CN117831660A

CN117831660A - Method, device, equipment and medium for evaluating gas-containing performance of tight reservoir

Info

Publication number: CN117831660A
Application number: CN202311586752.6A
Authority: CN
Inventors: 贾海燕; 于开斌; 赵国英; 李玉城; 唐钦锡; 朱新佳; 吴则鑫; 井元帅; 王�琦; 郝坤; 李爽; 王鹏飞; 曲成永; 靳辉; 任茵; 林赫; 刘利忱
Original assignee: China National Petroleum Corp; CNPC Great Wall Drilling Co
Current assignee: China National Petroleum Corp; CNPC Great Wall Drilling Co
Priority date: 2023-11-24
Filing date: 2023-11-24
Publication date: 2024-04-05

Abstract

The invention discloses a method, a device, equipment and a storage medium for evaluating gas-containing performance of a tight reservoir. The method comprises the following steps: acquiring control factor data of the gas content of the to-be-evaluated tight reservoir in a target area, and determining the gas content characteristics of the to-be-evaluated tight reservoir according to the control factor data; determining the gas content performance evaluation result of the compact reservoir to be evaluated according to the gas content characteristics and a pre-trained gas content performance prediction model; the gas-containing performance prediction model comprises a plurality of K neighbor sub-models, and the number of the K neighbor sub-models in the gas-containing performance prediction model is determined based on the test results of a preset number of candidate gas-containing performance prediction models. According to the technical scheme, the problems of low gas-containing performance evaluation efficiency, poor accuracy and the like of the tight reservoir are solved, and the gas-containing performance evaluation of the tight reservoir can be rapidly and accurately realized through the gas-containing performance prediction model.

Description

Method, device, equipment and medium for evaluating gas-containing performance of tight reservoir

Technical Field

The invention relates to the technical field of reservoir evaluation, in particular to a method and a device for evaluating gas-containing performance of a tight reservoir, electronic equipment and a storage medium.

Background

The dense sandstone gas refers to natural gas resources enriched in a low-permeability and ultra-low-permeability dense sandstone reservoir, is an important component of unconventional natural gas exploration and development, and has important strategic significance for improving energy structure and guaranteeing source safety.

At present, the existing gas-containing performance evaluation modes of the tight reservoir mainly comprise geological risk probability evaluation, fuzzy comprehensive evaluation, multiple regression evaluation, bayes and evidence weight evaluation, markov distance and support vector machine evaluation, oil and gas aggregation value evaluation and the like. The gas-containing performance of the tight reservoir is evaluated mainly through artificial experience, so that subjectivity is high, artificial interference degree is high, and the method is not suitable for actual exploration work. Quantitative description is carried out on the gas content according to data mining and comprehensive analysis of actual oil and gas exploration results by adopting methods such as multiple regression evaluation, bayesian and evidence weight evaluation, markov distance and support vector machine evaluation, oil and gas aggregation value evaluation and the like, the requirement on the condition of a data sample is high, continuous variables are needed, and the gas content performance evaluation effect is not ideal.

Therefore, there is a need for a method of evaluating the gas-containing properties of a tight reservoir to achieve a highly accurate assessment of the gas-containing properties of the tight reservoir.

Disclosure of Invention

The invention provides a method, a device, equipment and a storage medium for evaluating the gas-containing performance of a tight reservoir, which are used for solving the problems of low gas-containing performance evaluation efficiency, poor accuracy and the like of the tight reservoir, and can realize rapid and accurate gas-containing performance evaluation of the tight reservoir through a gas-containing performance prediction model.

According to an aspect of the present invention, there is provided a method of evaluating gas-containing properties of a tight reservoir, the method comprising:

acquiring control factor data of the gas content of the to-be-evaluated tight reservoir in a target area, and determining the gas content characteristics of the to-be-evaluated tight reservoir according to the control factor data;

determining the gas content performance evaluation result of the compact reservoir to be evaluated according to the gas content characteristics and a pre-trained gas content performance prediction model;

the gas-containing performance prediction model comprises a plurality of K neighbor sub-models, and the number of the K neighbor sub-models in the gas-containing performance prediction model is determined based on the test results of a preset number of candidate gas-containing performance prediction models.

According to another aspect of the present invention, there is provided a gas-containing property evaluation apparatus for a tight reservoir, the apparatus comprising:

the gas-containing characteristic determining module is used for acquiring control factor data of gas-containing performance of the to-be-evaluated compact reservoir in the target area and determining the gas-containing characteristic of the to-be-evaluated compact reservoir according to the control factor data;

The evaluation result determining module is used for determining the gas content performance evaluation result of the compact reservoir to be evaluated according to the gas content characteristics and a pre-trained gas content performance prediction model;

According to another aspect of the present invention, there is provided an electronic apparatus including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of evaluating gas-containing properties of a tight reservoir according to any of the embodiments of the present invention.

According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute the method for evaluating gas-containing properties of a tight reservoir according to any of the embodiments of the present invention.

According to the technical scheme, the gas-containing characteristic of the to-be-evaluated compact reservoir is determined according to control factor data by acquiring the control factor data of the gas-containing property of the to-be-evaluated compact reservoir in the target area; and determining the gas content performance evaluation result of the to-be-evaluated tight reservoir according to the gas content characteristics and a pre-trained gas content performance prediction model. According to the technical scheme, the problems of low gas-containing performance evaluation efficiency, poor accuracy and the like of the tight reservoir are solved, and the gas-containing performance evaluation of the tight reservoir can be rapidly and accurately realized through the gas-containing performance prediction model.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for evaluating the gas-containing properties of a tight reservoir according to a first embodiment of the present invention;

FIG. 2 is a flow chart of a method for evaluating the gas-containing properties of a tight reservoir provided according to a second embodiment of the present invention;

FIG. 3 is a schematic structural view of a gas-containing property evaluation apparatus for a tight reservoir according to a third embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device implementing the method for evaluating gas-containing properties of a tight reservoir according to an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations.

Example 1

Fig. 1 is a flowchart of a method for evaluating gas-containing performance of a tight reservoir according to an embodiment of the present invention, where the embodiment is applicable to a scenario for evaluating gas-containing performance of a tight reservoir. The method may be performed by a gas-containing property evaluation device of a tight reservoir, which may be implemented in hardware and/or software, which may be configured in an electronic device. As shown in fig. 1, the method includes:

s110, acquiring control factor data of the gas content of the to-be-evaluated tight reservoir in the target area, and determining the gas content characteristics of the to-be-evaluated tight reservoir according to the control factor data.

The scheme can be executed by electronic equipment such as a computer, a server and the like, and the electronic equipment can acquire control factor data of the gas-containing performance of the tight reservoir to be evaluated in the target area through the oil and gas exploration platform. It will be appreciated that the target zone may be a region of investigation of a hydrocarbon reservoir and the tight reservoir to be evaluated may be a tight reservoir where evaluation of gas bearing properties is desired. The control factor data for gas-bearing properties may be data characterizing tight reservoir conditions, and in particular, may include at least one of hydrocarbon formation strength, relative formation height, reservoir physical properties, and distance to adjust fracture.

The electronic device can determine control factor data for the gas-containing performance of the tight reservoir to be evaluated according to the logging data, the porosity data, the permeability data, the hydrocarbon-producing intensity data of the hydrocarbon source rock and the top surface construction height contour map of the target interval. Specifically, the electronic device may obtain logging interpretation reservoir data of a target zone tight sandstone gas reservoir target zone, for example, well location name, well location coordinates, reservoir top depth, reservoir bottom depth, and other data of each logging interpretation reservoir. After obtaining basic data such as well position names, well position coordinates, reservoir top depths, reservoir bottom depths and the like of each well logging, the electronic equipment can determine a hydrocarbon source rock horizon adjacent to the tight reservoir, and obtain a spreading rule of hydrocarbon production intensity of the hydrocarbon source rock horizon on a plane through a hydrocarbon production potential method, a basin simulation method and the like, so as to obtain hydrocarbon production intensity data of each tight sandstone gas reservoir target layer well logging interpretation reservoir. As can be readily understood, hydrocarbon production intensity refers to the hydrocarbon production per unit area of a set of hydrocarbon source rocks, which can effectively characterize the sufficiency of the hydrocarbon source in a tight sandstone gas reservoir.

The compact sandstone gas reservoirs are mainly distributed at the parts with lower relative structures, and the relative structure heights can represent the relative structure heights of a set of compact reservoirs, so that the development favorability of the compact reservoirs is evaluated. According to the target area destination layer top surface construction height value diagram, the electronic equipment can identify the destination layer construction heights of all well positions and determine the maximum construction height and the minimum construction height matched with all destination layers of all well positions. The calculation of the relative formation height of a tight reservoir can be expressed as:

Wherein H is _rgi Representing the relative formation height of a tight reservoir at a desired zone, H representing the formation height of a well site at the desired zone, H _min Indicating the minimum build height of the well position match at the destination layer, H _max Indicating the maximum formation height at which the well location matches at the destination.

It is understood that porosity and permeability can effectively characterize the reservoir space, the seepage capability, and the self-seal of a tight reservoir. The porosity and permeability of each tight sandstone gas reservoir objective layer log interpretation reservoir can be obtained from core test data or log interpretation data. Fracture activity after reservoir formation of tight sandstone gas reservoirs can damage tight sandstone gas reservoirs, and this portion of the fracture is referred to as a conditioning fracture. The distance between the tight sandstone gas reservoir and the fracture can be adjusted to effectively represent the preservation condition of the tight sandstone reservoir in the later period. According to the target zone target layer adjustment fracture distribution diagram, the electronic equipment can identify the adjustment fracture of the movement after the tight sandstone gas reservoir formation period, find out the adjustment fracture midpoint with the closest well position distance and calculate the distance, so as to obtain the reservoir distance adjustment fracture distance data of the tight sandstone gas reservoir target layer logging interpretation.

After the control factor data is obtained, the electronic device can perform preprocessing operations such as data cleaning, normalization and standardization on the control factor data, and generate the gas-containing characteristic of the compact reservoir to be evaluated according to the preprocessed control factor data, for example, generate a one-dimensional characteristic vector according to the control factor data of each well position at each destination layer.

And S120, determining the gas content performance evaluation result of the to-be-evaluated tight reservoir according to the gas content characteristics and a pre-trained gas content performance prediction model.

After obtaining the gas-containing characteristic of the to-be-evaluated compact reservoir, inputting the gas-containing characteristic into a gas-containing performance prediction model, and determining a gas-containing performance evaluation result of the to-be-evaluated compact reservoir by the electronic equipment according to output data of the gas-containing performance prediction model. The gas-containing performance prediction model comprises a plurality of K neighbor sub-models, and the number of the K neighbor sub-models in the gas-containing performance prediction model is determined based on the test results of a preset number of candidate gas-containing performance prediction models. The gas-containing property evaluation result may include an evaluation level of oil-containing gas, low oil-containing gas, and the like.

It can be understood that the gas-containing performance prediction model can be a multi-random K-nearest neighbor joint prediction model, namely, the gas-containing performance evaluation result of the to-be-evaluated tight reservoir is determined based on the gas-containing performance prediction results output by the K-nearest neighbor sub-models. The number of K neighbor sub-models in the gas-containing performance prediction model can be used as one parameter of the gas-containing performance prediction model, the electronic equipment can obtain a plurality of candidate gas-containing performance prediction models by setting different parameters, and the gas-containing performance prediction model is determined according to the test result of each candidate gas-containing performance prediction model.

Example two

Fig. 2 is a flowchart of a method for evaluating the gas-containing performance of a tight reservoir according to a second embodiment of the present invention, where the training process of the gas-containing performance prediction model is refined based on the above embodiment. As shown in fig. 2, the method includes:

s210, acquiring a K neighbor sub-model number value set, a K neighbor coefficient value set and control factor data of the gas-containing performance of a plurality of groups of evaluated tight reservoirs.

It can be understood that the gas-containing performance prediction model may include a number of K-nearest neighbor sub-models, a K-nearest neighbor coefficient, and other parameters, and the electronic device may select an optimal parameter combination from different parameter combinations to ensure that the prediction effect of the gas-containing performance prediction model is optimal. Specifically, the electronic device may obtain, in advance, a value range of the number of K-nearest neighbor sub-models and a value range of the K-nearest neighbor coefficients. Wherein, the number of K neighbor sub-models and the K neighbor coefficient are both nonzero natural numbers. The K-nearest neighbor coefficient value set may include all selectable values of the K-nearest neighbor coefficient within its value range.

S220, determining at least two parameter combinations according to the K neighbor sub-model number value set and the K neighbor coefficient value set.

The electronic device may combine the elements of the K-nearest neighbor coefficient value set with the elements of the K-nearest neighbor coefficient value set to obtain a plurality of parameter combinations. Wherein each parameter combination comprises a K neighbor sub-model number and a K neighbor coefficient.

S230, according to the gas-containing performance matched with each group of control factor data, sample division is carried out on each group of control factor data, and a sample set matched with each parameter combination is determined.

It will be appreciated that the electronics may obtain data of the production of each tight sandstone reservoir for the purpose of logging the formation to account for the reservoir, such as cumulative production per formation, daily production per formation, thickness of formation, etc. According to the productivity data of each tight sandstone gas reservoir target layer logging interpretation reservoir, the electronic equipment can determine to evaluate each tight reservoir to obtain the gas content performance of each tight sandstone gas reservoir target layer logging interpretation reservoir. Each tight sandstone reservoir logging interpretation reservoir may correspond to a set of control factor data and a gas-containing property. The gas-containing performance can be used as a label for logging and explaining control factor data of a reservoir stratum of the same tight sandstone gas reservoir, and can be oil-containing gas or low-oil-containing gas.

According to the gas-containing performance matched with each group of control factor data, the electronic equipment can divide the samples of each group of control factor data to determine the number of samples of each type of gas-containing performance. And determining a sample set matched with each parameter combination according to the number of samples of various gas-containing performances. Specifically, the sample set matched by each parameter combination may be determined according to the number of K-nearest neighbor sub-models in each parameter combination.

In this aspect, optionally, the gas-containing property includes a first gas-containing property and a second gas-containing property;

the gas-containing performance matching according to each group of control factor data divides each group of control factor data into samples, and determines a sample set matched with each parameter combination, which comprises the following steps:

determining a first sample set and a second sample set according to the gas-containing performance matched with each group of control factor data; wherein the first sample set comprises each set of control factor data having a first gas-containing property and the second sample set comprises each set of control factor data having a second gas-containing property;

and determining a sample set matched by each parameter combination according to the first sample set and the second sample set.

It is understood that the first gas-containing property may be oil-containing gas and the second gas-containing property may be low oil-containing gas. The electronic device may divide the data set formed by the control factor data and the tags thereof according to the tags of each group of control factor data, that is, the matched gas-containing performance, to obtain a first sample set and a second sample set. Wherein the first sample set includes sets of control factor data having a first gas-containing property and the second sample set includes sets of control factor data having a second gas-containing property. Based on the first sample set, the second sample set, and the number of K-nearest neighbor sub-models in each parameter combination, the electronic device may determine a sample set that matches each parameter combination.

In this embodiment, optionally, the determining, according to the first sample set and the second sample set, a sample set that matches each parameter combination includes:

if the first sample number in the first sample set is larger than the second sample number in the second sample set, determining the first sample set matched with each parameter combination according to the second sample number;

determining a sample set matched with each parameter combination according to the first sample set and the second sample set matched with each parameter combination;

or alternatively, the first and second heat exchangers may be,

if the first sample number in the first sample set is smaller than the second sample number in the second sample set, determining the first sample set matched with each parameter combination according to the first sample number;

and determining a sample set matched with each parameter combination according to the second sample set matched with each parameter combination and the first sample set.

It is readily understood that in the data set formed by the control factor data and the labels thereof, the number of the first samples in the first sample set may be equal to or different from the number of the second samples in the second sample set. If equal, the union of the first and second sample sets may be taken as the sample set for which each parameter combination matches. If the number of the sample sets is not equal, determining sample sets with more sample numbers in the first sample set and the second sample set, reserving all samples in the sample sets with less sample numbers, randomly selecting sample subsets with the same sample number as the sample sets with less sample numbers from the sample sets with more sample numbers, and taking the union set of the sample subsets and the sample sets with less sample numbers as the sample set matched by each parameter combination.

In the case where the number of samples in the first sample set and the second sample set is not equal, the number of sample sets to be matched by each parameter combination may be one or more, and the number of sample sets to be matched by each parameter combination may be the same or different. For example, the electronic device may use the number of K-nearest neighbor sub-models in each parameter combination as a random number, randomly select, from among the sample sets with a large number of samples, a plurality of sample subsets with the same number of samples as the sample set with a small number of samples according to the random number, and use the union set of the sample subset obtained randomly each time and the sample set with a small number of samples as the sample set matched with each K-nearest neighbor sub-model.

The method and the device can realize the maximum utilization of sample data, ensure the reliability and scene applicability of the gas-containing performance prediction model, and avoid the influence of abnormal data on the model test result.

S240, determining test results of candidate gas-containing performance prediction models matched with all parameter combinations according to sample sets matched with all parameter combinations, and determining target parameter combinations according to all test results.

After obtaining the sample set matched by each parameter combination, the electronic equipment can train each K neighbor sub-model based on the K neighbor algorithm according to the sample set matched by each parameter combination, and perform model test after each training to obtain the test result of the candidate gas-containing performance prediction model matched by each parameter combination. And determining a target parameter combination from the parameter combinations according to the test result of the candidate gas-containing performance prediction model matched with the parameter combinations.

In one possible implementation, optionally, the determining, according to the sample set matched by each parameter combination, a test result of the candidate gas-containing performance prediction model matched by each parameter combination includes:

sequentially determining each parameter combination as a current parameter combination;

determining a test sample from samples which are not tested in a sample set matched with the current parameter combination;

training other samples except the test sample in a sample set matched with the current parameter combination based on a K neighbor algorithm according to the current parameter combination to obtain a candidate gas-containing performance prediction model matched with the test sample;

determining the gas-containing performance prediction result of the test sample in the iteration according to the test sample and the candidate gas-containing performance prediction model matched with the test sample;

returning to execute the test sample determination in the samples which are not tested in the sample set matched with the current parameter combination until the samples which are not tested are not existed in the sample set matched with the current parameter combination, and outputting the gas-containing performance prediction result of each sample in the sample set;

and determining the test result of the candidate gas-containing performance prediction model matched with the current parameter combination according to the gas-containing performance prediction result of each sample.

In this scheme, the electronic device may sequentially determine each parameter combination as a current parameter combination, and randomly select one or more samples from samples of the sample set matched by each K-nearest neighbor model, where the samples are not tested, as test samples. Training other samples except the test sample in the sample set matched with each K neighbor model based on a K neighbor algorithm according to the K neighbor coefficient in the current parameter combination to obtain a test object of the test sample, namely a candidate gas-containing performance prediction model corresponding to the test sample. The candidate gas-containing performance prediction models comprise K neighbor sub-models matched with the current parameter combination. And inputting the test sample into the candidate gas-containing performance prediction model, and determining a gas-containing performance prediction result of the test sample by the electronic equipment according to output data of the candidate gas-containing performance prediction model.

After the test of the test sample is finished, the electronic equipment can return to execute the determination of the test sample in the samples which are not tested in the sample set matched with the current parameter combination until the samples which are not tested are not in the sample set matched with each K neighbor model, and the gas-containing performance prediction result of each sample is output. According to the gas-containing performance prediction results of the samples, the electronic equipment can determine the test results of the candidate gas-containing performance prediction models matched with the current parameter combination.

On the basis of the scheme, optionally, the candidate gas-containing performance prediction model matched with the current parameter combination comprises K neighbor sub-models of the number of K neighbor sub-models in the current parameter combination;

and determining a gas-containing performance prediction result of the test sample at the current iteration according to the test sample and the candidate gas-containing performance prediction model matched with the test sample, wherein the method comprises the following steps:

and inputting the test samples matched with each K-nearest neighbor sub-model into the K-nearest neighbor sub-model, and determining the gas-containing performance prediction result of the test samples matched with each K-nearest neighbor sub-model in the iteration.

It can be understood that the electronic device can generate the gas-containing characteristic corresponding to the test sample according to the test sample matched with each K-nearest neighbor sub-model, input the gas-containing characteristic matched with each K-nearest neighbor sub-model into the K-nearest neighbor sub-model, and determine the gas-containing performance prediction result output by each K-nearest neighbor sub-model, namely the gas-containing performance prediction result of the test sample matched with each K-nearest neighbor sub-model in the current iteration.

Based on the above embodiment, optionally, the outputting the gas content performance prediction result of each sample includes:

determining a first gas-containing energy amount and a second gas-containing energy amount in the gas-containing energy prediction result of each iteration of each sample;

And determining the gas content prediction result of each sample according to the first gas content amount and the second gas content amount.

After the gas-containing performance prediction results output by the K neighbor sub-models through iteration are obtained, the electronic equipment can count the gas-containing performance prediction results to obtain a first gas-containing performance quantity and a second gas-containing performance quantity. It can be appreciated that in the previous iterative training of each K-nearest neighbor model, there may be multiple gas-containing performance prediction results for each sample, and the electronic device may use the gas-containing performance prediction result with the most frequent occurrence of the gas-containing performance prediction results as the gas-containing performance prediction result for the sample. Specifically, the electronic device may compare the first gas-containing property amount and the second gas-containing property amount of each sample, and if the first gas-containing property amount is greater than the second gas-containing property amount, the first gas-containing property is used as a gas-containing property prediction result of the sample. If the second gas-containing property amount is greater than the first gas-containing property amount, the second gas-containing property is used as a gas-containing property prediction result of the sample.

In this scheme, optionally, the determining, according to the gas content performance prediction result of each sample, the test result of the candidate gas content performance prediction model matched with the current parameter combination includes:

Comparing the gas-containing performance prediction results of the samples with the gas-containing performance matched with the samples, and determining the prediction accuracy of the candidate gas-containing performance prediction model matched with the current parameter combination according to the comparison results;

the determining the target parameter combination according to each test result comprises the following steps:

and determining the target parameter combination according to the prediction accuracy of the candidate gas-containing performance prediction model matched with each parameter combination.

After the gas-containing property prediction results of the respective samples are obtained, one gas-containing property prediction result may be corresponding to each sample. The electronics can compare the predicted gas-containing properties for each sample to the gas-containing properties that match the sample. If the gas-containing property prediction result is the same as the sample label (gas-containing property of sample match), then the sample is determined to be correctly predicted. If the gas-containing property prediction result is different from the sample label (gas-containing property of sample matching), determining that the sample is mispredicted. Based on the number of samples predicted correctly and the total number of samples, the electronic device may calculate a prediction accuracy of the candidate gas-containing performance prediction model for which the current parameter combination matches.

After obtaining the prediction accuracy of the candidate gas-containing performance prediction model matched with each parameter combination, the electronic equipment can select the parameter combination corresponding to the candidate gas-containing performance prediction model with the highest prediction accuracy as the target parameter combination.

S250, determining a gas-containing performance prediction model according to the target parameter combination matched candidate gas-containing performance prediction model.

The electronic device can take the candidate gas-containing performance prediction model matched by the target parameter combination as a gas-containing performance prediction model for evaluating the gas-containing performance of the compact reservoir to be evaluated.

And S260, acquiring control factor data of the gas content performance of the to-be-evaluated tight reservoir in the target area, and determining the gas content characteristics of the to-be-evaluated tight reservoir according to the control factor data.

And S270, determining the gas content performance evaluation result of the to-be-evaluated tight reservoir according to the gas content characteristics and a pre-trained gas content performance prediction model.

Specific application scenario one

The present application scenario is a specific embodiment based on the above-described embodiment. In this scenario, the gas-containing performance evaluation of the tight reservoir included the following tasks:

(1) Construction of tight reservoir Condition data sets

Generally, the data that can characterize the conditions of a tight reservoir include at least: hydrocarbon production intensity of a tight reservoir adjacent to a hydrocarbon source rock, relative construction height of the tight reservoir, physical properties of the tight reservoir, and fracture distance adjustment of the tight reservoir. To construct the four indices, raw data including log data, porosity data, permeability data, hydrocarbon-producing intensity data of the source rock, and a top-surface formation elevation contour map of the target interval are collected.

a. Logging interpretation reservoir data for obtaining tight sandstone gas reservoir destination layer in research area

And acquiring logging interpretation reservoir data of a tight sandstone gas reservoir target layer in a research area, wherein the logging interpretation reservoir data comprises well position names, well position coordinates, reservoir top depths and reservoir bottom depths of each logging interpretation reservoir.

b. Acquiring hydrocarbon production intensity data of adjacent hydrocarbon source rocks of tight reservoir

The hydrocarbon generation intensity refers to the hydrocarbon generation amount of a set of hydrocarbon source rocks in unit area, and can effectively represent the sufficiency of hydrocarbon sources in the tight sandstone gas reservoir. Firstly, determining a hydrocarbon source rock horizon adjacent to a tight reservoir, and secondly, obtaining a spreading rule of hydrocarbon production intensity of the hydrocarbon source rock horizon on a plane through a hydrocarbon production potential method, a basin simulation method and the like, thereby obtaining hydrocarbon production intensity data of each tight sandstone gas reservoir target layer logging interpretation reservoir.

c. Acquiring relative formation height data for tight reservoirs

The tight sandstone gas reservoirs are mainly distributed at the low parts of the relative structures, and the relative structure heights can represent the relative structure heights of a set of tight reservoirs, so that the development beneficial degree of the tight reservoirs is evaluated. And identifying the construction height of each target layer of each well position and the local maximum construction height and the minimum construction height of the target layer according to the construction height value diagram of the top surface of the target layer of the research area. The relative construction height of the tight reservoir is the value obtained by subtracting the minimum construction height of the target layer from the construction height of each well position at the target layer and subtracting the minimum construction height from the maximum construction height of the target layer, and the relative construction height data of the tight sandstone gas reservoir target layer logging interpretation reservoir is obtained according to the calculation method.

d. Obtaining physical property data of tight reservoir

Porosity and permeability can effectively characterize the reservoir space, the seepage capability and the self-sealing of a tight reservoir. The porosity and permeability data of each tight sandstone gas reservoir target layer logging interpretation reservoir can be obtained by using actual core test or logging interpretation data.

e. Obtaining tight reservoir distance adjustment fracture distance data

Fracture activity after reservoir formation of tight sandstone gas reservoirs can damage tight sandstone gas reservoirs, and this portion of the fracture is referred to as a conditioning fracture. The distance between the tight sandstone gas reservoir and the fracture can be adjusted to effectively represent the preservation condition of the tight sandstone reservoir in the later period. And (3) identifying the movable adjustment fracture after the tight sandstone gas reservoir formation period according to the target layer adjustment fracture distribution map of the research area, finding out the closest adjustment fracture midpoint of each well position distance, calculating the size, and finally obtaining the logging interpretation reservoir distance adjustment fracture distance data of each tight sandstone gas reservoir target layer.

(2) Dividing gas content level

Firstly, collecting the production capacity data of each tight sandstone gas reservoir target layer logging interpretation reservoir, wherein the production capacity data comprise the accumulated production of each gas layer, daily production of each gas layer, thickness of each gas layer and the like, and can describe the indexes of the production capacity of the gas layer.

Secondly, the gas layers in the research area are divided into two categories of oil-gas and low oil-gas according to the capacity of each gas layer.

(3) Building training data sets

Firstly, obtaining hydrocarbon production intensity of adjacent hydrocarbon source rocks of each tight sandstone gas reservoir target layer logging interpretation reservoir, relative construction height of the tight reservoir, physical properties of the tight reservoir and adjustment fracture distance of the tight reservoir distance as attributes, carrying out normalization treatment, and establishing a data set by taking test result types of each test layer as labels.

(4) Building multi-random K-nearest neighbor joint prediction model

a. And determining the number n of the K neighbor sub-models contained in the multi-random K neighbor joint prediction model to be built.

b. And determining the number of samples in two categories of the gas bearing layer and the low gas bearing layer, and defining the large number of samples as multi-category data, and otherwise defining the large number of samples as less-category data. Samples consistent with the quantity of the few types of data are randomly extracted from the multiple types of data, and a new data set is constructed. The above procedure is repeated n times to build n datasets, each of which will be trained into an independent K-nearest neighbor model.

c. Determining the optimal adjacent coefficient (K) of the n K neighbor sub-models, and calculating the distance formula of any two sample points in the process as follows:

wherein,the distance from the ith sample point data to the jth sample point data is calculated for the ith sample point data; GI (GI) ⁱ 、RGI ⁱ 、Por ⁱ 、Perm ⁱ 、DBF ⁱ The hydrocarbon production intensity of the compact sandstone adjacent to the hydrocarbon source rock, the relative construction height of the compact reservoir, the physical properties of the compact reservoir and the distance between the compact reservoirs are respectively the data of the ith sample point; GI (GI) ^j 、RGI ^j 、Por ^j 、Perm ^j 、DBF ^j Hydrocarbon production intensity of compact sandstone adjacent to hydrocarbon source rock, relative construction height of compact reservoir, physical properties of compact reservoir, and adjustment of fracture distance of compact reservoir distance of jth sample point。

(5) Selecting optimal parameters of multi-random K-nearest neighbor joint prediction model

a. Constructing all possible parameter combinations of the model

The optimal parameters to be selected for the multi-random K-nearest neighbor joint prediction model comprise two parameters, namely: the number n of K-nearest neighbor sub-models and the optimal proximity coefficient K of each K-nearest neighbor sub-model. To construct all possible parameter combinations of the model, first, the value range [ K ] of the optimal neighboring coefficient K is set _min ,K _max ]Therein K, K _min And K _max All are non-zero natural numbers, and the value range [ n ] of the number n of the submodels is set _min ,n _max ]Wherein n, n _min 、n _max All the parameters are non-zero natural numbers, all combinations of K and n parameters are listed, all the combined parameters are substituted into the model, the performance of each prediction model is evaluated, and finally the parameter combination which enables the model performance to be optimal is found.

b. The specific way to substitute all the combination parameters into the model and evaluate the performance of each prediction model is as follows:

b1, optionally, the parameter combination deployment model is provided that the selected parameter combination is [ Ki, nj ], and then the parameters are substituted into (4) -a.

b2, randomly selecting one sample point data from the constructed data set to serve as test data to be reserved, and taking the rest sample point data as training data.

b3, substituting the training data into the (4) -b, substituting the classification parameters of the sub-models into Ki, and training the nj sub-models by using the divided nj data sets.

And b4, substituting the attribute of the reserved test model in the step b2 into the nj sub-model trained in the step b3 to obtain nj predicted results, and taking the predicted result with the largest occurrence frequency in the nj predicted results as a final predicted result.

And b5, cycling the steps b 2-b 4 to enable all data in the data set to be used as test data, obtaining a predicted result, comparing the real result of each test data to obtain the overall prediction accuracy of all test data, and taking the overall prediction accuracy as the accuracy of the prediction model constructed by the parameter combination selected in the step b 1.

b6, cycling b 1-b 5 to obtain the prediction accuracy of the prediction model constructed by all parameter combinations. And taking the parameter combination of the prediction model with highest prediction accuracy as the optimal parameter of the multi-random K-nearest neighbor joint prediction model.

(6) And (3) taking the optimal parameter model obtained in the step (5) as a prediction model for predicting the gas content of the unknown tight reservoir.

Example III

Fig. 3 is a schematic structural diagram of a device for evaluating gas-containing properties of a tight reservoir according to a third embodiment of the present invention. As shown in fig. 3, the apparatus includes:

the gas-containing characteristic determining module 310 is configured to obtain control factor data of gas-containing performance of the to-be-evaluated tight reservoir in the target area, and determine gas-containing characteristics of the to-be-evaluated tight reservoir according to the control factor data;

the evaluation result determining module 320 is configured to determine a gas content performance evaluation result of the tight reservoir to be evaluated according to the gas content characteristics and a pre-trained gas content performance prediction model;

In this solution, optionally, the apparatus further includes a model training module, where the model training module includes:

the parameter value acquisition unit is used for acquiring a K neighbor model number value set, a K neighbor coefficient value set and a plurality of groups of control factor data of the evaluated gas-containing performance of the tight reservoir;

The parameter combination determining unit is used for determining at least two parameter combinations according to the K neighbor sub-model number value set and the K neighbor coefficient value set; each parameter combination comprises a K neighbor sub-model number and a K neighbor coefficient;

the sample set determining unit is used for dividing samples of each group of control factor data according to the gas-containing performance matched with each group of control factor data, and determining a sample set matched with each parameter combination;

the target combination determining unit is used for determining test results of the candidate gas-containing performance prediction model matched with each parameter combination according to the sample set matched with each parameter combination, and determining target parameter combinations according to each test result;

and the prediction model determining unit is used for determining the gas-containing performance prediction model according to the candidate gas-containing performance prediction model matched with the target parameter combination.

In one possible approach, optionally, the gas-containing properties include a first gas-containing property and a second gas-containing property;

the sample set determining unit is specifically configured to:

In this aspect, optionally, the target combination determining unit includes:

a current combination determining subunit, configured to sequentially determine each parameter combination as a current parameter combination;

a test sample determining subunit, configured to determine a test sample from samples that are not tested by the sample set that is matched by the current parameter combination;

the candidate model determining subunit is used for training other samples except the test sample in the sample set matched with the current parameter combination based on the K nearest neighbor algorithm according to the current parameter combination to obtain a candidate gas-containing performance prediction model matched with the test sample;

the sample testing subunit is used for determining the gas-containing performance prediction result of the test sample in the iteration according to the test sample and the candidate gas-containing performance prediction model matched with the test sample;

the prediction result determining subunit is used for returning and executing to determine a test sample from samples which are not tested in the sample set matched with the current parameter combination until the samples which are not tested are not existed in the sample set matched with the current parameter combination, and outputting the gas-containing performance prediction result of each sample;

And the test result determining subunit is used for determining the test result of the candidate gas-containing performance prediction model matched with the current parameter combination according to the gas-containing performance prediction result of each sample.

In a preferred scheme, the candidate gas-containing performance prediction model matched with the current parameter combination comprises K neighbor sub-models of the number of K neighbor sub-models in the current parameter combination;

the sample testing subunit is specifically configured to:

On the basis of the above scheme, optionally, the prediction result determining subunit is specifically configured to:

In this embodiment, optionally, the test result determining subunit is specifically configured to:

The target combination determining unit is specifically configured to:

The gas-containing performance evaluation device of the tight reservoir provided by the embodiment of the invention can execute the gas-containing performance evaluation method of the tight reservoir provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

Example IV

Fig. 4 shows a schematic diagram of an electronic device 410 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 4, the electronic device 410 includes at least one processor 411, and a memory, such as a Read Only Memory (ROM) 412, a Random Access Memory (RAM) 413, etc., communicatively connected to the at least one processor 411, wherein the memory stores computer programs executable by the at least one processor, and the processor 411 may perform various suitable actions and processes according to the computer programs stored in the Read Only Memory (ROM) 412 or the computer programs loaded from the storage unit 418 into the Random Access Memory (RAM) 413. In the RAM 413, various programs and data required for the operation of the electronic device 410 may also be stored. The processor 411, the ROM 412, and the RAM 413 are connected to each other through a bus 414. An input/output (I/O) interface 415 is also connected to bus 414.

Various components in the electronic device 410 are connected to the I/O interface 415, including: an input unit 416 such as a keyboard, a mouse, etc.; an output unit 417 such as various types of displays, speakers, and the like; a storage unit 418, such as a magnetic disk, optical disk, or the like; and a communication unit 419 such as a network card, modem, wireless communication transceiver, etc. The communication unit 419 allows the electronic device 410 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.

The processor 411 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 411 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 411 performs the various methods and processes described above, such as the gas-bearing performance evaluation method of a tight reservoir.

In some embodiments, the gas-containing property evaluation method of a tight reservoir may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as storage unit 418. In some embodiments, some or all of the computer program may be loaded and/or installed onto the electronic device 410 via the ROM 412 and/or the communication unit 419. When the computer program is loaded into RAM 413 and executed by processor 411, one or more steps of the gas-containing property evaluation method of a tight reservoir described above may be performed. Alternatively, in other embodiments, processor 411 may be configured to perform the gas-bearing performance evaluation method of the tight reservoir in any other suitable manner (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems-on-a-chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable compact reservoir gas content evaluation device such that the computer programs, when executed by the processor, cause the functions/operations specified in the flowchart and/or block diagram block or blocks to be performed. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.

The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims

1. A method for evaluating the gas-containing properties of a tight reservoir, the method comprising:

2. The method of claim 1, wherein the training of the gas-containing performance prediction model comprises:

acquiring a K neighbor sub-model number value set, a K neighbor coefficient value set and control factor data of the gas-containing performance of a plurality of groups of evaluated tight reservoirs;

determining at least two parameter combinations according to the K neighbor sub-model number value set and the K neighbor coefficient value set; each parameter combination comprises a K neighbor sub-model number and a K neighbor coefficient;

according to the gas-containing performance of the matching of each group of control factor data, sample division is carried out on each group of control factor data, and a sample set of the matching of each parameter combination is determined;

according to the sample set matched with each parameter combination, determining the test result of the candidate gas-containing performance prediction model matched with each parameter combination, and determining the target parameter combination according to each test result;

and determining the gas-containing performance prediction model according to the candidate gas-containing performance prediction model matched with the target parameter combination.

3. The method of claim 2, wherein the gas-containing properties comprise a first gas-containing property and a second gas-containing property;

4. The method of claim 2, wherein determining test results for candidate gas-containing performance predictive models for which each parameter combination matches based on the sample set for which each parameter combination matches comprises:

returning to execute the test sample determination in the samples which are not tested in the sample set matched with the current parameter combination until the samples which are not tested are not existed in the sample set matched with the current parameter combination, and outputting the gas-containing performance prediction result of each sample;

5. The method of claim 4, wherein the candidate gas-containing performance prediction model for which the current parameter combination matches comprises a K-nearest neighbor model of the number of K-nearest neighbor models in the current parameter combination;

according to the test sample and the candidate gas-containing performance prediction model matched with the test sample, determining a gas-containing performance prediction result of the test sample in the iteration comprises the following steps:

6. The method of claim 5, wherein outputting the gas-containing property prediction result for each sample comprises:

7. The method of claim 6, wherein determining test results of candidate gas-containing property prediction models matched by the current parameter combination based on the gas-containing property prediction results of each sample comprises:

8. A gas-containing property evaluation device for a tight reservoir, comprising:

9. An electronic device, the electronic device comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of evaluating gas-bearing properties of a tight reservoir of any of claims 1-7.

10. A computer readable storage medium storing computer instructions for causing a processor to perform the method of evaluating gas-containing properties of a tight reservoir according to any of claims 1-7.