Oil gas parameter determination method and device

Abstract

The invention discloses an oil and gas parameter determination method and device. The method comprises the following steps: acquiring various historical oil and gas parameter data and a to-be-determined sample; the to-be-determined sample is real-time oil and gas parameter data with at least one missing data value; according to the to-be-determined sample, screening first oil and gas parameter data from the various historical oil and gas parameter data by using a correlation analysis algorithm, and determining the first oil and gas parameter data as a training sample; adding a label to the training sample according to the category of the data in the training sample; calculating the distance between the to-be-determined sample and the training sample added with the label by using a preset Gaussian kernel distance association relationship; determining a front preset number of training samples which are sequenced according to the distance from small to large as neighbor samples; and according to the neighbor sample, the to-be-determined sample and the preset Gaussian kernel standard deviation, determining the missing data value in the to-be-determined sample by using the KNN algorithm, so that the accuracy and efficiency of determining the missing oil and gas parameters can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas project geology and oil and gas reservoir evaluation, and in particular to a method and device for determining oil and gas parameters. Background Art

[0002] This section is intended to provide a background or context for embodiments of the present invention. No description herein is admitted to be prior art by virtue of its inclusion in this section.

[0003] When oil companies conduct oil and gas project transactions, they need to evaluate and assess the target blocks and target oil and gas fields. In the technical and economic evaluation process of new oil and gas projects, due to the different exploration levels of the blocks, incomplete oil and gas field data, inconsistent data formats, inconsistent standards, missing data items, and difficulty in data interpretation, etc. Among all the problems, the lack of parameters has a huge impact on the evaluation. For the missing parameters, it is currently mainly judged by the experience of petroleum geology and oil and gas reservoir engineering experts, which is not only time-consuming, but also highly deviates from the actual situation, and is greatly affected by subjective factors, making it difficult to meet the needs of efficient and accurate evaluation. Summary of the invention

[0004] An embodiment of the present invention provides a method for determining oil and gas parameters, which is used to improve the accuracy and efficiency of determining missing oil and gas parameters. The method includes:

[0005] Acquire a variety of historical oil and gas parameter data and samples to be determined; the samples to be determined are real-time oil and gas parameter data with at least one data value missing;

[0006] According to the sample to be determined, a correlation analysis algorithm is used to select first oil and gas parameter data from a plurality of historical oil and gas parameter data, and the first oil and gas parameter data is determined as a training sample; the first oil and gas parameter data is a plurality of oil and gas parameter data whose correlation with the sample to be determined is higher than a preset correlation threshold;

[0007] Add labels to training samples according to the categories of data in them;

[0008] Using the preset Gaussian kernel distance correlation relationship, calculate the distance between the sample to be determined and the training sample after adding labels;

[0009] Determine a preset number of training samples sorted in ascending order of distance as neighbor samples;

[0010] According to the neighbor samples, the samples to be determined and the preset Gaussian kernel standard deviation, the KNN (K Nearest Neighbors) algorithm is used to determine the missing data values ​​in the samples to be determined.

[0011] The embodiment of the present invention further provides an oil and gas parameter determination device for improving the accuracy and efficiency of determining missing oil and gas parameters, the device comprising:

[0012] An acquisition module is used to acquire a variety of historical oil and gas parameter data and samples to be determined; the samples to be determined are real-time oil and gas parameter data with at least one data value missing;

[0013] A training sample determination module is used to select first oil and gas parameter data from a plurality of historical oil and gas parameter data according to the sample to be determined by using a correlation analysis algorithm, and determine the first oil and gas parameter data as a training sample; the first oil and gas parameter data is a plurality of oil and gas parameter data whose correlation with the sample to be determined is higher than a preset correlation threshold;

[0014] A label adding module is used to add labels to training samples according to the categories of data in the training samples;

[0015] A distance determination module is used to calculate the distance between the sample to be determined and the training sample after adding labels by using the preset Gaussian kernel distance association relationship;

[0016] A neighbor sample determination module, used to determine a preset number of training samples sorted in ascending order of distance as neighbor samples;

[0017] The missing data determination module is used to determine the missing data values ​​in the sample to be determined by using the KNN algorithm based on the neighbor samples, the sample to be determined and the preset Gaussian kernel standard deviation.

[0018] Compared with the technical solution in the prior art in which petroleum geology and oil and gas reservoir engineering experts make empirical judgments, the embodiment of the present invention obtains a variety of historical oil and gas parameter data and samples to be determined; the samples to be determined are real-time oil and gas parameter data with at least one data value missing; according to the samples to be determined, a correlation analysis algorithm is used to screen the first oil and gas parameter data from the multiple historical oil and gas parameter data, and the first oil and gas parameter data is determined as a training sample; the first oil and gas parameter data is a variety of oil and gas parameter data whose correlation with the samples to be determined is higher than a preset correlation threshold; labels are added to the training samples according to the categories of the data in the training samples; the distance between the samples to be determined and the labeled training samples is calculated using a preset Gaussian kernel distance association relationship; the first preset number of training samples sorted in ascending order of distance are determined as neighbor samples; the missing data values ​​in the samples to be determined are determined using a KNN algorithm based on the neighbor samples, the samples to be determined and the preset Gaussian kernel standard deviation, so as to improve the accuracy and efficiency of determining the missing oil and gas parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0020] Figure 1 A flow chart of a method for determining oil and gas parameters provided in an embodiment of the present invention;

[0021] Figure 2 A flowchart of a specific example of a method for determining oil and gas parameters provided in an embodiment of the present invention;

[0022] Figure 3 Schematic diagram of the dimensional space of the training samples after the standardization process provided in the embodiment of the present invention Figure 1 ;

[0023] Figure 4 Schematic diagram of the dimensional space of the training samples after the standardization process provided in the embodiment of the present invention Figure 2 ;

[0024] Figure 5 A KNN algorithm calculation flow chart provided in an embodiment of the present invention;

[0025] Figure 6 A schematic diagram of an oil and gas parameter determination device provided in an embodiment of the present invention;

[0026] Figure 7 A schematic diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0028] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0029] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.

[0030] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps is not limited and can be appropriately adjusted as needed.

[0031] The purpose of the present invention is to provide an oil and gas parameter recommendation method based on the Gaussian kernel KNN algorithm, which provides resource and reserve evaluation parameters such as favorable area, reservoir thickness, resource abundance, reservoir lithology, etc. for oil and gas exploration, and provides development parameters such as porosity, permeability, oil saturation, effective thickness, and ground-saturation pressure difference for oil and gas field development, which significantly improves the accuracy of oil and gas reserves and production predictions, improves the objectivity and reliability of oil and gas field economic evaluation, and the reliability and scientificity of investment decisions.

[0032] When oil companies conduct oil and gas project transactions, they need to evaluate and assess the target blocks and target oil and gas fields. In the technical and economic evaluation process of new oil and gas projects, there are many problems, such as different exploration levels of blocks, incomplete oil and gas field data, inconsistent data formats, inconsistent standards, missing data items, and difficulty in data interpretation. Among all the problems, the lack of parameters has a huge impact on the evaluation. For the missing parameters, the current experience judgment of major petroleum geology and oil and gas reservoir engineering experts is not only time-consuming and highly deviates from the actual situation, but also is greatly affected by subjective factors, making it difficult to meet the needs of efficient and accurate evaluation. Therefore, how to use modern computing technology to improve the accuracy of parameter prediction has become a key technical problem that needs to be solved urgently.

[0033] In the current data analysis and prediction, if you want to roughly guess a person's monthly economic inflow, you can find some people with similar family background, education, work location, work content, and years of service from the database. According to the monthly economic inflow of these people, the weighted importance of each condition can be used to roughly estimate the range of this person's monthly economic inflow. The result of this calculation is based on real data and is much more reliable than subjective guesses. Similarly, based on historical experience, when evaluating new oil and gas resources, when evaluating oil and gas reservoirs with similar genesis, similar sedimentary strata, and oil and gas reservoir development methods, their related parameters also have a certain reference value.

[0034] The present invention is based on the above idea, and proposes a method for recommending parameter values ​​based on a proximity algorithm by taking the Gaussian kernel-based KNN algorithm in artificial intelligence as the core. The algorithm is based on a large amount of existing historical data, first finds out the parameters that are related to the missing parameters, and substitutes them into the model to construct the feature space, and then assigns most of the K most similar (i.e., the closest) samples in the feature space to a certain category, and then uses the Gaussian kernel function to calculate the weight of each feature (parameter) in this class, and gives the recommended value from the parameters of the same category according to the weight.

[0035] Figure 1 Flow chart of a method for determining oil and gas parameters provided in an embodiment of the present invention, such as Figure 1 As shown, the method may include:

[0036] Step 101, obtaining a variety of historical oil and gas parameter data and samples to be determined; the samples to be determined are real-time oil and gas parameter data with at least one data value missing;

[0037] Step 102, based on the sample to be determined, using a correlation analysis algorithm, screening first oil and gas parameter data from a plurality of historical oil and gas parameter data, and determining the first oil and gas parameter data as a training sample; the first oil and gas parameter data is a plurality of oil and gas parameter data whose correlation with the sample to be determined is higher than a preset correlation threshold;

[0038] Step 103, adding labels to the training samples according to the categories of the data in the training samples;

[0039] Step 104, using a preset Gaussian kernel distance association relationship, calculate the distance between the sample to be determined and the labeled training sample;

[0040] Step 105, determining a preset number of training samples sorted in ascending order of distance as neighbor samples;

[0041] Step 106, using the KNN algorithm to determine the missing data value in the sample to be determined according to the neighbor samples, the sample to be determined and the preset Gaussian kernel standard deviation.

[0042] When conducting oil and gas project evaluation, due to the limited exploration of the evaluation object or the limitation of data acquisition sources, some parameters are often missing, and the types of parameters provided for the acquired parameters cannot be determined. During the evaluation process, it is often necessary to infer and determine the possible values ​​of the qualitative parameters that have not been obtained based on the existing parameters.

[0043] This method adopts the latest supervised learning method widely used in artificial intelligence: the KNN algorithm based on Gaussian kernel is the core, and a parameter recommendation method based on proximity algorithm is proposed to solve the problem of missing parameter inference. The KNN algorithm with Gaussian kernel is applied to the data analysis of oil and gas exploration projects, and a parameter recommendation system based on geological characteristics and known oil and gas reservoir distribution is constructed, which effectively improves the effectiveness and accuracy of the evaluation of new oil and gas projects.

[0044] Figure 2 FIG. 1 is a flow chart of a specific example of a method for determining oil and gas parameters provided in an embodiment of the present invention. Figure 2 As shown in the figure, the construction of oil and gas exploration data analysis process involves the following steps: Data set establishment: Collect basic data including geological and geophysical data, such as seismic attributes, lithology, porosity, permeability, etc., and form these data into a data set to provide a data basis for the method; Data standardization and collation: Perform standardization processing to ensure that the data can be compared. Standardization processing usually includes steps such as normalization and scaling to make the data on the same scale for subsequent calculations; Data numerical transformation: Convert the type value, Boolean value (yes / no, male / female), location, geological level and other information in the data set into numbers; Model construction: Select those data items that are closely related to the recommended parameters as the data dimensions of the model to build the model. Feature selection can be carried out through correlation analysis, variance analysis, chi-square test, etc.; Model optimization: Check the integrity of the dimensional data and the data item data in the model, and eliminate the data without significance; K nearest neighbor calculation: For each area to be explored, calculate the Euclidean distance between it and the historical known oil and gas reservoir samples. Select the K samples closest to it as its K nearest neighbors; Calculate the recommended parameter values: Use the Gaussian kernel function to calculate the results for ranking evaluation. The Gaussian kernel function can effectively handle nonlinear problems by mapping the data into a high-dimensional feature space.

[0045] Specifically, in the first step, multi-source geological and geophysical data such as seismic attributes, lithology, porosity and permeability are integrated to form a comprehensive database. In the second step, data standardization, such as normalization and scaling, is performed to ensure the accuracy of cross-data comparison. In the third step, the data is converted into digital form, and each category and Boolean information is encoded into a digital format. In the fourth step, the model is constructed by screening key variables through correlation analysis, and feature selection is optimized using variance analysis and chi-square test. In the fifth step, the model is reviewed and insignificant features are eliminated for simplification. In the sixth step, the K nearest neighbor method is applied to calculate the Euclidean distance between each exploration area and the known oil and gas reservoir samples, and the nearest K samples are selected as neighbor references. In the seventh step, the Gaussian kernel function is adopted to implement the sorting evaluation, and its advantage in processing nonlinearity is used to map the data to a high-dimensional space, deeply explore the potential of oil and gas reservoirs, and realize an efficient recommendation evaluation system.

[0046] Each step is described below:

[0047] The entire algorithm is based on a large amount of data collected in the real evaluation process. These basic data include geological and geophysical data, such as seismic attributes, lithology, porosity, permeability, etc. These data constitute a data set and provide a data basis for this method.

[0048] In one embodiment, after acquiring a plurality of historical oil and gas parameter data, the method may further include: performing standardization processing on the plurality of historical oil and gas parameter data using a normalization algorithm, a unit conversion algorithm, and a classification algorithm respectively.

[0049] First, it is necessary to standardize the oil and gas parameter data, including historical oil and gas parameter data, to ensure that different features are comparable. During data preprocessing, these basic data come from different sources and need to be standardized to ensure that they are comparable. Standardization usually includes steps such as normalization, unit conversion, and classification unification to make the data on the same standard for subsequent calculations. The basic data types are different. The data have information such as numbers, types, Boolean values ​​(yes / no, male and female), locations, geological layers, etc. In order to ensure that they can be compared, all these data need to be converted into comparable numbers, such as addresses converted to postal codes, whether they are converted to 1 or 0, etc.

[0050] In this step, all oil and gas field parameters obtained from the new project data are taken as the research object, and the parameters of historical oil and gas field projects such as burial depth, permeability, thickness, porosity, saturation and other parameters are sorted out and entered into the oil and gas field evaluation model.

[0051] In one embodiment, in step 102, based on the sample to be determined, a correlation analysis algorithm is used to screen the first oil and gas parameter data from a plurality of historical oil and gas parameter data, and the first oil and gas parameter data is determined as a training sample, including: based on the sample to be determined, using a correlation analysis algorithm, respectively determining the correlation between a plurality of historical oil and gas parameter data and the sample to be determined; determining the data integrity of a plurality of oil and gas parameter data having a correlation higher than a preset correlation threshold; determining the oil and gas parameter data having a data integrity higher than a preset data integrity as the first oil and gas parameter data; and determining the first oil and gas parameter data as a training sample.

[0052] In one embodiment, the correlation analysis algorithm includes an analysis of variance algorithm and / or a chi-square test algorithm.

[0053] In step 102, for different oil and gas reservoir project data, based on domain knowledge, data items that are closely related to the recommended parameters are selected as data dimensions of the model to build the model. This helps to reduce the dimensionality curse and improve the efficiency of the model. Feature selection can be performed through correlation analysis, variance analysis, chi-square test, etc.

[0054] After selecting the data items closely related to the recommended parameters, you need to retrieve the completeness of the data in the database. If the completeness of this dimension is less than 30% among all the data, and a large amount of data is missing, then this dimension is unreliable and the data of this dimension needs to be removed from the model. If the data item is missing in the recommended parameter column, then the data item is invalid and needs to be deleted.

[0055] Based on domain knowledge, data items closely related to the recommended parameters are selected to reduce the curse of dimensionality and improve model efficiency. The selected features are introduced into the model to complete the model construction.

[0056] In this step, firstly, based on geological theory and prior knowledge, the features that are obviously unrelated to oil and gas accumulation are excluded, and statistical methods are used to evaluate the correlation between each feature and the presence or absence of oil and gas, and the features with low correlation are eliminated; finally, the selected set of features such as permeability, thickness, porosity, and saturation are submitted to geological experts for review to ensure that the feature selection is in line with geological principles and can effectively support the prediction objectives of the Gaussian kernel KNN artificial intelligence algorithm.

[0057] Therefore, four parameters were extracted from 20 data in the experiment, namely, burial depth, permeability, thickness, porosity, etc. as training samples. The data are shown in Table 1 below, where permeability, thickness, and porosity were normalized using the LN function:

[0058] Table 1

[0059]

[0060]

[0061]

[0062] For each area to be explored, the Euclidean distance between it and the historical known oil and gas reservoir samples is calculated. This is obtained by calculating the difference between the two samples on all selected features. Then, the K samples with the closest distance are selected as its K nearest neighbors. According to the labels of these K samples, the results of the Gaussian kernel function calculation are used to perform sorting evaluation. Based on the fact that the Gaussian kernel function can effectively handle nonlinear problems, by mapping the data to a high-dimensional feature space, the originally indistinguishable patterns become separable. Among them, the label is the target variable or category in the data set.

[0063] The K value in the KNN model is determined to be the smaller of the training sample size and 20 to avoid overfitting on small data sets while maintaining computational efficiency on large data sets. The Gaussian Kernel is introduced, which is a method to measure the similarity between two samples. It calculates a similarity score based on the Euclidean distance between samples and the preset standard deviation (here is 100).

[0064] The Gaussian kernel can take into account the nonlinear relationship of distance in the feature space, making similarity calculation more flexible and adaptable to complex distributed data. The KNN algorithm using the Gaussian kernel is combined with a custom similarity calculation method (here is the Gaussian kernel function) to sort all samples in the training set by their similarity to the sample to be predicted, so that the most similar samples will be placed at the front of the list.

[0065] For each area to be explored, the Euclidean distance between it and the historical known oil and gas reservoir samples is calculated. The nearest neighbor model is to normalize each parameter in the multidimensional space, project it into the coordinate system, and find the neighboring points, such as Figure 3 , Figure 4 As shown, the data are classified after normalization based on porosity and thickness, and each data is weighted and calculated once according to the four dimensions.

[0066] Figure 3 Schematic diagram of the dimensional space of the training samples after the standardization process provided in the embodiment of the present invention Figure 1 , Figure 3 The 45-degree angle scatter plot of the data provided by the present invention after dimension enhancement processing is shown. The angle is 30 degrees and the pitch angle is 30 degrees. Figure 4 Schematic diagram of the dimensional space of the training samples after the standardization process provided in the embodiment of the present invention Figure 2 , Figure 4 The scatter plot of the rotation angle of 30 degrees and the pitch angle of 30 degrees classified by the data provided by the present invention after the dimension enhancement processing is shown.

[0067] Figure 5A KNN algorithm calculation flow chart provided in an embodiment of the present invention is shown in FIG. Figure 5 As shown, after the start, the data is preprocessed and converted into a multidimensional data set in memory. The distance between the test data and each data in the model is calculated, sorted according to the increasing relationship of the distance, and the error between the expected output and the actual output is calculated. If the recommended parameters are appropriate, the process ends. If not, adjust the K value, improve the range, and return to the step of calculating the distance between the test data and each data in the model until the recommended parameters are determined to be appropriate.

[0068] After preprocessing, the data has been converted into a multidimensional data set in memory. At this time, when evaluating a new project M, facing a missing qualitative parameter X, we can calculate the underground temperature based on experience, such as formation thickness, and surface air temperature, surface air temperature, and lithology, and then we know what the temperature of the oil field is. Select several parameters related to the missing parameters in the acquired data and obtain recommended values ​​according to the following steps.

[0069] In one embodiment, the preset Gaussian kernel distance correlation relationship is:

[0070]

[0071] Among them, X t is the sample to be determined, X i is the i-th training sample, and the dimensions of the samples to be determined and the training samples are determined by the number of categories of the first oil and gas parameter data; d(X t ,X i ) is the distance between the sample to be determined and the training sample in the dimensional space; the dimension of the dimensional space is determined by the number of categories of the first oil and gas parameter data; X tj is the sample to be determined for the jth label; X ij is the i-th training sample of the j-th label; It represents the sum of the distances between the sample to be determined and all training samples on the jth label in the dimensional space; σ is the bandwidth of the Gaussian kernel function or the standard deviation of the Gaussian kernel; N is the number of training samples, and m is the number of labels.

[0072] In this example, the original data has n total samples, N training samples, and d-dimensional space. The training samples can be expressed as X = (X 1, X2,...,X n )∈R N×d ; n represents the total number of samples, N represents the training samples, and d represents the dimension of the dimensional space. Let C be the set of labeled training sets, and there are four labels in C, that is, C = (C1, C2, ..., C4), then each data in the data set can be expressed as follows: X j =(x j1 ,x j2 ,...,xjd ), j∈N;X j The label is C i , i=1,2,...,20. The algorithm proposed in this paper will be introduced in detail in the following section.

[0073] In this method, Euclidean distance is used to calculate spatial distance. Assume that X t is four-dimensional test data, then X i , i=1,2,...,20 is four-dimensional training data, and the distance calculation formula in the prior art is as follows:

[0074]

[0075] Among them, X t is the sample to be determined, X i is the i-th training sample; X tk is the tth training sample of the kth label; X ij is the i-th training sample of the j-th label; j, k = 1, 2, ..., m.

[0076] From the distance formula introduced above, it can be seen that the calculation of existing distance measurement methods is based on all features of the sample. Since it cannot be guaranteed that the relationship between all features and classifications is consistent, if the features are not given weights to calculate the distance, the distance between neighbors will be controlled by irrelevant features. The nearest neighbor method is highly sensitive to this situation. Therefore, the affinity distance function, a local distance function, is proposed. It considers local learning of features while considering distance. Its distance formula is:

[0077]

[0078] The formula of the weight function W in the formula is as follows:

[0079]

[0080] For the formula Represents the sum of the distances between the test point and all training points on the jth feature; It represents the training point X on the jth feature i The sum of the distances to the rest of the training points.

[0081] For weight functions, some are As a weight function, the Gaussian kernel function is a weight function, and its expression formula is as follows:

[0082]

[0083] Among them, Z is to make d(X t ,Xi ) is X t To be determined sample and X i The distance between two points of the i-th training sample, K(.) represents the kernel function.

[0084] The present invention proposes a distance function based on a Gaussian kernel, which is expressed in the dimensional space as follows:

[0085]

[0086] In the above formula, the first term on the right side of the equation is the absolute distance between the test point and the training point, and the second term is relative to the weight function in the above formula. t is the sample to be determined, X i is the i-th training sample, and the dimensions of the samples to be determined and the training samples are determined by the number of categories of the first oil and gas parameter data; d(X t ,X i ) is the distance between the sample to be determined and the training sample in the dimensional space; the dimension of the dimensional space is determined by the number of categories of the first oil and gas parameter data; X tj is the sample to be determined for the jth label; X ij is the i-th training sample of the j-th label; It represents the sum of the distances between the sample to be determined and all training samples in the dimensional space at the jth label; σ is the bandwidth of the Gaussian kernel function or the standard deviation of the Gaussian kernel.

[0087] The distance in the Gaussian kernel function in this paper is calculated using the absolute distance function. The second term in the numerator on the right side of the equation is Considered as a sample X to be determined t With training sample X i The projection value of the dimensional space on the jth feature, is the sample X to be determined on the jth feature t The projection value of a training sample with all training samples in the dimensional space, and the projection value of a training sample with the rest of the training samples in the high-dimensional space on the jth feature (j = 1, 2, ..., d) is Therefore, the numerator in the weight function can be regarded as the influence of the test data on the entire training data set in the dimensional space, and the denominator can be regarded as the influence between the related test data and the related training data in the high-dimensional space.

[0088] Select the K projects closest to project M and obtain the parameter X values ​​in these projects. Compare the parameter X values ​​of the K projects, and according to the principle of minority obeys majority, assign the parameter X value of project M to the value with the highest proportion in the K projects. Thus, parameter recommendations are obtained.

[0089] According to the labels of these k samples, the Gaussian kernel K-Nearest Neighbors (KNN) artificial intelligence algorithm is used to predict the recommended parameter values.

[0090] After obtaining 20 sample data and feature vectors and the corresponding formula of Gaussian kernel KNN artificial intelligence algorithm through the above three steps, the value of the parameter to be measured is predicted by program:

[0091] In step 1, a training data set is prepared, which includes multiple instances with features and labels, the test samples to be predicted are defined, the K value and the standard deviation of the Gaussian kernel are set, and the Gaussian kernel function is used to sort the training set according to the similarity with the test samples to obtain a list of stored data. The label values ​​of the first K neighbors are selected from the sorted neighbors and stored in the list.

[0092] Call the Gaussian kernel KNN algorithm method, pass in the neighbor label, neighbor sample list, Gaussian kernel standard deviation and the test sample itself, and calculate the weighted average predicted label. Finally, output the predicted label value: the porosity value of the inferred sample i: 18.377862427129372.

[0093] The embodiment of the present invention further proposes an oil and gas parameter determination device, the principle of which is similar to the oil and gas parameter determination method, and will not be repeated here.

[0094] Figure 6 Schematic diagram of an oil and gas parameter determination device provided in an embodiment of the present invention, such as Figure 6 As shown, the oil and gas parameter determination device may include:

[0095] The acquisition module 601 is used to acquire a variety of historical oil and gas parameter data and samples to be determined; the samples to be determined are real-time oil and gas parameter data with at least one data value missing;

[0096] The training sample determination module 602 is used to select first oil and gas parameter data from a plurality of historical oil and gas parameter data according to the sample to be determined by using a correlation analysis algorithm, and determine the first oil and gas parameter data as a training sample; the first oil and gas parameter data is a plurality of oil and gas parameter data whose correlation with the sample to be determined is higher than a preset correlation threshold;

[0097] A label adding module 603 is used to add labels to training samples according to the categories of data in the training samples;

[0098] The distance determination module 604 is used to calculate the distance between the sample to be determined and the labeled training sample by using the preset Gaussian kernel distance association relationship;

[0099] A neighbor sample determination module 605 is used to determine a preset number of training samples sorted in ascending order of distance as neighbor samples;

[0100] The missing data determination module 606 is used to determine the missing data value in the sample to be determined by using the KNN algorithm according to the neighbor samples, the sample to be determined and the preset Gaussian kernel standard deviation.

[0101] In one embodiment, the training sample determination module 602 is specifically configured to:

[0102] According to the samples to be determined, the correlation analysis algorithm is used to determine the correlation between various historical oil and gas parameter data and the samples to be determined;

[0103] Determining data integrity of multiple oil and gas parameter data having correlations above a preset correlation threshold;

[0104] Determining the oil and gas parameter data having a data integrity higher than a preset data integrity as the first oil and gas parameter data;

[0105] The first oil and gas parameter data is determined as a training sample.

[0106] In one embodiment, the preset Gaussian kernel distance correlation relationship is:

[0107]

[0108] Among them, X t is the sample to be determined, X i is the i-th training sample, and the dimensions of the samples to be determined and the training samples are determined by the number of categories of the first oil and gas parameter data; d(X t ,X i ) is the distance between the sample to be determined and the training sample in the dimensional space; the dimension of the dimensional space is determined by the number of categories of the first oil and gas parameter data; X tj is the sample to be determined for the jth label; X ij is the i-th training sample of the j-th label; It represents the sum of the distances between the sample to be determined and all training samples on the jth label in the dimensional space; σ is the bandwidth of the Gaussian kernel function or the standard deviation of the Gaussian kernel; N is the number of training samples, and m is the number of labels.

[0109] In one embodiment, the oil and gas parameter determination device further includes: a standardization processing module, which is used to:

[0110] Various historical oil and gas parameter data are standardized using normalization algorithm, unit conversion algorithm and classification algorithm.

[0111] In one embodiment, the correlation analysis algorithm includes an analysis of variance algorithm and / or a chi-square test algorithm.

[0112] Compared with the technical solution in the prior art in which petroleum geology and oil and gas reservoir engineering experts make empirical judgments, the embodiment of the present invention obtains a variety of historical oil and gas parameter data and samples to be determined; the samples to be determined are real-time oil and gas parameter data with at least one data value missing; according to the samples to be determined, a correlation analysis algorithm is used to screen the first oil and gas parameter data from the multiple historical oil and gas parameter data, and the first oil and gas parameter data is determined as a training sample; the first oil and gas parameter data is a variety of oil and gas parameter data whose correlation with the samples to be determined is higher than a preset correlation threshold; labels are added to the training samples according to the categories of the data in the training samples; the distance between the samples to be determined and the labeled training samples is calculated using a preset Gaussian kernel distance association relationship; the first preset number of training samples sorted in ascending order of distance are determined as neighbor samples; the missing data values ​​in the samples to be determined are determined using a KNN algorithm based on the neighbor samples, the samples to be determined and the preset Gaussian kernel standard deviation, so as to improve the accuracy and efficiency of determining the missing oil and gas parameters.

[0113] This method adopts the latest supervised learning method widely used in artificial intelligence: the KNN algorithm based on Gaussian kernel is the core, and a parameter recommendation method based on proximity algorithm is proposed to solve the problem of missing parameter inference. Its characteristic is that the KNN algorithm with Gaussian kernel is applied to the data analysis of oil and gas exploration projects, and a parameter recommendation system based on geological characteristics and known oil and gas reservoir distribution is constructed, thereby effectively improving the effectiveness and accuracy of the evaluation of new oil and gas projects.

[0114] The results of the present invention can obtain key parameters for oil and gas resource evaluation, oil and gas reserve evaluation, and oil and gas field development plan preparation. Compared with the current manual judgment based on experience, the prediction accuracy of key parameters has been significantly improved. The method is used to predict the parameters of known oil and gas fields. The results show that the porosity, oil saturation, oil and gas production and other parameters are more than 90% consistent with the actual operating parameters of the oil and gas fields, indicating that the method is reliable and easy to be widely promoted and applied, and can become an important basis for oil and gas field investment decisions.

[0115] An embodiment of the present invention further provides a computer device, Figure 7 It is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 700 includes a memory 710, a processor 720, and a computer program 730 stored in the memory 710 and executable on the processor 720. When the processor 720 executes the computer program 730, the above-mentioned oil and gas parameter determination method is implemented.

[0116] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned oil and gas parameter determination method is implemented.

[0117] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned oil and gas parameter determination method is implemented.

[0118] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0120] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0122] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for determining oil and gas parameters, characterized in that: include: Obtain various historical oil and gas parameter data and samples to be determined; The sample to be determined is real-time oil and gas parameter data with at least one data value missing; According to the sample to be determined, a correlation analysis algorithm is used to select first oil and gas parameter data from a plurality of historical oil and gas parameter data, and the first oil and gas parameter data is determined as a training sample; the first oil and gas parameter data is a plurality of oil and gas parameter data whose correlation with the sample to be determined is higher than a preset correlation threshold; Add labels to training samples according to the categories of data in them; Using the preset Gaussian kernel distance correlation relationship, calculate the distance between the sample to be determined and the training sample after adding labels; Determine a preset number of training samples sorted in ascending order of distance as neighbor samples; According to the neighbor samples, the samples to be determined and the preset Gaussian kernel standard deviation, the K-nearest neighbor KNN algorithm is used to determine the missing data values ​​in the samples to be determined.

2. The method according to claim 1, characterized in that According to the sample to be determined, a correlation analysis algorithm is used to select the first oil and gas parameter data from a variety of historical oil and gas parameter data, and the first oil and gas parameter data is determined as a training sample, including: According to the samples to be determined, the correlation analysis algorithm is used to determine the correlation between various historical oil and gas parameter data and the samples to be determined; Determining data integrity of multiple oil and gas parameter data having correlations above a preset correlation threshold; Determining the oil and gas parameter data having a data integrity higher than a preset data integrity as the first oil and gas parameter data; The first oil and gas parameter data is determined as a training sample.

3. The method according to claim 1, characterized in that The preset Gaussian kernel distance correlation relationship is: Among them, X t is the sample to be determined, X i is the i-th training sample, and the dimensions of the samples to be determined and the training samples are determined by the number of categories of the first oil and gas parameter data; d(X t ,X i ) is the distance between the sample to be determined and the training sample in the dimensional space; the dimension of the dimensional space is determined by the number of categories of the first oil and gas parameter data; X tj is the sample to be determined for the jth label; X ij is the i-th training sample of the j-th label; It represents the sum of the distances between the sample to be determined and all training samples on the jth label in the dimensional space; σ is the bandwidth of the Gaussian kernel function or the standard deviation of the Gaussian kernel; N is the number of training samples, and m is the number of labels.

4. The method according to claim 1, characterized in that After obtaining various historical oil and gas parameter data, it also includes: Various historical oil and gas parameter data are standardized using normalization algorithm, unit conversion algorithm and classification algorithm.

5. The method according to claim 1, characterized in that The correlation analysis algorithm includes a variance analysis algorithm and / or a chi-square test algorithm.

6. An oil and gas parameter determination device, characterized in that: include: An acquisition module is used to obtain various historical oil and gas parameter data and samples to be determined; The sample to be determined is real-time oil and gas parameter data with at least one data value missing; A training sample determination module is used to select first oil and gas parameter data from a plurality of historical oil and gas parameter data according to the sample to be determined by using a correlation analysis algorithm, and determine the first oil and gas parameter data as a training sample; the first oil and gas parameter data is a plurality of oil and gas parameter data whose correlation with the sample to be determined is higher than a preset correlation threshold; A label adding module is used to add labels to training samples according to the categories of data in the training samples; A distance determination module is used to calculate the distance between the sample to be determined and the training sample after adding labels by using the preset Gaussian kernel distance association relationship; A neighbor sample determination module, used to determine a preset number of training samples sorted in ascending order of distance as neighbor samples; The missing data determination module is used to determine the missing data values ​​in the sample to be determined by using the K-nearest KNN algorithm based on the neighbor samples, the sample to be determined and the preset Gaussian kernel standard deviation.

7. The device according to claim 6, characterized in that The training sample determination module is specifically used for: According to the samples to be determined, the correlation analysis algorithm is used to determine the correlation between various historical oil and gas parameter data and the samples to be determined; Determining data integrity of multiple oil and gas parameter data having correlations above a preset correlation threshold; Determining the oil and gas parameter data having a data integrity higher than a preset data integrity as the first oil and gas parameter data; The first oil and gas parameter data is determined as a training sample.

8. The device according to claim 6, characterized in that The preset Gaussian kernel distance correlation relationship is: Among them, X t is the sample to be determined, X i is the i-th training sample, and the dimensions of the samples to be determined and the training samples are determined by the number of categories of the first oil and gas parameter data; d(X t ,X i ) is the distance between the sample to be determined and the training sample in the dimensional space; the dimension of the dimensional space is determined by the number of categories of the first oil and gas parameter data; X tj is the sample to be determined for the jth label; X ij is the i-th training sample of the j-th label; It represents the sum of the distances between the sample to be determined and all training samples on the jth label in the dimensional space; σ is the bandwidth of the Gaussian kernel function or the standard deviation of the Gaussian kernel; N is the number of training samples, and m is the number of labels.

9. The device according to claim 6, characterized in that Also includes: Standardized processing modules for: Various historical oil and gas parameter data are standardized using normalization algorithm, unit conversion algorithm and classification algorithm.

10. The device according to claim 6, characterized in that The correlation analysis algorithm includes a variance analysis algorithm and / or a chi-square test algorithm.

11. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

13. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images