Carbon sink nonlinear trend extraction system and method based on ensemble empirical mode decomposition

Abstract

The invention relates to the technical field of carbon sink trend extraction, in particular to an extraction method performed by a carbon sink nonlinear trend extraction system based on ensemble empirical mode decomposition, which comprises the following steps: S1, acquiring forest carbon sink time sequence data to form a forest carbon sink time sequence data set, the forest carbon sink time sequence data are vegetation coverage rate VCR, soil coverage rate SCR, vegetation carbon content Cveg, soil carbon content Coil and total carbon storage capacity Ctotal at different seasons and times. According to the method, the overall carbon sink capacity of the forest can be evaluated more accurately by monitoring the carbon storage capacity in the vegetation and the soil, and how the carbon sink is influenced by climatic factors can be better understood by monitoring the changes according to the fact that the carbon storage capacity of the vegetation and the soil possibly fluctuates along with seasonal changes, so that the carbon sink capacity of the forest can be evaluated more accurately. Meanwhile, in the process of monitoring the change of the carbon storage capacity in different seasons, a long-term change trend chart of the forest carbon sink is extracted according to the rule of the influence of the non-linear change of the carbon storage capacity of the vegetation and the soil on the forest carbon sink capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon sink trend extraction, and in particular to a carbon sink nonlinear trend extraction system and method based on ensemble empirical mode decomposition. Background Art

[0002] Carbon sink refers to the absorption and storage of carbon dioxide in the atmosphere by natural ecosystems such as forests, oceans, and soils through biochemical processes such as photosynthesis. It plays a key role in the global carbon cycle and has a positive effect on curbing climate warming. However, the efficiency of carbon sinks is affected by many factors such as climate change, land use, and human activities. Therefore, monitoring and analyzing its dynamic changes is crucial.

[0003] After searching, it was found that a forest carbon stock and carbon sink value monitoring system and a dynamic assessment method were disclosed in the Chinese patent document with patent number CN202311214927.0. The above scheme divided the forest area to be monitored according to vegetation type, soil type and climatic conditions, thereby reducing the error in carbon stock assessment between different types of forest communities; and introduced sample site surveys and information collection on the basis of remote sensing image data inversion, and collected vegetation conditions at different levels in the sample sites and environmental data of the sample sites, which made up for the defect that it was difficult to fully reflect the forest structure when inverting remote sensing image data, and took environmental factors into consideration to correct the output results of the assessment model that may cause errors and fluctuations, thereby improving the accuracy of carbon stock and carbon sink value assessment.

[0004] However, in the process of forest carbon sink monitoring, the forest carbon storage is a static state, which describes the amount of carbon that has accumulated in the forest carbon sink system at a certain moment. The carbon fixed by plants is not only retained in the plants, but may also be transferred to the soil and become part of the soil organic carbon. As the seasons change, it will affect the growth of plants and the state of the surrounding soil, resulting in nonlinear changes in vegetation carbon storage and soil carbon storage on different time scales, and they are interrelated. The existing assessment system cannot extract the seasonal change trend of forest carbon sinks based on the nonlinear changes between vegetation carbon storage and soil carbon storage. Summary of the invention

[0005] Technical issues solved

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides a carbon sink nonlinear trend extraction system and method based on ensemble empirical mode decomposition, which can solve the problem in the prior art that the existing forest carbon sink monitoring system cannot extract the forest carbon sink change trend line graph through the vegetation carbon storage and soil carbon storage in the seasonal changes during the monitoring process.

[0007] Technical Solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] The present invention provides an extraction method based on a carbon sink nonlinear trend extraction system of ensemble empirical mode decomposition, comprising the following steps:

[0010] S1: Collect forest carbon sink time series data to form a forest carbon sink time series data set. The forest carbon sink time series data includes vegetation coverage rate VCR, soil coverage rate SCR, vegetation carbon content C in different seasons. veg , soil carbon content C soil , total carbon storage C total ;

[0011] S2: Preprocess the forest carbon sink time series data set based on forest carbon storage theory to obtain forest carbon sink C in different seasons Carbon Sinkl (t);

[0012] S3: Using the ensemble empirical mode decomposition model to Carbon Sinkl (t) Extract and obtain data set X features (t) and using the ANN model to features (t) processing to extract the nonlinear trend of forest carbon sink, wherein the ensemble empirical mode decomposition model uses ensemble empirical mode decomposition (EEMD) to calculate the C Carbon Sinkl (t) Process and extract C Carbon Sinkl (t) and extract the feature X through IMFs features (t).

[0013] Furthermore, in S1, the data acquisition module collects forest carbon sink time series data through remote sensing monitoring technology and ground measurement technology.

[0014] Furthermore, in S2, C Carbon Sinkl The calculation process of (t) is:

[0015] S201: Based on forest carbon storage theory, obtain the total carbon storage C in static time total , specifically:

[0016] C total =(α*VCR*C veg )+(β*SCR*C soil )+[λ*VCR*SCR*(C veg +C soil )]

[0017] in:

[0018] α*VCR*Cveg Indicates the contribution of vegetation coverage to vegetation carbon storage; β*SCR*C soil represents the contribution of soil coverage to soil carbon storage; λ*VCR*SCR*(C veg +C soil ) represents the additional effect of the interaction between vegetation and soil coverage on total carbon storage; α is the influence coefficient of vegetation coverage on vegetation carbon storage; β is the influence coefficient of soil coverage on soil carbon storage; γ is the influence coefficient of the interaction between vegetation and soil on total carbon storage; α, β, and γ are determined by nonlinear regression analysis;

[0019] S202: Based on the total carbon storage C total Calculate the forest carbon sink C at different times during seasonal changes Carbon Sinkl (t), specifically:

[0020] C Carbon Sink (t) = C total (t)+∫[I(t)-O(t)]dt

[0021] Among them, I(t) is the carbon input rate that changes with time, O(t) is the carbon output rate that changes with time, and ∫[I(t)-O(t)]dt represents the cumulative integral of carbon input minus carbon output from the initial time to time t, that is, the total carbon increase or decrease.

[0022] Furthermore, in S3, the C Carbon Sinkl (t)

[0023] The extraction is performed, comprising the following steps:

[0024] S301: Towards C Carbon Sinkl Add white noise to (t):

[0025] C Carbon Sinkl(i) (t) = C Carbon Sinkl (t)+αn i (t)

[0026] S302: For each signal C that introduces white noise Carbon Sinkl(i) (t) Performing empirical mode decomposition (EMD);

[0027] S303: Extraction of intrinsic mode function:

[0028]

[0029] Among them, IMF ij (t) indicates that this is the jth mode function obtained by decomposition after adding white noise for the i-th time;

[0031] S304:Xfeatures (t) Extraction:

[0032] X features (t)=[mean(IMF ij (t)), std(IMF ij (t)), skewness(IMF ij (t)), kurtosis (IMF ij (t)), frequency(IMF ij (t))]

[0033] in,

[0034]

[0035] n is the number of data points and is the value of the ith data point.

[0036] Furthermore, in S3, the ANN model is used to analyze the X features (t) Processing to extract forest

[0037] The nonlinear trend of carbon sinks is as follows:

[0038] According to X features (t) Building ANN model:

[0039] h t =σ(W*X features (t)+b)

[0040] h t is the output of the hidden layer, W is the weight matrix, b is the bias, and σ is the activation function;

[0041] The training formula of the ANN model is:

[0042]

[0043] ΔW is the weight update, η is the learning rate, L is the loss function, is the partial derivative of the weight with respect to the loss function;

[0044] The evaluation formula of the ANN model is:

[0045]

[0046] MSE is the mean square error, which is the actual value, y i is the actual value, is the predicted value;

[0047] The parameter tuning formula is:

[0048]

[0049] Δη is the learning rate update, α is the learning rate adjustment coefficient, is the partial derivative of the loss function with respect to the learning rate;

[0050] y predicted =h T ·softmax(W O ·h T +b O )

[0051] h T is a vector representing the output of the last layer (hidden layer) of the neural network;

[0052] W O is the output layer weight matrix, connecting the hidden layer and the output layer;

[0053] b O is the bias vector of the output layer;

[0054] Softmax is an activation function that converts the output of the network into a probability distribution, where the value of each element represents the probability of the corresponding category;

[0055] y predicted for the final nonlinear trend extraction results.

[0056] A carbon sink nonlinear trend extraction system based on ensemble empirical mode decomposition, comprising a data acquisition module and a data processing module;

[0057] The data acquisition module is used to collect forest carbon sink time series data to form a forest carbon sink time series data set. The forest carbon sink time series data set includes vegetation coverage rate VCR, soil coverage rate SCR, vegetation carbon content C in different seasons. veg , soil carbon content C soil , total carbon storage C total ;

[0058] The data processing module is used to preprocess the forest carbon sink time series data set to obtain forest carbon sink C in different seasons. Carbon Sinkl (t) and extracting the forest carbon sink time series data set according to the pre-constructed ensemble empirical mode decomposition model to obtain a data set X features (t), according to the pre-built ANN model, features (t) Processing to extract nonlinear trends in forest carbon sinks.

[0059] Beneficial Effects

[0060] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects:

[0061] 1. By monitoring the carbon storage in vegetation and soil, the overall carbon sink capacity of the forest can be assessed more accurately. Since the carbon storage in vegetation and soil may fluctuate with seasonal changes, by monitoring these changes, we can better understand how carbon sinks are affected by climate factors (such as temperature, precipitation and light). At the same time, in the process of monitoring the changes in carbon storage in different seasons, and based on the law of the impact of nonlinear changes in carbon storage in vegetation and soil on the forest carbon sink capacity, the long-term trend of forest carbon sink can be extracted.

[0062] 2. This scheme extracts forest carbon sink trends through EEMD and can process nonlinear and non-stationary data of forest carbon sinks, which is difficult to achieve with traditional extraction methods. This makes EEMD more effective in processing data with complex fluctuation patterns such as carbon sinks. In addition, EEMD can adaptively decompose data into multiple IMFs, each of which represents a different fluctuation scale, which helps to extract the multi-scale characteristics of carbon sink data and provide more details for in-depth analysis.

[0063] 3. In the process of extracting forest carbon sink trends by EEMD, this scheme adds an ANN model for optimization. ANN can use the multi-scale features obtained by EEMD decomposition to comprehensively consider the changes in carbon sinks at different time scales, improve the comprehensiveness of the prediction, reduce the risk of EEMD overfitting, and improve the generalization ability of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0065] Figure 1 Schematic diagram of the extraction system workflow in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0067] The present invention will be further described below in conjunction with the embodiments.

[0068] Example:

[0069] Please refer to the attached Figure 1This scheme proposes a carbon sink nonlinear trend extraction system based on ensemble empirical mode decomposition. According to the law of the impact of the nonlinear changes of vegetation carbon storage and soil carbon storage on the forest carbon sink capacity during the temporal changes, the seasonal change trend of forest carbon sink is extracted.

[0070] Specifically, the system includes a data acquisition module, a data processing module and a construction module.

[0071] First, the data acquisition module collects forest carbon sink time series data through remote sensing monitoring technology and ground measurement technology, including vegetation coverage rate VCR, soil coverage rate SCR, vegetation carbon content C veg , soil carbon content C soil , total carbon storage C total .

[0072] Then the forest carbon sink time series data is transmitted to the data processing module.

[0073] The data processing module preprocesses the forest carbon sink time series data and obtains the forest carbon sink C at different time scales during seasonal changes. Carbon Sinkl (t).

[0074] Then, based on the ensemble empirical mode decomposition (EEMD), the forest carbon sink C Carbon Sinkl (t) for processing.

[0075] First, for C Carbon Sinkl (t) performing pre-processing, including:

[0076] Data cleaning: remove missing values ​​and outliers, and perform necessary smoothing;

[0077] Detrending and demeaning: Detrending is performed to ensure that the data is more suitable for EEMD analysis.

[0078] Then to C Carbon Sinkl Add white noise to (t).

[0079] Then, EEMD is used to analyze each noise-introduced signal C Carbon Sinkl(i) (t) Perform empirical mode decomposition (EMD) and extract C Carbon Sinkl The intrinsic mode functions IMFs and residuals r in (t) i (t).

[0080] Extract feature X from multiple IMFs extracted by building modules features (t), and according to the extracted X features (t) Construct artificial neural network structure ANN.

[0081] Finally, the trained ANN model is used to perform C Carbon Sinkl(t) Trend prediction, and draw a trend line graph of the prediction results to visualize the nonlinear trend of the predicted forest carbon sink.

[0082] More specifically, the vegetation carbon content calculation formula is: C veg =A veg *B*CF

[0083] Among them, C veg is the carbon storage in vegetation; A veg It is the area of ​​vegetation coverage determined by remote sensing technology; B is the average biomass of vegetation per unit area, calculated by the correlation between the normalized difference vegetation index and ground measurement data; CF is the carbon fraction, which refers to the proportion of carbon in biomass, and plant dry weight is usually about 50% carbon.

[0084] Soil carbon content calculation formula: C soil =A soil *D*BD*SOC*(1-VCR)

[0085] Among them, C soil is soil carbon storage; A soil is the area covered by soil; D is the depth of the soil, because carbon storage is usually only calculated to a certain soil depth; BD is the soil bulk density, which represents the mass of unit volume of soil; SOC is the soil organic carbon content, which represents the proportion of organic carbon in unit mass of soil, and is obtained by taking soil samples for measurement.

[0086] The C obtained by the above calculation veg and C soil , the total carbon storage C can be calculated total , the calculation formula is as follows:

[0087] in:

[0088] α*VCR*C veg Indicates the contribution of vegetation coverage to vegetation carbon storage; β*SCR*C soil represents the contribution of soil coverage to soil carbon storage; λ*VCR*SCR*(C veg +C soil ) represents the additional effect of the interaction between vegetation and soil coverage on the total carbon storage; α is the influence coefficient of vegetation coverage on vegetation carbon storage; β is the influence coefficient of soil coverage on soil carbon storage; γ is the influence coefficient of the interaction between vegetation and soil on the total carbon storage; α, β, and γ are determined by nonlinear regression analysis.

[0089] Furthermore, according to C in static time total Calculate the forest carbon sink C at different times during seasonal changes Carbon Sinkl (t), the calculation formula is as follows: CCarbon Sink (t) = C total (t)+∫[I(t)-O(t)]dt

[0090] Among them, C total (t) is the total carbon storage over time; I(t) is the carbon input rate over time, including photosynthesis, carbon deposition or other sources. O(t) is the carbon output rate over time, which may include respiration, combustion, decay or other forms of carbon release; ∫[I(t)-O(t)]dt represents the cumulative integral of carbon input minus carbon output from the initial time to time t, that is, the total carbon increase or decrease.

[0091] The calculation formula of I(t) is: I(t) = P MAX *L(t)*T(t)*VCR(t)

[0092] Among them, P MAX is the maximum photosynthetic rate; L(t) is the light intensity function that changes with time; T(t) is the temperature function that changes with time; VCR(t) is the plant coverage that changes with time.

[0093] The calculation formula of O(t) is: O(t) = R base *T(t)*B(t)

[0094] Where: R base is the basal respiration rate; B(t) is the biomass function that changes with time.

[0095] The above conventional parameters are collected and calculated through ground measurement technology.

[0096] The difference is that in the C Carbon Sinkl (t) When adding white noise, multiple groups of white noise n are generated i (t) and added to the original signal. By setting different noise intensities α, the randomness of the signal is increased and the modal aliasing phenomenon is reduced. Carbon Sinkl (t) is calculated as:

[0097] C Carbon Sinkl(i) (t) = C Carbon Sinkl (t)+αn i (t)

[0098] In EEMD, for each noise-introduced signal C Carbon Sinkl(i) (t) When performing EMD, K modal functions are obtained, where K represents the total number of intrinsic modal functions extracted from the signal: The calculation formula is as follows:

[0099]

[0100] Among them, IMFij (t) indicates that this is the jth mode function obtained by decomposing after adding white noise for the i-th time. Each mode function represents a specific frequency component in the original signal, which helps to capture the local characteristics and changes of the signal.

[0101] It is worth noting that the extracted X features (t) includes the following: mean, standard deviation, skewness, kurtosis and frequency. Through effective feature extraction, the prediction performance and generalization ability of the model can be improved. The specific extraction formula is:

[0102] X features (t)=[mean(IMF ij (t)), std(IMF ij (t)), skewness(IMF ij (t)), kurtosis (IMF ij (t)), frequency(IMF ij (t))]

[0103] in,

[0104] n is the number of data points, χ i is the value of the ith data point;

[0105]

[0106] According to the extracted X features (t) Construct ANN model:

[0107] h t =σ(W*X features (t)+b)

[0108] h t is the output of the hidden layer, W is the weight matrix, b is the bias, and σ is the activation function.

[0109] Then train the ANN model:

[0110]

[0111] ΔW is the weight update, η is the learning rate, L is the loss function, is the partial derivative of the weight with respect to the loss function.

[0112] Then evaluate the ANN model:

[0113]

[0114] MSE is the mean square error, which is the actual value, y i is the actual value, is the predicted value;

[0115] And tune the parameters:

[0116]

[0117] Δη is the learning rate update, α is the learning rate adjustment coefficient, is the partial derivative of the loss function with respect to the learning rate;

[0118] Final result application:

[0119] y predicted =h T ·softmax(W O ·h T +b O )

[0120] h T It is a vector, representing the output of the last layer (hidden layer) of the neural network, which contains the result of the network's processing of the input data;

[0121] W O is the output layer weight matrix, which connects the hidden layer and the output layer. O Each element of determines the influence of the hidden layer features on the output results;

[0122] b O is the bias vector of the output layer, which has a separate bias term for each output category;

[0123] Softmax is an activation function that converts the output of the network into a probability distribution, where the value of each element represents the probability of the corresponding category;

[0124] y predicted is the final prediction result, which is the output of the softmax function multiplied by the hidden layer output h T This multiplication operation can be viewed as weighting each category, with the weight being the hidden layer output h T Decide.

[0125] Through this integrated approach, the powerful feature learning and pattern recognition capabilities of ANN can be utilized to optimize the decomposition results of EEMD, thereby improving the accuracy and efficiency of carbon sink data analysis.

[0126] According to the above content, this scheme has the following advantages in monitoring and predicting forest carbon sink trends:

[0127] 1. By monitoring the carbon storage in vegetation and soil, the overall carbon sink capacity of the forest can be assessed more accurately. Since the carbon storage in vegetation and soil may fluctuate with seasonal changes, by monitoring these changes, we can better understand how carbon sinks are affected by climate factors (such as temperature, precipitation and light). At the same time, in the process of monitoring the changes in carbon storage in different seasons, and based on the law of the impact of nonlinear changes in carbon storage in vegetation and soil on the forest carbon sink capacity, the long-term trend of forest carbon sink can be extracted.

[0128] 2. This scheme extracts forest carbon sink trends through EEMD and can process nonlinear and non-stationary data of forest carbon sinks, which is difficult to achieve with traditional extraction methods. This makes EEMD more effective in processing data with complex fluctuation patterns such as carbon sinks. In addition, EEMD can adaptively decompose data into multiple IMFs, each of which represents a different fluctuation scale, which helps to extract the multi-scale characteristics of carbon sink data and provide more details for in-depth analysis.

[0129] 3. In the process of extracting forest carbon sink trends by EEMD, this scheme adds an ANN model for optimization. ANN can use the multi-scale features obtained by EEMD decomposition to comprehensively consider the changes in carbon sinks at different time scales, improve the comprehensiveness of the prediction, reduce the risk of EEMD overfitting, and improve the generalization ability of the model.

[0130] The carbon sink nonlinear trend extraction method based on ensemble empirical mode decomposition includes the following steps:

[0131] Step 1: Collect carbon sink time series data through the data collection module;

[0132] Step 2: Process the collected carbon sink time series data through the data processing module, and calculate the C total Calculate the forest carbon sink C at different times during seasonal changes Carbon Sinkl (t);

[0133] Step 3: Based on the ensemble empirical mode decomposition (EEMD) Carbon Sinkl (t) is processed and C is extracted Carbon Sinkl (t) and extract the feature X from the extracted multiple IMFs. features (t);

[0134] Step 4: Construct modules based on the extracted X features (t) constructing artificial neural network structure ANN;

[0135] Step 5: C through ANN model Carbon Sinkl(t) Trend prediction, and extract the nonlinear trend of forest carbon sink based on the trend line graph of the prediction results.

[0136] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for extracting carbon sink nonlinear trend based on ensemble empirical mode decomposition, characterized in that: The following steps are involved: S1: Collect forest carbon sink time series data to form a forest carbon sink time series data set. The forest carbon sink time series data includes vegetation coverage rate VCR, soil coverage rate SCR, vegetation carbon content C in different seasons. veg , soil carbon content C soil , total carbon storage C total ; S2: Preprocess the forest carbon sink time series data set based on forest carbon storage theory to obtain forest carbon sink C in different seasons CarbonSinkl (t); S3: Using the ensemble empirical mode decomposition model to CarbonSinkl (t) Extract and obtain data set X features (t) and using the ANN model to features (t) processing to extract the nonlinear trend of forest carbon sink, wherein the ensemble empirical mode decomposition model uses ensemble empirical mode decomposition (EEMD) to calculate the C CarbonSinkl (t) Process and extract C CarbonSinkl (t) and extract the feature X through IMFs features (t).

2. The carbon sink nonlinear trend extraction method based on ensemble empirical mode decomposition according to claim 1 is characterized in that: In S1, the data acquisition module collects forest carbon sink time series data through remote sensing monitoring technology and ground measurement technology.

3. The carbon sink nonlinear trend extraction method based on ensemble empirical mode decomposition according to claim 2 is characterized in that: In S2, C CarbonSinkl The calculation process of (t) is: S201: Based on forest carbon storage theory, obtain the total carbon storage C in static time total , specifically: C total =(α*VCR*C veg )+(β*SCR*C soil )+[λ*VCR*SCR*(C veg +C soil )] in: α*VCR*C veg Indicates the contribution of vegetation coverage to vegetation carbon storage; β*SCR*C soil represents the contribution of soil coverage to soil carbon storage; λ*VCR*SCR*(C veg +C soil ) represents the additional effect of the interaction between vegetation and soil coverage on total carbon storage; α is the influence coefficient of vegetation coverage on vegetation carbon storage; β is the influence coefficient of soil coverage on soil carbon storage; γ is the influence coefficient of the interaction between vegetation and soil on total carbon storage; α, β, and γ are determined by nonlinear regression analysis; S202: Based on the total carbon storage C total Calculate the forest carbon sink C at different times during seasonal changes CarbonSinkl (t), specifically: C CarbonSink (t)=C total (t)+∫[I(t)-O(t)]dt Among them, I(t) is the carbon input rate that changes with time, O(t) is the carbon output rate that changes with time, and ∫[I(t)-O(t)]dt represents the cumulative integral of carbon input minus carbon output from the initial time to time t, that is, the total carbon increase or decrease.

4. The carbon sink nonlinear trend extraction method based on ensemble empirical mode decomposition according to claim 3 is characterized in that: In S3, the C CarbonSinkl (t) extracting, comprising the following steps: S301: Towards C CarbonSinkl Add white noise to (t): C CarbonSinkl ( i) (t)=C CarbonSinkl (t)+αn i (t) S302: For each signal C that introduces white noise CarbonSinkl(i) (t) Performing empirical mode decomposition (EMD); S303: Extraction of intrinsic mode function: Among them, IMF ij (t) indicates that this is the jth mode function obtained by decomposition after adding white noise for the i-th time; S304:X features (t) Extraction: in, n is the number of data points and is the value of the ith data point.

5. The carbon sink nonlinear trend extraction method based on ensemble empirical mode decomposition according to claim 4 is characterized in that: In S3, the ANN model is used to analyze the X features (t) Processing to extract nonlinear trends in forest carbon sinks, specifically: According to X features (t) Construct ANN model: h t =σ(W*X features (t)+b) h t is the output of the hidden layer, W is the weight matrix, b is the bias, and σ is the activation function; The training formula of the ANN model is: ΔW is the weight update, η is the learning rate, L is the loss function, is the partial derivative of the weight with respect to the loss function; The evaluation formula of the ANN model is: MSE is the mean square error, which is the actual value, y i is the actual value, is the predicted value; The parameter tuning formula is: Δη is the learning rate update, α is the learning rate adjustment factor, is the partial derivative of the loss function with respect to the learning rate; y predicted =h T ·softmax(W O ·h T +b O ) h T is a vector representing the output of the last layer (hidden layer) of the neural network; W O is the output layer weight matrix, connecting the hidden layer and the output layer; b O is the bias vector of the output layer; Softmax is an activation function that converts the output of the network into a probability distribution, where the value of each element represents the probability of the corresponding category; y predicted for the final nonlinear trend extraction results.

6. A carbon sink nonlinear trend extraction system based on ensemble empirical mode decomposition, applied to execute the nonlinear trend extraction method according to any one of claims 1 to 5, the system comprising a data acquisition module and a data processing module; The data acquisition module is used to collect forest carbon sink time series data to form a forest carbon sink time series data set. The forest carbon sink time series data set includes vegetation coverage rate VCR, soil coverage rate SCR, vegetation carbon content C in different seasons. veg , soil carbon content C soil , total carbon storage C total ; The data processing module is used to preprocess the forest carbon sink time series data set to obtain forest carbon sink C in different seasons. CarbonSinkl (t) and extracting the forest carbon sink time series data set according to the pre-constructed ensemble empirical mode decomposition model to obtain a data set X features (t), according to the pre-built ANN model, features (t) Processing to extract nonlinear trends in forest carbon sinks.

Images