0089] At the same time, the model code written according to the decision tree can be used to annotate the input data. According to the label number setting of {'drilling': '0', 'starting drilling': '1', 'drilling': '2', 'sliding drilling': '3', 'ream': '4', 'drilling tool connection': '5', 'cycle': '6', 'other': '7'}, a "TAG" column is added at the end, and the recognition results are stored in this column. The recognition results of the decision tree model are shown in Table 6.

[0090] Table 6 Recognition results

[0091]

[0092] Furthermore, to ensure the accuracy of the decision tree, the recognition results were compared with the operating condition table corresponding to the original data. At the same time, some data were randomly selected for manual verification according to the judgment method of the operating conditions in Table 3. The results showed that there was almost no misclassified data, thus verifying the accuracy of the decision tree classification results.

[0093] Another aspect of the present invention, as Fig.11 As shown, a decision tree drilling condition identification device (electronic device) based on time series characteristics is also provided, which includes a processor, a network interface and a memory, wherein the processor, the network interface and the memory are interconnected, wherein the memory is used to store a computer program, and the computer program includes program instructions, and the processor is configured to call the program instructions to execute the above-mentioned data subscription method.

[0094] Another aspect of the present invention further provides a computer storage medium, wherein the computer storage medium stores program instructions, and when the program instructions are executed by at least one processor, they are used to implement the above-mentioned data subscription method.

[0095] In the embodiments of the present invention, the processor may be an integrated circuit chip having the ability to process signals. The processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0096] The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor for execution, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The processor reads the information in the storage medium and completes the steps of the above method in combination with its hardware.

[0097] The storage medium may be a memory, which may be, for example, a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.

[0098] Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.

[0099] The volatile memory may be a random access memory (RAM) which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDRSDRAM), enhanced synchronous DRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus DRAM (DRRAM).

[0100] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0101] It should be understood that the system disclosed in the present invention can be implemented in other ways. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the communication connection between modules can be through some interfaces, indirect coupling or communication connection of servers or units, which can be electrical or other forms.

[0102] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each module may exist physically separately, or two or more modules may be integrated into one processing unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional unit.

[0103] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A decision tree drilling condition identification method based on time series characteristics, characterized in that: include: Constructing a first sample data set based on historical drilling data corresponding to the characteristic parameters; Calculating the time series mapping relationship between the characteristic parameter and the drilling condition; comprising: Select at least two time periods, and draw multiple time series variation diagrams of the characteristic parameter and the drilling condition in different time periods respectively; Obtaining a time-series mapping relationship between the characteristic parameter and the drilling condition based on a plurality of the time-series change graphs; Calculating the time series features of each sample data in the first sample data set based on the time series mapping relationship, the time series features include: the difference with the standard well depth of the previous sample data, the difference with the big hook load parameter of the sample data separated by a first preset time, the difference between the current standard well depth and the drill bit position, and the difference with the drill bit position parameter of the sample data separated by a second preset time; converting the first sample data set into the second sample data set based on the time series features; A decision tree algorithm is used to perform data mining on the second sample data set to obtain a drilling condition identification decision tree; the decision tree algorithm is an ID3 algorithm, and the decision tree creation steps are: Step 1: Treat all features as nodes one by one; Step 2: Traverse each segmentation method of the current feature and find the best segmentation point through information gain; Step 3: Use the second step to traverse all features, select the optimal feature, and the optimal partitioning method of the feature, and obtain the final child nodes N1, N2...Nm; Step 4: Continue to execute Step 2-Step 3 for child nodes N1, N2...Nm respectively until each final child node reaches the predetermined purity standard, and continuously select the parameters with the largest information gain as nodes to create branches. The parameters include timing parameters, so that each data contains timing information.

2. The decision tree drilling condition identification method based on time series characteristics according to claim 1 is characterized in that: Establishing a first sample data set based on the historical drilling data corresponding to the characteristic parameters includes: Original historical drilling data is acquired, data cleaning is performed on the original historical drilling data, invalid parameters other than the characteristic parameters in the original historical drilling data are filtered out, and the first sample data set is established based on the filtered historical drilling data.

3. The method for identifying drilling conditions based on a decision tree of time series features according to any one of claims 1 to 2, characterized in that: The characteristic parameters include: standard well depth, drill bit position, hook load, top drive speed, vertical pressure, and pump stroke; The drilling conditions include: pulling out the drill, drilling down the drill, drilling, sliding drilling, connecting the drill bit, reaming, and circulation.

4. The method for identifying drilling conditions based on a decision tree of time series characteristics according to claim 3 is characterized in that: The converting the first sample data set into a second sample data set based on the time series feature comprises: The time series feature is added to each sample data of the first sample data set, and the characteristic parameters of standard well depth, hook load and drill bit position in each sample data are deleted.

5. The method for identifying drilling conditions based on a decision tree of time series characteristics according to claim 4 is characterized in that: The first preset time is 3 minutes, and the second preset time is 8 minutes.

6. A decision tree drilling condition identification device based on time series characteristics, characterized in that: It includes a processor, a network interface and a memory, wherein the processor, the network interface and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the decision tree drilling condition identification method based on time series features as described in any one of claims 1-5.

Images