A data management system for forestry information entry platform

Abstract

The present invention discloses a forestry information entry platform data management system, which specifically relates to the field of forestry digitization and is used to solve the spatiotemporal differences in multi-source data and the collaborative optimization problem of felling and transportation. It eliminates the collaborative calculation obstacles caused by the differences in coordinate system and sampling frequency by constructing a unified spatiotemporal grid coding system; by designing a two-way parameter coupling rule, the economic timber distribution is converted into the time constraint of the transportation path, and the real-time road conditions are reversely mapped into the felling priority weight, thereby realizing the intelligent interaction of the prediction and planning model; by extracting the economic value consistency and the transportation structure stress index, combined with iterative optimization, a felling plan with high economic benefit and stable transportation is screened out, so that the comprehensive loss rate and transportation cost converge to a balanced state, thereby enhancing the system's adaptability to complex environments; avoiding the optimization deviation caused by data fragmentation in traditional systems, and comprehensively improving the collaborative benefits of felling and transportation.

Description

Technical Field

[0001] The present invention relates to the field of forestry digitization, and more specifically, to a forestry information entry platform data management system. Background Art

[0002] Current forestry digitalization systems generally employ a multi-tiered data collection architecture. Forest surveyors input plot measurement data via mobile devices, drone-generated aerial imagery is analyzed and integrated into a resource database via a GIS platform, and logging units report transport vehicle status parameters via IoT devices. While existing technologies have developed stock volume prediction models (based on a fusion of growth equations and machine learning algorithms) and path planning models (integrating Dijkstra optimization with real-time road condition sensing), the input data for these two models differ significantly in terms of temporal and spatial benchmarks: stock volume prediction relies on structured survey forms within the coordinate system of forest sub-division boundaries, while transportation route calculations require the use of time-stamped floating vehicle trajectory data and dynamic road damage events. Existing data entry platforms often employ a phased, independent form design, resulting in the discrete storage of timber type classification data, standing timber spatial distribution vector data, and transport equipment status data in heterogeneous databases. This prevents the dynamic spatial coupling of the economic timber distribution hotspots output by the prediction model with the road network capacity data used in real-time by the transport dispatcher.

[0003] The core bottleneck of the existing system lies in the lack of a multi-source data stream fusion architecture: while traditional data entry platforms enable electronic collection of survey data, they fail to establish data mapping rules with spatiotemporal synchronization constraints. Specifically, attribute data such as the diameter at breast height of standing trees and the value coefficient of timber species involved in stock volume forecasting lack strong association annotations with three-dimensional spatial coordinate systems (such as the topological grid for accessibility of logging machinery). Furthermore, the equipment operating data (such as truck load variation curves) and forest road status images required for transportation route planning cannot reversely correct the real-time priority weights of economic timber output. This fragmented storage mode of spatial attributes and business parameters forces the computational units of the "stock volume-transportation capacity" joint optimization model to perform cross-database redundant matching, causing the optimal solution for economic timber output to often deviate from the actual mechanical operating capacity and road network carrying capacity threshold.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a forestry information entry platform data management system, which eliminates the collaborative computing obstacles caused by differences in coordinate system and sampling frequency by constructing a unified space-time grid coding system; by designing a two-way parameter coupling rule, the economic timber distribution is converted into the time constraint of the transportation path, and the real-time road conditions are reversely mapped into the felling priority weight, thereby realizing the intelligent interaction of the prediction and planning model; by extracting the economic value consistency and transportation structure stress index, combined with iterative optimization, the felling scheme with high economic benefits and stable transportation is screened out, so that the comprehensive loss rate and transportation cost converge to a balanced state, thereby enhancing the system's adaptability to complex environments; avoiding the optimization deviation caused by data fragmentation in traditional systems, and comprehensively improving the collaborative benefits of felling and transportation to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A forestry information entry platform data management system, comprising the steps of:

[0008] Space-time weaving module, bipolar coupling module, deep digging and balancing module, and automatic closure module;

[0009] Spatiotemporal Networking Module: Based on the spatial topological relationship between forestry zoning and transportation road networks, it performs unified grid coding on stock volume, equipment trajectory, and road status data, and constructs a multi-source data fusion model with spatiotemporal benchmark alignment.

[0010] Bipolar coupling module: converts economic timber distribution data into a time constraint matrix for transportation routes, while reversely mapping real-time road condition characteristics into a harvesting priority weight matrix;

[0011] Deep mining and balancing module: This module conducts in-depth mining of the extracted matrix to extract key indicators reflecting economic value and transportation conditions, performs pre-screening of logging queues and pre-judgment of path risks, and iterates optimization until the comprehensive loss rate and transportation cost converge to a balance threshold.

[0012] Automatic closed-loop control module: Based on the real-time data stream from mechanical sensors and transport terminals, it dynamically modifies model parameters and triggers priority recalculation to form a self-optimizing closed-loop control link for the operation process.

[0013] In a preferred embodiment, the specific processing process of the spatiotemporal network module is as follows:

[0014] S1-1, based on the geographical scope of forestry zoning and the actual coverage area of ​​the transportation network, a regular two-dimensional grid division method is used to divide the entire forest area into grid cells of fixed size, and a spatiotemporal grid system covering the entire forest area is constructed;

[0015] S1-2, mapping the structured survey data within the boundaries of forest sub-groups into a spatiotemporal grid system according to spatial location;

[0016] S1-3, collect real-time trajectory data of transportation equipment and map it into the spatiotemporal grid system;

[0017] S1-4, extracting the state parameters of forest roads and mapping them to the spatiotemporal grid system;

[0018] S1-5, after completing the grid mapping of storage volume, equipment trajectory and road status data, the multi-source data of each grid unit are integrated to generate a unified spatiotemporal feature matrix.

[0019] In a preferred embodiment, the dipole coupling module includes the following:

[0020] S2-1, based on the constructed spatiotemporal feature matrix, extract standing timber attribute data from the stock volume feature vector of each grid cell and calculate the economic value score of economic timber; economic value score = timber species value coefficient × standing timber volume. The economic value scores of all grid cells are organized into a two-dimensional distribution matrix M1 by time slice;

[0021] S2-2, based on the economic value distribution matrix M1, combined with the transportation route planning requirements, the economic material distribution is converted into the time constraints of transportation, and the time constraint matrix M2 is generated. The transportation time weight W is set for the economic value score of each grid unit. t , the calculation formula is: W t =(V i ×V max )×T base , where V i is the economic value score of grid i, V max The maximum economic value score among all grids, T base is the basic time window for transportation; the time weight of each grid unit W t Fill in the time constraint matrix M2 and connect it to the transportation path planning model as a time window constraint to limit the priority of high-value areas during path optimization.

[0022] In a preferred embodiment, S2-3, obtain road state data from the spatiotemporal feature matrix and calculate the road traffic efficiency coefficient E of each grid unit. r , E r = Traffic capacity index / benchmark traffic capacity×(1-damage degree adjustment factor), and the traffic efficiency coefficient E of all grid cells is r Stored as matrix M3 by time slice;

[0023] S2-4, using the road traffic efficiency coefficient matrix M3, reversely adjust the harvesting priority of economic timber distribution to generate a dynamic weight coefficient W p , W p =V i ×E r , W of all grid cells p Organize into priority weight matrix M4 according to time slice;

[0024] S2-5, pass the time constraint matrix M2 to the transportation path planning model as the time window input of the Dijkstra algorithm; at the same time, feed the priority weight matrix M4 back to the accumulation prediction model through the same interface to update the sorting logic of the logging queue; after each update, the transportation model adjusts the path priority according to M2, and the accumulation model corrects the logging area weight according to M4, forming a closed loop of mutual penetration and correction of parameters.

[0025] In a preferred embodiment, the deep digging and balancing module includes the following contents:

[0026] S3-1, based on the priority weight matrix M4, extract the dynamic weight coefficient from each grid cell, combine the economic value distribution of adjacent grids, and calculate the economic value coherence index. The calculation formula is: in, Represents the dynamic weight coefficient of grid i represents the dynamic weight coefficient of the adjacent grid j; d ij represents the Euclidean distance between the center points of grid i and j; N i Represents the eight-neighborhood set of grid i; for each grid, after calculating its economic value continuity index value, it is stored in the matrix M5 according to grid code and time slice.

[0027] In a preferred embodiment, S3-2 calculates the transport structure stress index based on the spatiotemporal grid system and the time constraint matrix M2, combined with the road traffic efficiency coefficient and equipment trajectory data. The calculation formula is: Among them, T i represents the time-sensitive weight of grid i; T max Indicates the maximum timeliness weight among all grids; represents the road traffic efficiency coefficient of grid i; L i Represents the path length from grid i to the nearest transport destination; for each grid i, after calculating its transport structure stress index value, it is stored in the matrix M6 according to grid code and time slice.

[0028] In a preferred embodiment, S3-3 is to convert the economic value continuity index EVCI i The matrix M5 and the transport structure stress index TNSSIi The matrix M6 comprehensively calculates the judgment value J i , used for pre-screening and risk prediction of felling grids, and establishing candidate felling queues; setting a judgment threshold, if the judgment value is greater than or equal to the judgment threshold, the grid is included in the candidate felling queue C1 and sorted from high to low according to the judgment value; if the judgment value is less than the judgment threshold, it is marked as a risky grid, and extra attention needs to be paid to transportation resource allocation.

[0029] In a preferred embodiment, S3-4, based on the candidate logging queue C1, the comprehensive loss rate D and the transportation cost C are calculated, and iterative optimization is performed. The comprehensive loss rate formula is: Where N represents the number of candidate grids; the transportation cost formula is: C = ∑ i∈C1 (L i ×F×Q i ), where F represents the transportation cost per unit distance; Q i represents the harvesting amount of grid i;

[0030] If the comprehensive loss rate and transportation cost are both less than the corresponding critical value, the balance threshold is obtained. If they exceed the critical value, the grid with the lowest economic value continuity index or the highest transportation structure stress index is removed, the comprehensive loss rate and transportation cost are recalculated, and the iterative update queue is C2;

[0031] S3-5, after each iteration, check whether the comprehensive loss rate and transportation cost are both less than the corresponding critical values; if not converged, continue to dig deeper into the balancing module-4; if converged, stop the iteration, and the current queue C2 is the optimization result.

[0032] In a preferred embodiment, the automated closure module includes the following:

[0033] S4-1, using sensors installed on harvesting machinery and transport equipment to collect dynamic data during the operation in real time;

[0034] S4-2, mapping the collected real-time data to the spatiotemporal grid system and updating the feature matrix of each grid cell;

[0035] S4-3, based on the updated gridded characteristic matrix, dynamically modify the parameters of the storage volume prediction model and the transportation path planning model, and verify the reliability of the initial assumptions;

[0036] S4-4, based on the updated feature matrix, recalculate the time constraint matrix M2 and the priority weight matrix M4; then, re-screen the candidate logging queues according to the judgment value and update the queue C2; ​​at the same time, the transportation path planning model re-runs the Dijkstra algorithm based on the new M2 to adjust the path;

[0037] S4-5, sends the optimized felling queue C2 and transportation path to the felling machinery and transportation terminal; at the same time, continuously monitors the deviation between the real-time data stream and the optimization results, calculates the system performance indicators, and triggers an alarm and adjusts the cycle frequency if an anomaly is detected.

[0038] The technical effects and advantages of the forestry information entry platform data management system of the present invention are as follows:

[0039] The present invention eliminates the obstacles to collaborative calculation caused by differences in coordinate systems and sampling frequencies by constructing a unified spatiotemporal grid coding system and integrating multi-source data such as storage volume, equipment trajectory, and road status. By designing a two-way parameter coupling rule, the distribution of economic timber is converted into time constraints for transportation routes, while the real-time road conditions are reversely mapped into felling priority weights, thus realizing intelligent interaction between prediction and planning models. By extracting economic value consistency and transportation structure stress indicators and combining iterative optimization, felling plans with high economic benefits and stable transportation are screened out, so that the comprehensive loss rate and transportation cost converge to a balanced state. Based on the real-time data stream of mechanical sensors and transportation terminals, parameters are dynamically corrected and a self-optimizing closed loop is formed, thereby enhancing the system's adaptability to complex environments. These technical features work together to significantly improve the spatiotemporal consistency of forestry data management, the efficiency of economic value optimization, and the robustness of the operation process. They avoid the optimization deviation caused by data fragmentation in traditional systems and comprehensively improve the collaborative benefits of felling and transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a structural diagram of a forestry information entry platform data management system of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] Example 1: Figure 1 The present invention provides a forestry information entry platform data management system, comprising: a spatiotemporal network module, a bipolar coupling module, a deep digging and balancing module, and an automatic closing module;

[0043] Spatiotemporal networking module: Based on the spatial topological relationship between forestry zoning and transportation road networks, unified grid coding is performed on the storage volume, equipment trajectory and road status data to build a multi-source data fusion model with spatiotemporal benchmark alignment.

[0044] Bipolar coupling module: converts economic timber distribution data into a time constraint matrix for transportation routes, and reversely maps real-time road condition characteristics into a harvesting priority weight matrix.

[0045] Deep mining and balancing module: Deeply mine the extracted matrix to extract key indicators reflecting economic value and transportation conditions, perform pre-screening of logging queues and prediction of path risks, and iterate and optimize until the comprehensive loss rate and transportation cost converge to the balance threshold.

[0046] Automatic closed-loop control module: Based on the real-time data stream from mechanical sensors and transport terminals, it dynamically modifies model parameters and triggers priority recalculation to form a self-optimizing closed-loop control link for the operation process.

[0047] To address the challenges of collaborative computing caused by differences in spatiotemporal benchmarks and heterogeneous storage across multiple data sources in existing technologies, a unified grid coding system provides a solid data foundation for subsequent bidirectional parameter coupling and optimization calculations. This system integrates multidimensional information from forest surveys, equipment trajectories, and road conditions into a spatiotemporally aligned fusion model, eliminating the fragmentation inherent in traditional phased data entry and providing a reliable spatiotemporal container for cross-model interaction. This in turn enables the bipolar coupling module to deeply explore and dynamically correlate data value.

[0048] The specific processing process of the space-time weaving module is as follows:

[0049] S1-1, define a unified space-time grid system:

[0050] Based on the geographic scope of the forestry zoning and the actual coverage of the transportation network, a regular two-dimensional gridding method was used to divide the entire forest area into fixed-size grid cells. The grid cell side length was set to 50 meters by 50 meters (this can be adjusted based on the complexity of the forest terrain and the required operational accuracy; for example, it can be reduced to 30 meters in mountainous areas). Each grid cell was assigned a unique code identifier in the format of row R and column C (e.g., R001C001 represents row 1, column 1). The geometric center point of each grid cell was bound to longitude and latitude coordinates, recorded in the WGS84 coordinate system (e.g., the center point of R001C001 is at longitude 120.3456° and latitude 30.7890°). Furthermore, a time dimension was introduced for each grid cell, with a time resolution of 5 minutes, generating a timestamped four-dimensional grid identifier (e.g., R001C001-T202502201000 represents the grid cell at 10:00 on February 20, 2025). Through this method, a spatiotemporal grid system covering the entire forest area is constructed as a unified basis for multi-source data mapping.

[0051] S1-2, mapping storage data to a spatiotemporal grid system:

[0052] The structured survey data within the boundaries of the forest sub-group (including attributes such as standing tree diameter at breast height, tree height, and wood species value coefficient) are mapped to the spatiotemporal grid system according to spatial location. The specific process is: read the sampling point coordinates in the survey form (GPS point, such as longitude 120.3460°, latitude 30.7895°), and use the point-in-polygon algorithm to determine the grid cell (such as R001C001) into which the point falls. For each grid cell, the attribute data of all the sampling points within it are counted, and the average value is calculated as the accumulation characteristic of the corresponding grid. For example, average diameter at breast height = ∑ (diameter at breast height of each sampling point) / number of sampling points, average tree height = ∑ (tree height of each sampling point) / number of sampling points, and wood species proportion = number of sampling points of a certain wood species / total number of sampling points (such as pine wood accounts for 70%). If there is no sampling point in a grid, it is marked as a null value. In the time dimension, the collection timestamp of the survey data is aligned to the nearest 5-minute time slice (for example, the collection time 09:58 on February 20, 2025 is aligned to T202502201000), and a gridded stock feature vector is generated (for example, R001C001-T202502201000, [average diameter at breast height 15 cm, average tree height 12 m, pine wood proportion 70%]).

[0053] S1-3, mapping device trajectory data to a spatiotemporal grid system:

[0054] Using GPS positioning modules on transport equipment (such as trucks and logging machinery), real-time trajectory data (including parameters such as latitude and longitude, speed, and load) is collected and mapped onto a spatiotemporal grid system. The method involves linearly interpolating the trajectory points at predetermined intervals to calculate the grid cells traversed by the equipment within each time slice. For example, if a truck moves from coordinates (120.3450°, 30.7880°) to (120.3470°, 30.7900°) within time slice T202502201000, the intersection of the trajectory line and the grid boundaries indicates that it passed through grid cells R001C001 and R001C002. If the truck crosses multiple grid cells, parameters are allocated based on the proportion of time spent in each grid cell (e.g., if the truck spends 3 minutes in R001C001, 60% of a 5-ton load is counted as 3 tons). Finally, gridded equipment trajectory data is generated, such as R001C001-T202502201000, [truck ID001, load 3 tons, speed 20 km / h], to achieve temporal and spatial alignment with the accumulation data.

[0055] S1-4, mapping road status data to a spatiotemporal grid system:

[0056] Based on drone aerial imagery, GIS platform analysis results, and road sensor data, we extracted forest road status parameters (including road width, damage level, and water depth) and mapped them to a spatiotemporal grid system. The specific process involves performing a spatial intersection analysis between the road network vector data (linear features) and grid cells to determine the grid cell covered by each road segment (e.g., a road segment passes through R001C001 and R001C002). For each grid cell, the status parameters of the road segments within it are summarized to calculate a capacity index. The capacity index is calculated as follows: Capacity Index = Road Width × (1 - Damage Level) × Loadability Factor, where road width is expressed in meters (e.g., 3 meters), damage level is expressed as a percentage (e.g., 10% is 0.1), and the loadability factor is determined based on the water depth (1.0 for water depths <10 cm, 0.8 for 10-30 cm, and 0.5 for >30 cm). For example, if the road width in R001C001 is 3 meters, the damage level is 10%, and the water depth is 5 cm, then the capacity index = 3 × (1 - 0.1) × 1.0 = 2.7. In the temporal dimension, the road status is updated at 5-minute intervals (e.g., sensors report changes in water depth), generating dynamic data, such as R001C001-T202502201000, [capacity index 2.7, damage level 10%].

[0057] S1-5, build a multi-source data fusion model:

[0058] After completing the grid mapping of storage volume, equipment trajectory, and road status data, the multi-source data for each grid cell is integrated to generate a unified spatiotemporal feature matrix. The specific method is to construct a structure for each grid cell, containing three types of fields: storage volume fields (such as average diameter at breast height and wood species proportion), equipment trajectory fields (such as load and speed), and road status fields (such as capacity indicators and damage degree). These fields are organized into time series using timestamps as indexes. For example, the feature matrix for R001C001-T202502201000 is [Storage volume: [15cm, 12m, 70%], Equipment: [3 tons, 20km / h], Road: [2.7, 10%]]. This data is stored in a database that supports spatial queries (such as PostgreSQL + PostGIS), with the index fields including grid code and timestamp. This fusion model eliminates the problems of coordinate system differences and inconsistent sampling frequencies, providing spatiotemporal data support for subsequent steps.

[0059] By establishing a unified spatiotemporal grid coding system based on the spatial topological relationship between forestry zoning and transportation road networks, we achieved grid mapping and multi-source fusion of stock volume, equipment trajectory, and road status data, constructing a data model with consistent spatiotemporal datums. This process effectively eliminated computational obstacles associated with differences in coordinate systems and inconsistent sampling frequencies, providing high-precision, interactive data support for subsequent steps and ensuring efficient and collaborative operation of the system within a unified spatiotemporal framework.

[0060] Building on the spatiotemporal alignment of multi-source data fusion provided by the Spatiotemporal Networking Module, the Bipolar Coupling Module focuses on breaking down the parameter barriers between stock volume forecasting and transportation route planning. By designing bidirectional coupling rules, it tightly connects economic timber distribution with transportation constraints. This not only transforms static economic value forecasts into dynamic transportation time constraints, but also leverages real-time road conditions to reversely optimize harvesting priorities. This provides intelligent parameter interaction support for the joint optimization and screening of the Deep Digging Balance Module, ushering in the transition from data fusion to dynamic decision-making.

[0061] The bipolar coupling module includes the following:

[0062] S2-1, calculate the economic value score of economic material distribution and generate the initial distribution matrix:

[0063] Based on the constructed spatiotemporal feature matrix, the standing tree attribute data (such as average DBH, average tree height, and wood species proportion) are extracted from the stock volume feature vector of each grid unit to calculate the economic value score of economic timber. The specific calculation method is: economic value score = wood species value coefficient × standing tree volume, where the wood species value coefficient is determined by the preset forestry economic evaluation table (for example, 1.2 for pine and 1.0 for fir), and the standing tree volume is estimated by the volume formula of DBH and tree height (such as V = π × (DBH / 2)2 × tree height × shape coefficient, with the shape coefficient being 0.45). For example, if the average DBH of grid R001C001-T202502201000 is 15 cm, the average tree height is 12 m, and the pine wood proportion is 70%, then the volume of a single pine tree = 3.14 × (0.075)2 × 12 × 0.45 ≈ 0.095 m 3 , economic value score = 1.2 × 0.095 ≈ 0.114. The economic value scores of all grid cells are organized into a two-dimensional distribution matrix M1 (rows and columns correspond to grid codes, such as M1[R001C001,T202502201000] = 0.114) by time slice, which serves as the initial quantitative table for economic material distribution.

[0064] S2-2, convert the economic material distribution matrix into the time constraint matrix of the transportation path:

[0065] Based on the economic value distribution matrix M1, combined with the transportation route planning requirements, the economic material distribution is converted into the time constraint conditions of transportation to generate the time constraint matrix M2. In specific implementation, the transportation time constraint weight W is set for the economic value score of each grid unit. t , the calculation formula is: W t =(V i ×V max )×T base , where V i V is the economic value score of the grid. max The maximum economic value score among all grids, T base is the basic time window for transportation (the default is 2 hours, i.e. 120 minutes). For example, if V max =0.2, the economic value score of R001C001 is V i =0.114, then W t =(0.114 / 0.2)×120=68.4 minutes, which means that the economic materials of this grid need to be transported within 68.4 minutes. t Fill in the time constraint matrix M2 (the structure is the same as M1, such as M2[R001C001, T202502201000] = 68.4) and connect it to the transportation path planning model as a time window constraint to limit the priority of high-value areas during path optimization.

[0066] S2-3, extract real-time traffic characteristics and calculate road traffic efficiency coefficient:

[0067] Obtain road status data (such as capacity index and damage degree) from the spatiotemporal feature matrix and calculate the road efficiency coefficient E of each grid unit r , reflecting the impact of real-time road conditions on transportation capacity. The specific calculation formula is: E r = Capacity Index / Baseline Capacity × (1-Damage Adjustment Factor), where the capacity index has been calculated in the spatiotemporal network module (e.g., 2.7 for R001C001), the base capacity is the ideal road design value (e.g., 5.0), and the damage adjustment factor is assigned based on the water depth or damage level (e.g., 0 for water depth <10cm, 0.2 for 10-30cm, and 0.5 for >30cm). For example, if the capacity index of R001C001 is 2.7, the damage level is 10%, and the water depth is 5cm, then E r =2.7 / 5.0×(1-0)=0.54. The road traffic efficiency coefficient E of all grid cells is r The matrix M3 is stored according to the time slice (eg, M3[R001C001, T202502201000]=0.54), providing a quantitative basis for the real-time road condition for reverse mapping.

[0068] S2-4, reverse mapping of real-time road condition characteristics to dynamic weight coefficients of felling priority:

[0069] Using the road traffic efficiency coefficient matrix M3, reversely adjust the harvesting priority of economic timber distribution to generate a dynamic weight coefficient W p The specific calculation method is: W p =V i ×E r , where V i is the economic value score of the grid (from M1), E r is the road traffic efficiency coefficient (from M3). For example, for R001C001, V i =0.114, E r =0.54, then W p =0.114×0.54≈0.0616, which represents the dynamic weight of the logging priority of the grid after being affected by the road conditions. p The data are organized into a priority weight matrix M4 (e.g., M4[R001C001, T202502201000] = 0.0616) by time slices and fed back to the stocking prediction model to dynamically adjust the selection priority of the harvesting area to ensure that high-value areas match transportation accessibility.

[0070] S2-5, data interaction interface to realize bidirectional parameter coupling:

[0071] A dynamic parameter interaction interface is established between the stock volume prediction model and the transportation routing model to ensure real-time execution of bidirectional coupling rules. For implementation, a data transmission protocol is designed: the time constraint matrix M2 is passed to the transportation routing model via an API (e.g., a RESTful service) as the time window input for the Dijkstra algorithm. Simultaneously, the priority weight matrix M4 is fed back to the stock volume prediction model via the same interface to update the sorting logic of the harvesting queue. The interface update frequency coincides with the time slice (every 5 minutes). After each update, the transportation model adjusts the path priority based on M2, and the stock volume model modifies the harvesting area weights based on M4, forming a closed loop of parameter interpenetration and correction. For example, the time constraint weight of R001C001, 68.4 minutes, restricts the transportation route selection, while its priority weight of 0.0616 adjusts the harvesting order inversely, ensuring the coordinated optimization of economic value and transportation efficiency.

[0072] After each update, the transportation model adjusts the path priority according to M2, and the storage model modifies the harvesting area weight according to M4. The specific process is as follows:

[0073] After each update, the transportation model adjusts the path priority according to the time constraint matrix M2. The specific process is to use the time weight in M2 (such as 68.4 minutes in R001C001) as the time window constraint input of the Dijkstra algorithm. When calculating the optimal transportation path, the grid cells that meet the time window (i.e., areas with high economic value and requiring rapid transportation) are given priority, thereby generating the shortest path sequence from the felling point to the destination. At the same time, the accumulation model modifies the felling area weight according to the priority weight matrix M4. The specific method is to use the dynamic weight coefficient in M4 (such as 0.0616 in R001C001) as the basis for sorting the felling queue. Through the weighted sorting algorithm (such as weight-based heap sorting), the priority of the felling grid is rearranged, so that the grid with high road traffic efficiency and reasonable economic value score (such as E r Higher and V i Moderate areas) are selected first, while grids with blocked roads (E r The weight of the harvesting plan (lower) is reduced and moved back, forming a dynamic adjustment of the harvesting plan; these two processes are closed-loop through a parameter interaction interface every 5 minutes. M2 pushes the transportation path to tilt towards high-value areas, and M4 reversely optimizes the harvesting selection to match the transportation capacity, ensuring real-time coordinated optimization of economic benefits and transportation efficiency.

[0074] The bipolar coupling module converts economic timber distribution data into a time-constraint matrix by calculating economic value scores and road efficiency coefficients. It also maps real-time road condition characteristics into dynamic weight coefficients for harvesting priorities, establishing a bidirectional parameter coupling interface between the stocking volume and transportation models. This process enables the mutual penetration and correction of parameters in the two models, providing a precise input basis for subsequent optimization and ensuring real-time matching of economic benefits and transportation capacity.

[0075] Leveraging the bidirectional parameter coupling results of the Bipolar Coupling Module, the Deep Digging Equilibrium Module further advances system optimization. By introducing the Economic Value Coherence Index and the Transport Structure Stress Index, this module enables pre-screening of logging queues and predicting path risks. With the dual objectives of optimizing economic value and minimizing transport losses, iterative optimization resolves local conflicts, laying the foundation for a global equilibrium solution in the automated closure module's real-time adaptive control, marking a move from static planning to dynamic execution.

[0076] The deep digging and balancing module includes the following:

[0077] S3-1, based on the priority weight matrix M4, extract the dynamic weight coefficient from each grid cell (such as 0.0616 for R001C001), and combine it with the economic value distribution of adjacent grids to calculate the Economic Value Continuity Index (EVCI), which is used to measure the spatial continuity of the economic benefits of the harvesting area. The calculation formula is:

[0078]

[0079] ,in:

[0080] represents the dynamic weight coefficient of grid i (from M4), which represents the comprehensive weight of economic value and transportation accessibility.

[0081] Represents the dynamic weight coefficient of the adjacent grid j.

[0082] d ij Represents the Euclidean distance between the center points of grid cells i and j (unit: meter).

[0083] N i Represents the eight-neighborhood set of grid i (the surrounding 8 grids).

[0084] For each grid, the economic value continuity index is calculated and stored in matrix M5 by grid code and time slice. Matrix M5 is a two-dimensional array with rows corresponding to grid codes (e.g., R001C001) and columns corresponding to timestamps (e.g., T202502201000). Each element stores the economic value continuity index value for that grid at a specific time slice, which is used for subsequent logging queue screening and optimization analysis.

[0085] S3-2, based on the spatiotemporal grid system and the time constraint matrix M2, combined with the road traffic efficiency coefficient and equipment trajectory data, calculates the transport structure stress index (TNSSI) to assess the stability and risk of the transport route. The calculation formula is:

[0086]

[0087] ,in:

[0088] T i Represents the timeliness weight of grid i (from M2, unit: minute), which represents the transportation time window constraint.

[0089] T max Indicates the maximum timeliness weight among all grids (e.g. 120 minutes).

[0090] Represents the road traffic efficiency coefficient of grid i (from M3), ranging from 0 to 1, where 1 represents the best traffic state.

[0091] L i Represents the path length from grid i to the nearest transportation destination (unit: meter, calculated by Dijkstra algorithm).

[0092] For each grid i, the transport structure stress index value is calculated and stored in matrix M6 by grid code and time slice. Matrix M6 is a two-dimensional array with rows corresponding to grid codes (e.g., R001C001) and columns corresponding to timestamps (e.g., T202502201000). Each element stores the transport structure stress index value for that grid at a specific time slice, which is used for subsequent logging queue screening and optimization analysis.

[0093] S3-3, Economic Value Continuity Index EVCI i The matrix M5 and the transport structure stress index TNSSI i The matrix M6 comprehensively calculates the judgment value J i , used for pre-screening and risk prediction of felling grids, establishing candidate felling queues, and designing a judgment value J i , and its calculation formula is: Among them EVCI max is the maximum value in matrix M5, TNSSI max is the maximum value in the matrix M6, and α1 is the weight coefficient (ranging from 0 to 1, for example, 0.6, indicating that the economic value has a slightly higher priority).

[0094] The judgment process is as follows: set a judgment threshold. If the judgment value is greater than or equal to the judgment threshold, the grid is included in the candidate logging queue C1 and sorted from high to low according to the judgment value; if the judgment value is less than the judgment threshold, it is marked as a risky grid and requires additional attention to transportation resource allocation.

[0095] S3-4, based on the candidate logging queue C1, calculate the comprehensive loss rate D and transportation cost C, and perform iterative optimization. The comprehensive loss rate formula is: Where N represents the number of candidate grids.

[0096] The transportation cost formula is: C = ∑ i∈C1 (L i ×F×Q i );

[0097] L i Indicates the path length from grid i to the end point (unit: meter).

[0098] F represents the transportation cost per unit distance (unit: yuan / meter / ton, such as 0.5).

[0099] Q i represents the harvest volume of grid i (unit: tons, estimated based on standing wood volume).

[0100] If the comprehensive loss rate and transportation cost are both less than the corresponding critical value, the balance threshold is obtained. If they exceed the critical value, the grid with the lowest economic value continuity index or the highest transportation structure stress index is removed, the comprehensive loss rate and transportation cost are recalculated, and the iterative update queue is C2.

[0101] After each iteration, S3-5 checks whether the combined loss rate and transportation cost are both below their respective critical values. If convergence has not yet occurred, the optimization continues in Equilibrium Module-4. If convergence has occurred, iterations cease, and the current queue C2 is the optimization result. Output includes: a list of harvested grids, a transportation route (generated by Dijkstra, e.g., R001C001 → Destination A, 500 meters long), harvested volume, and cost. These are stored as structured data (e.g., JSON) and passed to the automated closure module. This process avoids locally greedy optimization and achieves global equilibrium.

[0102] The Deep Digging Balance Module dynamically calculates EVCI and TNSSI, combines them with judgment values ​​to select candidate harvesting grids, and iteratively optimizes the comprehensive loss rate and transportation costs, ultimately converging to a global equilibrium solution. This process not only selects harvesting areas with consistent economic value and stable transportation, but also outputs optimized harvesting queues and transportation routes, providing a reliable decision-making basis for subsequent real-time adjustments and avoiding the hidden losses caused by local greedy optimization.

[0103] Based on the harvesting and transportation plans optimized by the Deep Digging and Balancing Module, the Automated Closure Module incorporates real-time data streams from machine sensors and transport terminals to establish a self-optimizing closed-loop control chain to respond to dynamic changes in actual operations. This seamlessly integrates static models with field feedback, enhancing the system's adaptability to complex environments through parameter correction and priority calculation. This injects continuous optimization vitality into the entire forestry information entry platform, completing the closed-loop sublimation from theoretical design to practical application.

[0104] The automated closure module includes the following:

[0105] S4-1 uses sensors installed on logging machinery (such as felling machines) and transport equipment (such as trucks) to collect dynamic data during operations in real time. Specific data includes: operating status parameters of logging machinery (such as logging volume per minute, fuel consumption rate, and fault status signals, expressed in cubic meters / minute, liters / minute, and Boolean values, respectively), and operating parameters of transport terminals (such as real-time load, speed, and GPS location, expressed in tons, kilometers / hour, and longitude and latitude, respectively). Data collection is set to once every 5 seconds and transmitted to a central server via an IoT protocol (such as MQTT).

[0106] S4-2 maps the collected real-time data to the spatiotemporal grid system of the spatiotemporal network module and updates the feature matrix of each grid cell (including the storage volume, equipment trajectory, and road status fields). The specific method is as follows: for transport terminal data, the current grid is determined based on the GPS location and the equipment trajectory field is updated; for felling machine data, the storage volume field is updated based on the operation location, the remaining standing timber volume is calculated, and the economic value score is adjusted synchronously (recalculated according to the formula in step S2-1). If a fault signal is detected (such as felling machine fault = true), the grid is marked as non-operational. The updated feature matrix is ​​refreshed every 5 minutes to ensure that it reflects the latest operation status.

[0107] S4-3, based on the updated grid feature matrix, dynamically correct the parameters of the stock volume prediction model and the transportation route planning model, and verify the reliability of the initial assumptions. In specific implementation, for the stock volume prediction model, analyze the deviation between the real-time felling volume and the predicted felling volume, and adjust the parameters in the growth equation (such as the shape coefficient or growth rate) through the linear regression method: use the real-time felling volume as the dependent variable and the initial prediction parameters as the independent variables, refit the parameter values ​​to ensure that the model output is close to the actual operation data; at the same time, if the timber species value coefficient is based on the market price assumption, then according to the actual felling species ratio fed back by the transportation terminal (such as calculated by load and timber species identification), update the coefficient in proportion and weight. For the transportation route planning model, recalculate the road traffic efficiency coefficient based on the real-time road status data (such as the water depth reported by the sensor or the change in equipment load) by substituting the latest traffic capacity index and damage adjustment factor according to the formula in step S2-3 to update the coefficient value. In terms of hypothesis verification, a deviation threshold (e.g., the upper limit of the relative error between the forecast and the actual value) is set to test each key assumption (e.g., stable weather, fault-free equipment). If real-time data exceeds the threshold, the corresponding assumption is marked as unreliable, its impact on the model is eliminated (e.g., after removing the stable weather assumption, the dynamic weight of rainfall on road conditions is increased), and parameter recalculation is triggered. If all assumptions are reliable, the current parameters are retained. This ensures that model parameters are consistent with the real-time environment, improving the accuracy of forecasting and planning.

[0108] S4-4, using the revised model parameters, triggers the priority recalculation of the bipolar coupling module and the deep digging balance module to optimize the felling queue and transportation path. The specific process is: according to the updated feature matrix, recalculate the time constraint matrix M2 (for example, the time constraint weight of R001C001 is adjusted from 68.4 minutes to 72 minutes due to the increase in load) and the priority weight matrix M4 (for example, W p From 0.0616 to 0.058, due to E r Then, according to the judgment value J iThe candidate harvesting queues are re-screened, and queue C2 is updated. Simultaneously, the transport routing model reruns the Dijkstra algorithm based on the new M2, adjusting the route (for example, avoiding flooded sections). This recalculation occurs every five minutes to ensure that harvesting and transport plans adapt to real-time changes. For example, if R001C001 fails and is inoperable, it is removed from C2 and priority is transferred to R002C002.

[0109] S4-5, establish a real-time data closed-loop link between the physical device layer and the algorithm layer to achieve self-optimization control. In specific implementation, a feedback mechanism is designed: the optimized felling queue C2 and transportation path are sent to the felling machinery and transportation terminal through the API interface (such as WebSocket) to guide on-site operations; at the same time, the deviation between the real-time data stream and the optimization results (such as the actual execution rate of felling volume and the error in path travel time) is continuously monitored, and the system performance indicators (such as adaptability index = 1-deviation percentage, target > 0.9) are calculated. If an abnormality is detected (such as the adaptability index drops to 0.85 due to sudden weather changes), an alarm is triggered and the cycle frequency is adjusted (such as from 5 minutes to 2 minutes). For example, if the R001C001 path is blocked by waterlogging, the system automatically switches to the backup path and updates C2, forming a closed-loop adaptive control to ensure the robustness of the operation process in complex environments (such as equipment failure and rainfall).

[0110] The automated closed-loop module updates the grid feature matrix by collecting real-time data streams, dynamically correcting model parameters and triggering priority recalculation, optimizing logging queues and transportation routes. This ultimately forms a closed-loop control link between the physical device layer and the algorithm layer. This process significantly improves the system's robustness to complex environments such as sudden weather changes and equipment failures, ensuring that the economic benefits and transportation efficiency of the operation process are continuously optimized under dynamic conditions.

[0111] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0112] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0113] It should be noted that, in this document, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article or device. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0114] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A forestry information entry platform data management system, characterized in that: Including steps: Space-time weaving module, bipolar coupling module, deep digging and balancing module, and automatic closure module; Spatiotemporal Networking Module: Based on the spatial topological relationship between forestry zoning and transportation road networks, it performs unified grid coding on stock volume, equipment trajectory, and road status data, constructs a multi-source data fusion model with spatiotemporal benchmark alignment, and generates a unified spatiotemporal feature matrix. Bipolar coupling module: converts economic timber distribution data into a time constraint matrix for transportation routes, while reversely mapping real-time road condition characteristics into a harvesting priority weight matrix; The bipolar coupling module includes the following: S2-1, based on the constructed spatiotemporal feature matrix, extract standing timber attribute data from the stock volume feature vector of each grid cell and calculate the economic value score of economic timber; economic value score = timber species value coefficient × standing timber volume. The economic value scores of all grid cells are organized into a two-dimensional distribution matrix M1 by time slice; S2-2, based on the economic value distribution matrix M1, combined with the transportation route planning requirements, converts the economic material distribution into the time constraints of transportation, generates the time constraint matrix M2, and sets the transportation time weight for the economic value score of each grid unit. , the calculation formula is: ,in For Grid The economic value score of Score the maximum economic value among all grids, is the basic time window for transportation; the timeliness weight of each grid cell is Fill in the time constraint matrix M2 and connect it to the transportation route planning model as a time window constraint to limit the priority of high-value areas during route optimization; S2-3, obtain road status data from the spatiotemporal feature matrix and calculate the road traffic efficiency coefficient of each grid unit , = Traffic capacity index / benchmark traffic capacity×(1-damage degree adjustment factor), the traffic efficiency coefficient of all grid cells is calculated. Stored as matrix M3 by time slice; S2-4, using the road traffic efficiency coefficient matrix M3, reversely adjust the harvesting priority of economic timber distribution to generate a dynamic weight coefficient , , all grid cells Organize into priority weight matrix M4 according to time slice; S2-5: The time constraint matrix M2 is passed to the transportation route planning model as the time window input of the Dijkstra algorithm. At the same time, the priority weight matrix M4 is fed back to the storage volume prediction model through the same interface to update the sorting logic of the logging queue. After each update, the transportation model adjusts the path priority according to M2, and the storage volume model corrects the logging area weight according to M4, forming a closed loop of mutual penetration and correction of parameters. Deep mining and balancing module: This module conducts in-depth mining of the extracted matrix to extract key indicators reflecting economic value and transportation conditions, performs pre-screening of logging queues and pre-judgment of path risks, and iterates optimization until the comprehensive loss rate and transportation cost converge to the equilibrium threshold. The queue is then iteratively updated to C2. Automated closed-loop control module: Based on real-time data streams from mechanical sensors and transport terminals, it dynamically modifies model parameters and triggers priority recalculation, forming a self-optimizing closed-loop control link for the operation process. The automated closure module includes the following: S4-1, using sensors installed on harvesting machinery and transport equipment to collect dynamic data during the operation in real time; S4-2, mapping the collected real-time data to the spatiotemporal grid system and updating the feature matrix of each grid cell; S4-3, based on the updated gridded characteristic matrix, dynamically modify the parameters of the storage volume prediction model and the transportation path planning model, and verify the reliability of the initial assumptions; S4-4, based on the updated feature matrix, recalculate the time constraint matrix M2 and the priority weight matrix M4; then, re-screen the candidate logging queues according to the judgment value and update the queue C2; ​​at the same time, the transportation path planning model re-runs the Dijkstra algorithm based on the new M2 to adjust the path; S4-5, sends the optimized felling queue C2 and transportation path to the felling machinery and transportation terminal; at the same time, continuously monitors the deviation between the real-time data stream and the optimization results, calculates the system performance indicators, and triggers an alarm and adjusts the cycle frequency if an anomaly is detected.

2. A forestry information entry platform data management system according to claim 1, characterized in that: The specific processing process of the space-time weaving module is as follows: S1-1, based on the geographical scope of forestry zoning and the actual coverage area of ​​the transportation network, a regular two-dimensional grid division method is used to divide the entire forest area into grid cells of fixed size, and a spatiotemporal grid system covering the entire forest area is constructed; S1-2, mapping the structured survey data within the boundaries of forest sub-groups into a spatiotemporal grid system according to spatial location; S1-3, collect real-time trajectory data of transportation equipment and map it into the spatiotemporal grid system; S1-4, extracting the state parameters of forest roads and mapping them to the spatiotemporal grid system; S1-5, after completing the grid mapping of storage volume, equipment trajectory and road status data, the multi-source data of each grid unit are integrated to generate a unified spatiotemporal feature matrix.

3. A forestry information entry platform data management system according to claim 2, characterized in that: The deep digging and balancing module includes the following: S3-1, based on the priority weight matrix M4, extract the dynamic weight coefficient from each grid cell, combine the economic value distribution of adjacent grids, and calculate the economic value coherence index. The calculation formula is: ,in, Representation Grid Dynamic weight coefficient Represents adjacent grids Dynamic weight coefficient of Representation Grid and Euclidean distance between center points; Representation Grid For each grid, after calculating its economic value continuity index value, it is stored in the matrix M5 according to the grid code and time slice.

4. A forestry information entry platform data management system according to claim 3, characterized in that: S3-2, based on the space-time grid system and the time constraint matrix M2, combined with the road traffic efficiency coefficient and equipment trajectory data, calculate the transport structure stress index. The calculation formula is: ,in, Representation Grid The time weight of Indicates the maximum timeliness weight among all grids; Representation Grid Road traffic efficiency coefficient; Representation Grid The path length to the nearest transport destination; for each grid ,After calculating the transport structure stress index value, it is stored in the matrix M6 according to grid coding and time slice.

5. A forestry information entry platform data management system according to claim 4, characterized in that: S3-3 is the economic value consistency index Matrix M5 and transport structure stress index The matrix M6 is used to calculate the judgment value , used for pre-screening and risk prediction of felling grids, and establishing candidate felling queues; setting a judgment threshold, if the judgment value is greater than or equal to the judgment threshold, the grid is included in the candidate felling queue C1 and sorted from high to low according to the judgment value; if the judgment value is less than the judgment threshold, it is marked as a risky grid, and extra attention needs to be paid to transportation resource allocation.

6. A forestry information entry platform data management system according to claim 5, characterized in that: S3-4, calculate the comprehensive loss rate based on the candidate logging queue C1 and transportation costs , perform iterative optimization, and the comprehensive loss rate formula is: ,in represents the number of candidate grids; the transportation cost formula is: ,in, Indicates the transportation cost per unit distance; Representation Grid the amount of felling; If the comprehensive loss rate and transportation cost are both less than the corresponding critical value, the balance threshold is obtained. If they exceed the critical value, the grid with the lowest economic value continuity index or the highest transportation structure stress index is removed, the comprehensive loss rate and transportation cost are recalculated, and the iterative update queue is C2; S3-5, after each iteration, check whether the comprehensive loss rate and transportation cost are both less than the corresponding critical values; if not converged, continue to dig deeper into the balancing module-4; if converged, stop the iteration, and the current queue C2 is the optimization result.

Images