A method and system for identifying and warning of ecological carrying capacity transitions in hydropower projects

Abstract

This invention provides a method and system for identifying and warning of ecological carrying capacity transitions in hydropower projects, applicable to the field of data processing. This application focuses on the impact zone of hydropower projects, integrating multi-period data on ecological carrying capacity, engineering disturbance, construction restoration, and geographic environment. Through spatial registration, temporal matching, and unified coding preprocessing, a six-level carrying capacity system is constructed, generating a temporal level map. Based on inter-period pixel tracking, a transition matrix is ​​constructed, and flux and rate are calculated. A transition risk index is constructed by integrating core features, and multiple risk zones are identified by associating engineering units. Graded early warning and risk tracing are implemented, and verification and correction are performed using measured data, balancing accuracy, efficiency, and adaptability to form a comprehensive ecological carrying capacity transition identification and risk warning technology system.

Description

Technical Field

[0001] This invention relates to the field of data processing, and in particular to a method and system for identifying and warning of the leap in the ecological carrying capacity level of hydropower projects. Background Technology

[0002] While existing technologies for assessing the ecological carrying capacity and providing early warning of risks in hydropower projects can achieve static evaluation and macro-trend analysis, they cannot meet the needs of refined and dynamic risk management throughout the entire life cycle of hydropower projects. Their core shortcomings are as follows: Existing technologies tend to focus on single-period or annual static evaluations, making it difficult to track the dynamic evolution of ecological carrying capacity levels during the entire life cycle of hydropower projects, including upgrades, downgrades, and leaps. They also cannot distinguish the differentiated leap characteristics of construction disturbances, natural restoration, and secondary disturbances, making it difficult to accurately identify the trajectory of risk evolution.

[0003] Existing area comparisons and simple transfer analyses are all macro-level total statistics, which cannot achieve precise temporal matching and cross-period tracking at the 30m raster pixel level; they are difficult to reveal the spatial reorganization process of "unchanged total amount but local deterioration", and cannot pinpoint the precise location of single-point severe degradation or cross-level degradation.

[0004] Existing technologies lack refined indicators such as the extent of degradation, cross-level degradation, high-risk inflows, and leap rates: they cannot quantify the severity of ecological degradation, nor can they identify the rapid deterioration window during construction peaks and new construction start-up periods, resulting in insufficient timeliness of early warnings.

[0005] The existing early warning results are not spatially correlated with specific engineering disturbance units such as construction roads, spoil disposal sites, and tunnel entrances; it is impossible to locate the source of risk, quantify the risk contribution of each unit, and the early warning results are difficult to directly translate into targeted engineering control measures.

[0006] The existing early warning thresholds are singular and the classification is simple, failing to distinguish between severe degradation at a single point, concentrated degradation in a localized area, and overall risk in a region; it cannot achieve a four-level refined early warning system and is difficult to match the hierarchical control requirements of "single-point rectification, localized control, and regional prevention and control".

[0007] Existing technologies cannot identify the effectiveness of ecological restoration and improvement and the risk of re-disturbance in both directions; they cannot distinguish between restoration and upgrading during the operation period and secondary downgrading during the new construction start-up period, making it difficult to support the assessment of restoration effectiveness and the early warning of secondary disturbances in cascade projects.

[0008] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore includes information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0009] According to one aspect of this application, a method for identifying and risk warning of ecological carrying capacity level transitions in hydropower projects is provided, comprising: acquiring multi-period ecological carrying capacity evaluation results, spatial data of project disturbance, construction stage information, ecological restoration area and auxiliary geographical environment data of the hydropower project's impact area; completing data preprocessing based on spatial registration, temporal matching and unified level coding, constructing a six-category ecological carrying capacity level system through level classification rules, and generating multi-period ecological carrying capacity level maps; establishing inter-period tracking relationships with pixels as units, constructing a level transition matrix and calculating upgrading, downgrading, stability, cross-level transition and high-risk inflow / outflow flux indicators, incorporating time interval parameters to calculate the level transition rate; constructing pixel-level and regional-level transition risk indices, and fusing downgrading through a linear weighted model. The system identifies key characteristics such as magnitude, high-risk transition, and cross-level downgrade to generate a risk feature set for ecological carrying capacity level transitions. Based on spatial overlay and buffer zone analysis, the transition results are correlated with engineering disturbance units such as construction roads, spoil heaps, and tunnel entrances to identify degradation zones, restoration and improvement zones, stability maintenance zones, and re-disturbance risk zones. A multi-threshold grading mechanism is used to classify four early warning levels, and a risk tracing and control suggestion generation module is introduced. The system combines the risk contribution of engineering units to output early warning results and control strategies. Based on data from all stages of the process and actual monitoring data, accuracy verification and dynamic correction are performed. A multi-index comprehensive evaluation algorithm is used to balance the requirements of transition identification accuracy, early warning response efficiency, and engineering adaptability, generating ecological carrying capacity level transition identification and risk early warning results suitable for the entire process of hydropower projects.

[0010] Another aspect of this application discloses a system for identifying and warning of ecological carrying capacity level transitions in hydropower projects, comprising: a multi-source data acquisition module for acquiring multi-period ecological carrying capacity evaluation results, spatial data of engineering disturbances, construction stage information, ecological restoration area and auxiliary geographical environment data of the hydropower project impact area, and generating basic dataset information; a data preprocessing and level construction module for performing spatial registration, temporal matching and unified level coding, constructing six types of ecological carrying capacity levels according to level classification rules, and generating multi-period ecological carrying capacity level map information; a level transfer and rate calculation module for establishing inter-period tracking relationships at the pixel level, constructing a level transfer matrix, calculating upgrade, downgrade, stability, cross-level transition and high-risk inflow / outflow flux, incorporating time interval parameters to calculate level transfer rate, and generating transfer rate calculation result information; and a transition risk index construction module for constructing pixel-level and regional-level transition risk indices, and fusing downgrade magnitudes through a linear weighted model. The system comprises four modules: a high-risk transition and a cross-level downgrade core characteristic module to generate a risk feature set for ecological carrying capacity level transitions; an engineering association and zoning identification module to associate transition results with engineering disturbance units such as construction roads, spoil heaps, and tunnel entrances based on spatial overlay and buffer zone analysis, identifying degradation zones, restoration and improvement zones, stability maintenance zones, and re-disturbance risk zones, and generating zoning identification results; an early warning classification and control generation module to divide four early warning levels using a multi-threshold classification judgment mechanism, introducing a risk tracing and control suggestion generation module, and combining the risk contribution of engineering units to output early warning results and control strategies, generating early warning and control result information; and a verification, correction, and result output module to conduct accuracy verification and dynamic correction based on full-process stage data and actual monitoring data, balancing transition identification accuracy, early warning response efficiency, and engineering adaptability requirements through a multi-index comprehensive evaluation algorithm, generating ecological carrying capacity level transition identification and risk early warning result information suitable for the entire process of hydropower projects.

[0011] According to another aspect of this application, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a second processor, implements the above-described method for identifying and warning of the transition of ecological carrying capacity levels in hydropower projects.

[0012] The beneficial effects of this application are as follows: It integrates multi-period data on ecological carrying capacity, engineering disturbance, construction restoration, and geographic environment; through spatial registration, temporal matching, and unified coding preprocessing, it constructs six carrying capacity levels and generates a temporal level map; it constructs a transfer matrix based on inter-period pixel tracking to calculate flux and rate; it integrates the characteristics of degradation magnitude, high-risk transition, and cross-level degradation to construct a transition risk index; it identifies four types of risk zones by associating engineering disturbance units; it provides graded early warning and traces the source of risks; and it combines measured data for verification and correction, balancing accuracy, efficiency, and adaptability to form a full-process dynamic identification and accurate early warning technology system.

[0013] This application aims to achieve full-cycle dynamic identification, covering the entire process from pre-construction, construction, operation, and new construction startup. It accurately depicts the trajectory of bearing capacity level transitions, overcoming the limitations of static evaluation. Employing pixel-level tracking combined with matrix calculations, it quantifies the intensity and rate of transitions, precisely locating degraded patches and rapid deterioration windows. Furthermore, it correlates with disturbance units such as construction roads and spoil heaps, tracing the source of risks and outputting tiered control strategies to support precise on-site management. Through measured verification and dynamic correction, it balances identification accuracy, response efficiency, and engineering adaptability, ensuring reliable and implementable early warning results.

[0014] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0015] Figure 1 This invention provides a flowchart of a method for identifying and warning of ecological carrying capacity transitions in hydropower projects, according to an embodiment of this application. Figure 2 This paper presents a schematic diagram of the structure of a hydropower project ecological carrying capacity level transition identification and risk warning system according to an embodiment of this application. Detailed Implementation

[0016] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0017] The following is combined Figure 1 This application describes a method for identifying and warning of ecological carrying capacity level transitions in hydropower projects according to exemplary embodiments: S101, obtain multi-phase ecological carrying capacity assessment results, spatial data of engineering disturbance, construction stage information, ecological restoration area and auxiliary geographical environment data of the hydropower project impact area.

[0018] In one implementation, ecological carrying capacity evaluation results for different years and construction stages in the impact area of ​​a hydropower project are obtained. This includes annual / period-by-period ecological carrying capacity index raster data, preliminary carrying capacity level assessment results, and calculation results of basic evaluation indicators. This ensures that the data timeline covers the entire process from pre-construction, construction disturbance period, operation recovery period, to the start-up period of new construction, meeting the data continuity requirements for time-series comparison and transition analysis. Taking a high-altitude hydropower project as an example, five periods of ecological carrying capacity index raster data are obtained: 2015 (pre-construction baseline period), 2017 (construction disturbance period), 2019 (initial operation period), 2022 (operation recovery period), and 2024 (start-up period of new construction). Each period's data includes calculation results of basic indicators such as land cover, vegetation index, humidity index, surface temperature, aridity index, topographic indicators, and bare soil index, as well as preliminary carrying capacity level classification results. This ensures that multi-period data covers the entire life cycle of the project, providing a time-series basis for subsequent level transition tracking.

[0019] Spatial vector / raster data of all types of disturbance units in hydropower projects are acquired, covering construction roads, spoil heaps, tunnel entrances, mixing systems, construction platforms, construction camps, riverbank disturbance zones, quarrying areas, temporary stockpiles, and construction boundaries. The spatial location, boundary extent, disturbance type, and disturbance intensity of each disturbance unit are clearly defined to ensure complete disturbance information and accurate spatial positioning, supporting subsequent spatial correlation and risk tracing analysis. For the target hydropower project, linear vector data of construction roads, surface boundary data of each spoil heap, point location data of tunnel entrances, surface distribution data of mixing systems and construction platforms, and strip vector data of riverbank disturbance zones are collected one by one. Simultaneously, attribute information such as construction time, disturbance range, excavation / stockpiling scale, and disturbance intensity of each disturbance unit is recorded to ensure that all engineering disturbance units are included and that spatial coordinates accurately match the study area, providing basic spatial elements for subsequent buffer zone analysis and risk correlation.

[0020] This study obtains information on the entire lifecycle of a hydropower project, including the start and end times, construction content, disturbance intensity levels, and key events for the pre-construction, construction disturbance, operation recovery, and new project initiation phases. It clarifies the time boundaries and disturbance characteristics of each phase, supporting time-series matching, phase comparison, and transition rate calculation, ensuring the temporal logic of dynamic analysis. The construction timeline of the target hydropower project is analyzed as follows: 2013–2015 is the pre-construction baseline period, with no large-scale engineering disturbances; 2016–2018 is the construction disturbance period, involving core disturbance activities such as dam site excavation, road construction, spoil disposal, and tunnel excavation; 2019–2021 is the operation recovery period, implementing ecological restoration measures such as slope protection, spoil disposal site revegetation, and riverbank restoration; and 2022 to the present is the new project initiation period, involving preliminary construction of cascade projects. Key construction nodes and disturbance intensity levels for each phase are recorded simultaneously, providing a temporal basis for subsequent time-series matching, phase bearing capacity comparison, and transition rate calculation.

[0021] Acquire spatial data related to ecological restoration in the impact area of ​​hydropower projects, including spatial boundaries, restoration types, implementation times, restoration measures, and restoration scope of restoration units such as spoil heap treatment areas, slope protection areas, vegetation restoration areas, riverbank restoration areas, soil and water conservation areas, and ecological corridors. Clarify the spatial distribution and timing of restoration areas to support the identification of improved restoration areas and the evaluation of restoration effectiveness. Collect areal vector data of vegetation restoration areas, excavated slope protection areas, riverbank ecological restoration zones, and soil and water conservation areas in the construction area of ​​the target hydropower project spoil heap. Simultaneously record the restoration start time, restoration measures adopted (such as hydroseeding, shrub planting, and drainage works), restoration area, and vegetation restoration targets for each restoration area. Ensure that the restoration area data accurately covers the scope of engineering disturbance restoration, providing spatial support for subsequent identification of improved ecological restoration areas and effectiveness assessment.

[0022] Acquire basic geographic and environmental background data for the hydropower project's impact area, including administrative boundaries, watershed boundaries, digital elevation models (DEMs), slope and aspect data, meteorological data (temperature and precipitation), soil types, river system distribution, protected area boundaries, and current land use status. This data will provide a foundational map and environmental constraints for data preprocessing, spatial registration, background correction, and zoning analysis. Collect vector data of the watershed boundaries and county-level administrative boundaries where the target hydropower project is located, a 30-meter resolution DEM and derived slope and aspect raster data, monthly temperature and precipitation meteorological observation data for the past 10 years, soil type raster data, river system vector data, nature reserve / ecological red line boundary data, and annual land use status data. Ensure that auxiliary data covers the entire study area, providing fundamental support for subsequent spatial registration, temporal matching, and background interference correction.

[0023] S102, based on spatial registration, temporal matching and unified level coding, completes data preprocessing, constructs a six-category ecological carrying capacity level system through level classification rules, and generates multi-period ecological carrying capacity level maps.

[0024] In one implementation, based on the fundamental standards of unified coordinates, unified spatial resolution, and unified raster range, standardized preprocessing is carried out on multi-period ecological carrying capacity raster data. First, invalid images outside the study area, such as water bodies, cloud shadows, shadows, high-reflectivity bare rocks, and invalid boundaries, are removed. Second, masking removal or neighboring pixel interpolation is performed on the missing measurement areas. Then, pixel positions are aligned grid by grid. Finally, the spatial registration, temporal matching, and unified level coding of multi-period data are completed, ensuring that the data of each period are strictly consistent in spatial, temporal, and coding dimensions, and can be directly used for cross-period pixel-level level tracking and transfer matrix construction.

[0025] Ecological carrying capacity raster data from 2015, 2017, and 2019 for the impact area of ​​a high-altitude hydropower project were selected. The original resolutions were 25 meters, 30 meters, and 50 meters, respectively, and the original coordinate systems were WGS84, CGCS2000, and Gauss-Kruger 3-degree zones, respectively. During preprocessing, the data were uniformly converted to a 30-meter resolution and a Gauss-Kruger 3-degree zone (central meridian 99°). The raster area was clipped to extend 5 kilometers beyond the project boundary, and the number of rows and columns was uniformly set to 1500 rows × 1200 columns. Subsequently, water body pixels (NDWI > 0.3), shadow pixels (brightness value < 50), and invalid boundary pixels were removed, with a removal rate of approximately 7.5%. For missing measurement areas caused by cloud cover or sensor anomalies, 3×3 window neighbor effective value interpolation was used to fill the gaps, reducing the missing measurement rate to below 0.8%. Finally, pixel-by-pixel coordinate matching was used to achieve precise alignment of the three phases of raster data, ensuring that the same geographic location in the three phases of data corresponds to a unique pixel, and unifying the six types of carrying capacity level codes (1 to 6), forming a three-phase standard raster data with consistent time sequence, spatial matching, and unified coding, which meets the accuracy requirements for subsequent pixel cross-phase tracking and level transfer matrix construction.

[0026] A carrying capacity index threshold system is determined based on preset standards, natural discontinuities, quantiles, cloud models, and expert experience. A rule is adopted where smaller codes indicate higher risk, constructing six carrying capacity levels: multiple overloads, critical warning, sub-suitable buffer, balanced pressure, resilient surplus, and ideal sustainability. The ecological carrying capacity index grading threshold system is determined using various methods, including preset standards, natural discontinuities, quantiles, cloud models, and expert experience. Following the core rule that smaller codes indicate higher ecological risk, six carrying capacity levels are classified: multiple overloads, critical warning, sub-suitable buffer, balanced pressure, resilient surplus, and ideal sustainability. This achieves standardized conversion from continuous carrying capacity indices to discrete risk levels, adapting to different ecological risk management needs of hydropower projects.

[0027] The threshold determination methods are as follows: **Preset Standard Method:** Fixed thresholds are set based on industry ecological assessment standards and hydropower project environmental impact assessment standards, applicable to ecological grading in conventional areas. **Natural Breakpoint Method:** Based on the numerical distribution characteristics of the regional carrying capacity index, abrupt numerical change nodes are automatically identified as grading boundaries, conforming to the natural laws of regional data. **Quantile Method:** Thresholds are divided according to the 20%, 40%, 60%, 80%, and 90% quantiles of the carrying capacity index, ensuring a relatively balanced distribution of pixels at each level. **Cloud Model Fuzzy Grading Method:** A fuzzy membership cloud of ecological carrying capacity is constructed, transforming deterministic indices into fuzzy risks, resolving the ambiguity of grading boundaries, and adapting to hydropower project areas with high altitudes and complex terrain. **Expert Experience Method:** Thresholds are calibrated by ecological engineering experts, combining the regional ecological background and the disturbance characteristics of hydropower projects, conforming to the actual risk characteristics of the projects.

[0028] Based on the principle that the smaller the coded value, the higher the risk, the ecological carrying capacity index (range 0-1, with higher values ​​indicating stronger carrying capacity) of high-altitude hydropower project areas is classified into levels. The specific thresholds and their corresponding levels are as follows: Level 1 (Multiple Overloads): Carrying capacity index 0 ≤ E < 0.3, the ecosystem is completely collapsed and cannot withstand engineering disturbances, belonging to extremely high risk; Level 2 (Critical Warning): Carrying capacity index 0.3 ≤ E < 0.5, the ecosystem is fragile and easily degraded by slight disturbances, belonging to high risk; Level 3 (Subsuitable Buffer): Carrying capacity index... Level 1 (0.5 ≤ E < 0.65): Medium ecological carrying capacity, able to withstand minor disturbances, classified as medium to high risk; Level 2 (Balanced Pressure): 0.65 ≤ E < 0.8: Stable ecosystem, adaptable to conventional engineering disturbances, classified as medium risk; Level 3 (Resilient Surplus): 0.8 ≤ E < 0.9: Strong ecological carrying capacity, capable of recovering from disturbances, classified as low risk; Level 4 (Equilibrium Pressure): 0.65 ≤ E < 0.8: Stable ecosystem, adaptable to conventional engineering disturbances, classified as medium risk; Level 5 (Resilient Surplus): 0.8 ≤ E < 0.9: Relatively strong ecological carrying capacity, capable of recovering from disturbances, classified as low risk; Level 6 (Ideal Sustainability): 0.9 ≤ E ≤ 1.0: Healthy and stable ecosystem, capable of withstanding high-intensity disturbances, classified as low risk. A carrying capacity index sample of 10,000 raster pixels from a high-altitude hydropower project area was selected to construct a normal cloud model, with cloud expectation Ex = 0.6, entropy En = 0.15, and hyperentropy He = 0.02. The membership degree of each index was calculated, and the six-level thresholds were fuzzily matched to determine the final classification boundary, avoiding the absolute bias of a single threshold and conforming to the ecological heterogeneity characteristics of high-altitude areas.

[0029] After completing spatial registration, temporal matching, invalid image removal, missing measurement interpolation, and pixel alignment, the multi-period ecological carrying capacity raster data is integrated with preset grading rules. Grading is assigned to each pixel to generate multi-period ecological carrying capacity grading raster results that are temporally aligned, spatially consistent, have uniform resolution, and uniform coding standards. This ensures that the grading data of each period are fully comparable and can directly support subsequent analyses such as pixel-level cross-period tracking, grading transition matrix construction, and transition flux calculation.

[0030] Three pre-processed raster data of the ecological carrying capacity index for 2015, 2017, and 2019 were selected from the impact area of ​​a high-altitude hydropower project. All three data points were standardized to 30-meter resolution, Gauss-Kruger 3-degree coordinates, and a raster layout of 1500 rows × 1200 columns. Invalid images and missing data areas were removed and interpolated. The carrying capacity index for each pixel was assigned a six-level grading threshold: 0 ≤ E < 0.3 for multiple overloads, 0.3 ≤ E < 0.5 for critical warning, 0.5 ≤ E < 0.65 for sub-suitable buffer, 0.65 ≤ E < 0.8 for balanced pressure, 0.8 ≤ E < 0.9 for excess resilience, and 0.9 ≤ E ≤ 1.0 for ideal sustainability. Each pixel was assigned a grade code from 1 to 6. The output is a three-phase graded raster map. The spatial boundaries of the three phases of data completely overlap, the resolution is 30 meters, and the grade coding rules are completely consistent. It can be directly used for subsequent inter-phase pixel tracking, grade transition matrix construction, and transition index calculation.

[0031] S103 establishes inter-period tracking relationships using pixels as units, constructs a grade transition matrix, and calculates indicators for upgrades, downgrades, stability, cross-grade transitions, and high-risk inflow / outflow fluxes, incorporating time interval parameters to calculate grade transition rates.

[0032] In one implementation, a raster pixel is used as the unique spatial tracking unit. Following the principle of unique spatial location matching, a temporal correspondence relationship is established for the same geographical location in different evaluation periods. This achieves accurate spatial matching of raster data for ecological carrying capacity levels across multiple periods, ensuring that the data in each period are completely consistent in spatial location and can be directly compared across periods. In specific implementation, the raster division standard of the study area is first unified, dividing the hydropower project impact area into 30m × 30m regular raster pixels. The raster coordinate system uniformly adopts the Gauss-Kruger 3-degree zone (central meridian 99°). The raster range is clipped to 5 kilometers beyond the project boundary, and the number of rows and columns is uniformly 1500 rows × 1200 columns, ensuring that the size, position, and coordinates of the raster are completely consistent throughout the entire period.

[0033] We selected three pre-processed raster data of ecological carrying capacity levels in the hydropower project impact area in 2017, 2019, and 2022. Using geographic coordinates (X, Y) as the sole matching basis, we established a one-to-one mapping relationship between pixels at the same location in the three raster data. That is, the pixel in row i and column j in 2017 corresponds to the pixel in the same row and column with the same geographic coordinate in 2019 and 2022, thus achieving accurate spatial alignment of the three data.

[0034] During the matching process, invalid boundary images, water body images, and cloud images are removed to ensure that all pixels participating in the matching are valid land units. For missing locations, 3×3 window neighbor valid value interpolation is used to complete the data, ensuring that the number and location of pixels in the three periods are completely consistent, without misalignment, omission, or redundancy. This provides a unique and reliable spatial benchmark for subsequent pixel-level cross-period grade tracking, transfer matrix construction, and transition index calculation.

[0035] Based on multi-period ecological carrying capacity level raster data with precise temporal matching, consistent spatial range, and unified coding standards, three types of matrices are constructed: a matrix of the number of levels transitioned, an area matrix, and a probability matrix. The matrix rows and columns are used to systematically count the number of pixels that transitioned into, out of, or remained between each carrying capacity level, the corresponding actual area, and the probability of transition. This comprehensively quantifies the inter-period change pattern of ecological carrying capacity status and provides basic data support for the analysis of level transition characteristics.

[0036] Three raster data points for the ecological carrying capacity levels of the hydropower project's impact area in 2017, 2019, and 2024 were selected. All three data points were 30-meter resolution, Gauss-Kruger 3-degree coordinates, with a uniform row and column count of 1500 rows × 1200 columns, and an actual pixel area of ​​900 square meters (30m × 30m). Three matrices were constructed for the two time periods of 2017–2019 and 2019–2024. Rows in the matrices represent the earlier carrying capacity levels, and columns represent the later carrying capacity levels. The level codes are 1 (multiple overloads), 2 (critical warning), 3 (sub-suitable buffer), 4 (balanced pressure), 5 (resilient surplus), and 6 (ideal sustainability), with each matrix having a 6×6 dimension.

[0037] The grade transfer quantity matrix uses the number of pixels as the statistical unit, for a time period. to Construct a matrix of level transition quantities: ,in, Indicates the time period Belongs to the level During the period Transfer to level The number of pixels, i.e. ,in, This indicates the number of pixels that meet the conditions. Let i be the ecological level of the i-th pixel in time period t. Based on the hydropower project impact area in 2017 (… From 2019 to 2019 For example, let's assume... =1250, indicating a time period. Level 2 (critical alert), time period The number of pixels transferred to Level 1 (Multiple Overload) is 1250.

[0038] If the area of ​​each cell is The corresponding level transition area matrix is ,in, Based on the above =1250, therefore, =1250×900=1,125,000m2, that is, the total area transferred from level 2 to level 1 is 112.5 hectares.

[0039] Furthermore, construct the level transition probability matrix: ,in, , Indicates the original level is The pixels are transferred to the next time period as levels The probability of. Assuming a time period. Total number of pixels in Tier 2 =25800, then = ≈4.84% means that the probability of transitioning from level 2 to level 1 is 4.84%. In the above matrix: when When, it indicates that the level is stable; when When, it indicates an upgrade in the ecological carrying capacity; when When this occurs, it indicates a downgrade in the ecological carrying capacity; when... When, it indicates a shift to a high-risk level; when When r=4 and s=1, it indicates that a step-wise degradation has occurred. s=3≥2 indicates that a cross-level degradation has occurred, directly degrading from the balanced bearing capacity level to the multiple overload level.

[0040] To further quantify changes in ecological carrying capacity, this invention calculates six core level-transfer flux indicators based on a level-transfer quantity matrix, including stable flux, upgrading flux, downgrading flux, cross-level downgrading flux, high-risk level inflow flux, and high-risk level outflow flux, comprehensively depicting the scale characteristics of dynamic changes in ecological carrying capacity levels from the perspective of pixel proportion.

[0041] Stable flux represents the proportion of pixels whose carrying capacity remains constant. ;in, For time period arrive In the diagram, the proportion of pixels whose carrying capacity level remains unchanged; M represents the total number of ecological carrying capacity levels. For time period For level r, time period The number of cells that remain at level r (the number of cells whose level has not changed). For time period For level r, time period The number of pixels transferred to level s; taking the transfer matrix from 2017 to 2019 as an example, the number of stable pixels from level 1 to level 6 are 1800, 4200, 7500, 8920, 6800, and 5100 respectively, with a total of 45000 pixels. = ≈76.27%.

[0042] Upgrade flux represents the proportion of pixels whose carrying capacity is shifted to a better level. ; The proportion of pixels that were transferred from the current state to a better level; from 2017 to 2019, a total of 4,800 pixels were upgraded between levels. ≈10.67%.

[0043] Degradation flux represents the proportion of pixels whose load conditions are shifted to a worse level. From 2017 to 2019, a total of 5,880 pixels were downgraded between different classes. ≈13.07%.

[0044] Cross-level degradation flux is used to represent severe degradation processes that decline by two or more levels. ; The proportion of severely degraded pixels that decreased by two or more levels; from 2017 to 2019, a total of 1,250 pixels were downgraded across levels. ≈2.78%.

[0045] Multiple overloads (Level 1) and critical alert (Level 2) are defined as the set of high-risk levels. The inflow flux at the high-risk level is This indicator represents the proportion of pixels that were not originally classified as high-risk, but subsequently entered the high-risk category in the next period, reflecting the scale of newly added high-risk areas. From 2017 to 2019, a total of 2,100 pixels that were not classified as high-risk (3-6) were transferred to the high-risk category (1-2). ≈4.67%.

[0046] High-risk level outflow flux is This indicator is used to characterize the proportion of high-risk areas that have exited the high-risk level after ecological restoration or natural recovery, reflecting the scale of restoration and improvement in high-risk areas. From 2017 to 2019, a total of 1,350 pixels were downgraded from high-risk level (1-2) to non-high-risk level (3-6). =3.00%.

[0047] To identify the window period of rapid deterioration of ecological carrying capacity, this invention introduces the time interval parameter between adjacent evaluation periods. Combined with the calculated downgrade flux, cross-level downgrade flux, and high-risk level inflow flux, the transfer rate of the three levels is calculated respectively, quantifying the speed of ecological carrying capacity deterioration and providing key time dimension indicators for dynamic early warning.

[0048] Let two adjacent evaluation periods be the early stage. and later The time interval between the two is The unit is years, representing the actual time between two assessments, and serves as the benchmark for rate normalization. The degradation rate characterizes the proportion of pixels whose ecological carrying capacity has degraded per unit time, reflecting the overall degradation intensity. ,in, This represents the degradation flux for the corresponding time period. The cross-class degradation rate characterizes the percentage of pixels experiencing cross-class degradation per unit time, reflecting the risk of severe degradation. ,in, This represents the cross-level downgrade flux for the corresponding time period. The high-risk level inflow rate characterizes the proportion of newly added pixels entering the high-risk level per unit time, reflecting the speed of new risk. ,in, This refers to the high-risk inflow flux for the corresponding time period.

[0049] Two typical time periods for hydropower projects were selected: 2017–2019 (peak construction disturbance period, Δt = 2 years) and 2019–2022 (operation recovery period, Δt = 3 years). Flux values ​​were substituted into these periods to calculate the rate. For the 2017–2019 period (Δt = 2 years), if the flux is downgraded... =13.07%, then the degradation rate =213.07%≈6.54% / year; if the flux is downgraded across levels =2.78%, then the rate of downgrade is... =22.78%≈1.39% / year; if high-risk inflow flux =4.67%, then the high-risk inflow rate =24.67%≈2.34% / year. If the preset degradation rate warning threshold is 5% / year, then the degradation rate during this period exceeds the threshold, indicating that the ecological carrying capacity is rapidly deteriorating in a short period of time, and an early warning needs to be triggered.

[0050] From 2019 to 2022 (Δt = 3 years), if the flux is downgraded... =4.83%, then the degradation rate =34.83%≈1.61% / year; if the flux is downgraded across levels =0.92%, then the rate of downgrade is... =30.92%≈0.31% / year; if high-risk inflow flux =1.55%, then the high-risk inflow rate =31.55%≈0.52% / year. During this period, all three rates were below the warning threshold, indicating that ecological restoration measures effectively curbed the degradation trend.

[0051] The aim is to identify critical windows of rapid degradation in ecological carrying capacity during construction preparation, peak construction periods, or the initiation phase of new projects. By calculating rates and comparing them with thresholds, the system can accurately pinpoint the periods of fastest ecological deterioration, providing a basis for prioritizing engineering management and ecological protection measures.

[0052] S104 constructs pixel-level and regional-level transition risk indices, and uses a linear weighted model to integrate the core features of downgrade magnitude, high-risk transition, and cross-level downgrade to generate a feature set of ecological carrying capacity level transition risks.

[0053] In one implementation, based on three core features—degradation magnitude, high-risk transition markers, and cross-level degradation markers—and following the construction principles of risk correlation, weight adaptability, and hierarchical differentiation, the weight coefficients and index calculation rules are jointly determined by a linear weighted model and a flux fusion model. This invention further constructs an ecological carrying capacity level transition risk index to comprehensively characterize the intensity of carrying capacity level degradation, the inflow of high-risk levels, and the degree of cross-level degradation, achieving a quantitative assessment of single-pixel ecological risk. The pixel-level transition risk index is... ,in: For pixels exist to The transition risk index during the period; This refers to the extent of the downgrade; This is a high-risk transfer indicator; This is a downgrade indicator for cross-level classification; For the weighting coefficients, satisfying + + =1.

[0054] downgrade range Indicates the pixel in the time period to The formula for calculating the decrease in bearing capacity level during the period is as follows: ,in, For pixel i in The carrying capacity level at any time, for The level of time. When A value ≥1 indicates a downgrade has occurred; A score of ≥2 indicates a downgrade across risk levels. (High-risk transfer indicator) If we define multiple overloads (Level 1) and critical alert (Level 2) as the high-risk level set H={1,2}, then the high-risk transition identifier is: When a pixel transitions from a non-high-risk level to a high-risk level, this flag is set to 1; otherwise, it is set to 0. (This refers to a flag indicating a cross-level downgrade.) Indicates whether a pixel has experienced a degradation of two or more levels across different levels. When the downgrade magnitude is greater than or equal to 2, this flag is set to 1; otherwise, it is set to 0.

[0055] Taking the surrounding pixels of a hydropower project's spoil disposal site from 2017 to 2019 as an example: Pixel i in =The 2017 grade was 4 (balanced pressure). =2019 rating was 1 (multiple overloading); downgraded by =4 1=3; High-risk transfer indicator =1 (transfer from non-high-risk to high-risk); downgrade indicator =1 ( ≥2); Set weighting coefficients =0.2、 =0.4、 =0.4; Substitute into the formula to calculate the pixel-level transition risk index. =0.2×3+0.4×1+0.4×1=0.6+0.4+0.4=1.4. This index comprehensively quantifies the risk of ecological carrying capacity leap in a single pixel by linearly weighting three core characteristics: the magnitude of downgrade, high-risk transition, and cross-level downgrade, providing a standardized quantitative basis for subsequent risk classification, zoning identification, and early warning.

[0056] The method employs a tiered approach that integrates single-pixel feature aggregation, regional feature averaging, and risk level stratification, assigning high-risk levels 1 and 2 for priority configuration. , Weight greater than To emphasize the importance of high-risk inflows and cross-level downgrades, the requirements for quantifying risk are aligned with both pixel-level and regional-level risk assessments. To highlight the significance of high-risk inflows and cross-level downgrades, the weights of the pixel-level risk index are differentiated. ,in, To evaluate the area or the affected area of ​​a specific engineering disturbance unit, The number of pixels is defined as follows: the weight α2 for high-risk transfer and the weight α3 for cross-level degradation are set to be greater than the degradation magnitude weight α1, prioritizing the contribution of severe degradation risk. The weight configuration is as follows: α1=0.2, α2=0.4, α3=0.4. At this point, α2, α3>α1, and the sum of the weights is 1.

[0057] The overall risk level of the region is obtained by averaging the risk indices of all pixels within the region, using the following formula: The 800m buffer zone of the spoil heap contains 1000 valid pixels. The sum of the risk indices of all pixels is 692. Therefore, the regional risk index is: =0.692.

[0058] The risk index is constructed directly using regional-level transfer flux, and the formula is as follows: ,in: To reduce flux, For cross-level downgrade flux, For high-risk inflow flux, High-risk outflow flux; , , , The weighting coefficients are such that the sum is 1; the high-risk outflow flux is preceded by a negative sign, reflecting its mitigating effect on regional risks.

[0059] Flux data for the spoil heap buffer zone: Degradation flux =12%, Cross-level degradation flux =3%, high-risk inflow flux =5%, high-risk outflow flux =2%; weight =0.3, =0.3, =0.3, =0.1, then =0.3×0.12+0.3×0.03+0.3×0.05 0.1 × 0.02 = 0.068. Its core function is to integrate multiple risk level transition characteristics into a unified risk indicator, making early warning judgments more intuitive and quantifiable. By combining pixel-level and regional-level indices, it can both pinpoint severe degradation risks at single points and assess the overall risk level of engineering areas, providing a standardized basis for subsequent risk classification, zoning identification, and the formulation of control measures.

[0060] S105, based on spatial overlay and buffer analysis, associates the transition results with engineering disturbance units of construction roads, spoil heaps, and tunnel entrances, and identifies degradation zones, repair and improvement zones, stability maintenance zones, and re-disturbance risk zones.

[0061] In one implementation, based on the spatial attributes of engineering disturbance units, the raster characteristics of transition results, spatial overlay matching rules, and buffer analysis logic, and following the construction criteria of spatial location correlation, influence distance adaptability, risk contribution separability, and zone identification accuracy, the buffer range, spatial matching parameters, and zone determination conditions are determined collaboratively by distance thresholds and spatial overlay algorithms. Taking engineering disturbance units such as hydropower construction roads, spoil heaps, tunnel entrances, mixing systems, and construction platforms as objects, and combining spatial attributes such as unit vector boundaries, distribution range, size, and disturbance intensity, the raster characteristics of ecological carrying capacity level transition results are matched. Following the four construction criteria of spatial location correlation, influence distance adaptability, risk contribution separability, and zone identification accuracy, the buffer range, spatial matching parameters, and zone determination conditions are determined collaboratively by distance thresholds and spatial overlay algorithms, achieving precise spatial correlation between engineering disturbance units and ecological risks.

[0062] Regarding spatial correlation, the closer the risk and disturbance units are spatially, the higher the correlation probability, and neighboring grids are prioritized for matching. Regarding the adaptability of impact distance, differentiated impact distances are set according to the intensity and scale of the disturbance, with strong disturbance units having a larger impact range. Regarding the separability of risk contribution, the risk contribution of single disturbances and superimposed disturbances is distinguished to avoid confusion of risks from multiple disturbances. Regarding the accuracy of zoning identification, matching parameters and judgment conditions are quantifiable to ensure clear and non-overlapping boundaries between degradation and improvement zones.

[0063] Distance thresholds are set differently based on the type and scale of disturbance: large spoil heaps (storage area ≥ 10 hectares): 800 meters; medium-sized spoil heaps (3-10 hectares): 500 meters; construction roads (main lines): 500 meters; construction roads (branch lines): 300 meters; tunnel entrances: 300 meters; mixing systems: 400 meters. The spatial overlay algorithm rules are as follows: for matching accuracy parameters, a match is determined when the grid center point falls into the buffer zone, with a matching accuracy of 100%. The zoning criteria are as follows: high-risk grids (risk index ≥ 0.5) accounting for ≥ 30%: construction disturbance degradation zone; risk index decrease ≥ 0.2: ecological restoration and improvement zone; risk index fluctuation < 0.1: stable maintenance zone; initial improvement followed by later degradation: re-disturbance risk zone.

[0064] A medium-sized spoil heap (5.2 hectares) from a high-altitude hydropower project was selected. The vector boundary was clear, and the disturbance intensity was high. A 500-meter buffer zone was set for the medium-sized spoil heap. Spatial overlay matching: the spoil heap vector was overlaid with risk grids, and all grids within 500 meters (892 in total) were extracted. Zoning was as follows: 325 high-risk grids (RI≥0.5), accounting for 36.4% (≥30%), were identified as construction disturbance degradation zones. The average risk of the degradation zone was 0.68, highly correlated with spoil heap disturbance, contributing 62% to the risk.

[0065] The method adopts an integrated approach that combines unit impact zone construction, risk indicator correlation, and zoning type determination. It sets distance thresholds based on the intensity of disturbance, impact range, and ecological sensitivity for typical engineering disturbance units such as construction roads, spoil heaps, tunnel entrances, mixing systems, and construction platforms. Through buffer zone analysis, the impact range of each unit is accurately constructed, achieving precise matching between the transition risk index and the spatial location of the engineering unit, providing a spatial correlation basis for subsequent risk tracing and zoning identification.

[0066] Based on the intensity, scale, and ecological impact of the engineering disturbance, the distance thresholds for typical disturbance units are defined as follows: Construction roads (main lines): wide disturbance range and continuous impact zone, a distance threshold of 500 meters is set; Spoil disposal sites (medium-sized, 3-10 hectares): strong disturbance from stockpiling and large impact area, a distance threshold of 800 meters is set; Tunnel entrances: concentrated excavation disturbance and small impact range, a distance threshold of 300 meters is set; Mixing systems: concentrated point disturbance, a distance threshold of 400 meters is set; Construction platforms: localized area disturbance, a distance threshold of 350 meters is set.

[0067] The vector layer of each unit buffer is spatially overlaid with the 30-meter resolution transition risk index grid. The determination rule is: if the center point of the grid falls within the buffer area, it is determined to be the associated grid of that unit, ensuring that the matching accuracy is without deviation or redundancy.

[0068] Taking a medium-sized spoil heap (5.2 hectares) of a high-altitude hydropower project as an example: A distance threshold of 800 meters is set to generate a spoil heap buffer vector; within the buffer zone, 892 effective 30m × 30m grids are matched; the transition risk index of each grid (values ​​0-2) is extracted to obtain the buffer zone risk index set { , ,..., Average risk index of the area associated with the spoil disposal site: Construction roads (main roads): 1240 grids are matched in a 500-meter buffer zone, with an average risk index of 0.58; tunnel entrances: 416 grids are matched in a 300-meter buffer zone, with an average risk index of 0.75.

[0069] The buffer zone and matching results are validated for reasonableness, invalid matching grids at the boundaries are removed, and the continuity of the risk index distribution is verified to ensure that the impact range of the engineering unit is consistent with the actual range of ecological disturbance. It is adapted to the needs of pixel-level and regional-level risk analysis, and provides accurate spatial correlation data for subsequent identification of degradation areas and improvement areas and risk tracing.

[0070] After completing the spatial matching of engineering disturbance units and transition risk indices, the association results, zoning rules, and distance thresholds were simultaneously verified. Verification was conducted item by item from three aspects: spatial matching degree, risk contribution, and zoning boundaries. Invalid data and weakly correlated information were eliminated. Finally, the unit association risk index, high-risk area, and proportion were calculated to ensure the accuracy and reliability of the association results. The matching position between the buffer zone and the risk grid was verified to ensure its rationality, with a focus on identifying boundary overflows, cross-terrain mismatches, and invalid overlapping areas. Invalid matching grids within 10 meters of the buffer zone edge, invalid grids in non-land areas such as water bodies / bare rock, and misaligned matching grids across valleys were eliminated. For a medium-sized spoil heap, 892 grids were matched in a 500-meter buffer zone. After verification, 36 invalid grids at the edge and 18 invalid grids in water bodies were eliminated, leaving 838 valid matching grids, a matching efficiency of 94%.

[0071] The correlation between the transition risk index and engineering disturbance units was verified, and weakly correlated and randomly correlated data were filtered to ensure a strong correlation between risk and disturbance. A correlation threshold was set: a mean risk index within a unit ≥ 0.4 was considered a valid correlation, while < 0.4 was considered a weak correlation and was filtered out. A 500-meter buffer zone of a construction road matched 1240 grids, with a mean risk index of 0.58 ≥ 0.4, indicating a valid correlation. A buffer zone at a remote tunnel entrance, with a mean risk index of 0.29 < 0.4, was determined to have a weak correlation, and the matching results for this unit were filtered out.

[0072] Check that the boundaries of the zones are clear, non-overlapping, and free of fragmentation to avoid mixing across different types. An overlap width of ≤5 meters between adjacent zones is acceptable; overlaps greater than 5 meters require correction. Within a single zone, the proportion of fragmented grid cells ≤10% is acceptable. The overlap width between the degraded zone and the surrounding stable zone in the spoil heap is 3 meters, which meets the requirements; there are 78 locally fragmented grid cells, accounting for 9.3% of the effective grid cells, which do not require correction.

[0073] Integrating and verifying spatial correlation results, and simultaneously matching four types of information—risk index, distance threshold, unit type, and zoning type—the system accurately identifies four functional zones based on risk level, trend, and spatial location: construction disturbance degradation zone, ecological restoration and improvement zone, stability maintenance zone, and re-disturbance risk zone. The zoning boundaries are clear, and the judgment criteria are quantified. For construction disturbance degradation zone identification, the average unit correlation risk is ≥0.5, and the distance from the disturbance unit is ≤ the distance threshold, with the risk index showing a downward trend. The average risk of a medium-sized spoil heap is 0.68 ≥0.5, located within the 500-meter threshold range. The risk index decreased by 0.32 from 2017 to 2019, classifying it as a construction disturbance degradation zone, covering an area of ​​27.45 hectares. For ecological restoration and improvement zone identification, the risk index increased by ≥0.2 compared to the previous period, and it is located within the ecological restoration zone (spoil heap revegetation area and slope protection area). The downstream 300-meter restoration zone of the spoil heap had a risk index of 0.52 in 2017 and 0.28 in 2019, an increase of 0.24 (≥0.2), and was identified as an ecological restoration and improvement zone, covering an area of ​​12.8 hectares. It was identified as a stable maintenance zone, with a risk index fluctuation of <0.1, no obvious escalation / degradation trend, and far from the core disturbance unit. The area 500-800 meters outside the spoil heap had a risk index of 0.35-0.42 from 2017 to 2019, a fluctuation of 0.07 (<0.1), and was identified as a stable maintenance zone, covering an area of ​​45.2 hectares. It was identified as a re-disturbance risk zone, with the risk index increasing by ≥0.1 in the early stage (2015-2017) (restored and improved), and decreasing by ≥0.1 in the later stage (2017-2019) (re-degraded). The area surrounding the branch road under construction was identified as a re-disturbance risk zone, with a risk level of 0.48 in 2015, 0.35 in 2017 (an increase of 0.13), and 0.49 in 2019 (a decrease of 0.14). The area covers 8.5 hectares.

[0074] In another implementation, based on pixel-level rank changes and transition risk indices, the study area is divided into different types of ecological carrying capacity change zones, transforming the rank transition results into spatial classification results with clear management significance. A pixel can be identified as a construction-induced degradation zone when it meets one of the following conditions: or ,in, For the preset risk index threshold (such as =0.5). If the pixel is also close to engineering disturbance units such as construction roads, spoil heaps, tunnel entrances, and mixing systems, it is further marked as an engineering-related degradation zone. A pixel near a spoil heap, with a grade of 4 in 2017 and a grade of 1 in 2019, meets the following criteria. =1 4= 3 < 0, and the risk index =1.4≥0.5, and is located within the 800-meter influence range of the spoil disposal site, therefore it was determined to be a project-related degradation zone.

[0075] When a cell satisfies: If a pixel has a rating >0 and is located within an ecological restoration area, revegetation area, slope protection area, or spoil disposal site remediation area, it is considered an ecological restoration and improvement area. A pixel within the revegetation belt downstream of a spoil disposal site, with a rating of 2 in 2017 and 4 in 2019, meets the following criteria: =4 Since 2=2>0 and is located within the vector range of the revegetation area, it is therefore identified as an ecological restoration and improvement area.

[0076] When a cell satisfies: =0 or the change range does not exceed the preset threshold (e.g., |Δ) When |<0.5 (meaning the grading level has not substantially changed), it is considered a stable region. A pixel in the background region far from the disturbing cell, whose grading level was 5 in 2017 and remained 5 in 2019, meets the following condition: =0, therefore it is determined to be in the stable region.

[0077] For a given pixel, if it experiences an upgrade in the previous stage but then reverts to a downgrade in the next stage, it is identified as a re-perturbation risk zone. For example: >0 and: A value <0 indicates that the pixel has undergone a "recovery followed by degradation" process and should be a key monitoring target during the initial stage of new construction projects or secondary construction disturbances. A pixel near a construction road saw its grade rise from 3 to 4 (ΔL>0, indicating improvement) between 2015 and 2017, but its grade dropped back to 2 (ΔL<0, indicating further degradation) between 2017 and 2019, thus being identified as a re-disturbance risk zone. Transforming the abstract grade transition results into spatial classification results with clear management significance provides a direct basis for developing differentiated control measures for different types of areas.

[0078] S106 adopts a multi-threshold hierarchical judgment mechanism to divide the four-level early warning level, introduces a risk source tracing and control suggestion generation module, and outputs early warning results and control strategies in combination with the risk contribution of engineering units.

[0079] In one implementation, a multi-threshold grading mechanism is adopted, comprehensively considering core indicators such as the transition risk index, degradation rate, and high-risk inflow flux. Multiple grading thresholds are set to classify risks into four levels: low risk, risk of concern, warning risk, and severe risk, achieving precise risk level classification. Based on risk data from the initial construction phase of a hydropower project, three core indicators—regional transition risk index, cross-level degradation rate, and high-risk inflow flux—are selected, and three grading thresholds are set: a risk index below 0.3 and a degradation rate less than 0.1 years. - ¹ represents low risk; risk index 0.3-0.5, downgrade rate 0.1-0.2 years. - ¹To focus on risk; risk index 0.5-0.7, downgrade rate 0.2-0.3 years. - ¹This indicates a risk warning; the risk index is above 0.7 and the downgrade rate is greater than 0.3 years. - ¹For severe risks, a precise four-level early warning system is established.

[0080] A risk tracing module is introduced to link information such as risk level transfer paths, high-risk sources, and types of engineering disturbance units, tracing the core causes, spatial locations, and key influencing factors of risk generation, and clarifying the risk causal chain. For a severely warned area surrounding a spoil heap, the tracing module shows that the risk originates from a leap from the equilibrium pressure level to the critical warning level and multiple overload levels. The core cause is the disturbance of spoil heaps, compounded by rainfall runoff. The affected area is concentrated in the 500-meter buffer zone downstream of the spoil heap, accurately pinpointing the risk source and impact path.

[0081] A control recommendation generation module is introduced, which combines early warning level, risk causes, and risk contribution of engineering units, and matches hierarchical control logic to generate targeted and implementable control measures, forming a standardized control strategy system. For the warning risk zone around construction roads, considering that road disturbance accounts for over 60% of the risk, control recommendations are generated: strengthen temporary protection of road slopes, add drainage ditches, regularly clean loose debris from the road surface, and strictly control the disturbance range during construction. For the severely risky waste disposal site, recommendations include optimizing storage zoning, adding seepage prevention and drainage works, and accelerating topsoil backfilling and vegetation restoration.

[0082] By integrating risk contribution data from various engineering units, linking early warning levels, tracing causes, and control recommendations, the system outputs tiered early warning results, risk tracing reports, and differentiated control strategies, achieving precise integration of risk early warning and engineering control. It summarizes risk contribution data from units such as construction roads, spoil heaps, and tunnel entrances, and combines four levels of early warning with tracing results to output early warning reports: A tunnel entrance is classified as a warning risk, caused by vegetation degradation due to excavation disturbance; the control strategy is to slow down excavation and implement slope protection simultaneously. A spoil heap is classified as a serious risk, caused by exposed stockpile and rainwater erosion; the control strategy is to immediately stop new stockpiling and accelerate revegetation, forming a complete early warning and control outcome.

[0083] S107 uses data from all stages of the process and actual monitoring data to verify accuracy and make dynamic corrections. It balances the accuracy of transition identification, the efficiency of early warning response, and the requirements of engineering adaptability through a multi-index comprehensive evaluation algorithm, and generates ecological carrying capacity level transition identification and risk early warning results that are suitable for the entire process of hydropower projects.

[0084] In one implementation, based on the temporal characteristics of the entire life cycle of a hydropower project—before construction, during construction, during operation recovery, and during the new construction start-up phase—and combined with monitoring data verification standards, transition identification accuracy requirements, and engineering adaptation constraints, the system follows four construction principles: temporal consistency, data authenticity, controllable accuracy, and complete adaptation. It integrates data from all stages of the process, measured monitoring data, and simulation results through a multi-source data fusion mechanism, and collaboratively determines verification dimensions, dynamic correction rules, and balance parameters using a multi-index comprehensive evaluation algorithm. This achieves a balance between the accuracy of ecological carrying capacity transition identification and engineering adaptation throughout the entire process.

[0085] The time-series characteristics of the entire life cycle of hydropower projects are clearly defined. Specifically, before construction (baseline period): ecological stability, monitoring mainly focuses on baseline background data; during construction (disturbance period): strong disturbance, monitoring focuses on real-time disturbance data; during operation and recovery period: ecological restoration, monitoring focuses on restoration effectiveness data; during new construction and start-up period: secondary disturbance, monitoring covers preliminary preparation data. The four construction principles are specifically adapted as follows: full-process time-series alignment, ensuring the continuity and lack of gaps in the time-series data of multiple periods (2015 / 2017 / 2019 / 2022 / 2024); priority is given to measured monitoring data, eliminating invalid data from simulation distortion and abnormal disturbances; the transition identification error is controlled within ≤5%, meeting the accuracy requirements for project risk identification; and the entire project constraint is covered, including high altitude, strong disturbance, and cascade development.

[0086] Taking a high-altitude hydropower project as an example, this study integrates three core data types: full-process time-series data, measured monitoring data, and simulation results. The full-process time-series data includes ecological carrying capacity level grids for 2015 (before construction), 2017 (construction period), 2019 (restoration period), and 2022 (new construction period). The measured monitoring data includes on-site monitoring values ​​of vegetation cover, soil erosion, water quality, and surface temperature. The simulation results include carrying capacity index and simulated values ​​of the level transition model.

[0087] By fusing multi-source data, four verification dimensions were determined: Temporal matching dimension: verifying the temporal alignment accuracy of multi-period data (error ≤ 1 pixel is acceptable); Data error dimension: verifying the deviation between measured and simulated data (deviation ≤ 5% is acceptable); Accuracy threshold dimension: verifying the accuracy of transition recognition (≥ 95% is acceptable); Adaptability dimension: verifying the matching degree between the verification results and engineering disturbances, terrain, and climate (≥ 90% is acceptable). Verification dimensions, correction rules, and balancing parameters were performed simultaneously. For temporal deviations exceeding the threshold, the temporal granularity was refined and realigned; for large data errors, the simulation model was optimized and the weight of measured data was strengthened; for insufficient accuracy, feature weights were increased and algorithm parameters were optimized; for insufficient adaptation, terrain / climate constraints were supplemented and thresholds were adjusted. The final output is a temporally consistent, data-reliable, accuracy-balanced, and fully adapted ecological carrying capacity transition recognition result, providing accurate support for subsequent early warning.

[0088] An integrated approach is adopted, encompassing time-series data alignment, monitoring data benchmarking, accuracy index calculation, and adaptability assessment. Differentiated time-series verification standards are matched for different stages: pre-construction, construction, operation recovery, and new construction startup. Dynamic correction coefficients are set based on the deviation between measured monitoring data and simulation results. Balance weights are configured according to the triple requirements of transition identification accuracy, early warning response efficiency, and engineering adaptability. Targeting the characteristics of different stages of hydropower projects, the pre-construction phase emphasizes benchmark time-series alignment and historical monitoring benchmarking; the construction phase strengthens disturbance time-series matching and real-time monitoring benchmarking; the operation recovery phase focuses on the continuity of repair time-series and effectiveness monitoring benchmarking; and the new construction startup phase matches the connection of early-stage time-series and preparatory monitoring benchmarking. Dynamic correction coefficients are set at gradients of 0.1, 0.3, and 0.5 for measured-to-simulated deviations of 0-5%, 5-10%, and above 10%, respectively. Balance weights of 0.4, 0.3, and 0.3 are configured based on the principles of prioritizing identification accuracy, considering efficiency, and providing a safety net for adaptability.

[0089] Simultaneous verification was conducted on the verification dimensions, correction rules, and balancing parameters. The timing alignment strategy was adjusted to address timing deviations, the dynamic correction coefficients were optimized to address data errors, and the balancing weights were reconfigured to address indicator conflicts, resulting in multi-dimensional verification and dynamic correction results. Verification revealed that timing deviations during the construction period exceeded the threshold, leading to adjustments in the timing alignment strategy and refinement of the timing matching granularity for construction nodes. Large discrepancies between measured and simulated data were found in some areas, prompting optimization of the correction coefficient gradient and the addition of extreme deviation correction levels. Conflicts between accuracy and efficiency indicators were identified, leading to a reconfiguration of the balancing weights, increasing the accuracy weight to 0.5 and decreasing the efficiency weight to 0.2. The multi-dimensional verification results, after timing alignment, data correction, and weight optimization, were output.

[0090] By integrating data from all stages of the hydropower project, multi-dimensional verification results, dynamically adjusted parameters, and balancing weights, the project outputs a final result that is consistent in time series, reliable in data, balanced in accuracy, and adaptable to the entire hydropower project process, identifying the ecological carrying capacity level transition and providing risk warnings. It summarizes the time series data from the four stages of the hydropower project, the time series / data / accuracy / adaptability verification results, gradient correction coefficients, and optimized balancing weights, integrating them to generate an ecological carrying capacity level transition map, risk index map, warning level map, and control strategies that cover the entire life cycle, are time-series consistent, have authentic data, meet accuracy standards, and are closely aligned with the actual project situation, forming a complete final result.

[0091] In another implementation, to transform the risk results of grade transition into engineering control objects, this invention spatially correlates the transition risk results with engineering disturbance units. Let the set of engineering disturbance units be... ,in, This can represent construction roads, spoil heaps, tunnel entrances, mixing systems, construction platforms, riverbank disturbance zones, etc. For each engineering disturbance unit... Construct its area of ​​influence: ,in, Represents a cell To engineering disturbance unit distance, The disturbance impact distance threshold is used. The engineering disturbance element is calculated. The associated risk index: Simultaneously calculate the high-risk transition area of ​​the disturbance element in this project. And the proportion of high-risk jumps: .like , or If the corresponding threshold is exceeded, the project disturbance unit is determined to be an ecological risk contribution unit. The ecological carrying capacity level transition results are linked to specific construction units, enabling early warning results to directly serve on-site project management.

[0092] The warning levels are classified based on the transition risk index, degradation rate, high-risk inflow flux, and engineering disturbance unit correlation results. There are four warning levels: Level I, corresponding to low risk, is judged by the level being basically stable, with low degradation flux and risk index; Level II, corresponding to risk of concern, is judged by localized minor degradation requiring continuous monitoring; Level III, corresponding to risk of alert, is judged by significant high-risk inflow or an increase in cross-level degradation; and Level IV, corresponding to severe risk, is judged by large-scale degradation, rapid transition, or significant engineering unit correlation risk.

[0093] In one possible implementation, it can be based on a regional risk index. Make a judgment: ;in, The warning threshold is determined using the following methods: statistical analysis of historical risk events; quantile method; expert experience; engineering environmental management standards; or adaptive determination based on multi-year ecological carrying capacity transition results. The results of the grade transition analysis are then transformed into warning levels that can be used for management decision-making.

[0094] After triggering an early warning, this invention further generates risk sources and control recommendations. For a given early warning area, the system identifies the following information: the main source of the downgrade, such as a shift from excess resilience to sub-suitable buffer, or from balanced pressure to critical alert; the main type of high-risk level transition, such as critical alert or multiple overloads; the type of adjacent engineering disturbance unit, such as construction roads, spoil heaps, tunnel entrances, or mixing systems; the main spatial location and area; whether it is a cross-level downgrade; whether it is a high-risk level inflow; and whether it is a re-disturbance after recovery.

[0095] Based on the above identification results, corresponding control recommendations can be generated. For example: if the risk originates from a rapid decline in the bearing capacity around the construction road, it is recommended to strengthen construction boundary control, road drainage, slope protection, and temporary covering; if the risk originates from the expansion of high-risk levels around the spoil heap, it is recommended to strengthen the zoned storage of spoil, interception and drainage works, topsoil covering, and revegetation measures; if the risk originates from the water and sediment response of the riverbank disturbance zone, it is recommended to strengthen riverbank protection, construction drainage control, and sediment interception; if the risk originates from the mixing system or construction platform, it is recommended to optimize construction organization, reduce hardening expansion, and strengthen dust / sewage control; if identified as an ecological restoration and improvement zone, it can be used as an effective area for restoration measures; if identified as a re-disturbance risk zone, it should be used as a key monitoring area during the initial phase of new construction projects. The aim is to achieve the transformation from "risk identification" to "risk management recommendations."

[0096] In one implementation, such as Figure 2 As shown, this application also provides a system for identifying and warning of the transition of ecological carrying capacity levels in hydropower projects, including: The multi-source data acquisition module 201 is used to acquire multi-phase ecological carrying capacity evaluation results, engineering disturbance spatial data, construction stage information, ecological restoration area and auxiliary geographic environment data of the hydropower project impact area, and generate basic dataset information; The data preprocessing and grading module 202 is used to carry out spatial registration, temporal matching and unified grading coding, construct six types of ecological carrying capacity grading according to the grading rules, and generate multi-period ecological carrying capacity grading map information. The grading transfer and rate calculation module 203 is used to establish inter-period tracking relationships with pixels as units, construct a grading transfer matrix, calculate the inflow / outflow flux of upgrading, downgrading, stability, cross-grading, and high-risk, incorporate time interval parameters to calculate the grading transfer rate, and generate transfer rate calculation result information. The leap risk index construction module 204 is used to construct pixel-level and regional-level leap risk indices. It integrates the core features of downgrade magnitude, high-risk transition, and cross-level downgrade through a linear weighted model to generate ecological carrying capacity level leap risk feature set information. The engineering association and zoning identification module 205 is used to associate the transition results with the engineering disturbance units of construction roads, spoil disposal sites and tunnel entrances based on spatial overlay and buffer analysis, identify degradation zones, repair and improvement zones, stability maintenance zones and re-disturbance risk zones, and generate zoning identification result information. The early warning classification and control generation module 206 is used to divide the four early warning levels using a multi-threshold classification judgment mechanism, introduce the risk source tracing and control suggestion generation module, and combine the risk contribution of engineering units to output early warning results and control strategies, and generate early warning and control result information. The verification, correction, and result output module 207 is used to conduct accuracy verification and dynamic correction based on data from all stages of the process and actual monitoring data. It balances the accuracy of transition identification, early warning response efficiency, and engineering adaptability requirements through a multi-index comprehensive evaluation algorithm, and generates ecological carrying capacity level transition identification and risk early warning result information that is suitable for the entire process of hydropower projects.

[0097] All embodiments in this application are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the embodiments of the method, electronic device, electronic device, and readable storage medium for assessing the ecological carrying capacity level transition and risk warning of hydropower projects are basically similar to the above-described embodiments of the method for assessing the ecological carrying capacity level transition and risk warning of hydropower projects, so the description is relatively simple. Relevant parts can be referred to in the descriptions of the above-described embodiments of the method for assessing the ecological carrying capacity level transition and risk warning of hydropower projects.

Claims

1. A method for identifying and risk warning of ecological carrying capacity level transitions in hydropower projects, characterized in that, include: Obtain multi-phase ecological carrying capacity assessment results, spatial data of engineering disturbance, construction stage information, ecological restoration area and auxiliary geographical environment data of the hydropower project impact area; Data preprocessing was completed based on spatial registration, temporal matching and hierarchical unified coding. A six-category ecological carrying capacity level system was constructed through level classification rules, and multi-period ecological carrying capacity level maps were generated. Establish inter-period tracking relationships using pixels as units, construct a grade transition matrix, and calculate indicators for upgrade, downgrade, stability, cross-grade transition, and high-risk inflow / outflow fluxes. Incorporate time interval parameters to calculate the grade transition rate. Construct pixel-level and regional-level transition risk indices, and integrate the core features of downgrade magnitude, high-risk transition, and cross-level downgrade through a linear weighted model to generate a set of ecological carrying capacity level transition risk features; Based on spatial overlay and buffer analysis, the transition results are associated with engineering disturbance units of construction roads, spoil heaps, and tunnel entrances to identify degradation zones, repair and improvement zones, stability maintenance zones, and re-disturbance risk zones. A multi-threshold hierarchical judgment mechanism is adopted to divide the four-level early warning level, and a risk source tracing and control suggestion generation module is introduced. The early warning results and control strategies are output in combination with the risk contribution of engineering units. Accuracy verification and dynamic correction are performed based on data from all stages of the process and actual monitoring data. A multi-index comprehensive evaluation algorithm is used to balance the accuracy of transition identification, early warning response efficiency, and engineering adaptability requirements, and to generate ecological carrying capacity level transition identification and risk early warning results that are suitable for the entire process of hydropower projects.

2. The method as described in claim 1, characterized in that, Data preprocessing was completed based on spatial registration, temporal matching, and unified hierarchical coding. A six-category ecological carrying capacity level system was constructed using hierarchical classification rules, generating multi-period ecological carrying capacity level maps, including: Based on a unified coordinate system, spatial resolution, and raster range rules, combined with invalid image removal, missing area masking / interpolation, and pixel-to-pixel correspondence operations, multi-phase ecological carrying capacity data preprocessing is completed through spatial registration, temporal matching, and unified level coding. Based on preset / natural breakpoints / quantiles / cloud models / expert experience, a carrying capacity index threshold system is determined. The rule that the smaller the code, the higher the risk is used to construct six carrying capacity levels: multiple overload, critical warning, sub-suitable buffer, balanced pressure, resilient surplus, and ideal sustainability. Generate multi-period ecological carrying capacity level raster maps with temporal matching, spatial registration, and unified coding.

3. The method as described in claim 1, characterized in that, Establish inter-period tracking relationships using pixels as units, construct a grading transition matrix, and calculate indices for upgrades, downgrades, stability, cross-grading transitions, and high-risk inflow / outflow fluxes. Incorporate time interval parameters to calculate grading transition rates, including: Using pixels as the sole spatial tracking unit, a temporal cross-period mapping relationship is established for the same location at different times, accurately matching the ecological carrying capacity level of each period; Construct a matrix of number of transfers, an area matrix, and a probability matrix for each level, and count the number of pixels transferred in, transferred out, and retained, the corresponding area, and the transfer probability by row and column. The six core indicators are calculated according to matrix statistical rules: stable flux, upgraded flux, downgraded flux, cross-level downgrade flux, high-risk inflow flux, and high-risk outflow flux. By introducing the time interval parameter between adjacent evaluation periods, the degradation rate, cross-level degradation rate, and high-risk inflow rate are calculated respectively to quantify the rate of deterioration of the ecological carrying capacity.

4. The method as described in claim 1, characterized in that, A pixel-level and regional-level transition risk index is constructed. A linear weighted model is used to integrate core features such as the magnitude of downgrade, high-risk transition, and cross-level downgrade to generate a feature set of ecological carrying capacity level transition risks, including: Based on three core characteristics—the extent of downgrade, the high-risk transfer indicator, and the cross-level downgrade indicator—and following the construction principles of risk correlation, weight adaptability, and hierarchical differentiation, the weight coefficients and index calculation rules are determined collaboratively by the linear weighted model and the flux fusion model. The method employs a tiered approach that integrates single-pixel feature aggregation, regional feature averaging, and risk level stratification, assigning high-risk levels 1 and 2 for priority configuration. , Weight greater than To highlight the importance of high-risk transfers and cross-level downgrades, and to match the requirements for risk quantification at the pixel level and regional level; Simultaneous verification of weight configuration, feature contribution, and index threshold is performed. Multi-dimensional verification is conducted on feature completeness, weight rationality, and index discrimination to generate two types of results: pixel-level linear weighted risk index and regional-level flux-weighted risk index. By integrating three core features and simultaneously matching feature types, weight coefficients, contribution ratios, index levels, and high-risk definition information, a feature set for the risk of ecological carrying capacity level transition is generated.

5. The method as described in claim 1, characterized in that, Based on spatial overlay and buffer analysis, the transition results are correlated with engineering disturbance units of construction roads, spoil heaps, and tunnel entrances to identify degradation zones, repair and improvement zones, stability maintenance zones, and re-disturbance risk zones, including: Based on the spatial attributes of engineering disturbance units, the raster characteristics of transition results, spatial superposition matching rules and buffer analysis logic, and following the construction criteria of spatial location correlation, influence distance adaptability, risk contribution separability and partition identification accuracy, the buffer range, spatial matching parameters and partition judgment conditions are determined by the distance threshold and spatial superposition algorithm in collaboration. An integrated approach is adopted, which includes unit impact zone construction, risk indicator correlation, and zoning type determination. Distance thresholds are set according to unit types such as construction roads, spoil heaps, and tunnel entrances. The unit impact range is constructed through buffer zone analysis, and the transition risk index is matched with the unit spatial location. The association results, partitioning rules, and distance thresholds are simultaneously verified. The location rationality is verified for spatial matching degree, the association validity is verified for risk contribution degree, and the partitioning accuracy is checked for partition boundaries to form unit association risk index, high-risk area and proportion. By integrating spatial correlation results and simultaneously matching risk index, distance threshold, unit type, and zoning type information, we can identify construction disturbance degradation zones, ecological restoration and improvement zones, stability maintenance zones, and re-disturbance risk zones.

6. The method as described in claim 5, characterized in that, Based on data from all stages of the process and actual monitoring data, accuracy verification and dynamic correction are performed. A multi-index comprehensive evaluation algorithm balances the accuracy of transition identification, early warning response efficiency, and engineering adaptability requirements, generating ecological carrying capacity level transition identification and risk early warning results suitable for the entire process of hydropower projects. These results include: Based on the time-series characteristics of the entire hydropower project, the monitoring data verification standards, the accuracy requirements for transition identification, and the engineering adaptation constraints, and following the construction principles of time-series consistency, data authenticity, accuracy controllability, and adaptation integrity, the verification dimensions, correction rules, and balance parameters are determined collaboratively by a multi-source data fusion mechanism and a multi-index comprehensive evaluation algorithm. It adopts an integrated setting method of time series data alignment, monitoring data benchmarking, accuracy index calculation, and adaptability evaluation. It matches time series verification standards according to the pre-construction, construction period, operation recovery period, and new construction start-up period, sets dynamic correction coefficients according to the measured-simulated deviation, and configures balanced weights according to identification accuracy, response efficiency, and engineering adaptability. Simultaneous verification of verification dimensions, correction rules, and balancing parameters; adjustment of alignment strategies for time series deviations; optimization of correction coefficients for data errors; and reconfiguration of balancing weights for indicator conflicts, resulting in multi-dimensional verification and dynamic correction results. By integrating data from the entire process, verification results, correction parameters, and balancing weights, the system outputs a final result that is consistent in timing, reliable in data, balanced in accuracy, and adaptable to the entire engineering process, enabling the identification and risk warning of ecological carrying capacity level transitions.

7. A system for identifying and warning of ecological carrying capacity transitions in hydropower projects, characterized in that, The system includes: The multi-source data acquisition module is used to acquire multi-phase ecological carrying capacity evaluation results, spatial data of engineering disturbance, construction stage information, ecological restoration area and auxiliary geographic environment data of the hydropower project impact area, and generate basic dataset information; The data preprocessing and grading module is used to carry out spatial registration, temporal matching and unified grading coding, and to construct six types of ecological carrying capacity grading according to the grading rules, generating multi-period ecological carrying capacity grading map information. The grading transfer and rate calculation module is used to establish inter-period tracking relationships at the pixel level, construct a grading transfer matrix, calculate the inflow / outflow flux of upgrading, downgrading, stabilization, cross-grading, and high-risk, incorporate time interval parameters to calculate the grading transfer rate, and generate transfer rate calculation results information. The transition risk index construction module is used to construct pixel-level and regional-level transition risk indices. It integrates the core features of downgrade magnitude, high-risk transition, and cross-level downgrade through a linear weighted model to generate a set of ecological carrying capacity level transition risk features. The engineering association and zoning identification module is used to associate the transition results with the engineering disturbance units of construction roads, spoil disposal sites, and tunnel entrances based on spatial overlay and buffer analysis, identify degradation zones, repair and improvement zones, stability maintenance zones, and re-disturbance risk zones, and generate zoning identification result information. The early warning classification and control generation module is used to classify four early warning levels using a multi-threshold classification judgment mechanism. It introduces the risk source tracing and control suggestion generation module, combines the risk contribution of engineering units to output early warning results and control strategies, and generates early warning and control result information. The verification, correction, and result output module is used to conduct accuracy verification and dynamic correction based on data from all stages of the process and actual monitoring data. Through a multi-index comprehensive evaluation algorithm, it balances the requirements of transition identification accuracy, early warning response efficiency, and engineering adaptability, and generates ecological carrying capacity level transition identification and risk early warning results information that are suitable for the entire process of hydropower projects.

8. An electronic device, characterized in that, include: First processor; The processor also includes a memory for storing executable instructions of the first processor; wherein the first processor is configured to execute the method for identifying and warning of the transition of ecological carrying capacity levels in hydropower projects according to any one of claims 1 to 6 by executing the executable instructions.

Images