Method and system for extracting key parameters related to drill jamming

Abstract

The invention discloses a method and system for extracting key parameters related to drilling tool jamming. The method comprises the steps that a drilling tool jamming sample set and a normal sample set are collected from a logging instrument and preprocessed; according to the drill jamming sample set, calculating the self anomaly degree and the association anomaly degree of each parameter; and respectively comparing the self abnormity degree and the association abnormity degree of each parameter with corresponding abnormity degree threshold values, and judging whether the current parameter is a drilling tool sticking association key parameter or not, so as to finish the judgment of all the parameters. According to the method, the abnormal change trend of the parameters in the same period of time and the abnormal change of the correlation degree between the parameters are referred at the same time, the method is not limited by the increase of parameter types, and effective utilization of the data is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil drilling engineering, and in particular, to a method and system for extracting key parameters related to stuck pipe. Background Art

[0002] Oil is one of the very important chemical energy sources. The exploitation of oil mainly relies on drilling technology. However, drilling engineering is an operation with high risks, multiple types of work, and multiple parameters affecting it. There are problems such as stuck pipe accidents that seriously affect the normal progress of drilling operations and also increase the cost of oil exploitation. Due to the concealment of drilling operations underground and the multi-dimensional and complex nature of drilling parameter types, it brings great difficulties for people to extract relevant parameters related to stuck pipe. At present, in engineering, the extraction of key related parameters for stuck pipe mainly relies on expert experience, and the extracted key parameters are only limited to traditional parameters such as hook load, drilling speed, torque, weight on bit, and mud pit volume.

[0003] However, with the increase in the types of sensors in engineering applications, the dimension of data in drilling engineering is getting higher and higher, and new parameters that map the occurrence of stuck pipe accidents are also generated. Improving the utilization rate of parameters and excavating more parameters closely related to stuck pipe accidents is of great significance for subsequent analysis of the causes of stuck pipe accidents and early warning of stuck pipe accidents.

[0004] In the prior art, the extraction of key parameters for stuck pipe mostly focuses on a single piece of data itself and ignores the changes in data over a long time series. Summary of the Invention

[0005] The purpose of the present invention is to provide a solution for extracting key parameters related to stuck pipe considering the changes in the long-time series state.

[0006] To solve the above technical problems, an embodiment of the present invention provides a method for extracting key parameters related to stuck pipe, including: collecting a stuck pipe sample set and a normal sample set from a logging instrument and performing preprocessing; calculating the self-abnormality degree and the associated abnormality degree of each parameter according to the stuck pipe sample set; comparing the self-abnormality degree and the associated abnormality degree of each parameter with the corresponding abnormality threshold respectively to determine whether the current parameter is a key parameter related to stuck pipe, thereby completing the determination of all parameters.

[0007] Preferably, the self-abnormality threshold and the associated abnormality threshold are calculated according to the following steps: according to the data sequence of each parameter in the normal sample set, according to the normal distribution criterion, the abnormality threshold of the data sequence of each parameter under the normal distribution is calculated, wherein the abnormality threshold λ=μ+3δ, μ is the average value of the current parameter, and δ is the standard deviation of the data sequence of the current parameter; according to the data sequence of each parameter in the normal sample set, the dynamic time bending distance between each parameter and other parameters is calculated, thereby obtaining the associated abnormality threshold of the parameter by calculating the average value of several dynamic time bending distances corresponding to each parameter.

[0008] Preferably, the stuck drill sample set and the normal sample set are preprocessed respectively according to the following steps: remove the columns in the sample set whose data is always 0 or remains unchanged; remove the abnormal point data in the sample set according to the Grubbs criterion; and calculate the correlation between any two parameters using the Spearman correlation analysis method according to each parameter in the sample set from which the abnormal points are removed, so that only one parameter with strong correlation is retained, wherein when the correlation coefficient is higher than a preset correlation threshold, it is determined that the current two parameters have a strong correlation, and the preset correlation threshold is 0.8.

[0009] Preferably, the process of collecting the stuck drill sample set and the normal sample set includes: extracting historical data under normal drilling conditions from the logging instrument based on daily drilling information to form the normal sample set, and extracting historical data within a specified time period before the stuck drill condition to form the stuck drill sample set.

[0010] Preferably, before the step of calculating the own abnormality and associated abnormality of each parameter according to the stuck drill sample set, the method further comprises: normalizing the preprocessed data in the stuck drill sample set; and normalizing the preprocessed data in the normal sample set.

[0011] Preferably, the process of calculating the self-abnormality and the associated abnormality of each parameter includes: calculating the self-abnormality of each data point in each parameter according to the data sequence of each parameter in the stuck drill sample set by using a local abnormality factor detection algorithm; calculating the dynamic time bending distance between each parameter and other parameters according to the data sequence of each parameter in the stuck drill sample set, so as to obtain the associated abnormality of the parameter by calculating the average value of several dynamic time bending distances corresponding to each parameter.

[0012] Preferably, in the process of calculating the self-anomaly degree and the associated anomaly degree of each parameter, it further includes: defining a preset time window and a unit step length, continuously moving the preset time window according to the unit step length to intercept multiple time periods from the sticking sample set, and calculating the self-anomaly degree and the associated anomaly degree of each parameter within each time period.

[0013] Preferably, in the process of calculating the self-anomaly degree of each data point, the value of k is determined according to the total number of data points of each parameter within the same time period.

[0014] Preferably, in the process of comparing the self-anomaly degree and the associated anomaly degree of each parameter with the corresponding anomaly degree threshold respectively to determine whether the current parameter is a key parameter associated with sticking, it includes: comparing the self-anomaly degree of each data point of the current parameter with the self-anomaly degree threshold of the corresponding parameter respectively. If there is a situation where a certain self-anomaly degree exceeds the corresponding threshold, then carry out the comparison of the associated degree and enter the next step; compare the associated anomaly degree of the current self-anomaly parameter with the associated anomaly degree threshold of the corresponding parameter. If the current associated anomaly degree exceeds the associated anomaly degree threshold of the corresponding parameter, then determine the current parameter as a key parameter associated with sticking.

[0015] On the other hand, an embodiment of the present invention provides a system for extracting key parameters associated with sticking, including: a sample set collection module configured to collect a sticking sample set and a normal sample set from a logging instrument and perform preprocessing; a data set processing module configured to calculate the self-anomaly degree and the associated anomaly degree of each parameter according to the sticking sample set; a key parameter extraction module configured to compare the self-anomaly degree and the associated anomaly degree of each parameter with the corresponding anomaly degree threshold respectively to determine whether the current parameter is a key parameter associated with sticking, so as to complete the judgment of all parameters.

[0016] Compared with the prior art, one or more embodiments of the above solution may have the following advantages or beneficial effects:

[0017] The present invention proposes a method and a system for extracting key parameters associated with sticking. For each column of parameters after preprocessing the data in the logging instrument, in the form of the established time window, the self-anomaly degree and the associated anomaly degree are calculated, so as to extract the key parameters associated with sticking at different stages before the occurrence of sticking, and obtain the key parameters associated with sticking that have not been discovered by previous experience. The present invention simultaneously refers to the self-anomaly change trend of the parameters within the same period of time and the abnormal change of the correlation degree between the parameters, is not limited by the increase in the types of parameters, realizes the effective utilization of the data, and provides strong support for subsequent sticking warning.

[0018] Other features and advantages of the present invention will be described in the following specification, and, in part, will be apparent from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained by the structures specifically pointed out in the specification, claims, and drawings. Description of the Drawings

[0019] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0020] Figure 1 It is a schematic diagram of the steps of the method for extracting key parameters related to stuck pipe in the embodiment of the present application.

[0021] Figure 2 It is a specific flowchart of the method for extracting key parameters related to stuck pipe in the embodiment of the present application.

[0022] Figure 3 It is a schematic diagram of the structure of the system for extracting key parameters related to stuck pipe in the embodiment of the present application. Detailed Description of the Embodiments

[0023] The following will describe in detail the embodiments of the present invention in combination with the drawings and embodiments, so as to fully understand how the present invention uses technical means to solve technical problems and the implementation process of achieving technical effects and implement accordingly. It should be noted that as long as there is no conflict, the various embodiments in the present invention and the various features in each embodiment can be combined with each other, and the formed technical solutions are all within the protection scope of the present invention.

[0024] In addition, the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0025] The terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms "a" and "an" used herein are also intended to include the plural. It should also be understood that the terms "comprises" and / or "comprising" used herein specify the presence of the stated features, integers, steps, operations, units, and / or components, and do not exclude the presence or addition of one or more other features, integers, steps, operations, units, components, and / or combinations thereof.

[0026] Petroleum is one of the very important chemical energy sources. The exploitation of petroleum mainly relies on drilling process technologies. However, drilling engineering is an operation with high risks, involving multiple types of work and affected by multiple parameters. There are problems such as stuck pipe accidents that seriously affect the normal progress of drilling operations and also increase the cost of petroleum exploitation. Due to the concealment of drilling operations underground, as well as the multi-dimensional and complex nature of drilling parameter types, it brings great difficulties for people to extract the correlation parameters related to stuck pipe. Currently, in engineering, the extraction of key correlation parameters for stuck pipe mainly relies on expert experience, and the extracted key parameters are only limited to traditional parameters such as hook load, drilling speed, torque, weight on bit, and mud pit volume.

[0027] However, with the increase in the types of sensors in engineering applications, the dimension of data in drilling engineering is getting higher and higher, and new parameters that map the occurrence of stuck pipe accidents are also generated. Improving the utilization rate of parameters and excavating more parameters closely related to stuck pipe accidents are of great significance for subsequent analysis of the causes of stuck pipe accidents and early warning of stuck pipe accidents.

[0028] In the prior art, the extraction of key parameters for stuck pipe mostly focuses on a single piece of data itself, while ignoring the changes in data over a long time series.

[0029] To solve the technical problems in the above background art, the embodiments of the present application propose a method and system for extracting key correlation parameters for stuck pipe. This method and system consider that the occurrence of stuck pipe not only depends on the abnormal changes of parameters at a single time point, but also depends on the abnormal change trends of parameters over a period of time and the abnormal changes in the correlation degrees between parameters. The present invention aims to analyze the data abnormality degree of a single piece of data within each time window and its correlation abnormality degree with other data by selecting a sliding time window, and then compare with two set abnormality thresholds to extract all key parameters related to stuck pipe.

[0030] Figure 1 It is a schematic diagram of the steps of the method for extracting key correlation parameters for stuck pipe according to the embodiments of the present application. Figure 2 It is a specific process schematic diagram of the method for extracting key correlation parameters for stuck pipe according to the embodiments of the present application. The following combines Figure 1 and Figure 2 to illustrate the specific step process of the method for extracting key correlation parameters for stuck pipe (also referred to as the "stuck pipe key parameter extraction method") described in the embodiments of the present invention.

[0031] Step S110 collects a stuck pipe sample set and a normal sample set from the mud logging instrument and performs preprocessing.

[0032] According to the drilling daily report information, historical logging data representing the normal drilling state is extracted from the logging instrument to form a normal sample set, and historical logging data representing a specified time period before the stuck pipe condition occurs is extracted from the logging instrument to form a stuck pipe sample set. Specifically, according to the stuck pipe time recorded in the drilling daily report information, historical logging data within a preset time period before the stuck pipe time point is extracted as the stuck pipe sample set. At the same time, historical logging data within a preset time period before normal drilling of the same well as the stuck pipe sample set is extracted as the normal sample set. For example, the preset time period is about 1 hour.

[0033] After generating the stuck pipe sample set and the normal sample set, step S110 also preprocesses these two sample sets respectively. It should be noted that in the embodiments of the present invention, the two sample sets are preprocessed in the same way.

[0034] In one embodiment, for both the stuck pipe sample set and the normal sample set, it is necessary to remove the invariant values that have no influence on the extraction of key correlation parameters, the measurement outliers caused by sensor measurement or other reasons, and remove the strongly correlated data obtained through correlation analysis (where only one item of the strongly correlated data needs to be retained), so as to realize the preprocessing of the sample data in the two sample sets.

[0035] Specifically, the data columns in each sample set that are always 0 or remain unchanged are removed respectively; according to the Grubbs criterion, the outlier data in each sample set after removing the data values with no influence is removed; according to each parameter in each sample set after removing the outliers, the Spearman correlation analysis method is used to calculate the correlation between any two parameters (in the same sample set), so as to retain only one parameter of the strongly correlated parameters. Among them, when the correlation coefficient is higher than the preset correlation threshold, it is determined that the current two parameters have strong correlation. The preset correlation threshold is 0.8.

[0036] Specifically, the columns in the normal sample set and the stuck pipe sample set that are always 0 or remain unchanged are removed first.

[0037] Then, the Grubbs criterion is used to remove the outlier data from the stuck pipe sample set and the normal sample set data respectively (the operation is repeated until there are no outliers). The Grubbs criterion is a method for removing outliers from samples close to a normal distribution, and its operation process is as follows:

[0038] Select the significance level α, which is the probability of misjudging the data as abnormal data using the Grubbs criterion;

[0039] Calculate where σ represents the standard deviation of the sample, and x n represents the nth data in the sample set;

[0040] Look up the table to obtain T according to the values of n and α 0 (n, α) value;

[0041] If T ≥ T 0 (n, α), then the data x n is abnormal data and should be excluded; otherwise, it is normal data and should be retained;

[0042] Repeat the operation until there is no abnormal data.

[0043] Finally, perform Spearman correlation analysis on the data in the stuck-pipe sample set and the normal sample set after excluding abnormal points respectively. For data columns with strong correlation, only retain one column. Use Spearman correlation analysis to draw the correlation heat map of all parameters. For parameters with strong correlation between the two, only retain one column and exclude the other column. Further, define that strongly correlated parameters refer to the correlation between two parameters being greater than 0.8.

[0044] To avoid errors in the operation process caused by differences in the order of magnitude of different parameters, it is necessary to perform normalization processing on the data in the two preprocessed sample sets respectively.

[0045] That is to say, before step S120, the stuck-pipe key parameter extraction method described in the embodiments of the present invention further includes: performing normalization processing on the data in the preprocessed stuck-pipe sample set; and performing normalization processing on the data in the preprocessed normal sample set.

[0046] Perform normalization processing on the data in each preprocessed sample set according to the following expressions:

[0047]

[0048] Among them, x k represents the k-th data in the sample set to be processed before normalization processing, x k ’ represents the normalization processing result of the k-th data in the sample set to be processed, x max , x min respectively represent the maximum and minimum values corresponding to each column parameter in the two sample sets.

[0049] Step S120 calculates the self-abnormality and associated abnormality of each parameter in the stuck-pipe sample set according to the preprocessed and normalized stuck-pipe sample set obtained in step S110. Among them, the self-abnormality represents the abnormality degree of each data point in the sample data set of each parameter; the associated abnormality represents the average value of the correlation degree between each parameter and other parameters.

[0050] In addition, before calculating the self-anomaly degree and the associated anomaly degree of each (mud logging) parameter in the embodiments of the present invention, it is also necessary to define a preset time window and a unit step length, and continuously move the preset (sliding) time window according to the unit step length to intercept multiple time periods (time series segments) from the stuck pipe sample set, and calculate the self-anomaly degree and the associated anomaly degree of each parameter within each time period.

[0051] The selection of the length of the preset sliding time window is directly related to the subsequent calculation of the associated anomaly degree between parameters. In the embodiments of the present invention, the length of the sliding time window should not be too long or too short. If it is too long, it is easy to cause the time series segments in which the associated anomaly degree between parameters changes to be included in one window, masking the change trend characteristics of the associated anomaly degree of the parameters; if it is too short, it will increase the unnecessary calculation amount.

[0052] In one embodiment, the determination of the time window length is closely related to the length of the stuck pipe sample set extracted. For example, for data with a stuck pipe sample set length of 1 hour, the sample set is usually divided into 10-12 segments, that is, the length of each time window is approximately 5-6 minutes.

[0053] Within each time window, it is necessary to calculate the self-correlation degree of each mud logging parameter within the corresponding time period: according to the data sequence of each parameter within each time period in the stuck pipe sample set, the local outlier factor detection algorithm is used to calculate the self-anomaly degree of each data point within the corresponding parameter within the corresponding time period.

[0054] The calculation of the parameter self-data anomaly degree is mainly based on the local outlier factor detection algorithm (LOF algorithm). The LOF algorithm is a density-based method. This algorithm characterizes the anomaly degree of each data point in the sample set by calculating the "local outlier factor". The larger the local outlier factor, the sparser the surrounding of the data point, and the greater the possibility that the data point is an outlier; on the contrary, the smaller the local outlier factor, the denser the surrounding of the data point, and the smaller the possibility that the data point is an outlier.

[0055] Further, define the distance between the points p(x 1 , y 1 , z 1 ) and q(x 2 , y 2 , z 2 ) in the feature space as At the same time, define the k-distance of the data point p for which the self-anomaly degree is to be calculated as: d k (p) = d(p, q), which satisfies:

[0056] There are at least k points o in the set that do not include the p point, such that d(p, o) ≤ d(p, q);

[0057] There are at most k - 1 points o in the set, excluding p, such that d(p, o) < d(p, q);

[0058] Then the k - th reachable distance is reach_dist k (p, q) = max{d k (p), d(p, q)};

[0059] Then the local reachability density of point p is where k(p) represents the k - nearest neighbor range of point p; then the local outlier factor of point q is

[0060] In one embodiment, the value of k in the local outlier factor detection algorithm is determined according to the total number of data points of each parameter within the same time period.

[0061] In the process of calculating the correlation anomaly degree of each mud logging parameter within each time period, according to the data sequence of each parameter corresponding to each time period in the stuck - pipe sample set, calculate the dynamic time warping distance between each (mud logging) parameter and other (mud logging) parameters (within the same time period), so as to obtain the correlation anomaly degree of each drilling parameter within that time period by calculating the average value of several dynamic time warping distances corresponding to each parameter.

[0062] Similarly, within each time window, calculate the correlation anomaly degree between the remaining parameters after data pre - processing and other parameters. The calculation of the correlation anomaly degree between parameters is mainly based on the dynamic time warping distance calculation method (DTW algorithm).

[0063] The DTW algorithm is used to calculate the dynamic time warping distance between two time series. For two time series X = <x 1 , x 2 , …, x n > and Y = <y 1 , y 2 , …, y n >, where the distance formula matrix between every two data points is: The cumulative distance matrix is: In the formula, D(n, n) is the dynamic time warping distance between two time series, and its calculation formula is:

[0064] Furthermore, within each time window, calculate the dynamic time warping distance between each mud logging parameter after pre - processing and other mud logging parameters, and take the average value of these dynamic time warping distances as the dynamic time warping distance of the corresponding mud logging parameter under the corresponding time window, which is the correlation anomaly degree between the mud logging parameter and other parameters.

[0065] In step S130, the self - abnormality degree and the associated abnormality degree of each logging parameter are respectively compared with the corresponding abnormality degree thresholds to determine whether the current parameter is a key parameter associated with stuck pipe, thus completing the determination of all parameters.

[0066] Before implementing step S130, first, according to the pre - processed normal sample set, the self - abnormality degree threshold and the associated abnormality degree threshold of each (logging) parameter are calculated respectively.

[0067] When calculating the self - abnormality degree threshold, it is considered that when the data volume is large enough, the characteristic distribution of the data points of each parameter in the normal sample set over time follows a normal distribution. According to the 3δ criterion of the normal distribution, the data within the interval (μ - 3δ, μ + 3δ) is normal data. Therefore, λ = μ + 3δ is selected as the self - abnormality degree threshold.

[0068] In one embodiment, according to the data sequence of each (logging) parameter in the pre - processed and normalized normal sample set, based on the normal distribution criterion, the abnormality degree threshold under the normal distribution of the data sequence of each (logging) parameter is calculated. Among them, the abnormality degree threshold λ is: λ = μ + 3δ. Where μ is the average value of the current parameter, and δ is the standard deviation of the data sequence of the current parameter.

[0069] When calculating the associated abnormality degree threshold, the normal sample set is selected as the calculation sample, and the dynamic time distance between each parameter and other parameters is calculated; similarly, the average value of the dynamic time warping distances of each parameter is taken as the abnormality degree threshold of the associated abnormality degree calculated for this parameter in the stuck - pipe sample set.

[0070] In one embodiment, according to the data sequence of each (logging) parameter in the pre - processed and normalized normal sample set, the dynamic time warping distance between each (logging) parameter and other parameters is calculated, so as to obtain the associated abnormality degree threshold of this (logging) parameter by calculating the average value of several dynamic time warping distances corresponding to each (logging) parameter.

[0071] In step S130, within each time window, the self - abnormality - related degree data of each data point within each logging parameter is respectively compared with the self - abnormality degree threshold of the corresponding logging parameter to determine whether there is a situation where the self - abnormality degree of a certain data point exceeds the self - abnormality degree threshold of the logging parameter to which it belongs during the current period. In one embodiment, if there is a situation where the self - abnormality degree of a certain data point exceeds the self - abnormality degree threshold of the corresponding parameter, then the correlation degree comparison is carried out, thus entering the next step to conduct the correlation degree comparison work for the current time window (period).

[0072] During the correlation degree comparison process of the corresponding time window, the correlation anomaly degree of each item (one or more) of its own abnormal mud logging parameters is compared with the correlation anomaly degree threshold of the corresponding item parameter to determine whether the correlation anomaly degree of the mud logging parameter being compared exceeds the correlation anomaly degree threshold of the corresponding mud logging parameter during the current period. In one embodiment, if the correlation anomaly degree of the current own abnormal mud logging parameter exceeds the correlation anomaly degree threshold of the corresponding parameter, the current mud logging parameter is determined as a key parameter associated with stuck pipe.

[0073] When, within each time window, the self-anomaly degree of a certain data point exceeds the self-anomaly degree threshold of the corresponding parameter, the correlation anomaly degree of the drilling parameter to which the data belongs is compared within the current time window to determine whether it exceeds the correlation anomaly degree threshold of the corresponding parameter. If the correlation anomaly degree of the mud logging parameter to which the self-abnormal data belongs within the window also exceeds the correlation anomaly degree threshold of the corresponding mud logging parameter, it can be determined that this mud logging parameter is a key parameter associated with stuck pipe within the window time. Thus, it is determined that during the period when the self-anomaly degree and the correlation anomaly degree of a certain mud logging parameter jointly exceed the corresponding thresholds, this mud logging parameter is a key parameter associated with stuck pipe.

[0074] In this way, after completing the comparison and evaluation of the self-anomaly degree and the correlation anomaly degree of each mud logging parameter in the stuck pipe sample set within each time window, the key parameters associated with stuck pipe within the corresponding time window are extracted.

[0075] Based on the above method for extracting key parameters related to stuck pipe, the present invention also provides a system for extracting key parameters related to stuck pipe (also referred to as the "system for extracting key parameters related to stuck pipe"). This system for extracting key parameters related to stuck pipe is used to implement the method for extracting key parameters related to stuck pipe as described above.

[0076] Figure 3 It is a schematic structural diagram of the system for extracting key parameters related to stuck pipe according to an embodiment of the present application. As Figure 3 shown, the system for extracting key parameters related to stuck pipe described in the embodiment of the present invention includes: a sample set collection module 301, a data set processing module 302, and a key parameter extraction module 303.

[0077] The sample set collection module 301 is implemented according to the method described in step S110 above, and is configured to collect the stuck pipe sample set and the normal sample set from the mud logger and perform preprocessing; the data set processing module 302 is implemented according to the method described in step S120 above, and is configured to calculate the self-anomaly degree and the correlation anomaly degree of each parameter according to the stuck pipe sample set; the key parameter extraction module 303 is implemented according to the method described in step S130 above, and is configured to compare the self-anomaly degree and the correlation anomaly degree of each parameter with the corresponding anomaly degree threshold respectively to determine whether the current parameter is a key parameter associated with stuck pipe, thereby completing the determination of all parameters.

[0078] The present invention discloses a method and a system for extracting key parameters related to stuck pipe. For each column of parameters after data preprocessing in a logging instrument, in the form of a set time window, the method and system calculate their own anomaly degree and associated anomaly degree, so as to extract key parameters related to stuck pipe at different stages before the occurrence of stuck pipe, and obtain key parameters related to stuck pipe that have not been discovered by past experience. The present invention also refers to the trend of its own anomaly change of parameters and the anomaly change of the correlation degree between parameters within the same period of time, is not limited by the increase in the type of parameters, realizes the effective utilization of data, and provides strong support for subsequent stuck pipe warning.

[0079] As mentioned above, only the preferred specific embodiments of the present invention are described, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0080] In the description of the present invention, unless otherwise specified, the meaning of "a plurality of" is two or more; the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0081] In the description of the present invention, it should be noted that unless otherwise clearly specified and limited, the terms "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0082] It should be understood that the embodiments disclosed by the present invention are not limited to the specific structures, processing steps or materials disclosed herein, but should extend to equivalent alternatives of these features understood by those of ordinary skill in the relevant art. It should also be understood that the terms used herein are only for the purpose of describing specific embodiments and do not mean to limit.

[0083] As used herein, the term "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the phrases "one embodiment" or "an embodiment" that appear throughout the specification do not necessarily all refer to the same embodiment.

[0084] The embodiments of the present invention are provided for purposes of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are chosen and described in order to best explain the principles of the invention and its practical application, and to enable those of ordinary skill in the art to understand the invention and design various embodiments adapted to particular uses with various modifications.

[0085] Although the embodiments disclosed in the present invention are as described above, the above content is only an embodiment adopted for the convenience of understanding the present invention, and is not intended to limit the present invention. Any person skilled in the art within the technical field to which the present invention pertains may make any modifications and variations in the form and details of the implementation without departing from the spirit and scope disclosed by the present invention. However, the scope of patent protection of the present invention shall still be subject to the scope defined by the appended claims.

Claims

1. A method for extracting key parameters related to stuck pipe, characterized in that, it includes: Collecting a stuck pipe sample set and a normal sample set from a logging instrument and performing preprocessing; Calculating the self-abnormality degree and the associated abnormality degree of each parameter according to the stuck pipe sample set; Comparing the self-abnormality degree and the associated abnormality degree of each parameter with the corresponding abnormality degree threshold respectively to determine whether the current parameter is a key parameter related to stuck pipe, thereby completing the judgment of all parameters.

2. The method according to claim 1, characterized in that, The self-abnormality degree threshold and the associated abnormality degree threshold are calculated according to the following step process: According to the data sequence of each parameter in the normal sample set, based on the normal distribution criterion, calculate the abnormality degree threshold under the normal distribution of the data sequence of each parameter, where the abnormality degree threshold λ = μ + 3δ, μ is the average value of the current parameter, and δ is the standard deviation of the data sequence of the current parameter; According to the data sequence of each parameter in the normal sample set, calculate the dynamic time warping distance between each parameter and other parameters, and thus obtain the associated abnormality degree threshold of each parameter by calculating the average value of several dynamic time warping distances corresponding to each parameter.

3. The method according to claim 1 or 2, characterized in that, The stuck pipe sample set and the normal sample set are preprocessed respectively according to the following step process: Removing the columns in the sample set where the data is always 0 or remains unchanged; Eliminating the outlier data in the sample set according to the Grubbs criterion; According to each parameter in the sample set after eliminating outliers, using the Spearman correlation analysis method to calculate the correlation between any two parameters, and thus only retaining one parameter among the parameters with strong correlation. Among them, when the correlation coefficient is higher than the preset correlation threshold, it is determined that the current two parameters have strong correlation, and the preset correlation threshold is 0.

8.

4. The method according to any one of claims 1 to 3, characterized in that, During the process of collecting the stuck pipe sample set and the normal sample set, it includes: Extracting the historical data under normal drilling conditions from the logging instrument according to the drilling daily report information to form the normal sample set, and extracting the historical data within a specified time period before the stuck pipe condition to form the stuck pipe sample set.

5. The method according to any one of claims 1 to 4, characterized in that, Before the step of calculating the self-abnormality degree and the associated abnormality degree of each parameter according to the stuck pipe sample set, the method further includes: Normalizing the data in the preprocessed stuck pipe sample set; and Normalizing the data in the preprocessed normal sample set.

6. The method according to any one of claims 1 to 5, characterized in that, During the process of calculating the self-abnormality degree and the associated abnormality degree of each parameter, it includes: According to the data sequence of each parameter in the stuck pipe sample set, using the local outlier factor detection algorithm to calculate the self-abnormality degree of each data point within each parameter; According to the data sequences of each parameter in the stuck-pipe sample set, calculate the dynamic time warping distances between each parameter and other parameters, so as to obtain the correlation anomaly degree of each parameter by calculating the average value of several dynamic time warping distances corresponding to each parameter.

7. The method according to claim 6, wherein, in the process of calculating the self-anomaly degree and the correlation anomaly degree of each parameter, it further includes: Define a preset time window and a unit step size, and continuously move the preset time window according to the unit step size to intercept multiple time periods from the stuck-pipe sample set, and calculate the self-anomaly degree and the correlation anomaly degree of each parameter within each time period.

8. The method according to claim 7, wherein, in the process of calculating the self-anomaly degree of each data point, the value of k is determined according to the total number of data points of each parameter within the same time period.

9. The method according to any one of claims 6 to 8, wherein, in the process of comparing the self-anomaly degree and the correlation anomaly degree of each parameter with the corresponding anomaly degree thresholds respectively to determine whether the current parameter is a key parameter related to stuck-pipe, it includes: Compare the self-anomaly degrees of each data point of the current parameter with the self-anomaly degree threshold of the corresponding parameter respectively. If there is a situation where a certain self-anomaly degree exceeds the corresponding threshold, then carry out the correlation degree comparison and enter the next step; Compare the correlation anomaly degree of the current self-anomaly parameter with the correlation anomaly degree threshold of the corresponding parameter. If the current correlation anomaly degree exceeds the correlation anomaly degree threshold of the corresponding parameter, then determine the current parameter as a key parameter related to stuck-pipe.

10. A system for extracting key parameters related to stuck-pipe, wherein, it includes: A sample set collection module configured to collect a stuck-pipe sample set and a normal sample set from a logging instrument and perform preprocessing; A data set processing module configured to calculate the self-anomaly degree and the correlation anomaly degree of each parameter according to the stuck-pipe sample set; A key parameter extraction module configured to compare the self-anomaly degree and the correlation anomaly degree of each parameter with the corresponding anomaly degree thresholds respectively to determine whether the current parameter is a key parameter related to stuck-pipe, so as to complete the judgment of all parameters.

Images