Tree growth parameter prediction method and system based on cascade loop network
Through a method based on a cascaded recurrent network, combined with multi-layer cascade structure and GIS technology, the problem of unconsidered environmental factors in the prediction of tree growth parameters is solved, the accuracy and practicality of the prediction are improved, and the transmission line is optimized. Maintenance and security management.
Patent Information
- Application Number
- CN202410957582.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-12-10
AI Technical Summary
The existing technology has not fully considered the comprehensive impact of various environmental factors in the prediction of tree growth parameters, and the model has shortcomings in the combination of practical application scenarios and data preprocessing, model structure design and prediction performance evaluation, which limits its accuracy and Application scope.
Using a cascading recurrent network method, integrating more comprehensive environmental data, designing a multi-layer cascading recurrent network structure, combining geographic information system (GIS) for risk assessment and early warning of transmission lines, improves the accuracy and practicality of predictions value.
It significantly improves the accuracy and reliability of tree growth parameter prediction, optimizes the maintenance of transmission lines, reduces faults caused by tree growth, and enhances the safety and stability of transmission lines.
Smart Images
Figure CN119106278A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of environmental science and engineering technology, and in particular to a tree growth parameter prediction method and system based on a cascade cycle network. Background Art
[0002] Accurate prediction of tree growth parameters has always been an important research topic in the fields of urban planning, forestry management, and power transmission line safety. Traditional tree growth monitoring methods rely on regular physical measurements, which are not only time-consuming and labor-intensive, but also unable to provide real-time data or predict future growth trends. With the development of information technology, especially the advancement of machine learning and deep learning technologies, data-driven tree growth prediction models have become a hot topic of research. These models can use historical growth data to predict future growth parameters of trees, such as tree height, breast diameter, and crown width. In particular, recurrent neural networks (RNNs) and their variants, such as long short-term memory networks (LSTMs), have been widely used in such prediction tasks due to their superior performance in processing time series data.
[0003] Although existing technologies are able to predict tree growth to a certain extent, there are still some shortcomings. First, most existing models fail to fully consider the various environmental factors that affect tree growth, such as the combined effects of climate change, soil conditions, and tree species differences. Second, models in existing technologies often ignore the combination of prediction results with actual application scenarios, such as the maintenance and management of transmission lines, which limits the practicality and effectiveness of the models in actual operations. In addition, existing prediction models need to be optimized in terms of data preprocessing, model structure design, and prediction performance evaluation. These technical bottlenecks restrict the accuracy and application scope of the models.
[0004] In view of the above shortcomings, the present invention proposes a tree growth parameter prediction method and system based on a cascaded cyclic network, which significantly improves the accuracy and reliability of the prediction by integrating more comprehensive environmental data and adopting a multi-layer cascaded cyclic network structure. The cascade structure allows the model to extract and utilize information in the time series at different levels, thereby better simulating the combined influence of multiple factors on tree growth. In addition, the present invention greatly improves the practical value of the model by combining the prediction results with the geographic information system (GIS) and directly using them for risk assessment and early warning of transmission lines. This not only optimizes the maintenance of transmission lines and reduces accidents caused by tree growth, but also provides scientific data support for urban greening management and demonstrates a new paradigm for ecological monitoring and management in complex environments. Summary of the invention
[0005] In view of the above-mentioned problems, the present invention is proposed.
[0006] Therefore, the technical problem solved by the present invention is: how to accurately predict tree growth parameters in real time by integrating deep learning models and geographic information system (GIS) technology, and use tree growth parameters to perform risk assessment and early warning of transmission lines to improve the safety management and maintenance efficiency of transmission lines.
[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a tree growth parameter prediction method based on a cascaded cyclic network, which comprises the following steps:
[0008] Collect historical growth parameters of trees and perform preprocessing; build a tree growth parameter prediction model based on a cascade loop network; train the prediction model with the preprocessed historical tree growth data; predict tree growth parameters with the trained tree growth parameter prediction model; combine the predicted tree growth parameters with GIS to provide early warning for transmission lines.
[0009] As a preferred solution of the tree growth parameter prediction method based on cascaded cyclic network described in the present invention, the historical growth parameters include tree height, breast diameter and crown width.
[0010] As a preferred embodiment of the tree growth parameter prediction method based on cascaded cyclic network described in the present invention, the preprocessing includes cleaning the collected historical growth parameters of tree growth, removing outliers and missing values, and normalizing the growth parameters.
[0011] As a preferred solution of the tree growth parameter prediction method based on cascaded recurrent network described in the present invention, wherein: the construction of the tree growth parameter prediction model includes selecting LSTM recurrent units according to the type of tree growth parameters, designing a multi-layer cascaded recurrent network structure, including the first layer, the second layer, the third layer and the fully connected layer, wherein the first layer is responsible for capturing the impact of weather conditions changing over time on tree growth, inputting historical growth parameters and climate data, and outputting hidden states and cell status The second layer receives the output of the first layer and soil data, further analyzing the growth response of trees under specific climate and soil conditions, and outputting hidden states and cell status The third layer input and species characteristics t , use specific gating to adjust the influence of tree species characteristics and output hidden state and cell status The fully connected layer transforms the output of the third layer into predicted growth parameters.
[0012] As a preferred embodiment of the tree growth parameter prediction method based on cascaded cyclic network described in the present invention, the training of the prediction model includes sorting the breast diameter, tree height and crown width of each tree according to the collection time to form a time series of growth parameters of a single tree, taking the first 60% of the time series of each tree species as a training set and the last 40% of the time series as a test set, adjusting the parameters of the prediction model using the training set data, and using the determination coefficient on the test set to evaluate the performance of the prediction model.
[0013] As a preferred embodiment of the tree growth parameter prediction method based on cascaded cyclic network described in the present invention, the predicted tree growth parameter includes the historical growth parameter represented as
[0014]
[0015] in, Indicates the historical tree height, represents the historical breast diameter, Indicates historical crown width.
[0016] The effect of changing weather conditions over time on tree growth is expressed as,
[0017]
[0018] Among them, s t represents seasonal gating, W s represents the weight matrix, b s represents the bias term, and σ represents the activation function.
[0019] The cell state It is expressed as,
[0020]
[0021] in, Represents the current candidate value, expressed as,
[0022]
[0023] Among them, W c represents the weight matrix, b c represents the bias term, x t Represents an input collection.
[0024] The analysis shows that the growth response of trees under specific climate and soil conditions is expressed as,
[0025]
[0026] Among them, W i 、b irepresents the weight matrix and bias term, and α represents a scalar.
[0027] The cell state It is expressed as,
[0028]
[0029] Among them, β represents the weight factor.
[0030] The climate data are expressed as
[0031] Among them, temp t Indicates temperature, hum t Indicates humidity, rain t Indicates rainfall.
[0032] The soil data is expressed as
[0033] Among them, soil-type t Indicates soil type, pH t Indicates pH value, nutrients t Indicates nutritional content;
[0034] The effect of using specific gating to adjust tree species characteristics is expressed as,
[0035] g t =σ(W g species t +b g )
[0036] Among them, g t represents specificity gating, W g represents the weight matrix, b g Represents the bias term.
[0037] The cell state It is expressed as,
[0038]
[0039] Among them, f t represents the forget gate, i t Represents the input gate.
[0040] The predicted growth parameter is expressed as
[0041] in, represents the predicted tree height, represents the predicted DBH, Represents the predicted crown width.
[0042] As a preferred solution of the tree growth parameter prediction method based on the cascade cycle network described in the present invention, the early warning of the transmission line includes establishing a buffer zone for the transmission line, importing the predicted growth parameters into the GIS system, creating a new layer in the GIS to represent different predicted growth parameters, integrating the predicted growth parameters to obtain a risk score represented as,
[0043]
[0044] in, represents the maximum possible value of the tree height, The maximum possible value of DBH, represents the maximum possible value of the crown width, G represents the growth rate of the tree, G max represents the maximum growth rate of trees, P represents the distance between trees and the nearest power transmission line, and E represents the environmental impact factor.
[0045] When the Risk Score is greater than the threshold, it means that any predicted growth parameter will enter the buffer zone, and targeted measures will be taken for different growth parameters.
[0046] The targeted measures include carrying out top pruning in advance when the predicted tree height enters the buffer zone and installing protective nets near transmission lines.
[0047] When the predicted DBH enters the buffer zone, a structural integrity assessment is conducted to check whether there is internal decay. If decay exists and is within the bearing range, supports or cables are installed. If it is not within the bearing range, the decayed parts are removed.
[0048] When the predicted crown width enters the buffer zone, guided pruning is used to change the growth direction of branches, and sensors and cameras are installed to monitor the dynamics of the crown trees.
[0049] Another object of the present invention is to provide a tree growth parameter prediction system based on a cascade cyclic network, which can simulate the tree growth process on different time scales by dynamically integrating multi-source environmental and growth data, thereby solving the problems of existing technologies in processing complex ecological data and making long-term predictions.
[0050] To solve the above technical problems, the present invention provides the following technical solutions: a tree growth parameter prediction system based on a cascaded cyclic network, comprising a data collection module, a cascaded cyclic network construction module, a training and prediction module, and a GIS integration and early warning module.
[0051] The data collection module is responsible for collecting historical growth parameter data of trees and performing data preprocessing;
[0052] The cascaded recurrent network construction module constructs a multi-layer cascaded recurrent network structure based on the LSTM recurrent unit.
[0053] The training and prediction module uses historical data to train the cascade recurrent network model, applies the trained model to the prediction stage, and generates prediction results of tree height, breast diameter and crown width.
[0054] The GIS integration and early warning module integrates the predicted growth parameters into the GIS system, creates new layers in the GIS system to represent different predicted growth parameters, and calculates risk scores. When the risk score exceeds a set threshold, corresponding early warning measures are triggered to ensure the safe operation of the transmission line.
[0055] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the tree growth parameter prediction method based on a cascaded cyclic network when executing the computer program.
[0056] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the tree growth parameter prediction method based on a cascaded recurrent network as described above.
[0057] The beneficial effects of the present invention are as follows: by using the cascade recurrent neural network (CRNN) model combined with the long short-term memory network (LSTM) technology, the dynamic changes of tree growth and its response to environmental factors can be effectively captured and analyzed, providing more accurate growth parameter predictions. In addition, integrating these prediction data into the geographic information system (GIS) not only improves the risk management efficiency of transmission lines, but also makes preventive measures more targeted and timely, significantly reduces transmission line failures and maintenance costs caused by trees, and enhances the safety and stability of transmission lines. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0059] Figure 1 This is an overall flow chart of a tree growth parameter prediction method based on a cascaded cyclic network provided in the first embodiment of the present invention.
[0060] Figure 2 The overall framework diagram of the tree growth parameter prediction system based on the cascaded cyclic network provided in the second embodiment of the present invention. DETAILED DESCRIPTION
[0061] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.
[0062] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a tree growth parameter prediction method based on a cascaded recurrent network, characterized in that:
[0063] S1: Collect historical growth parameters of trees and perform preprocessing.
[0064] Historical growth parameters include tree height, DBH, and crown width.
[0065] The preprocessing includes cleaning the collected historical growth parameters of tree growth, removing outliers and missing values, and normalizing the growth parameters.
[0066] S2: Construct a tree growth parameter prediction model based on a cascade recurrent network.
[0067] The construction of the tree growth parameter prediction model includes selecting LSTM recurrent units according to the type of tree growth parameters, designing a multi-layer cascade recurrent network structure, including the first layer, the second layer, the third layer and the fully connected layer. The first layer is responsible for capturing the impact of weather conditions changing over time on tree growth, inputting historical growth parameters and climate data, and outputting hidden states. and cell status The second layer receives the output of the first layer and soil data, further analyzing the growth response of trees under specific climate and soil conditions, and outputting hidden states and cell status The third layer input and species characteristics t , use specific gating to adjust the influence of tree species characteristics and output hidden state and cell status The fully connected layer transforms the output of the third layer into predicted growth parameters.
[0068] Furthermore, in this step, by selecting LSTM recurrent units and designing a multi-layer cascaded recurrent network structure, the present invention can accurately capture and analyze a variety of dynamic factors that affect tree growth, such as weather conditions and soil conditions. The cascade structure allows the model to process and learn data at different levels, so that each layer can respond to the specific information it receives. For example, the first layer processes the impact of climate data on tree growth, while subsequent layers combine the output of the previous layer with new inputs (such as soil data and tree species characteristics) to further deepen the analysis. This layered processing greatly enhances the learning ability and prediction accuracy of the model, making the prediction system not only responsive but also more refined.
[0069] S3: Train the prediction model using the preprocessed historical tree growth data.
[0070] Training the prediction model includes sorting the breast diameter, tree height and crown width of each tree according to the collection time to form a time series of growth parameters of a single tree. The first 60% of the time series of each tree species is used as the training set, and the last 40% of the time series is used as the test set. The parameters of the prediction model are adjusted using the training set data, and the determination coefficient is used on the test set to evaluate the performance of the prediction model.
[0071] S4: predicting tree growth parameters using the trained tree growth parameter prediction model;
[0072] The predicted tree growth parameters including historical growth parameters are expressed as
[0073] in, Indicates the historical tree height, represents the historical breast diameter, Indicates historical crown width.
[0074] The effect of changing weather conditions over time on tree growth is captured as,
[0075]
[0076] Among them, s t represents seasonal gating, W s represents the weight matrix, b s represents the bias term, and σ represents the activation function.
[0077] Cell state It is expressed as,
[0078]
[0079] in, Represents the current candidate value, expressed as,
[0080]
[0081] Among them, W c represents the weight matrix, b c represents the bias term, x t Represents an input collection.
[0082] The analysis of the growth response of trees under specific climate and soil conditions is expressed as,
[0083]
[0084] Among them, W i 、b i represents the weight matrix and bias term, and α represents a scalar.
[0085] Cell state It is expressed as,
[0086]
[0087] Among them, β represents the weight factor.
[0088] Climate data are expressed as
[0089] Among them, temp t Indicates temperature, hum t Indicates humidity, rain t Indicates rainfall.
[0090] Soil data are represented as
[0091] Among them, soil-type t Indicates soil type, pH t Indicates pH value, nutrients t represents the nutritional content. The effect of using specific gating to adjust tree species characteristics is expressed as,
[0092] g t =σ(W g species t +b g )
[0093] Among them, g t represents specificity gating, W g represents the weight matrix, b g Represents the bias term.
[0094] Cell state It is expressed as,
[0095]
[0096] Among them, ft represents the forget gate, i t Represents the input gate.
[0097] The predicted growth parameters are expressed as
[0098] in, represents the predicted tree height, represents the predicted DBH, Represents the predicted crown width.
[0099] S5: Combine the predicted tree growth parameters with GIS to provide early warning for transmission lines.
[0100] Early warning of transmission lines includes establishing a buffer zone for the transmission lines, importing the predicted growth parameters into the GIS system, creating a new layer in the GIS to represent different predicted growth parameters, and integrating the predicted growth parameters to obtain a risk score represented as:
[0101]
[0102] in, represents the maximum possible value of the tree height, The maximum possible value of DBH, represents the maximum possible value of the crown width, G represents the growth rate of the tree, G max represents the maximum growth rate of trees, P represents the distance between trees and the nearest power transmission line, and E represents the environmental impact factor.
[0103] When the Risk Score is greater than the threshold, it means that any predicted growth parameter will enter the buffer zone, and targeted measures will be taken for different growth parameters.
[0104] Targeted measures include top pruning in advance when the predicted tree height enters the buffer zone and installing protective nets near transmission lines.
[0105] When the predicted DBH enters the buffer zone, a structural integrity assessment is conducted to check whether there is internal decay. If decay exists and is within the bearing range, supports or cables are installed. If it is not within the bearing range, the decayed parts are removed.
[0106] When the predicted crown width enters the buffer zone, guided pruning is used to change the growth direction of branches, and sensors and cameras are installed to monitor the dynamics of the crown trees.
[0107] Example 2, reference Figure 2 , which is an embodiment of the present invention, provides a system for predicting tree growth parameters based on cascaded cyclic network, characterized by comprising a data collection module, a cascaded cyclic network construction module, a training and prediction module, and a GIS integration and early warning module.
[0108] The data collection module is responsible for collecting historical growth parameter data of trees and performing data preprocessing;
[0109] The cascaded recurrent network building module builds a multi-layer cascaded recurrent network structure based on the LSTM recurrent unit.
[0110] The training and prediction module uses historical data to train the cascade recurrent network model, and applies the trained model to the prediction stage to generate prediction results for tree height, breast diameter and crown width.
[0111] The GIS integration and early warning module integrates the predicted growth parameters into the GIS system, creates new layers in the GIS system to represent different predicted growth parameters, and calculates the risk score. When the risk score exceeds the set threshold, the corresponding early warning measures are triggered to ensure the safe operation of the transmission line.
[0112] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.
[0113] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.
[0114] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.
[0115] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0116] Embodiment 3: In this embodiment, in order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments. This embodiment experiments the existing traditional method and the method of this embodiment respectively.
[0117] Traditional method specific plan:
[0118] Data preparation: Collect historical growth parameter data of trees (tree height, breast diameter and crown width).
[0119] Model building: Use a linear regression model to predict future growth based on historical data.
[0120] Model Evaluation: A simple R2 (coefficient of determination) was used to evaluate the predictive performance of the model.
[0121] Our specific invention:
[0122] Data preparation: The historical growth parameters of the trees were also collected and preprocessed, including removing outliers, handling missing values, and normalizing.
[0123] Model construction: Based on the cascade recurrent neural network (CRNN), in particular, the use of multi-layer LSTM units to build a prediction model that can take into account the impact of climate change, soil conditions and tree species characteristics.
[0124] Model evaluation: using more than just R 2 The evaluation model also includes the mean square error (MSE) and the response time of the model to fully evaluate the efficiency and accuracy of the model.
[0125] The specific experimental table is shown in Table 1:
[0126] Table 1 Experimental effect comparison chart
[0127] Experimental indicators Traditional methods Method of the present invention <![CDATA[R 2 (coefficient of determination)]]> 0.65 0.90 MSE (mean square error) 15.2 5.8 Response time (seconds) 0.5 0.7 Prediction accuracy (%) 70 95
[0128] Compared with traditional methods that usually rely on a single type of historical growth data, our invention integrates multi-source data (including climate data, soil data and tree species characteristics) for model training and prediction. This not only improves the generalization ability of the prediction model, but also makes the prediction results more comprehensive and accurate. In particular, by introducing climate and soil data, the model can more comprehensively understand and predict the impact of environmental factors on tree growth, improving the model's adaptability to external changes and the reliability of predictions.
[0129] Compared with simple models, our invention integrates more types of input data (such as climate data, soil data, and tree species characteristics) in the prediction process, and each type of data needs to be processed and analyzed in the model. Although the integration of multi-source data can improve the accuracy and robustness of the prediction, it also increases the complexity and computational workload of data processing.
[0130] It should be noted that 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A tree growth parameter prediction method based on a cascaded recurrent network, characterized in that: include: Collect historical growth parameters of trees and perform preprocessing; Constructing a tree growth parameter prediction model based on cascade recurrent networks; The prediction model is trained using preprocessed historical tree growth data; Predicting tree growth parameters using the trained tree growth parameter prediction model; The predicted tree growth parameters are combined with GIS to provide early warning for transmission lines.
2. The tree growth parameter prediction method based on cascaded cyclic network according to claim 1, characterized in that: The historical growth parameters include tree height, breast diameter and crown width.
3. The tree growth parameter prediction method based on cascaded cyclic network as claimed in claim 2, characterized in that: The preprocessing includes cleaning the collected historical growth parameters of tree growth, removing outliers and missing values, and normalizing the growth parameters.
4. The tree growth parameter prediction method based on cascaded cyclic network as claimed in claim 3, characterized in that: The tree growth parameter prediction model is constructed by selecting LSTM recurrent units according to the tree growth parameter type, designing a multi-layer cascade recurrent network structure, including the first layer, the second layer, the third layer and the fully connected layer, wherein the first layer is responsible for capturing the impact of weather conditions changing over time on tree growth, inputting historical growth parameters and climate data, and outputting hidden states. and cell status The second layer receives the output of the first layer and soil data, further analyzing the growth response of trees under specific climate and soil conditions, and outputting hidden states and cell status The third layer input and species characteristics t , use specific gating to adjust the influence of tree species characteristics and output hidden state and cell status The fully connected layer transforms the output of the third layer into predicted growth parameters.
5. The tree growth parameter prediction method based on cascaded cyclic network according to claim 4, characterized in that: The training of the prediction model includes sorting the breast diameter, tree height and crown width of each tree according to the collection time to form a time series of growth parameters of a single tree, using the first 60% of the time series of each tree species as a training set and the last 40% of the time series as a test set, using the training set data to adjust the parameters of the prediction model, and using the determination coefficient on the test set to evaluate the performance of the prediction model.
6. The tree growth parameter prediction method based on cascaded cyclic network according to claim 5, characterized in that: The predicted tree growth parameters include the historical growth parameters expressed as in, Indicates the historical tree height, represents the historical breast diameter, Indicates the historical crown width; The effect of changing weather conditions over time on tree growth is expressed as, Among them, s t represents seasonal gating, W s represents the weight matrix, b s represents the bias term, σ represents the activation function; The cell state It is expressed as, in, Represents the current candidate value, expressed as, Among them, W c represents the weight matrix, b c represents the bias term, x t represents the input set; The analysis shows that the growth response of trees under specific climate and soil conditions is expressed as, Among them, W i 、b i represents the weight matrix and bias term, α represents a scalar; The cell state It is expressed as, Among them, β represents the weight factor; The climate data are expressed as Among them, temp t Indicates temperature, hum t Indicates humidity, rain t Indicates rainfall; The soil data is expressed as Among them, soil-type t Indicates soil type, pH t Indicates pH value, nutrients t Indicates nutritional content; The effect of using specific gating to adjust tree species characteristics is expressed as, g t =σ(W g species t +b g ) Among them, g t represents specificity gating, W g represents the weight matrix, b g represents the bias term; The cell state It is expressed as, Among them, f t represents the forget gate, i t represents the input gate; The predicted growth parameter is expressed as in, represents the predicted tree height, represents the predicted DBH, Represents the predicted crown width.
7. The tree growth parameter prediction method based on cascaded cyclic network according to claim 6, characterized in that: The early warning of the transmission line includes establishing a buffer zone for the transmission line, importing the predicted growth parameters into the GIS system, creating a new layer in the GIS to represent different predicted growth parameters, and integrating the predicted growth parameters to obtain a risk score represented as: in, represents the maximum possible value of the tree height, The maximum possible value of DBH, represents the maximum possible value of the crown width, G represents the growth rate of the tree, G max represents the maximum growth rate of trees, P represents the distance between trees and the nearest power transmission line, and E represents the environmental impact factor; When the Risk Score is greater than the threshold, it means that any predicted growth parameter will enter the buffer zone, and targeted measures will be taken for different growth parameters; The targeted measures include pre-emptive top pruning when the predicted tree height enters the buffer zone and installing protective nets near power transmission lines; When the predicted DBH enters the buffer zone, a structural integrity assessment is conducted to check whether there is internal decay. If decay exists and is within the tolerance range, supports or cables are installed. If it is not within the tolerance range, the decayed parts are removed. When the predicted crown width enters the buffer zone, guided pruning is used to change the growth direction of branches, and sensors and cameras are installed to monitor the dynamics of the crown trees.
8. A system using the tree growth parameter prediction method based on cascaded cyclic network as claimed in any one of claims 1 to 7, characterized in that: It includes data collection module, cascade cycle network construction module, training and prediction module, and GIS integration and early warning module; The data collection module is responsible for collecting historical growth parameter data of trees and performing data preprocessing; The cascaded recurrent network building module builds a multi-layer cascaded recurrent network structure based on the LSTM recurrent unit; The training and prediction module uses historical data to train the cascade recurrent network model, applies the trained model to the prediction stage, and generates prediction results of tree height, diameter at breast height, and crown width; The GIS integration and early warning module integrates the predicted growth parameters into the GIS system, creates new layers in the GIS system to represent different predicted growth parameters, and calculates risk scores. When the risk score exceeds a set threshold, corresponding early warning measures are triggered to ensure the safe operation of the transmission line.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the tree growth parameter prediction method based on the cascade recurrent network according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the tree growth parameter prediction method based on a cascaded recurrent network as claimed in any one of claims 1 to 7 are implemented.