Single-tree carbon sequestration capacity prediction method, device and acquisition equipment

Abstract

The present invention relates to the technical field of carbon sequestration in standing trees, and particularly to a method, device and acquisition equipment for predicting the carbon sequestration capacity of a single standing tree. The method includes: obtaining the stem moisture data and microenvironment parameter data of the single standing tree to be predicted; preprocessing the stem moisture data and microenvironment parameter data to obtain principal component data; inputting the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain the prediction result of the carbon sequestration capacity of the single standing tree to be predicted. The purpose of the present invention is to solve the problems in the prior art that predicting the carbon sequestration capacity of a single standing tree is likely to cause damage to plants, with large prediction errors, resulting in low accuracy of predicting the carbon sequestration capacity of a single standing tree.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon sequestration by standing trees, and particularly to a method, a device, and a collection device for predicting the carbon sequestration capacity of a single standing tree. Background Art

[0002] The greenhouse effect and environmental problems caused by the rising atmospheric carbon dioxide concentration are becoming increasingly prominent. Fixing and sequestering carbon dioxide through ecosystems formed by standing trees, that is, the carbon sequestration of a single standing tree is mainly the process of fixing organic matter through photosynthesis, subtracting its own respiratory consumption, and obtaining the net increment required for growth, development, and reproduction. This process is not only related to the microenvironmental parameters of the area where the standing tree is located, but also closely related to the internal water transport and other physiological activities of the tree. Predicting the carbon sequestration capacity of a single standing tree is of great significance for clarifying and understanding plant growth, development, and carbon sequestration by forest trees, and then guiding carbon sink afforestation from small to large.

[0003] Currently, in the research on the carbon sequestration capacity of forest trees, existing methods include the biomass method, the eddy covariance method, and the remote sensing estimation method, etc. Among them, the biomass method obtains measured data through a large number of field surveys of felled forest trees and diameter at breast height measurement for each tree, and constructs a model between factors such as diameter at breast height, tree height, and terrain and biomass; however, the construction time of the model is long, and the growth and carbon sink accumulation of forest trees are comprehensively determined by multiple factors such as tree species, age, growth vigor, internal water storage, site conditions, and climate. Using the existing constructed model to detect the carbon sequestration capacity of a single standing tree under different spatio-temporal conditions will result in large errors. The equipment used in the eddy covariance method is relatively expensive, and there are high requirements for terrain factors. During prediction, it is easy to cause damage to standing trees. The models constructed by the remote sensing estimation method are usually for estimating the forest tree carbon sink on a large scale such as regions, countries, and even the world, and the estimated results have large errors and are not universal. Summary of the Invention

[0004] The present invention provides a method for predicting the carbon sequestration capacity of a single standing tree to solve the problems in the prior art that predicting the carbon sequestration capacity of forest trees is easy to cause damage to plants, the prediction error is large, and the accuracy of predicting the carbon sequestration capacity of a single standing tree is low.

[0005] The present invention provides a method for predicting the carbon sequestration capacity of a single standing tree, including:

[0006] Obtaining the stem water data and microenvironmental parameter data of the single standing tree to be predicted;

[0007] Preprocessing the stem water data and microenvironmental parameter data to obtain principal component component data;

[0008] Inputting the principal component component data into a preset carbon sequestration capacity prediction model for prediction to obtain a prediction result of the carbon sequestration capacity of the single standing tree to be predicted;

[0009] Among them, the preset carbon sequestration capacity prediction model is obtained by training based on a neural network model, with the sample data of individual standing tree samples as input and the true value of the carbon sequestration capacity data corresponding to the sample data as labels. The sample data includes: the principal component component data after preprocessing of the stem moisture sample data and the microenvironment parameter sample data.

[0010] According to a method for predicting the carbon sequestration capacity of individual standing trees provided by the present invention, preprocessing the stem moisture data and the microenvironment parameter data to obtain the principal component component data, including:

[0011] Using a moving average filtering method to perform filtering processing on the stem moisture data and the microenvironment parameter data to obtain the filtered stem moisture data and the microenvironment parameter data;

[0012] Using the principal component analysis method to perform principal component analysis processing on the filtered stem moisture data and the microenvironment parameter data to obtain the principal component component data.

[0013] According to a method for predicting the carbon sequestration capacity of individual standing trees provided by the present invention, after the step of using a moving average filtering method to perform filtering processing on the stem moisture data and the microenvironment parameter data to obtain the filtered stem moisture data and the microenvironment parameter data, the following steps are further included:

[0014] Performing standardization processing on the filtered stem moisture data and the microenvironment parameter data to obtain the standardized stem moisture data and the microenvironment parameter data.

[0015] According to a method for predicting the carbon sequestration capacity of individual standing trees provided by the present invention, before the step that the preset carbon sequestration capacity prediction model is obtained by training based on a neural network model, with the sample data of individual standing tree samples as input and the true value of the carbon sequestration capacity data corresponding to the sample data as labels, and the sample data includes the principal component component data after preprocessing of the stem moisture sample data and the microenvironment parameter sample data, the step of training the neural network model is further included, including:

[0016] Obtaining the stem moisture sample data, the microenvironment parameter sample data of the individual standing tree samples, and the true value of the carbon sequestration capacity data corresponding to the stem moisture sample data and the microenvironment parameter sample data of the individual standing tree samples;

[0017] Preprocessing the stem moisture sample data and the microenvironment parameter sample data of the individual standing tree samples to obtain the principal component component data of the individual standing tree samples;

[0018] Dividing the sample set composed of the principal component component data of the individual standing tree samples and the true value of the carbon sequestration capacity data into a training set and a test set;

[0019] Taking the principal component component data of the individual tree samples in the training set as the input and the true value of the carbon sequestration capacity data in the training set as the label, training the neural network model to obtain the trained neural network model for the carbon sequestration capacity of individual trees;

[0020] Inputting the principal component component data of the individual tree samples in the test set into the trained neural network model for the carbon sequestration capacity of individual trees for detection to obtain the prediction result of the carbon sequestration capacity of individual trees in the test set;

[0021] Using the true value of the carbon sequestration capacity data in the test set to verify the prediction result of the carbon sequestration capacity of individual trees in the test set. After passing the verification, a preset carbon sequestration capacity prediction model is obtained.

[0022] According to a method for predicting the carbon sequestration capacity of individual trees provided by the present invention, the microenvironment parameter data includes:

[0023] Air temperature data, relative humidity data, soil temperature data, soil humidity data, stem temperature data, photosynthetically active radiation data, light intensity data, saturation vapor pressure deficit data, atmospheric pressure, and carbon dioxide concentration.

[0024] The present invention also provides a device for predicting the carbon sequestration capacity of individual trees, including:

[0025] A receiving module for obtaining the stem moisture data and microenvironment parameter data of the individual tree to be predicted;

[0026] A processing module for preprocessing the stem moisture data and microenvironment parameter data to obtain the principal component component data;

[0027] A prediction module for inputting the principal component component data into a preset carbon sequestration capacity prediction model for prediction to obtain the prediction result of the carbon sequestration capacity of the individual tree to be predicted;

[0028] Wherein, the preset carbon sequestration capacity prediction model is obtained by training based on a neural network model with the sample data of individual tree samples as the input and the true value of the carbon sequestration capacity data corresponding to the sample data as the label. The sample data includes the principal component component data obtained by preprocessing the stem moisture sample data and microenvironment parameter sample data.

[0029] According to a device for predicting the carbon sequestration capacity of individual trees provided by the present invention, the processing module includes a filtering module and an analysis module;

[0030] The filtering module is used to perform filtering processing on the stem moisture data and microenvironment parameter data by using a moving average filtering method to obtain the filtered stem moisture data and microenvironment parameter data;

[0031] The analysis module is used to perform principal component analysis on the stem moisture data and microenvironment parameter data after filtering processing by using the principal component analysis method, so as to obtain principal component component data.

[0032] The present invention also provides a collection device for predicting the carbon sequestration capacity of a single standing tree. The collection device is used to obtain the stem moisture sample data, microenvironment parameter sample data and the true value of the carbon sequestration capacity data in the above-mentioned method for predicting the carbon sequestration capacity of a single standing tree, including:

[0033] An assimilation box for obtaining the parameters for calculating the true value of the carbon sequestration capacity data;

[0034] A box cover for opening or closing the assimilation box;

[0035] A push rod connected to the assimilation box. In the working state, it is used to push the box cover to open the assimilation box;

[0036] A microenvironment parameter sensor for collecting microenvironment parameter data;

[0037] A stem moisture sensor for collecting stem moisture data; and

[0038] A data collector electrically connected to the microenvironment parameter sensor and the stem moisture sensor, for receiving the microenvironment parameter data and the stem moisture data of a single standing tree, and for calculating the true value of the carbon sequestration capacity data according to the parameters.

[0039] According to the provided collection device, it further includes: a fan, a conduit and a controller;

[0040] The fan is installed inside the assimilation box or at the bottom of the box cover;

[0041] The conduit is installed on the box cover and communicated with the assimilation box;

[0042] The controller is electrically connected to the push rod and is used to control the push rod to push the box cover.

[0043] The present invention also provides an electronic device, including a memory, a processor and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements any one of the above-mentioned methods for predicting the carbon sequestration capacity of a single standing tree.

[0044] A method, device and acquisition device for predicting the carbon sequestration capacity of a single standing tree provided by the present invention. After obtaining the stem moisture data and microenvironment parameter data of the single standing tree to be predicted through the method for predicting the carbon sequestration capacity of a single standing tree, the principal component component data obtained after preprocessing the stem moisture data and microenvironment parameter data is input into a preset carbon sequestration capacity prediction model for prediction. The prediction is more convenient, and the carbon sequestration capacity of a single standing tree can be obtained more quickly and accurately. At the same time, during the prediction, only the stem moisture data and microenvironment parameter data of the single standing tree that are easy to measure are obtained, the data acquisition time is shorter, and the damage to plants can also be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0046] Figure 1 is a schematic flow chart of a method for predicting the carbon sequestration capacity of a single standing tree provided by the present invention;

[0047] Figure 2 is a schematic structural diagram of a device for predicting the carbon sequestration capacity of a single standing tree provided by the present invention;

[0048] Figure 3 is a schematic structural diagram of an acquisition device for predicting the carbon sequestration capacity of a single standing tree provided by the present invention;

[0049] Figure 4 is a schematic structural diagram of an electronic device provided by the present invention.

[0050] REFERENCE NUMERALS:

[0051] 210: receiving module; 220: processing module; 230: prediction module; 300 single standing tree; 310: assimilation box; 320: box cover; 330: push rod; 340: microenvironment parameter sensor; 350: stem moisture sensor; 360: fan; 370: conduit; 380: data collector; 390: controller. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0053] The following combines Figure 1 to describe a method for predicting the carbon sequestration capacity of a single standing tree according to the present invention, including the following steps:

[0054] S1. Obtain the stem water data and microenvironment parameter data of the single standing tree to be predicted. Specifically, obtain the stem water data and microenvironment parameter data of the single standing tree to be predicted collected by the collection device.

[0055] S2. Preprocess the stem water data and microenvironment parameter data to obtain the principal component data. In this embodiment, the principal component data is obtained by performing principal component analysis on these two types of data, namely the stem water data and each microenvironment data, removing the correlation between the data, reducing redundant information, and obtaining the main information of each data.

[0056] Specifically, the microenvironment parameter data includes:

[0057] Air temperature data, relative humidity data, soil temperature data, soil humidity data, stem temperature data, photosynthetically active radiation data, light intensity data, saturation vapor pressure deficit data, atmospheric pressure, and carbon dioxide concentration. Among them, the saturation vapor pressure deficit data is calculated from the air temperature data and relative humidity data.

[0058] S3. Input the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain the prediction result of the carbon sequestration capacity of the single standing tree to be predicted.

[0059] Among them, the preset carbon sequestration capacity prediction model is obtained by training based on a neural network model with the sample data of single standing tree samples as input and the true value of the carbon sequestration capacity data corresponding to the sample data as the label. The sample data includes the principal component data obtained by preprocessing the stem water sample data and microenvironment parameter sample data.

[0060] Specifically, the stem water sample data, microenvironment parameter sample data, and label of the single standing tree samples are all obtained by the collection device. Preferably, the stem water sample data, microenvironment parameter sample data, and label of the single standing tree samples are obtained by the same collection device.

[0061] The method for predicting the carbon sequestration capacity of a single standing tree in the present invention, after obtaining the stem water data and microenvironment parameter data of the single standing tree to be predicted, inputs the principal component component data obtained after preprocessing the stem water data and microenvironment parameter data into a preset carbon sequestration capacity prediction model for prediction. The prediction is more convenient and improves the accuracy of the carbon sequestration capacity of a single standing tree. At the same time, during prediction, only by obtaining the stem water data and microenvironment parameter data of the single standing tree that are easy to measure, and with less data obtained, the prediction time is shorter, reducing the felling of single standing trees and the use of assimilation chambers, thereby being able to reduce damage to plants.

[0062] In this embodiment, step S2: Preprocess the stem water data and microenvironment parameter data to obtain principal component component data, including:

[0063] Adopt the moving average filtering method to filter the stem water data and microenvironment parameter data to obtain the filtered stem water data and microenvironment parameter data. This realizes the removal of burrs and noise in the stem water data and microenvironment parameter data, avoiding the interference of burrs and noise.

[0064] At the same time, after the step of adopting the moving average filtering method to filter the stem water data and microenvironment parameter data to obtain the filtered stem water data and microenvironment parameter data, the following is also included:

[0065] Perform standardization processing on the processed stem water data and microenvironment parameter data to obtain the standardized stem water data and microenvironment parameter data.

[0066] Since the obtained microenvironment parameter data and the stem water data of the single standing tree are at different orders of magnitude, performing standardization processing on the stem water data and microenvironment parameter data enables each data to have the same dimension, avoiding the smaller order of magnitude data being overwhelmed by the larger data. In this embodiment, first assume that the original data, that is, the stem water data and microenvironment parameter data, have n samples, and each sample has p data, forming an n×p-dimensional matrix X, as shown in Equation (1):

[0067]

[0068] Perform standardization processing on the matrix X using the Z-SCORE method to obtain the standardized matrix Z. Among them:

[0069]

[0070] In Equation (2), X ij refers to a value in the i-th row and j-th column of the X matrix, and Z ijis the value of the i-th row and j-th column in the standardized matrix Z after standardization processing (i = 1, 2, …, n; j = 1, 2, …, p) is the arithmetic mean of the j-th variable, σ j is the standard deviation of the j-th variable.

[0071] Using the principal component analysis method, perform principal component analysis processing on the standardized stem moisture data and microenvironment parameter data to obtain principal component component data. Using the dimensionality reduction method to convert the microenvironment parameter data and stem moisture data into mutually independent principal component components can remove the correlation between the microenvironment parameter data and stem moisture data, making it easier to obtain the main information and improving the data analysis efficiency.

[0072] Specifically, in step S3: The preset carbon sequestration capacity prediction model takes the sample data of a single-tree sample as input and the true value of the carbon sequestration capacity data corresponding to the sample data as the label, and is obtained after training based on the neural network model. Before the step of the sample data including the principal component component data obtained by preprocessing the stem moisture sample data and microenvironment parameter sample data, there is also a step of training the neural network model, including:

[0073] Step 1, obtain the stem moisture data and microenvironment parameter data of a single-tree sample and the carbon sequestration capacity data corresponding to the stem moisture data and microenvironment parameter data of the single-tree sample.

[0074] The true value of the carbon sequestration capacity data of the single-tree is the net carbon dioxide (CO2) exchange amount of the single-tree, and is specifically calculated by the following formula (3):

[0075]

[0076] where Fc is the true value of the carbon sequestration capacity data of the single-tree, V is the volume of the assimilation chamber, P av is the average atmospheric pressure in the assimilation chamber during the detection period, W av is the average water vapor concentration in the assimilation chamber during the detection period, R is the ideal gas constant, S is the bottom area of the assimilation chamber, T av is the average temperature in the assimilation chamber during the detection period, and dc / dt is the carbon dioxide concentration change rate during the detection period.

[0077] Specifically, Pav is obtained by the atmospheric pressure sensor, Wav is the average water vapor concentration during the detection period, and is first obtained by the air temperature sensor and humidity sensor for the temperature T and humidity RH, and then calculated by formula (4):

[0078] Wav = 6.1078 * exp((17.269 * T) / (273.15 + T - 35.86)) * RH / 100 (4).

[0079] R is the ideal gas constant, 8.314 J mol -1 K -1 。

[0080] dc / dt is the rate of change of carbon dioxide concentration. For example, if the concentration is 500 at the start of the detection and becomes 480 after a period of detection, then the rate of change dc / dt = (500 - 480) / detection time.

[0081] Step 2: Preprocess the stem moisture sample data and microenvironment parameter sample data of the single-tree sample to obtain the principal component component data of the single-tree sample.

[0082] Step 3: Divide the sample set composed of the principal component component data of the single-tree sample and the true value of the carbon sequestration capacity data into a training set and a test set. Among them, the ratio of the training set to the test set is 3:1.

[0083] Step 4: Use the principal component component data of the single-tree sample in the training set as the input and the true value of the carbon sequestration capacity data in the training set as the label to train the neural network model, and obtain the trained neural network model for the carbon sequestration capacity of single trees. In this step, the neural network model selects the BP neural network model optimized by the genetic algorithm (GA-BP neural network model). The GA-BP neural network model is a BP neural network optimized by the genetic algorithm. After operations such as selection, crossover, and mutation, it can effectively avoid the possibility of the network falling into local extrema. Use the optimal solution obtained by the genetic algorithm as the initial weights and thresholds of the BP neural network to achieve accurate prediction of the carbon sequestration capacity of single trees.

[0084] Step 5: Input the principal component component data of the single-tree sample in the test set into the trained neural network model for the carbon sequestration capacity of single trees for detection, and obtain the prediction results of the carbon sequestration capacity of single trees in the test set.

[0085] Step 6: Use the true value of the carbon sequestration capacity data in the test set to verify the prediction results of the carbon sequestration capacity of single trees in the test set, that is, use the true value of the carbon sequestration capacity data in the test set to compare with the prediction results of the carbon sequestration capacity of single trees in the test set to verify the accuracy of the trained neural network model for the carbon sequestration capacity of single trees. After passing the verification, obtain the preset carbon sequestration capacity prediction model.

[0086] Through the preset carbon sequestration capacity prediction model, input the principal component component data of the single tree to be predicted into the preset carbon sequestration capacity prediction model for prediction, to achieve a more rapid and accurate prediction of the carbon sequestration capacity of single trees, and it can also avoid putting the single tree to be predicted into the assimilation chamber for detection, reducing damage to the single tree.

[0087] The single-tree carbon sequestration capacity prediction device provided by the present invention will be described below. The single-tree carbon sequestration capacity prediction device described below can be correspondingly referred to the single-tree carbon sequestration capacity prediction method described above.

[0088] Please refer to Figure 2 , a single-tree carbon sequestration capacity prediction device, including a receiving module 210, a processing module 220, and a prediction module 230.

[0089] The receiving module 210 is used to obtain the stem moisture data and microenvironment parameter data of the single tree to be predicted.

[0090] The processing module 220 is used to preprocess the stem moisture data and microenvironment parameter data to obtain the principal component data.

[0091] In this embodiment, the processing module 220 includes a filtering module and an analysis module.

[0092] The filtering module is used to filter the stem moisture data and microenvironment parameter data by using the moving average filtering method to obtain the filtered stem moisture data and microenvironment parameter data. To remove the burrs and noises in the stem moisture data and microenvironment parameter data.

[0093] The analysis module is used to perform principal component analysis on the processed stem moisture data and microenvironment parameter data by using the principal component analysis method to obtain the principal component data. To remove the correlation between the stem moisture data and microenvironment parameter data, it is easier to obtain the main information and improve the data analysis efficiency.

[0094] The prediction module 230 is used to input the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain the prediction result of the carbon sequestration capacity of the single tree to be predicted.

[0095] Among them, the preset carbon sequestration capacity prediction model takes the sample data of the single-tree sample as the input, and the true value of the carbon sequestration capacity data corresponding to the sample data as the label, and is obtained after training based on the neural network model. The sample data includes: the principal component data after preprocessing of the stem moisture sample data and microenvironment parameter sample data.

[0096] In the single-tree carbon sequestration capacity prediction device of the present invention, the receiving module obtains the stem moisture data and microenvironment parameter data of the single tree to be predicted. After the processing module preprocesses the stem moisture data and microenvironment parameter data, the prediction module inputs the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain the prediction result of the carbon sequestration capacity of the single tree to be predicted. Thus, the carbon sequestration capacity of a single tree can be obtained more quickly and accurately.

[0097] The following will Figure 3 describe a collection device for predicting the carbon sequestration capacity of a single standing tree according to the present invention.

[0098] The collection device in this embodiment is used to obtain the stem moisture data, microenvironment parameter data and the label of a single standing tree in the above single standing tree carbon sequestration capacity prediction method.

[0099] The collection device includes an assimilation box body 310, a box cover 320, a push rod 330, a microenvironment parameter sensor 340, a stem moisture sensor 350 and a data collector 380. Among them, the assimilation box body 310 is made of acrylic plates with a light transmittance of 92% and a thickness of 3 mm, and the opening of the assimilation box body is opened at its top. The assimilation box body 310 is 1 meter high, and the upper and lower bottom surfaces are 50 cm wide and 50 cm long. The length, width and height of the assimilation box body are selected according to the actual situation of the single standing tree 300. Among them, the assimilation box body 310 is used to collect the parameters for calculating the true value of the carbon sequestration capacity data, that is, the assimilation box body 310 has a parameter sensor, and the parameter sensor is used to obtain the parameters for the true value of the carbon sequestration capacity data.

[0100] Specifically, the box cover 320 is installed at the top of the assimilation box body through a hinge, and the box cover 320 is used to open or close the assimilation box body 310. Among them, a compressible sealing strip is provided on the box cover 320 so that when the box cover 320 is closed on the assimilation box body, the assimilation box body can be sealed.

[0101] The push rod 330 is connected to the assimilation box body 310, and in the working state, it is used to push the box cover 320 to open the assimilation box body 310. In this embodiment, the push rod 330 is an electric push rod, and the electric push rod is placed inside the box, and the electric push rod is used to push the box cover to open the assimilation box body.

[0102] The microenvironment parameter sensor 340 is installed inside the assimilation box body 310 and is used to collect microenvironment parameter data. The microenvironment parameter sensor 340 can be used as a parameter sensor in the assimilation box body 310. That is, the microenvironment parameter sensor 340 is used to obtain the air temperature, relative humidity, soil temperature, soil humidity, stem temperature, photosynthetically active radiation, light intensity, saturation vapor pressure difference, atmospheric pressure and carbon dioxide concentration inside the assimilation box body 310.

[0103] The stem moisture sensor 350 is installed inside the assimilation box body 310 and is used to collect stem moisture data. In this embodiment, the stem moisture sensor 350 adopts a bimetallic ring structure sensor, and the stem moisture sensor 350 is installed at the stem of the single standing tree 300 through a hoop to obtain the stem moisture data of the single standing tree.

[0104] A data collector 380 is installed on one side of the assimilation box body 310 and is electrically connected to the microenvironment parameter sensor 340 and the stem moisture sensor 350. It is used to receive the microenvironment parameter data 340 and the stem moisture data of a single standing tree 350, and is used to calculate the true value of the carbon sequestration capacity data according to the parameters. Among them, the data collector 380 calculates the true value of the carbon sequestration capacity data according to the parameters through the above formula (2).

[0105] When the acquisition device is in use, open the box cover, place a single standing tree in the assimilation box body, and tightly wrap the tree surface soil when collecting data of the single standing tree to eliminate the influence of soil respiration. Control the electric push rod to descend, and the box cover covers the assimilation box body under its own gravity or cover the box cover on the assimilation box body, and the microenvironment parameter sensor and the stem moisture sensor start to collect, so as to realize the collection of microenvironment parameter data and stem moisture data. Preferably, the collection time of the microenvironment parameter sensor and the stem moisture sensor is 5 minutes.

[0106] At the same time, in order to better collect data of a single standing tree, in this embodiment, the acquisition device further includes: a fan 360, a conduit 370 and a controller 390.

[0107] The fan 360 is installed inside the assimilation box body 310 or at the bottom end of the box cover. In this embodiment, two fans 360 are selected. Two fans 360 with a diameter of 15 cm are installed along the diagonal direction of the assimilation box body 310. One fan 360 is installed at the bottom end inside the assimilation box body 310, and the other fan 360 is installed at the bottom end of the box cover. The fan 360 is used to stir the gas inside the assimilation box body 310 when the box cover 320 is closed to realize uniform gas distribution inside the assimilation box body 310.

[0108] The conduit 370 is installed on the box cover 320 and communicates with the assimilation box body 310. Among them, the conduit 370 is a Teflon tube, and a Teflon tube about 1 m long is installed on the box cover 320 to keep the atmospheric pressure inside and outside the assimilation box body balanced during operation.

[0109] The controller 380 is installed on one side of the assimilation box body 310. The controller 390 is electrically connected to the push rod 330 and is used to control the push rod 330 to push the box cover 320. The controller 390 uses a development board with a model STM32F103ZET6 single-chip microcomputer as the core to control the electric push rod to push the box cover 320 within a certain time, so that the box cover 320 can rise or fall.

[0110] Figure 4 Illustrates a schematic physical structure diagram of an electronic device, such as Figure 4As shown in the figure, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. Among them, the processor 410, the communications interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logical instructions in the memory 430 to execute a method for predicting the carbon sequestration capacity of a single standing tree, including:

[0111] S1. Obtain the stem moisture data and microenvironment parameter data of the single standing tree to be predicted;

[0112] S2. Preprocess the stem moisture data and microenvironment parameter data to obtain the principal component data;

[0113] S3. Input the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain the prediction result of the carbon sequestration capacity of the single standing tree to be predicted;

[0114] Among them, the preset carbon sequestration capacity prediction model takes the principal component data of the single standing tree sample as the input and the carbon sequestration capacity data corresponding to the principal component data of the single standing tree sample as the output, and is obtained after training based on a neural network model.

[0115] In addition, when the logical instructions in the above-mentioned memory 430 are implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention essentially or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0116] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for predicting the carbon sequestration capacity of a single standing tree provided by the above-mentioned various methods. The method includes:

[0117] S1. Obtain the stem moisture data and microenvironment parameter data of the single standing tree to be predicted;

[0118] S2. Preprocess the stem moisture data and microenvironment parameter data to obtain principal component data;

[0119] S3. Input the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain a prediction result of the carbon sequestration capacity of the single standing tree to be predicted;

[0120] Wherein, the preset carbon sequestration capacity prediction model takes the principal component data of the single standing tree sample as input, takes the carbon sequestration capacity data corresponding to the principal component data of the single standing tree sample as output, and is obtained after training based on a neural network model.

[0121] On the other hand, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for predicting the carbon sequestration capacity of a single standing tree provided by the above-mentioned various methods. The method includes:

[0122] S1. Obtain the stem moisture data and microenvironment parameter data of the single standing tree to be predicted;

[0123] S2. Preprocess the stem moisture data and microenvironment parameter data to obtain principal component data;

[0124] S3. Input the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain a prediction result of the carbon sequestration capacity of the single standing tree to be predicted;

[0125] Wherein, the preset carbon sequestration capacity prediction model takes the principal component data of the single standing tree sample as input, takes the carbon sequestration capacity data corresponding to the principal component data of the single standing tree sample as output, and is obtained after training based on a neural network model.

[0126] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0127] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the carbon sequestration capacity of a single standing tree, characterized in that Including: Obtain the stem moisture data and microenvironment parameter data of the single tree to be predicted; Preprocess the stem moisture data and microenvironment parameter data to obtain the principal component data; Input the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain the prediction result of the carbon sequestration capacity of the single tree to be predicted; Among them, the preset carbon sequestration capacity prediction model takes the sample data of the single tree sample as input and the true value of the carbon sequestration capacity data corresponding to the sample data as the label, and is obtained after training based on the neural network model. The sample data includes: the principal component data after preprocessing the stem moisture sample data and microenvironment parameter sample data.

2. The method for predicting the carbon sequestration capacity of a single standing tree according to claim 1, wherein Preprocessing the stem moisture data and microenvironment parameter data to obtain the principal component data includes: Adopt the moving average filtering method to filter the stem moisture data and microenvironment parameter data to obtain the filtered stem moisture data and microenvironment parameter data; Adopt the principal component analysis method to perform principal component analysis on the filtered stem moisture data and microenvironment parameter data to obtain the principal component data.

3. The method for predicting the carbon sequestration capacity of a single standing tree according to claim 2, wherein After the step of adopting the moving average filtering method to filter the stem moisture data and microenvironment parameter data to obtain the filtered stem moisture data and microenvironment parameter data, it further includes: Perform standardization processing on the filtered stem moisture data and microenvironment parameter data to obtain the standardized stem moisture data and microenvironment parameter data.

4. The method for predicting the carbon sequestration capacity of a single standing tree according to any one of claims 1 to 3, characterized in that, Before the step that the preset carbon sequestration capacity prediction model takes the sample data of the single tree sample as input and the true value of the carbon sequestration capacity data corresponding to the sample data as the label, and is obtained after training based on the neural network model. The sample data includes: the principal component data after preprocessing the stem moisture sample data and microenvironment parameter sample data, it further includes the step of training the neural network model, including: Obtain the stem moisture sample data, microenvironment parameter sample data of the single tree sample, and the true value of the carbon sequestration capacity data corresponding to the stem moisture sample data and microenvironment parameter sample data of the single tree sample; Preprocess the stem moisture sample data and microenvironment parameter sample data of the single tree sample to obtain the principal component data of the single tree sample; Divide the sample set composed of the principal component data of the single tree sample and the true value of the carbon sequestration capacity data into a training set and a test set; Take the principal component data of the single tree sample in the training set as input and the true value of the carbon sequestration capacity data in the training set as the label to train the neural network model to obtain the trained neural network model of the carbon sequestration capacity of the single tree; Input the principal component data of the single tree sample in the test set into the trained neural network model of the carbon sequestration capacity of the single tree for detection to obtain the prediction result of the carbon sequestration capacity of the single tree in the test set; Use the true value of the carbon sequestration capacity data in the test set to verify the prediction result of the carbon sequestration capacity of the single tree in the test set. After passing the verification, obtain the preset carbon sequestration capacity prediction model.

5. The method for predicting the carbon sequestration capacity of a single standing tree according to claim 1, wherein The microenvironment parameter data includes: Air temperature data, relative humidity data, soil temperature data, soil moisture data, stem temperature data, photosynthetically active radiation data, light intensity data, saturation vapor pressure deficit data, atmospheric pressure, and carbon dioxide concentration.

6. A prediction device for the carbon sequestration capacity of a single standing tree, characterized in that, Including: A receiving module for obtaining the stem moisture data and microenvironment parameter data of a single standing tree to be predicted; A processing module for preprocessing the stem moisture data and microenvironment parameter data to obtain principal component data; A prediction module for inputting the principal component data into a preset carbon sequestration capacity prediction model for prediction to obtain a prediction result of the carbon sequestration capacity of the single standing tree to be predicted; Wherein, the preset carbon sequestration capacity prediction model is obtained by training based on a neural network model with the sample data of single standing tree samples as input and the true value of the carbon sequestration capacity data corresponding to the sample data as the label, and the sample data includes: the principal component data obtained by preprocessing the stem moisture sample data and microenvironment parameter sample data.

7. The single-tree carbon sequestration capacity prediction device according to claim 6, characterized in that The processing module includes a filtering module and an analysis module; The filtering module is used to perform filtering processing on the stem moisture data and microenvironment parameter data by using a moving average filtering method to obtain the filtered stem moisture data and microenvironment parameter data; The analysis module is used to perform principal component analysis processing on the filtered stem moisture data and microenvironment parameter data by using principal component analysis to obtain principal component data.

8. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, it implements the method for predicting the carbon sequestration capacity of a single standing tree according to any one of claims 1 to 5.

Images