Carbon emission prediction method and device for mine target area, medium and product
Through remote sensing image processing and neural network model, combined with CNN-GRU network and attention mechanism, the problem of low carbon emission prediction accuracy in mine repair areas is solved, and more accurate carbon emission prediction is achieved, providing science for carbon emission control. in accordance with.
Patent Information
- Application Number
- CN202411314957.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2025-01-07
AI Technical Summary
In the prior art, the prediction accuracy of carbon emission prediction models in mine restoration areas is not high, and it is difficult to effectively control and optimize carbon emissions.
By obtaining remote sensing images of the target area of the mine, dividing the land type, obtaining characteristic data such as land area, temperature, humidity and atmospheric pressure, and using the trained neural network model to predict carbon emissions, combining the CNN-GRU network and attention mechanism , improve prediction accuracy.
The prediction accuracy of carbon emissions in the target area of the mine has been improved, more accurate carbon emission data is provided, and a scientific basis for the control and optimization of carbon emissions in the mine restoration area.
Smart Images
Figure CN119272989A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of carbon emission prediction, and in particular to a method, device, medium and product for predicting carbon emissions in a target area of a mine. Background Art
[0002] Land use carbon emissions refer to the process of releasing carbon into the atmosphere when land is intervened by human society to achieve production, ecological and social functions. Land use carbon emissions can be divided into direct carbon emissions and indirect carbon emissions. Direct carbon emissions refer to carbon emissions generated in the process of changing the purpose or method of land use; indirect carbon emissions refer to carbon emissions caused by human development and utilization of land. As mining and restoration of mines proceed, land use types will also slowly change. Different land use types will cause changes in carbon emissions.
[0003] Predicting carbon emissions in target areas after mine restoration can provide a reasonable basis for controlling and optimizing carbon emissions from mine restoration. However, currently, there is a lack of prediction models for carbon emissions in mine restoration areas, and the prediction accuracy is low, which is not conducive to controlling and optimizing carbon emissions from mine restoration in advance. Therefore, a carbon emission prediction method is urgently needed to address the problems existing in existing technologies. Summary of the Invention
[0004] The purpose of the present invention is to provide a method, device, medium and product for predicting carbon emissions in a target mine area, which can improve the prediction accuracy of carbon emissions in the target restoration area.
[0005] To achieve the above object, the present invention provides the following solutions:
[0006] A method for predicting carbon emissions in a target area of a mine, the method comprising:
[0007] Acquiring a current remote sensing image of the target mine area;
[0008] Determining the classification results of various land types in the target mine area according to the current remote sensing image;
[0009] Obtaining characteristic data corresponding to the classification results of each land type within a statistical time period; the characteristic data includes land area, temperature, humidity, wind speed and atmospheric pressure;
[0010] Normalizing the feature data to obtain normalized feature data;
[0011] Inputting the normalized characteristic data into a carbon emission prediction model to obtain carbon emission prediction results for each land type in the target mine area within a statistical time period; the carbon emission prediction model is obtained by training a neural network using a training data set; the training data set is a sample of normalized characteristic data for each land type and a corresponding sample of carbon emission results within a preset time period;
[0012] The carbon emission prediction results of each land type in the target mine area during the statistical time period are added together to obtain the carbon emission prediction results of the target mine area during the statistical time period.
[0013] Optionally, before executing the step of “determining the classification result of each land type in the target mine area according to the current remote sensing image”, the method further includes:
[0014] The current remote sensing image is preprocessed; the preprocessing operations include color correction, denoising and image enhancement.
[0015] Optionally, determining the division result of each land type in the target mine area according to the current remote sensing image specifically includes:
[0016] Construct a land type identification model;
[0017] Inputting the current remote sensing image into the land type recognition model to obtain an initial land type recognition result;
[0018] Calculate the similarity between the initial land type recognition result and the corresponding standard template of each land type to obtain a similarity result;
[0019] When the similarity result is greater than or equal to a preset similarity threshold, the initial land type recognition result is used as a division result;
[0020] When the similarity result is less than the preset similarity threshold, the initial land type recognition result corresponding to the similarity result is input into the land type recognition model to obtain a corresponding secondary recognition result;
[0021] Calculating the similarity between the secondary recognition result and the corresponding standard template of each land type to obtain a secondary similarity result;
[0022] When the secondary similarity result is greater than or equal to a preset similarity threshold, the secondary recognition result is used as a segmentation result;
[0023] When the secondary similarity result is less than the preset similarity threshold, a final land type corresponding to the secondary recognition result is obtained through investigation, and the final land type is used as a division result.
[0024] Optionally, inputting the current remote sensing image into the land type recognition model to obtain an initial land type recognition result specifically includes:
[0025] extracting an initial image map from the current remote sensing image using a corresponding conformation equation;
[0026] Cropping the initial image to obtain an orthophoto image;
[0027] The orthophoto map is input into the target neural network model to obtain an initial land type recognition result.
[0028] Optionally, the neural network is a long short-term memory network.
[0029] Optionally, the neural network is a CNN-GRU network that introduces an attention mechanism; the CNN-GRU network includes a CNN network and a GRU network connected in sequence;
[0030] The CNN network includes a convolutional layer, a pooling layer and a fully connected layer;
[0031] The GRU network includes a GRU network layer and a Dense layer;
[0032] The fully connected layer of the CNN network is connected to the GRU network layer of the GRU network;
[0033] An attention mechanism is introduced before the convolutional layer of the CNN network, and the attention weight distribution value is updated through the output of the GRU network layer.
[0034] Optionally, the process of constructing the carbon emission prediction model includes:
[0035] The neural network is trained using the normalized characteristic data samples of each land type as input and the corresponding carbon emission result samples within a preset time period as output to obtain a carbon emission prediction model corresponding to each land type.
[0036] A computer device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-mentioned methods for predicting carbon emissions in a target mine area.
[0037] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements any of the above-mentioned methods for predicting carbon emissions in a target mine area.
[0038] A computer program product comprises a computer program, which, when executed by a processor, implements any of the above-mentioned methods for predicting carbon emissions in a target area of a mine.
[0039] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0040] The present invention discloses a method, device, medium and product for predicting carbon emissions in a target mine area. The method determines the division results of each land type in the target mine area through the current remote sensing image of the target mine area, and obtains the characteristic data corresponding to the division results of each land type within a statistical time period; the characteristic data includes land area, temperature, humidity, wind speed and atmospheric pressure; the present invention can obtain more carbon emission information by obtaining data related to carbon emissions, further normalizes the characteristic data, inputs the normalized characteristic data into a carbon emissions prediction model, and obtains the target mine area. The carbon emission prediction results of each land type in the region within the statistical time period; the present invention can obtain more accurate carbon emission prediction values for the land type by constructing carbon emission prediction models corresponding to different land types, and the input data of the carbon emission prediction model is data related to carbon emissions, so that more accurate carbon emission data of each land type can be obtained. Finally, the carbon emission prediction results of each land type in the target mine area within the statistical time period are added together to obtain the carbon emission prediction results of the target mine area within the statistical time period, thereby improving the prediction accuracy of carbon emissions in the target restoration area. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0042] Figure 1 A schematic flow chart of a method for predicting carbon emissions in a target mine area provided in Example 1 of the present invention;
[0043] Figure 2 This is a diagram of the internal structure of a computer device. DETAILED DESCRIPTION
[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0045] The purpose of the present invention is to provide a method, device, medium and product for predicting carbon emissions in a target mine area, aiming to improve the prediction accuracy of carbon emissions in the target restoration area.
[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0047] Example 1
[0048] like Figure 1 As shown, the carbon emission prediction method for a target mine area in this embodiment includes:
[0049] Step S1: Acquire a current remote sensing image of the target mine area.
[0050] In practical applications, current remote sensing images can be obtained through open platforms such as the Digital Earth Open Platform, or by using drones carrying scanning equipment to scan target areas of mines to obtain current remote sensing images.
[0051] Step S2: Determine the division results of various land types in the target mine area based on the current remote sensing image.
[0052] S2 specifically includes:
[0053] Step S21: Constructing a land type recognition model.
[0054] Step S22: inputting the current remote sensing image into the land type recognition model to obtain an initial land type recognition result.
[0055] S22 specifically includes:
[0056] Step S221: extracting an initial image from the current remote sensing image using a corresponding conformation equation.
[0057] Step S222: cropping the initial image to obtain an orthophoto image.
[0058] Step S223: Input the orthophoto image into the target neural network model to obtain an initial land type recognition result.
[0059] The target neural network model is a common neural network classification model. The initial land type recognition result includes the land type and the corresponding land area.
[0060] Step S23: performing similarity calculation on the initial land type recognition result and the corresponding standard template of each land type to obtain a similarity result.
[0061] The similarity calculation refers to the similarity calculation of land types, and determines the similarity between the land type obtained by the land type recognition model and the standard land type.
[0062] Furthermore, the remote sensing image of the initial land type identification result is segmented, and each segmented land block is numbered, resulting in a total of n land blocks. The first quantitative feature of each land block in the initial land type identification result and the second quantitative feature of the corresponding land standard type are extracted, and the similarity is calculated. The calculation formula is as follows:
[0063]
[0064] Among them, S is the similarity; t i is the first quantitative feature of the i-th land parcel in the initial land type identification result of land type j; t j is the second quantitative characteristic of the land with the corresponding land type being land standard type j.
[0065] In practical applications, the standard templates for each land type refer to the characteristics of land types such as cultivated land, forest land, grassland, construction land, water area and unused land.
[0066] Step S24: When the similarity result is greater than or equal to a preset similarity threshold, the initial land type recognition result is used as a division result.
[0067] In practical applications, when the similarity result is greater than or equal to a preset similarity threshold, the initial land type recognition result of the land type is correct.
[0068] Step S25: When the similarity result is less than the preset similarity threshold, the initial land type recognition result corresponding to the similarity result is input into the land type recognition model to obtain a corresponding secondary recognition result.
[0069] In actual applications, if the similarity result is less than the preset similarity threshold, it is considered that the initial land type recognition result of the land type may be problematic and needs further confirmation. Therefore, the remote sensing image corresponding to the initial land type recognition result is input into the land type recognition model to obtain the corresponding secondary recognition result.
[0070] Step S26: Calculate the similarity between the secondary recognition result and the corresponding standard template of each land type to obtain a secondary similarity result.
[0071] This calculation method is similar to the calculation method in step S23 and will not be described again here.
[0072] Step S27: When the secondary similarity result is greater than or equal to a preset similarity threshold, the secondary recognition result is used as a segmentation result.
[0073] Step S28: When the secondary similarity result is less than the preset similarity threshold, a final land type corresponding to the secondary recognition result is obtained through investigation, and the final land type is used as the division result.
[0074] In actual applications, obtaining the final land type corresponding to the secondary identification result through investigation refers to the review of the identification results by technical personnel, that is, technical personnel verify the land type identification results on site and determine the final land type through investigation, inquiry, and evidence collection.
[0075] Step S3: Obtain characteristic data corresponding to the classification results of each land type within a statistical time period; the characteristic data include land area, temperature, humidity, wind speed and atmospheric pressure.
[0076] Step S4: normalizing the feature data to obtain normalized feature data.
[0077] Step S5: Input the normalized characteristic data into the carbon emission prediction model to obtain the carbon emission prediction results of each land type in the target area of the mine within the statistical time period; the carbon emission prediction model is obtained by training the neural network with a training data set; the training data set is the normalized characteristic data samples of each land type and the corresponding carbon emission result samples within a preset time period.
[0078] As a specific implementation, the neural network is a long short-term memory network.
[0079] As a specific implementation, the neural network is a CNN-GRU network that introduces an attention mechanism; the CNN-GRU network includes a CNN network and a GRU network connected in sequence.
[0080] The CNN network includes a convolutional layer, a pooling layer and a fully connected layer.
[0081] The GRU network includes a GRU network layer and a Dense layer.
[0082] The fully connected layer of the CNN network is connected to the GRU network layer of the GRU network.
[0083] An attention mechanism is introduced before the convolutional layer of the CNN network, and the attention weight distribution value is updated through the output of the GRU network layer.
[0084] Step S6: Adding the carbon emission prediction results of each land type in the target mine area within the statistical time period to obtain the carbon emission prediction result of the target mine area within the statistical time period.
[0085] Before executing step S2, the method for predicting carbon emissions in a target mine area provided by the present invention further includes:
[0086] The current remote sensing image is preprocessed; the preprocessing operations include color correction, denoising and image enhancement.
[0087] As a specific implementation method, the process of constructing the carbon emission prediction model includes:
[0088] The neural network is trained with the normalized characteristic data samples of each land type as input and the corresponding carbon emission result samples within a preset time period as output to obtain a carbon emission prediction model corresponding to each land type.
[0089] The present invention has the following advantages:
[0090] Before inputting data into the GRU network, the input space features are extracted through the CNN network, which improves the prediction model's utilization of historical carbon emissions data. In addition, the attention mechanism is combined with the CNN-GRU model, which not only improves the model's utilization of historical information, but also increases the difference in feature weights, thereby improving the prediction accuracy of the carbon emission prediction model.
[0091] Example 2
[0092] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the method for predicting carbon emissions in a target mine area in Example 1.
[0093] Example 3
[0094] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for predicting carbon emissions in a target mine area in Example 1.
[0095] Example 4
[0096] A computer program product includes a computer program, which, when executed by a processor, implements the method for predicting carbon emissions in a target mine area in Example 1.
[0097] Example 5
[0098] A computer device, which may be a database, may have an internal structure as shown in FIG. Figure 2As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store pending transactions. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements the carbon emissions prediction method for the target area of the mine in Example 1.
[0099] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.
[0100] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, etc., but are not limited to these.
[0101] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0102] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.
Claims
1. A method for predicting carbon emissions in a target area of a mine, characterized in that: The method comprises: Acquiring a current remote sensing image of the target area of the mine; Determining the division results of various land types in the target mine area according to the current remote sensing image; Acquire characteristic data corresponding to the division results of each land type within a statistical time period; the characteristic data includes land area, temperature, humidity, wind speed and atmospheric pressure; Normalizing the feature data to obtain normalized feature data; The normalized characteristic data is input into the carbon emission prediction model to obtain the carbon emission prediction results of each land type in the target area of the mine within the statistical time period; the carbon emission prediction model is obtained by training the neural network with a training data set; the training data set is the normalized characteristic data samples of each land type and the corresponding carbon emission result samples within a preset time period; The carbon emission prediction results of each land type in the target mine area within the statistical time period are added together to obtain the carbon emission prediction results of the target mine area within the statistical time period.
2. The method for predicting carbon emissions in a target mine area according to claim 1, characterized in that: Before executing the step of "determining the division results of various land types in the target mine area according to the current remote sensing image", it also includes: The current remote sensing image is preprocessed; the preprocessing operation includes color correction, denoising and image enhancement.
3. The method for predicting carbon emissions in a target mine area according to claim 1, characterized in that: Determine the division results of various land types in the target mine area according to the current remote sensing image, specifically including: Constructing a land type identification model; Inputting the current remote sensing image into the land type recognition model to obtain an initial land type recognition result; Calculate the similarity between the initial land type recognition result and the corresponding standard template of each land type to obtain a similarity result; When the similarity result is greater than or equal to a preset similarity threshold, the initial land type recognition result is used as a division result; When the similarity result is less than the preset similarity threshold, the initial land type recognition result corresponding to the similarity result is input into the land type recognition model to obtain a corresponding secondary recognition result; Calculate the similarity between the secondary recognition result and the corresponding standard template of each land type to obtain a secondary similarity result; When the secondary similarity result is greater than or equal to a preset similarity threshold, the secondary recognition result is used as a segmentation result; When the secondary similarity result is less than the preset similarity threshold, the final land type corresponding to the secondary recognition result is obtained through investigation, and the final land type is used as the division result.
4. The method for predicting carbon emissions in a target mine area according to claim 3, characterized in that: Inputting the current remote sensing image into the land type recognition model to obtain an initial land type recognition result specifically includes: Extracting an initial image map from the current remote sensing image using a corresponding conformation equation; Cropping the initial image to obtain an orthophoto image; The orthophoto map is input into the target neural network model to obtain an initial land type recognition result.
5. The method for predicting carbon emissions in a target mine area according to claim 1, characterized in that: The neural network is a long short-term memory network.
6. The method for predicting carbon emissions in a target mine area according to claim 1, characterized in that: The neural network is a CNN-GRU network that introduces an attention mechanism; the CNN-GRU network includes a CNN network and a GRU network that are connected in sequence; The CNN network includes a convolutional layer, a pooling layer and a fully connected layer; The GRU network includes a GRU network layer and a Dense layer; The fully connected layer of the CNN network is connected to the GRU network layer of the GRU network; An attention mechanism is introduced before the convolutional layer of the CNN network, and the attention weight distribution value is updated through the output of the GRU network layer.
7. The method for predicting carbon emissions in a target mine area according to claim 1, characterized in that: The construction process of the carbon emission prediction model includes: The neural network is trained with the normalized characteristic data samples of each land type as input and the corresponding carbon emission result samples within a preset time period as output to obtain a carbon emission prediction model corresponding to each land type.
8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for predicting carbon emissions in a target area of a mine as described in any one of claims 1 to 7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting carbon emissions in a target area of a mine as described in any one of claims 1 to 7 is implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for predicting carbon emissions in a target area of a mine as described in any one of claims 1 to 7 is implemented.