An ARCGIS-based regional heat reserve modeling method

Abstract

The application provides an ARCGIS-based regional heat reserve modeling method, which comprises the following steps: marking geographical regions of geothermal data to obtain a marked data set; performing spatial data analysis based on an ARCGIS platform to obtain spatial analysis information; correcting geographical information in the marked data set by using the spatial analysis information; calculating the corrected data set by using a data-driven segmented affine autoregressive method to obtain a feature vector set; dividing the corrected data set of a specified research region of each land type according to a clustering algorithm, a scope boundary estimation algorithm and the feature vector set to obtain a plurality of differential working domains; establishing a dynamic regional heat reserve model of the specified research region of each land type according to the differential working domains; and cascading the dynamic regional heat reserve models of the specified research regions of each land type to establish a regional heat reserve model. The data-driven multi-working domain division method can improve the accuracy of regional heat reserve modeling.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data model construction, in particular to a regional thermal reserve modeling method based on ARCGIS. BACKGROUND

[0002] GIS is based on computer and software technology, and collects, stores, manages, retrieves, reproduces and comprehensively analyzes various types of spatial data. The deep meaning of GIS is to correlate different aspects, different levels and different time real object data, give the "static" real object data "dynamic" meaning (law), and use coding space to quickly and repeatedly reproduce and reproduce the long-term historical evolution of nature, so as to reveal the internal relationship and evolution law of individual.

[0003] Whether the results obtained by the spatial information model of GIS are scientific depends on the level of human cognition, the relationship between individual evolution law and internal factors, and the accuracy and completeness of the obtained data. Through the spatial information analysis model constructed by GIS, the evolution result is compared with the real data for many times, so that it is further modified and improved in the time and space information mode. The advantage of GIS is that various spatial data of different entities are mapped to a unified geographic coordinate system, and all attributes and spatial data are organically integrated. The same entity is stored in the form of table (layer), and such information expression is closer to the real natural state, which cannot be replaced by other methods. The topological overlap feature of GIS makes it easy to analyze multiple spatial layers of objects. The data analysis model of GIS, such as DEM, network analysis model and distance search model, is designed to solve these problems, and a complex spatial analysis model can be constructed by secondary development. GIS system has good drawing and output capability, which can mark spatial database, analysis result and chart. Compared with ordinary information system, GIS has the following characteristics: using spatial data and attribute data for analysis and application, providing new ideas for people's understanding and application of geographical problems; ordinary information management system only manages attribute database, even if the image is saved, it is often in the form of document and mechanical processing of spatial data, which cannot be used for spatial query, retrieval, adjacent analysis and other spatial data operations, and cannot be used for spatial analysis; GIS emphasizes spatial analysis and uses spatial analysis mode to analyze spatial data, and the popularization of GIS cannot be separated from effective analysis and design.

[0004] However, there is no method of using ARCGIS to model geothermal reserves at present, which leads to the inability to improve resource evaluation, development, management and monitoring. SUMMARY

[0005] In order to overcome the deficiencies of the prior art, the purpose of the present application is to provide an ARCGIS-based regional heat reserve modeling method.

[0006] To achieve the above-mentioned purpose, the present application provides the following scheme:

[0007] An ARCGIS-based regional heat reserve modeling method, comprising:

[0008] According to the preset requirement, the geothermal data of the specified research area is obtained, and the geothermal data is marked in the geographical area to obtain a marked data set;

[0009] Based on the ARCGIS platform, the spatial data analysis of the specified research area is carried out to obtain spatial analysis information;

[0010] The spatial analysis information is used to correct the geographical information in the marked data set to obtain a corrected data set;

[0011] The data-driven segmented affine autoregressive method is used to calculate the corrected data set to obtain a feature vector set;

[0012] According to the clustering algorithm, the scope boundary estimation algorithm and the feature vector set, the corrected data set of the specified research area of each land type is divided to obtain a plurality of difference working domains;

[0013] According to the difference working domain, the dynamic regional heat reserve model of the specified research area of each land type is established;

[0014] The dynamic regional heat reserve models of the specified research areas of each land type are cascaded to establish a regional heat reserve model.

[0015] Preferably, according to the preset requirement, the geothermal data of the specified research area is obtained, and the geothermal data is marked in the geographical area to obtain a marked data set, comprising:

[0016] According to the research report of the Bureau of Geology on geothermal drilling, the geothermal data of the specified research area is counted; the geothermal data includes drilling information and geological data;

[0017] The geographical area information of the specified research area is obtained;

[0018] According to the geographical area information, the geothermal data is marked to obtain the marked data set.

[0019] Preferably, the drilling information and geological data include:

[0020] Drilling hole number and name, drilling geographical position, drilling longitude and latitude information, drilling depth, outlet water temperature, static water level, precipitation depth, water yield and stratum distribution.

[0021] Preferably, the spatial data analysis of the specified research area is carried out based on the ARCGIS platform to obtain spatial analysis information, including:

[0022] A spatial information analysis model is constructed through the ARCGIS platform;

[0023] Data mapping is performed on the specified research area according to the spatial information analysis model, so that the spatial data of various different entities of the specified research area are mapped into a unified geographic coordinate system to obtain coordinate data;

[0024] The spatial analysis information is obtained by analyzing the coordinate data according to the spatial information analysis model.

[0025] Preferably, the feature vector set is obtained by calculating the corrected data set based on a data-driven piecewise affine autoregressive method, including:

[0026] The input order and the output order of the dynamic regional heat reserve model are determined according to the characteristics of the regional heat reserve;

[0027] Based on the input vector architecture, the model output of the dynamic regional heat reserve model is determined according to the disturbance input of the dynamic regional heat reserve model, and the input vector of the dynamic regional heat reserve model is determined according to the control input, system output and disturbance input of the dynamic regional heat reserve model;

[0028] Taking any one data point in the corrected data set as a data center, the Euclidean distance between the input vector of each data point except the data center and the input vector of the data center is calculated, and the c-1 points with the smallest Euclidean distance are selected to form adjacent data points;

[0029] A plurality of local data sets are established according to the adjacent data points;

[0030] Based on the local data sets, the parameter vector of the local data sets is calculated by using the least square calculation formula, and the feature vector set is constructed according to the parameter vector and the input vector mean in the local data set.

[0031] Preferably, the corrected data set of the specified research area of each land type is divided to obtain a plurality of difference working domains according to a clustering algorithm, a scope boundary estimation algorithm and the feature vector set, including:

[0032] Based on the Gaussian distribution algorithm, the covariance of the feature vectors in the feature vector set is calculated;

[0033] According to the covariance and Gaussian distribution algorithm, the confidence of the mean of the feature vector is evaluated with a unified standard;

[0034] Based on the K-Means algorithm, each feature vector is clustered, and the modified data set corresponding to the feature vector is divided into a corresponding working domain by the number of clusters, and the steady-state data in each working domain is stored;

[0035] Based on the support vector machine, the hyperplane equation of each working domain is obtained;

[0036] According to the hyperplane equation and the steady-state data in each working domain, the differential working domain is determined.

[0037] Preferably, the dynamic regional heat reserve model includes an HSM model, an LSTM model or an ARX model.

[0038] According to the specific embodiments of the present application, the following technical effects are provided:

[0039] The present application provides a regional heat reserve modeling method based on ARCGIS, comprising: obtaining geothermal data of a specified research region according to a predetermined requirement, and labeling the geothermal data according to geographical regions to obtain a labeled data set; performing spatial data analysis on the specified research region based on the ARCGIS platform to obtain spatial analysis information; correcting the geographical information in the labeled data set using the spatial analysis information to obtain a modified data set; calculating the modified data set based on a data-driven piecewise affine autoregressive method to obtain a feature vector set; dividing the modified data set of the specified research region of each land type according to a clustering algorithm, a scope boundary estimation algorithm and the feature vector set to obtain a plurality of differential working domains; establishing a dynamic regional heat reserve model of the specified research region of each land type according to the differential working domains; and cascading the dynamic regional heat reserve models of the specified research regions of each land type to establish a regional heat reserve model. The present application can improve the accuracy of regional heat reserve modeling based on the data-driven multi-working domain division method. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0041] Figure 1 A flowchart of a regional heat reserve modeling method based on ARCGIS provided by the embodiments of the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be apparently and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0043] The present application aims to provide an ARCGIS-based regional heat reserve modeling method, which can improve the accuracy of regional heat reserve modeling based on a data-driven multi-working domain division method.

[0044] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0045] Figure 1 A flow chart of an ARCGIS-based regional heat reserve modeling method provided by an embodiment of the present application is shown in FIG. 1, and the present application provides an ARCGIS-based regional heat reserve modeling method, which comprises the following steps. Figure 1

[0046] Step 100: Obtain geothermal data of a specified research region according to a preset requirement, and perform geographic region labeling on the geothermal data to obtain a labeled data set;

[0047] Step 200: Perform spatial data analysis on the specified research region based on an ARCGIS platform to obtain spatial analysis information;

[0048] Step 300: Correct the geographic information in the labeled data set by using the spatial analysis information to obtain a corrected data set;

[0049] Step 400: Calculate the corrected data set based on a data-driven segmented affine autoregressive method to obtain a feature vector set;

[0050] Step 500: Divide the corrected data set of the specified research region of each land type according to a clustering algorithm, a scope boundary estimation algorithm and the feature vector set to obtain a plurality of differential working domains;

[0051] Step 600: Establish a dynamic regional heat reserve model of the specified research region of each land type according to the differential working domains;

[0052] Step 700: Cascade the dynamic regional heat reserve models of the specified research regions of each land type to establish a regional heat reserve model.

[0053] ​Preferably, geothermal data of a specified research area is obtained according to preset requirements, and the geothermal data is geographically labeled to obtain a labeled data set, including:

[0054] According to a research report on geothermal drilling holes provided by the geological bureau, geothermal data of the specified research area is counted; the geothermal data includes drilling hole information and geological data;

[0055] Geographical area information of the specified research area is obtained;

[0056] The geothermal data is labeled according to the geographical area information to obtain the labeled data set.

[0057] Specifically, in the embodiment, the geographical area information can be used to perform initial geographical labeling on the geothermal data, so that the accuracy of the initial data for modeling can be improved, and the geographical information can be integrated and labeled in the geothermal data to obtain initial data with higher information integration.

[0058] Preferably, the drilling hole information and the geological data include:

[0059] Drilling hole number and name, drilling hole geographical position, drilling hole longitude and latitude information, drilling hole depth, outlet water temperature, static water level, precipitation depth, water yield, and stratum distribution.

[0060] Further, in the embodiment, according to the research report on geothermal drilling holes provided by the geological bureau, drilling hole information and geological data of the research area are comprehensively counted to establish a geothermal attribute database, including drilling hole number and name, drilling hole geographical position, drilling hole longitude and latitude information, drilling hole depth, outlet water temperature, static water level, precipitation depth, water yield, stratum distribution, and surface and stratum temperature distribution data at each depth. Since the report provided by the geological bureau is not complete, the geothermal gradient calculation method is used to comprehensively enrich the geological drilling hole information statistical report.

[0061] Preferably, spatial data analysis of the specified research area is performed based on an ARCGIS platform to obtain spatial analysis information, including:

[0062] A spatial information analysis model is constructed through the ARCGIS platform;

[0063] Data of the specified research area is mapped according to the spatial information analysis model, so that spatial data of various entities of the specified research area are mapped to a unified geographical coordinate system to obtain coordinate data;

[0064] The coordinate data is analyzed according to the spatial information analysis model to obtain the spatial analysis information.

[0065] The role of GIS in this embodiment is to collect, judge, judge, process and other five types of problems for all data:

[0066] (1) Location, that is, where the problem is. The location can be expressed by place name, zip code, geographic coordinates, etc.

[0067] (2) Situation, that is, where a unit meets certain conditions. For example, find a place where the temperature does not exceed 40 degrees Celsius.

[0068] (3) Tendency prediction, that is, a change in a certain location and different periods of time.

[0069] (4) Type, that is, the allocation of spatial entities in a certain location. Through the study of the morphology of geographical entities, their spatial relationship can be seen.

[0070] (5) Simulation, that is, some problems will occur under certain conditions. GIS simulation is based on modeling.

[0071] Further, as a discipline based on complex hydrogeological environment, geothermal energy is based on complex geological environment and a large amount of data, so the application of GIS technology in the field of geothermal energy is a very useful method. GIS has strong geographic information expression ability, and the application of GIS technology to the development and management of geothermal areas can better master the distribution of geothermal resources, scientific evaluation, development of resources, dynamic image display, expression and other aspects. GIS technology is an upgrade of traditional database and computer mapping technology, and if it is used for management, especially regional geothermal fields, it can better utilize GIS technology to realize the improvement of resource evaluation, development, management and monitoring.

[0072] Preferably, a data-driven piecewise affine autoregressive method is used to calculate the modified data set to obtain a feature vector set, including:

[0073] According to the characteristics of the regional heat reserves, the input order and the output order of the dynamic regional heat reserves model are determined;

[0074] Based on the input vector architecture, the model output of the dynamic regional heat reserves model is determined according to the disturbance input of the dynamic regional heat reserves model, and the input vector of the dynamic regional heat reserves model is determined according to the control input, system output and disturbance input of the dynamic regional heat reserves model;

[0075] Taking any one data point in the modified data set as a data center, calculating the Euclidean distance between the input vector of each data point except the data center and the input vector of the data center, and selecting c-1 points with the smallest Euclidean distance to form neighboring data points;

[0076] According to the neighboring data points, a plurality of local data sets are established;

[0077] Based on the local data sets, the parameter vector of the local data set is calculated by using the least square calculation formula, and the feature vector set is constructed according to the parameter vector and the mean of the input vector in the local data set.

[0078] Preferably, according to the clustering algorithm, the scope boundary estimation algorithm and the feature vector set, the modified data set of each land type designated research area is divided to obtain a plurality of difference working domains, including:

[0079] Based on the Gaussian distribution algorithm, the covariance of the feature vector in the feature vector set is calculated;

[0080] According to the covariance and the Gaussian distribution algorithm, the confidence of the mean of the feature vector is evaluated by a unified standard;

[0081] Based on the K-Means algorithm, each feature vector is clustered, and the modified data set corresponding to the feature vector is divided into a corresponding working domain by the number of clusters, and the steady-state data in each working domain is stored;

[0082] Based on the support vector machine, the hyperplane equation of each working domain is obtained;

[0083] According to the hyperplane equation and the steady-state data in each working domain, the difference working domain is determined.

[0084] Preferably, the dynamic area heat reserve model includes HSM model, LSTM model or ARX model.

[0085] Specifically, in the embodiment, a multiple working domain-autoregressive exogenous (MWD-ARX) black box model and a multiple working domain-long short-term memory (MWD-LSTM) neural network black box model are established. The MWD-ARX modeling method is a black box parameterization modeling method based on parameter estimation of actual operation data of each working domain, which can establish an autoregressive model to represent the local characteristics of the system. According to the input and output order, the difference model architecture of each working domain is determined, and the model parameters are estimated according to the actual operation data, so as to establish the MWD-ARX model. In recent years, the application of data-driven machine learning algorithms in nonlinear system modeling practice has become increasingly rich. Only the input and output variables of the system need to be determined, and the corresponding neural network is created, and the model can be directly trained.

[0086] In addition, in addition to the black box modeling method, the nonlinear ordinary differential equation model of the system can also be determined through mechanism analysis, and the appropriate state variable is selected to convert the model into a linear state space equation. The mechanism representation of the gray box model is usually less than the black box model. The linear state space equation model that can work in each domain is established, and the multiple working domain-hybrid semi-mechanism (MWD-HSM) model with infinite approximation ability is established by combining the machine learning algorithm. It can combine the partial transparency of the mechanism model and approximate the unmodeled dynamic uncertainty in a black box way at low cost, and has the characteristics of low complexity and high precision.

[0087] The beneficial effects of the present application are as follows:

[0088] The multiple working domain division method based on data driving of the present application can improve the accuracy of regional thermal reserve modeling.

[0089] Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same and similar parts between each embodiment can be referred to each other.

[0090] The principles and implementation modes of the present application are described by applying specific examples in this paper, and the above description of the embodiments is only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In view of the above, the content of the specification should not be understood as a limitation of the present application.

Claims

1. An ARCGIS-based regional heat reserve modeling method, characterized in that, The method comprises the following steps: According to the preset requirements, the geothermal data of the specified research area is obtained, and the geothermal data is labeled in the geographical region to obtain a labeled data set; Based on the ARCGIS platform, spatial data analysis is performed on the specified research area to obtain spatial analysis information; Using the spatial analysis information, the geographical information in the labeled data set is corrected to obtain a corrected data set; Based on the data-driven segmented affine autoregressive method, the corrected data set is calculated to obtain a feature vector set; According to the clustering algorithm, the scope boundary estimation algorithm and the feature vector set, the corrected data set of the specified research area of each land type is divided to obtain a plurality of difference working domains; According to the difference working domain, a dynamic regional geothermal reserve model of the specified research area of each land type is established; The dynamic regional geothermal reserve models of the specified research areas of each land type are cascaded to establish a regional geothermal reserve model; Based on the data-driven segmented affine autoregressive method, the corrected data set is calculated to obtain a feature vector set, which comprises: According to the characteristics of the regional geothermal reserve, the input order and the output order of the dynamic regional geothermal reserve model are determined; Based on the architecture of the input vector, the model output of the dynamic regional geothermal reserve model is determined according to the disturbance input of the dynamic regional geothermal reserve model, and the input vector of the dynamic regional geothermal reserve model is determined according to the control input, system output and disturbance input of the dynamic regional geothermal reserve model; Taking any one data point in the corrected data set as a data center, the Euclidean distance between the input vector of each data point except the data center and the input vector of the data center is calculated, and the c-1 points with the smallest Euclidean distance are selected to form adjacent data points; According to the adjacent data points, a plurality of local data sets are established; Based on the local data set, the parameter vector of the local data set is calculated by using the least square calculation formula, and the feature vector set is constructed according to the parameter vector and the input vector mean in the local data set; According to the clustering algorithm, the scope boundary estimation algorithm and the feature vector set, the corrected data set of the specified research area of each land type is divided to obtain a plurality of difference working domains, which comprises: Based on the Gaussian distribution algorithm, the covariance of the feature vectors in the feature vector set is calculated; According to the covariance and the Gaussian distribution algorithm, the confidence of the mean of the feature vectors is evaluated by a unified standard; Based on the K-Means algorithm, each feature vector is clustered, and the corrected data set corresponding to the feature vector is divided into a corresponding working domain by the number of clusters, and the steady-state data in each working domain is stored; Based on the support vector machine, the hyperplane equation of each working domain is obtained; According to the hyperplane equation and the steady-state data in each working domain, the difference working domain is determined.

2. The ARCGIS-based regional reserve modeling method according to claim 1, characterized in that, According to the preset requirements, the geothermal data of the specified research area is obtained, and the geothermal data is labeled in the geographical region to obtain a labeled data set, which comprises: According to a survey report on geothermal drilling of the Bureau of Geology and Mineral Resources, geothermal data of a specified research area are counted, and the geothermal data include drilling information and geological data; Geographical area information of the specified research area is obtained; The geothermal data are labeled according to the geographical area information, and a labeled data set is obtained.

3. The ARCGIS-based regional reserve modeling method according to claim 2, characterized in that, The drilling information and the geological data include: Drilling hole number and name, drilling geographical position, drilling longitude and latitude information, drilling depth, outlet water temperature, static water level, precipitation depth, water yield and stratum distribution.

4. The ARCGIS-based regional reserve modeling method of claim 1, wherein, Spatial data analysis of the specified research area is performed based on an ARCGIS platform, and spatial analysis information is obtained, including: A spatial information analysis model is constructed through the ARCGIS platform; Data of the specified research area are mapped according to the spatial information analysis model, so that spatial data of various different entities of the specified research area are mapped into a unified geographical coordinate system, and coordinate data are obtained; The coordinate data are analyzed according to the spatial information analysis model, and the spatial analysis information is obtained.

5. The ARCGIS-based regional reserve modeling method of claim 1, wherein, The dynamic regional geothermal reserve model includes an HSM model, an LSTM model or an ARX model.

Images