An intelligent decision-making method for river and lake health recovery scheme based on progressive ecological restoration theory

Abstract

The application discloses a kind of river and lake health recovery scheme intelligent decision-making method based on progressive ecological restoration theory, belong to environmental management field, this method includes: the target steady state in river and lake ecosystem is decomposed into multiple ecological parameter balance states, and the corresponding response protocol is preset for each ecological parameter balance state, and the collection of all response protocols constitutes ecological response protocol cluster;According to the real-time state of river and lake ecosystem, activate and arrange protocol execution sequence from ecological response protocol cluster, construct flexible failure boundary model for each ecological parameter balance state, and the flexible failure boundary model is a dynamic state envelope;It can be realized by constructing dynamic flexible failure boundary model, combined with disturbance memory energy field, accurately identify resilience challenge events, solve the limitations of fixed repair scheme, effectively adapt to the complex scene of river and lake ecosystem disturbed by climate change, human activities and the like, and improve the anti-interference ability of repair scheme.

Description

Technical Field

[0001] This invention relates to the field of environmental governance, and more specifically, to an intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration. Background Technology

[0002] In actual river and lake ecological restoration work, the restoration process is often full of uncertainties due to external disturbances such as climate change and human activities. Most of the current mainstream restoration methods are implemented according to a pre-set fixed plan. However, the cycle of river and lake restoration projects is usually very long, ranging from several years to several decades. During this process, once unexpected disturbances such as sudden changes in climate patterns or sudden pollution occur, the original plan is often difficult to deal with effectively. Ultimately, the preset restoration path gradually deviates from the expected goal in actual implementation, or even becomes completely ineffective, which fully exposes the obvious shortcomings of fixed plans in dealing with dynamically changing environments.

[0003] If the pre-set gradual restoration plan is strictly followed, any unexpected situation may occur along the way, affecting the final restoration effect and causing the plan to gradually become ineffective in actual implementation. Existing decision-making methods excessively pursue the optimal solution of a single-stage restoration plan. This one-sided optimization orientation weakens the overall resilience of the restoration system in the face of continuous uncertainty interference and cannot adapt to the needs of multi-stage advancement and dynamic adjustment of gradual restoration work. At present, the intelligent decision-making methods for river and lake ecological restoration plans still have obvious deficiencies in uncertainty management, long-term effect evaluation and feedback, and dynamic adaptation capabilities. These problems affect the effectiveness of river and lake health restoration and the sustainable advancement of restoration work. Summary of the Invention

[0004] To address the problems existing in the prior art, the present invention aims to provide an intelligent decision-making method for river and lake health restoration schemes based on the theory of progressive ecological restoration. This method can accurately identify resilience challenge events by constructing a dynamic elastic failure boundary model and combining it with a perturbation memory energy field, thereby overcoming the limitations of fixed restoration schemes and effectively adapting to complex scenarios where river and lake ecosystems are subject to dynamic disturbances such as climate change and human activities, thus improving the anti-interference capability of restoration schemes.

[0005] To solve the above problems, the present invention adopts the following technical solution:

[0006] A smart decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, the method comprising:

[0007] Step 1: Deconstruct the target steady state in the river and lake ecosystem into multiple ecological parameter equilibrium states, and pre-set a corresponding response protocol for each ecological parameter equilibrium state. The set of all response protocols constitutes an ecological response protocol cluster. Based on the real-time state of the river and lake ecosystem, activate and arrange the protocol execution sequence from the ecological response protocol cluster.

[0008] Step 2: Construct an elastic failure boundary model for each ecological parameter equilibrium state. The elastic failure boundary model is a dynamic state envelope surface. Monitor the position and trajectory of the ecological parameter in the state envelope surface. When the trajectory touches and attempts to cross the current elastic failure boundary with a preset angle and acceleration, it is determined as a resilience challenge event, triggering the activation process of the corresponding protocol sub-cluster in the ecological response protocol cluster.

[0009] Step 3: When multiple response protocols are triggered, based on the perturbation pattern represented by the perturbation memory energy site, synthesize multiple simulated perturbation time sequences of ecological parameter equilibrium states within a preset future time period; enable all triggered candidate protocols and their combinations to synchronously perform stress tests of the perturbation time sequence in a parallel virtual river and lake model, and evaluate the offset distance between the ecological parameter trajectory and the failure boundary of each candidate protocol or combination under each perturbation time sequence; based on the evaluation results, select the execution right.

[0010] Furthermore, it also includes:

[0011] Step 4: Track the actual execution process and results of the combination of agreements that have obtained the right to execute, calculate the elastic capital consumed in the response and the ecological technology debt generated, and generate the accounting results including immediate benefits, elastic consumption and technology debt;

[0012] Step 5: Dynamically revise the elastic failure boundary model based on the calculation results;

[0013] Step 6: Periodically integrate the execution records of steps 1 to 5 to generate a resilience evolution map; based on the resilience evolution map, generate new response protocols and add them to the ecological response protocol cluster.

[0014] Furthermore, the target steady state in river and lake ecosystems is deconstructed into multiple ecological parameter equilibrium states, including:

[0015] Step 11: Define the target steady state as a complex consisting of structural steady state, process steady state, and functional steady state;

[0016] Step 12: Based on structural steady state, process steady state and functional steady state, perform control factor mapping for each dimension to identify the corresponding ecological parameters for each dimension;

[0017] Step 13: Define the conditional domains of the ecological parameters under multiple different baseline scenarios.

[0018] Furthermore, an elastic failure boundary model is constructed for the equilibrium state of each ecological parameter, including:

[0019] Step 21: Extract perturbation imprints from the time series of each ecological parameter, identify deviation events that exceed their conditional state domain, quantify the deviation magnitude, duration, and recovery rate of each deviation event; convert each deviation event into a perturbation energy unit with decay characteristics, and superimpose all historical perturbation energy units to form a perturbation memory energy field.

[0020] Step 22: Based on the real-time status of the perturbation memory energy field, seasonal phase signal, and associated ecological parameters, perform multi-factor coupling operation to calculate the dynamic boundary base value;

[0021] Step 23: Based on the dynamic boundary baseline value, elastic bandwidth modulation is performed according to the state of the perturbation memory energy field and associated ecological parameters to generate an elastic failure boundary jointly defined by the dynamic boundary baseline value and the dynamic elastic bandwidth.

[0022] Furthermore, the determination of resilience challenge events includes:

[0023] Step 24: Track the movement of ecological parameters in real time within the state envelope defined by the elastic failure boundary, and continuously calculate the trajectory dynamic vector, which includes the pointing component, the approaching component, and the inertial component.

[0024] Step 25: Perform joint analysis of the trajectory dynamic vector and the dynamic characteristics of the current elastic failure boundary, perform crossing risk prediction, and calculate the instantaneous crossing risk index;

[0025] Step 26: Perform pattern recognition on the time series of instantaneous traversal risk index. When a pulse impact pattern or continuous pressure pattern that matches the predefined pattern is identified, it is determined to be a resilience challenge event, and the corresponding protocol sub-cluster is activated from the ecological response protocol cluster to start the competition process.

[0026] Furthermore, the simulated perturbation time series of multiple ecological parameter equilibrium states over a future preset period are synthesized, including:

[0027] Step 31: Analyze historical disturbance data, perform disturbance event slicing and pattern primitive extraction, and construct a disturbance pattern primitive library containing three dimensions: intensity envelope, spectral features, and covariance relationship.

[0028] Step 32: Generate primitive selection and combination rules based on the current perturbation memory energy field, elastic failure boundary, and resilience challenge event pattern;

[0029] Step 33: Based on the combination rules, multiple future perturbation time series are synthesized using the path-dependent Monte Carlo method.

[0030] Furthermore, stress tests were performed on the perturbation time series, and the offset distance between the ecological parameter trajectory and the failure boundary of each candidate protocol or combination under each perturbation time series was evaluated, including:

[0031] Step 34: Create an independent virtual river and lake model copy for each triggered candidate protocol, using the state including real-time values ​​of ecological parameters, elastic failure boundary and disturbance memory energy field as a baseline snapshot, and input the protocol preloaded influence field according to the mechanism of action of the candidate protocol.

[0032] Step 35: Apply future perturbation timings in parallel to each virtual river and lake model replica, so that the protocol responds dynamically according to conditions and action modes, and records the evolution trajectory of ecological parameters.

[0033] Step 36: Calculate the offset distance spectrum based on the evolution trajectory and the dynamic elastic failure boundary at the corresponding time. The offset distance spectrum includes the minimum offset distance, the duration of offset danger, and the offset trend integral.

[0034] Furthermore, the actual execution process and results of the portfolio of agreements for which enforcement rights have been obtained are tracked, and the elastic capital consumed and the eco-technology debt generated in response are calculated, including:

[0035] Step 41: During and after the implementation of the protocol combination, the actual observation trajectory of ecological parameters is recorded synchronously, and the baseline prediction trajectory of natural fluctuations during the same period is generated. The net effect of intervention is obtained by comparison.

[0036] Step 42: Based on the net effect of the intervention, calculate the state maintenance energy and the process acceleration energy, and sum them up to obtain the elastic capital consumption;

[0037] Step 43: After the agreement is executed, compare the actual recovery trajectory of the parameters with the expected natural recovery trajectory to identify hysteresis distortions and calculate the value of eco-technical debt.

[0038] Step 44: Generate a three-dimensional accounting table. The three-dimensional accounting table records the immediate benefits, resilient capital consumption, and ecological technology debt value calculated based on the improvement of the ecological parameter trajectory challenged in the resilience challenge event relative to its resilient failure boundary after the protocol is implemented.

[0039] Furthermore, based on the calculation results, the elastic failure boundary model is dynamically revised, including:

[0040] Step 51: Analyze the three-dimensional accounting table, establish the association mapping between protocol type and vulnerability, and determine the association between vulnerability and elastic failure boundary;

[0041] Step 52: Based on the correlation mapping, extract the cost efficiency ratio, debt lag index and sensitivity changes of correlation parameters from the three-dimensional accounting table as quantitative influencing factors.

[0042] Step 53: Based on the quantified influence factor, correct the modulation coefficient used to generate the dynamic elastic bandwidth, the decay time constant of the perturbation memory energy field, and the weights of the multi-factor coupling rule in the elastic failure boundary model.

[0043] Furthermore, a resilience evolution map is generated. Based on the resilience evolution map, new response protocols are generated and added to the ecological response protocol cluster, including:

[0044] Step 61: Periodically extract the execution records from Step 1 to Step 5, perform time axis alignment and state space reconstruction, identify turning points and extract patterns from the trajectory segments between turning points to generate a structured resilience evolution knowledge network.

[0045] Step 62: Based on the structured resilience evolution knowledge network, perform pattern diagnosis and root cause correlation analysis to obtain the deterioration trajectory pattern and root cause hypothesis;

[0046] Step 63: Based on the deterioration trajectory pattern and root cause hypothesis, generate strategic-level protocols through reverse design and input them into the ecological response protocol cluster.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] (1) This scheme solves the limitations of fixed restoration schemes by constructing a dynamic elastic failure boundary model and combining it with the perturbation memory energy field to accurately identify resilience challenge events. It effectively adapts to the complex scenarios of dynamic disturbances such as climate change and human activities in river and lake ecosystems, and improves the anti-interference ability of the restoration scheme.

[0049] (2) This scheme conducts multi-disturbance time-series stress tests through parallel virtual river and lake models, and evaluates the candidate protocols and their combined effects by combining offset distance spectrum quantification, so as to achieve the scientific allocation of the optimal protocol execution right, ensure the accuracy and operability of the repair decision, and reduce the ineffective resource input.

[0050] (3) This plan continuously explores the evolutionary laws of the resilience of river and lake ecosystems through process tracking, three-dimensional accounting and dynamic model correction, dynamically enriches the ecological response protocol suite, and ensures the long-term sustainable advancement of ecological restoration work.

[0051] (4) This plan balances the immediate benefits and long-term potential impacts of the restoration plan by quantifying the consumption of flexible capital and ecological technology debt, avoiding the weakening of system resilience caused by single-stage optimization, and improving the overall benefits and ecological security of river and lake health restoration. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0053] Figure 1 This is a flowchart of the intelligent decision-making method for river and lake health restoration programs according to the present invention. Detailed Implementation

[0054] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0055] Please see Figure 1 A smart decision-making method for river and lake health restoration based on the theory of gradual ecological restoration, the method includes:

[0056] Step 1: Deconstruct the target steady state in the river and lake ecosystem into multiple ecological parameter equilibrium states, and pre-set a corresponding response protocol for each ecological parameter equilibrium state. The set of all response protocols constitutes an ecological response protocol cluster. Based on the real-time state of the river and lake ecosystem, activate and arrange the protocol execution sequence from the ecological response protocol cluster. The specific operations are as follows:

[0057] River and lake ecosystems, as complex dynamic systems in existing technologies, are organic wholes composed of water bodies, aquatic biological communities, surrounding terrestrial ecosystems, and various environmental factors. Their integrity depends on the coordinated balance of factors such as the spatiotemporal variability of hydrological conditions, the spatial heterogeneity of river and lake landforms, the three-dimensional connectivity of river and lake systems, the range of suitable physicochemical properties of water bodies for biological survival, food web structure, and biodiversity. These factors interact to support core ecological processes such as material cycling, energy flow, and information transmission within the system, thereby ensuring the stable output of ecological services such as water purification, flood control, and habitat provision. First, based on the above-mentioned composition and characteristics of river and lake ecosystems, it is necessary to systematically deconstruct their target steady state, breaking it down into multiple independently monitorable and quantifiable ecological parameter equilibrium states. Each ecological parameter equilibrium state corresponds to the stable state of a certain dimension of the river and lake ecosystem, ensuring that all characteristics of the target steady state can be obtained through specific ecological parameters. To demonstrate this, for each ecological parameter equilibrium state, a corresponding response protocol needs to be pre-defined. The response protocol should clearly define the countermeasures, execution process, triggering conditions, and expected effects when the parameter equilibrium state deviates. All pre-defined response protocols are compiled into an ecological response protocol cluster, which covers response plans for various possible parameter deviation scenarios, forming a complete response strategy system. In practical application, various status data of river and lake ecosystems are collected in real time through deployed monitoring equipment, including water quality, hydrology, biological community, and other related indicators. Based on these real-time data, the current real-time status of the river and lake ecosystem is determined, and then the response protocol that matches the current status is selected from the ecological response protocol cluster and activated. After activation, the execution sequence of the relevant protocols needs to be arranged. According to the priority, correlation, and time sequence of protocol execution, the start time and execution order of each protocol are determined to ensure that multiple protocols work together to efficiently guarantee the evolution of the river and lake ecosystem towards the target steady state.

[0058] The process of deconstructing the target steady state in river and lake ecosystems into multiple ecological parameter equilibrium states also includes the following steps:

[0059] Step 11: Define the target steady state as a complex consisting of structural steady state, process steady state, and functional steady state. The specific operations are as follows:

[0060] The target steady state is not a single-dimensional stable state, but an organic complex composed of structural steady state, process steady state, and functional steady state. These three are interconnected and mutually influential, jointly maintaining the overall stability of the river and lake ecosystem. Structural steady state focuses on the stability of the physical structure and biological community structure of the river and lake ecosystem, including riverbed morphology, shoreline structure, aquatic plant community composition, fish and microbial community structure, etc. Its role is to ensure the integrity and rationality of the spatial structure of the ecosystem and provide the basic carrier for the development of various ecological processes. Process steady state focuses on the core processes of material cycling, energy flow, and information transmission within the river and lake ecosystem. The stable operation of ecological processes, including the cycling and transformation of nutrients such as nitrogen and phosphorus, the balance of dissolved oxygen in water bodies, and predation and symbiosis among organisms, is crucial for the proper functioning of an ecosystem. Functional homeostasis emphasizes the stable output of various ecological services provided by river and lake ecosystems, including water purification, flood control, habitat provision, and climate regulation. Its stability directly relates to the supporting role of river and lake ecosystems in human society and the natural environment. By clarifying the three-dimensional composition of target homeostasis, a comprehensive definition of target homeostasis can be achieved, avoiding the one-sidedness of management caused by considering only one dimension.

[0061] Step 12: Based on structural, process, and functional steady states, perform control factor mapping for each dimension to identify the corresponding ecological parameters for each dimension. The specific operations are as follows:

[0062] For the structural stability dimension, control factor mapping needs to be performed to identify key control factors affecting the stability of the physical and biological community structures of river and lake ecosystems, such as riverbed slope, shoreline vegetation cover, aquatic plant biomass, fish species richness, and microbial community diversity. These control factors are then transformed into quantifiable and monitorable ecological parameters, each of which can intuitively reflect a specific aspect of structural stability. For the process stability dimension, control factor mapping is also performed, focusing on control factors of ecological processes such as material cycling and energy flow, such as nitrogen and phosphorus nutrient concentrations, dissolved oxygen content in water, etc. pH value, water flow velocity, biological metabolic rate, etc., these ecological parameters can be monitored to keep track of the operation status of various ecological processes in real time. In the functional homeostasis dimension, the control factor mapping needs to be carried out around the output effect of the ecosystem service function, identify the key control factors affecting water purification function, flood storage function, etc., and convert them into corresponding ecological parameters, such as chemical oxygen demand (COD) removal rate, ammonia nitrogen removal rate, water storage capacity, habitat suitability index, etc. Through the above mapping process, it is ensured that each homeostasis dimension has clear ecological parameters as support, so as to achieve the quantitative characterization of the target homeostasis.

[0063] Step 13: Define the conditional state domains of ecological parameters under multiple different baseline scenarios. The specific operations are as follows:

[0064] First, based on the actual conditions of river and lake ecosystems, multiple typical baseline scenarios need to be established. These scenarios should cover common scenarios such as natural conditions, mild disturbance, and moderate disturbance. Examples include natural background scenarios (no human activity disturbance, stable climate, etc.), mild disturbance scenarios due to agricultural non-point source pollution, moderate disturbance scenarios due to urban domestic sewage, and short-term disturbance scenarios due to extreme rainfall. This ensures the comprehensiveness and representativeness of the baseline scenarios. For each ecological parameter, its conditional domain under each baseline scenario needs to be determined based on historical monitoring data, indoor simulation experimental results, relevant industry standards, and literature. The definition of the conditional domain needs to clearly define the reasonable range of parameter values, including upper and lower threshold values. The value and optimal range of values ​​are determined by comprehensively considering the impact of the parameter on the corresponding steady-state dimension and the self-regulation capacity of the river and lake ecosystem. For example, for the ecological parameter of dissolved oxygen content in water, under natural background scenarios, its conditional domain may be set to 5 to 9 mg / L, with an optimal range of 6 to 8 mg / L. Under the scenario of mild disturbance from agricultural non-point source pollution, considering the consumption of dissolved oxygen by pollution, the conditional domain may be adjusted to 4 to 8 mg / L, with an optimal range of 5 to 7 mg / L. By setting a clear conditional domain for each ecological parameter under different baseline scenarios, it is possible to clearly determine whether the parameter is in a stable state under the current scenario.

[0065] In a preferred embodiment of the present invention, step 2 is further included: constructing an elastic failure boundary model for each ecological parameter equilibrium state, wherein the elastic failure boundary model is a dynamic state envelope surface; monitoring the position and trajectory of the ecological parameter in the state envelope surface; when the trajectory touches and attempts to cross the current elastic failure boundary with a preset angle and acceleration, it is determined as a resilience challenge event, triggering the activation process of the corresponding protocol sub-cluster in the ecological response protocol cluster, the specific operation is as follows:

[0066] For each ecological parameter equilibrium state obtained in step 1, an independent elastic failure boundary model needs to be constructed. This model is not a fixed threshold, but a state envelope that can be dynamically adjusted according to changes in the state of the river and lake ecosystem and the impact of historical disturbances. Its function is to define the safe operating range of the ecological parameter equilibrium state, while reserving a certain elastic adjustment space to adapt to system fluctuations. Then, the elastic failure boundary model is constructed. After the model is constructed, dynamic data of the corresponding ecological parameters are continuously collected through real-time monitoring. Based on this data, the specific position of the parameter within the state envelope is located in real time, and the motion trajectory of the parameter is generated by the trajectory fitting algorithm. The trajectory needs to reflect the changing trend and dynamic characteristics of the parameters over time. During trajectory monitoring, the motion posture of the trajectory is continuously analyzed, and attention is paid to the angle and acceleration of the trajectory approaching the elastic failure boundary. When the motion trajectory is detected to touch the elastic failure boundary at a preset angle and acceleration, and there is a tendency to cross the boundary, a resilience challenge event is determined to have occurred. This determination process needs to combine preset angle thresholds and acceleration thresholds to ensure the accuracy and reliability of the event determination. Once a resilience challenge event is determined, the protocol sub-cluster in the ecological response protocol cluster corresponding to the equilibrium state and disturbance type of the ecological parameter is immediately triggered, and the activation process of the protocol sub-cluster is initiated.

[0067] The process of constructing an elastic failure boundary model for the equilibrium state of each ecological parameter also includes the following steps:

[0068] Step 21: Extract perturbation imprints from the time series of each ecological parameter to identify deviation events that exceed their conditional state domain, and quantify the deviation magnitude, duration, and recovery rate of each deviation event; convert each deviation event into a perturbation energy unit with decay characteristics, and superimpose all historical perturbation energy units to form a perturbation memory energy field. The specific operations are as follows:

[0069] For each ecological parameter, time-series data generated from long-term monitoring is first collected. This time-series data must cover parameter values ​​under different scenarios to ensure data integrity and representativeness. Based on the conditional domains defined in step 13 for each baseline scenario, perturbation imprints are extracted from the time-series data. The parameter value at each moment in the time series is compared with the threshold of the corresponding conditional domain. Deviation events where the parameter value exceeds the upper or lower limit of the conditional domain are identified. The identification of deviation events must be based on the boundary threshold of the conditional domain; only when the parameter value exceeds the threshold range can it be considered a deviation event. For each identified deviation event, its characteristic parameters are quantified, including deviation magnitude, duration, and recovery rate. The deviation magnitude is calculated by the difference between the actual parameter value and the boundary threshold of the conditional domain. The duration is the parameter value at each moment. The duration of the value outside the conditional state domain is defined as the recovery rate, which is the rate at which the parameter reverts from the deviation state back to the conditional state domain, calculated using the slope of the regression process. Subsequently, each deviation event is transformed into a perturbation energy unit. The value of the energy unit is positively correlated with the deviation amplitude and duration, and negatively correlated with the recovery rate. At the same time, the energy unit is given a decay characteristic, meaning that as time goes on, the impact of historical perturbation events gradually weakens, and the value of the energy unit gradually decreases according to a preset decay law. Finally, all the perturbation energy units transformed from historical deviation events are superimposed along the time dimension. The superposition process needs to consider the decay state of each energy unit. The overall energy field formed after superposition is the perturbation memory energy field. This energy field can intuitively reflect the comprehensive impact of historical perturbations on ecological parameters, providing input data for the boundary base value calculation and bandwidth modulation of the elastic failure boundary model.

[0070] Step 22: Based on the real-time state of the perturbation memory energy field, seasonal phase signal, and associated ecological parameters, perform multi-factor coupling operation to calculate the dynamic boundary basis value. The specific operation is as follows:

[0071] The dynamic boundary baseline value serves as the benchmark parameter for the elastic failure boundary model. Its calculation requires comprehensive consideration of three core factors: the perturbation memory energy field, the seasonal phase signal, and the real-time state of other related ecological parameters. The perturbation memory energy field is constructed using the results from step 21, extracting the energy field intensity value at the current moment. This value reflects the comprehensive impact of historical perturbations on the current parameter state. The seasonal phase signal is obtained through analysis of the seasonal characteristics of the river and lake ecosystem region. Combined with the influence of seasonal changes on ecological parameters, the seasonal characteristics are transformed into quantified phase signals. Different seasons correspond to different phase signal values ​​to reflect the influence of seasonal factors on the parameter stability range. The real-time state of other related ecological parameters is obtained through real-time monitoring. Related ecological parameters need to be selected and determined based on the characteristics of the target ecological parameter, i.e., parameters that interact and influence each other with the target parameter. Their real-time state is directly obtained and quantified through monitoring data. After obtaining the quantified data of the three types of factors, a multi-factor coupling operation is performed. During the operation, each factor is assigned a corresponding weight coefficient based on its influence weight on the equilibrium state of the target ecological parameter. The weight coefficients are determined through historical data regression analysis to ensure the rationality of the weight allocation. The coupling operation adopts a combination of linear and nonlinear methods, comprehensively considering the synergistic and independent effects of each factor. The numerical value obtained through the operation is the dynamic boundary base value. This base value is adjusted in real time according to the changes of the three types of factors to ensure that it can adapt to the dynamic changes of the system and provide stable and accurate benchmark support for the subsequent elastic bandwidth modulation of the elastic failure boundary model.

[0072] Step 23: Based on the dynamic boundary baseline value, elastic bandwidth modulation is performed according to the state of the perturbation memory energy field and associated ecological parameters to generate an elastic failure boundary jointly defined by the dynamic boundary baseline value and the dynamic elastic bandwidth. The specific operation is as follows:

[0073] The dynamic boundary baseline value is only a reference for the elastic failure boundary model. The elastic bandwidth needs to be modulated in conjunction with the actual system state and disturbance characteristics to form a state envelope that can adapt to system fluctuations, ultimately constituting a complete elastic failure boundary model. The modulation of the elastic bandwidth is based on the current strength of the disturbance memory energy field and the real-time state of the associated ecological parameters. When the disturbance memory energy field strength is high, it indicates that historical disturbances have a significant impact on the system, and the system's current anti-interference capability is relatively weak. In this case, the elastic bandwidth needs to be increased to reserve more adjustment space for parameter fluctuations. When the disturbance memory energy field strength is low, the system's anti-interference capability is strong, and the elastic bandwidth can be appropriately reduced to improve the accuracy of parameter control. At the same time, the bandwidth is fine-tuned in conjunction with the real-time state of the associated ecological parameters. If the associated ecological parameters are in a stable state, the current bandwidth setting can be maintained. If the associated ecological parameters show a fluctuation trend, the bandwidth needs to be adjusted accordingly to avoid the risk of cooperative disturbances. The modulation of the elastic bandwidth needs to be implemented through a preset modulation algorithm. The algorithm embeds the threshold conditions and adjustment amplitude rules for bandwidth adjustment to ensure the standardization and repeatability of the modulation process. After modulation, the range of dynamic elastic bandwidth is expanded upwards and downwards from the dynamic boundary base value. The resulting upper and lower boundaries together constitute the elastic failure boundary, thus completing the establishment of the elastic failure boundary model. The state envelope surface defined by this model can adapt to changes in system state and disturbance characteristics in real time.

[0074] The determination of resilience challenge events also includes the following steps:

[0075] Step 24: Track the motion of ecological parameters in real time within the state envelope defined by the elastic failure boundary, and continuously calculate the trajectory dynamic vector. The trajectory dynamic vector includes pointing component, approaching component, and inertial component. The specific operation is as follows:

[0076] Based on continuous ecological parameter data collected through real-time monitoring, the trajectory of the parameters within the state envelope defined by the elastic failure boundary is tracked in real time. The trajectory tracking employs continuous point fitting, determining the instantaneous position and direction of motion of the trajectory by the parameter values ​​at adjacent moments, ensuring that the trajectory reflects the real-time dynamic changes of the parameters. During trajectory tracking, the trajectory dynamic vector is continuously calculated. This vector is the core parameter characterizing the trajectory's attitude and trend, containing three dimensions: pointing component, approaching component, and inertial component. The pointing component reflects the current direction of the trajectory relative to the elastic failure boundary, calculated by the angle between the instantaneous tangent direction of the trajectory and the normal direction of the boundary. It can characterize whether the parameters tend to approach the boundary; the approaching component reflects the rate characteristics of the trajectory approaching the elastic failure boundary, which is obtained by calculating the rate of change of the distance between the parameter and the boundary at adjacent time points, and can quantify the speed at which the parameter approaches the boundary; the inertial component reflects the inertial characteristics of the trajectory motion, which is derived from the acceleration change law of the parameter motion, and can characterize the continuity and stability of the trajectory motion. The solution of the trajectory dynamic vector needs to be implemented by the dynamic analysis algorithm. The algorithm is based on the classical kinematics principle and combines the time series data of the parameters to perform vector decomposition and calculation, ensuring that the vector solution result at each time point can accurately reflect the dynamic characteristics of the trajectory. The solution process is carried out in real time and synchronously.

[0077] Step 25: Perform joint analysis of the trajectory dynamic vector and the dynamic characteristics of the current elastic failure boundary, perform crossing risk prediction, and calculate the instantaneous crossing risk index. The specific operation is as follows:

[0078] First, the dynamic characteristic parameters of the elastic failure boundary at the current moment are extracted, including the current position of the boundary, the fluctuation rate, and the adjustment trend. These parameters are output in real time by the elastic failure boundary model, reflecting the dynamic change state of the boundary. The trajectory dynamic vector obtained in step 24 is jointly analyzed with the boundary dynamic characteristic parameters. The analysis focuses on the relative relationship between the pointing component and the boundary, the matching degree between the approaching component and the boundary fluctuation rate, and the influence of the inertial component on the trajectory motion trend. The possibility of the trajectory crossing the boundary is judged through comprehensive analysis. To accurately quantify the crossing risk, an instantaneous crossing risk index is introduced. The calculation of this index requires combining the vector components and the parameters of the boundary dynamic characteristics. The calculation formula is constructed based on the risk assessment logic. The derivation logic is: the stronger the trend of the trajectory approaching the boundary, the faster the approach rate, and the smaller the safety range caused by the boundary's own fluctuation, the higher the crossing risk. Based on this, each influencing factor is quantified and a linear coupling formula is constructed to intuitively reflect the risk level. The calculation formula of the instantaneous crossing risk index is:

[0079] ;

[0080] in The instantaneous crossing risk index at time t, with a value ranging from 0 to 1; For the component weighting coefficient, To approximate the component weighting coefficients, Let be the boundary fluctuation rate weighting coefficients, all three of which are non-negative real numbers and satisfy . The value conditions are based on the prediction model of resilient challenge events and the dynamic bandwidth state division of elastic failure boundaries, and are consistent with the response characteristics of river and lake ecosystems to different types of disturbances. The specific value conditions and corresponding values ​​are as follows:

[0081] (1) When the predicted resilience challenge event is a pulse impact mode, such as sudden pollution or extreme rainfall, the trajectory of ecological parameters tends to quickly point to and approach the elastic failure boundary, with the boundary fluctuation rate having a secondary impact. The value can be taken as follows: , , .

[0082] (2) When the predicted resilience challenge event is a sustained pressure pattern, such as long-term agricultural non-point source pollution or slow climate change, the sustained fluctuation of the elastic failure boundary has a more significant impact on the crossing risk, while the short-term impact of the trajectory approach rate is reduced. The value can be taken as follows: , , .

[0083] (3) When the elastic failure boundary is in a narrow bandwidth state, the system has weak anti-interference capability and low safety redundancy. It is necessary to focus on preventing and controlling the crossing risk caused by boundary fluctuations, and at the same time strictly control the trend of the trajectory pointing to the boundary. The value can be , , .

[0084] (4) When the elastic failure boundary is in a wide bandwidth state, the system has strong anti-interference ability and large safety redundancy. The crossing risk is mainly determined by the direction and approach characteristics of the trajectory itself, and the influence of boundary fluctuations can be weakened. The value can be: , , .

[0085] in addition, Let be the normalized value of the pointing component at time t. This refers to the maximum normalized value of the pointing component; The standardized value of the approaching component at time t. To approximate the maximum standardized value of the component; Let be the standardized value of the boundary fluctuation rate at time t. The instantaneous crossing risk index, calculated using the formula based on the maximum standardized value of the boundary fluctuation rate, can accurately quantify the risk level of the parameter trajectory crossing the elastic failure boundary at the current moment.

[0086] Step 26: Perform pattern recognition on the time series of the instantaneous traversal risk index. When a predefined pulse impact pattern or continuous pressure pattern is identified, it is determined to be a resilience challenge event, and the corresponding protocol sub-cluster is activated from the ecological response protocol cluster to initiate the competition process. The specific operation is as follows:

[0087] First, time-series data of the instantaneous crossing risk index are continuously collected to construct a risk index time-series sequence. The time interval of the time-series sequence is consistent with the parameter monitoring time interval to ensure a complete reflection of the risk index's changing trend. Based on a preset pattern recognition algorithm, pattern analysis is performed on the risk index time-series sequence. Pattern recognition compares the changing characteristics of the time-series sequence with the feature parameters of two predefined risk modes: a pulse impact mode and a continuous pressure mode. The pulse impact mode is characterized by the instantaneous crossing risk index rapidly climbing above a preset threshold within a short period of time and maintaining this level for a short duration, consistent with the impact characteristics of sudden disturbances on parameters. The continuous pressure mode is characterized by the instantaneous crossing... The risk index remains near or above a preset threshold and shows a slow upward trend, which is consistent with the characteristics of the impact of persistent disturbance on the parameter. In the pattern recognition process, it is necessary to set clear feature judgment thresholds, including risk index threshold, rise rate threshold, duration threshold, etc. By comparing the degree of matching between the time series and the feature thresholds, it is determined whether it conforms to the predefined pattern. When the risk index time series is identified as conforming to either the pulse impact pattern or the persistent pressure pattern, it is determined that a resilience challenge event has occurred. After the event is determined, the protocol sub-cluster corresponding to the equilibrium state and disturbance pattern of the ecological parameter is immediately selected from the ecological response protocol cluster, and the activation process of the protocol sub-cluster is initiated, while the protocol competition process is initiated.

[0088] In a preferred embodiment of the present invention, step 3 is further included: after multiple response protocols are triggered, based on the perturbation pattern represented by the perturbation memory energy site, multiple simulated perturbation time sequences of ecological parameter equilibrium states within a preset future time period are synthesized; all triggered candidate protocols and their combinations are simultaneously subjected to stress tests of the perturbation time sequences in a parallel virtual river and lake model, and the offset distance between the ecological parameter trajectory and the failure boundary of each candidate protocol or combination under each perturbation time sequence is evaluated; based on the evaluation results, the execution right is selected, and the specific operations are as follows:

[0089] After step 2 identifies the resilience challenge event and triggers multiple response protocols, the analysis is first conducted based on the perturbation memory energy field. The perturbation pattern characteristics represented by this energy field are extracted. These characteristics cover the intensity variation patterns, duration of impact, and evolution trends of historical and current perturbations, providing a reference for simulating future perturbation scenarios. Based on this perturbation pattern, combined with preset future time period parameters—which are determined according to the phased goals of river and lake ecological restoration, the perturbation response cycle, and actual management needs, typically including two dimensions: short-term (a few hours to a few days) and medium-term (a few weeks to a few months)—multiple simulated perturbation time series adapted to the ecological parameter equilibrium state are then synthesized. Each time series corresponds to… The system will assess a possible disturbance evolution scenario within a predetermined timeframe to ensure comprehensive coverage of all potential disturbances. Subsequently, a virtual river and lake model highly equivalent to the actual river and lake ecosystem will be constructed. This model must accurately reproduce the hydrological conditions, water quality characteristics, biological community structure, and various ecological processes of the actual river and lake, and support parallel operation of multiple replicas to improve testing efficiency. For each triggered candidate protocol and all possible valid combinations of these protocols, an independent virtual river and lake model replica will be created to ensure that the testing environments for each protocol and combination are independent and interference-free. The real-time values ​​of ecological parameters, the current parameters of the elastic failure boundary model, and the state of the disturbance memory energy field will be used as... A baseline snapshot is fully imported into each virtual river and lake model replica. Simultaneously, based on the mechanism of action of each candidate protocol, the preloaded impact of the protocol on the ecosystem is transformed into a quantified protocol preloaded impact field and input into the corresponding model replica. Specific examples of the protocol's preloaded impact on the ecosystem include adjustments to pollutant degradation coefficients corresponding to water purification protocols and flow velocity regulation parameters corresponding to hydrological control protocols. Multiple synthesized simulated perturbation time sequences are simultaneously applied to all virtual river and lake model replicas, initiating parallel stress testing. During the test, each candidate protocol and its combination dynamically responds to the perturbation time sequences according to its preset trigger conditions and action execution logic, recording the changes in ecological parameters over time in real time. After stress testing, for each candidate protocol and combination under each simulated disturbance time series, the positional relationship between the evolution trajectory of ecological parameters and the dynamic elastic failure boundary at the corresponding moment is compared one by one. The offset distance between the two is quantitatively evaluated. The evaluation of offset distance needs to take into account both the instantaneous deviation degree and the continuous deviation characteristics. Finally, based on the offset distance evaluation results under all simulated disturbance scenarios, combined with auxiliary evaluation indicators such as protocol execution cost, engineering feasibility, and ecological compatibility, a comprehensive scoring system is constructed to rank the comprehensive performance of each candidate protocol and combination. According to the ranking results, the protocol or protocol combination with the best comprehensive performance is selected and given actual execution rights.

[0090] The process of synthesizing the simulated perturbation time series of multiple ecological parameter equilibrium states within a preset future time period also includes the following steps:

[0091] Step 31: Analyze historical perturbation data, perform perturbation event slicing and mode primitive extraction, and construct a perturbation mode primitive library containing three dimensions: intensity envelope, spectral features, and covariance relationships. The specific operations are as follows:

[0092] First, long-term historical disturbance data for various ecological parameters of river and lake ecosystems are collected. This data needs to cover parameter change records corresponding to disturbance events of different types and intensities, including natural disturbances such as extreme rainfall and sudden temperature changes, and anthropogenic disturbances such as sudden pollution and water resource development. This ensures the completeness, diversity, and representativeness of the data. The collected historical disturbance data is preprocessed to remove outliers and complete the data, ensuring data quality. Then, a disturbance event slicing operation is performed. Taking a single complete disturbance event as the unit, the historical disturbance data is divided into independent disturbance event segments along the time dimension. Each segment must completely cover the entire process of the disturbance event's occurrence, development, decay, and termination. The segment boundary is determined by the parameter's stable state before the disturbance and the parameter's return to a stable state after the disturbance. For each segmented disturbance event segment, pattern primitive extraction is performed. Each disturbance event is extracted as a feature dimension, specifically including three dimensions: intensity envelope, spectral features, and covariance. The intensity envelope represents the boundary of the disturbance event's intensity change over time, obtained by fitting the upper and lower boundaries of the disturbance intensity time series curve, reflecting the peak and decay patterns of the disturbance intensity. The spectral features are obtained by performing a Fourier transform on the disturbance intensity time series data, representing the frequency distribution characteristics of the disturbance event, reflecting the periodicity and volatility of the disturbance. The covariance represents the interaction between the disturbance event and other related ecological parameter disturbances, obtained by calculating the correlation coefficient and lag correlation coefficient of different parameter disturbance time series. The three dimensions of features extracted from each disturbance event segment are integrated to form standardized disturbance model primitives. After all disturbance model primitives are collected, a disturbance model primitive library is constructed. The primitive library needs to establish an indexing mechanism to facilitate the rapid retrieval and retrieval of subsequent primitives.

[0093] Step 32: Based on the current perturbation memory energy field, elastic failure boundary, and resilience challenge event pattern, generate primitive selection and combination rules. The specific operations are as follows:

[0094] First, three types of input information are extracted at the current moment: the real-time intensity and distribution characteristics of the perturbation memory energy field; the current dynamic parameters of the elastic failure boundary model, including the dynamic boundary base value and dynamic elastic bandwidth; and the resilience challenge event mode determined in step 2, namely, the pulse impact mode or the continuous pressure mode. The three types of input information are then analyzed collaboratively to clarify the influence logic of each information on the perturbation time series synthesis: when the perturbation memory energy field intensity is high, it indicates that historical perturbations have a significant impact on the current system, and the future perturbation time series should prioritize incorporating high-intensity perturbation mode characteristics; when the current elastic bandwidth of the elastic failure boundary is narrow, it indicates that the system's current anti-interference capability is weak, and the future perturbation time series should also consider small-amplitude fluctuation scenarios; when the resilience challenge event is a pulse impact mode, the future perturbation time series should include short-term high-intensity perturbation characteristics, and when it is a continuous pressure mode, it should include long-term medium- and low-intensity perturbation characteristics. Based on the above... The analysis results generate primitive selection rules, clarifying the selection priority of various primitives in the primitive library. For example, for scenarios with pulse-impact type resilience challenge events and high perturbation memory energy field intensity, perturbation mode primitives with high intensity envelope peaks and short durations are prioritized; for scenarios with continuous compression type resilience challenge events and narrow elastic bandwidth, perturbation mode primitives with gentle intensity fluctuations and long durations are prioritized. Simultaneously, primitive combination rules are generated, clarifying the combination compatibility and order of different types of primitives. Combination compatibility must be determined based on the spectral characteristics and covariance relationships of the primitives to ensure that the combined perturbation modes conform to the actual perturbation evolution laws of river and lake ecosystems. The combination order must be combined with the occurrence logic of the perturbation event to ensure the rationality of the timing. Primitive selection and combination rules must be based on quantification thresholds. All rules must be standardized logically expressed to facilitate subsequent algorithmic implementation and invocation.

[0095] Step 33: According to the combination rules, multiple future perturbation time series are synthesized using the path-dependent Monte Carlo method. The specific operations are as follows:

[0096] First, the logic of path dependence must be clarified: the evolution trend of future disturbance time series must rely on historical disturbance trajectories and the current disturbance state, and cannot be randomly generated without considering the actual disturbance patterns of the system. This ensures the rationality and relevance of the synthesized time series. The application of the Monte Carlo method needs to be combined with this path dependence logic, generating multiple time series through multiple rounds of random sampling and combination operations. In specific operations, the primitive selection and combination rules generated in step 32 are used as constraints. Disturbance mode primitives that meet the rules are randomly selected from the disturbance mode primitive library constructed in step 31. The selection process needs to consider the primitive selection priority; primitives with higher priority have a correspondingly higher probability of being selected. The selected primitives are then processed according to the combination rules. The sequence combination process must ensure that the spectral characteristics and covariance relationships between the primitives match to avoid logically contradictory disturbance patterns. Each primitive extraction and combination generates a simulated disturbance time series for a preset future time period. The disturbance intensity value at each time node in the time series is calculated by interpolation of the combined primitive features. The extraction and combination process is repeated until a preset number of simulated disturbance time series are generated. The preset number needs to be determined based on the complexity of the disturbance scenario to ensure comprehensive coverage of all potential future disturbance situations, while also taking into account the efficiency of subsequent stress testing. Each synthesized simulated disturbance time series must be verified for rationality, and abnormal time series that do not conform to the system disturbance rules are removed.

[0097] The process of performing stress tests on perturbation timelines and evaluating the offset distance between the ecological parameter trajectories and failure boundaries for each candidate protocol or combination under each perturbation timeline also includes the following steps:

[0098] Step 34: Create an independent virtual river / lake model copy for each triggered candidate protocol. Use the state, including real-time values ​​of ecological parameters, elastic failure boundaries, and perturbation memory energy fields, as a baseline snapshot. Input the protocol preloaded influence field according to the mechanism of action of the candidate protocol. The specific operations are as follows:

[0099] First, a basic virtual river and lake model is constructed. This model must be based on the hydrological, water quality, biological, and geomorphological characteristics of the actual river and lake ecosystem. It uses numerical modeling methods to recreate key ecological processes within the system, such as material cycling, energy flow, and biological interactions. The model's parameter settings must be calibrated and matched with actual monitoring data to ensure the equivalence of the model to the actual system. For each candidate protocol triggered in step 2, and all possible valid combinations of these candidate protocols, combinations must be determined based on the compatibility of protocol effects, excluding conflicting protocol combinations. Independent copies of the basic virtual river and lake model are created, each allocated independent computing resources to avoid mutual interference during testing. A baseline snapshot is imported into each model copy. The baseline snapshot must fully record the system state information at the current moment, specifically including the actual values ​​of all ecological parameters. The system monitors real-time monitoring values, current dynamic parameters of the elastic failure boundary model (i.e., dynamic boundary base values ​​and dynamic elastic bandwidth), and real-time intensity and distribution data of the perturbation memory energy field. This ensures that the initial states of all model replicas are completely consistent and synchronized with the current state of the actual river and lake ecosystem. Subsequently, based on the mechanism of action of each candidate protocol or protocol combination, the system analyzes the mode and scope of the protocol's impact on the river and lake ecosystem. This impact is then transformed into a quantified protocol preloaded impact field. For example, for the aquatic plant remediation protocol, its preloaded impact field can be represented by the adjustment of the aquatic plant biomass growth rate parameter in the model; for the pollutant interception protocol, its preloaded impact field can be represented by the adjustment of the pollutant diffusion coefficient and degradation coefficient in the model. The protocol preloaded impact field is then accurately input into the corresponding model replica to complete the initialization settings of the model replica.

[0100] Step 35: Apply future perturbation time series in parallel to each virtual river and lake model replica, enabling the protocol to respond dynamically according to conditions and action modes, and record the evolution trajectory of ecological parameters. The specific operations are as follows:

[0101] First, a parallel computing framework is built, employing multi-threaded or distributed computing to achieve synchronous operation of all virtual river and lake model replicas. This ensures that each model replica undergoes perturbation testing at the same time dimension, guaranteeing the comparability of test results. Multiple simulated perturbation time sequences synthesized in step 33 are synchronously input into all model replicas. The application process must strictly adhere to the time nodes of the time sequence to ensure that each model replica receives the same perturbation intensity input at the same moment, achieving synchronous perturbation application. During stress testing, candidate protocols or protocol combinations within each model replica dynamically respond to the input perturbation time sequences according to their preset trigger conditions and action execution logic. For example, when the value of a certain ecological parameter in a model replica approaches the elastic failure boundary, the corresponding... The response protocol is automatically triggered and executed, and the parameters are regulated by adjusting the relevant ecological process parameters. During the test, the evolution trajectory of the target ecological parameters in each model replica is recorded in real time. The evolution trajectory must include the specific value of the parameter at each time point, as well as the slope and curvature of the parameter change over time, to fully reflect the change law of the parameter under the dual effects of disturbance and protocol response. At the same time, the response characteristic data such as the trigger time, execution duration, and execution intensity of each candidate protocol or combination are recorded. The runtime of the stress test is consistent with the future disturbance time series duration preset in step 33 to ensure that the entire disturbance response process within the preset future time period can be fully covered. After the test, the parameter evolution trajectory and response characteristic data of each model replica are sorted to form a standardized test dataset.

[0102] Step 36: Based on the evolution trajectory and the dynamic elastic failure boundary at the corresponding moment, calculate the offset distance spectrum. The offset distance spectrum includes the minimum offset distance, the duration of offset danger, and the offset trend integral. The specific operation is as follows:

[0103] First, extract the ecological parameter evolution trajectory of each model replica recorded in step 35, and the dynamic boundary parameters output by the elastic failure boundary model at the corresponding time point, namely the dynamic boundary base value and the dynamic elastic bandwidth. Ensure that the trajectory data and boundary parameters are fully aligned in the time dimension, with each time point having a corresponding parameter value and boundary range. For each candidate protocol or combination under each simulated perturbation time series, calculate the offset distance between the ecological parameter evolution trajectory and the corresponding dynamic elastic failure boundary. The offset distance is calculated based on the shortest distance between the parameter value and the elastic failure boundary. When the parameter value is inside the elastic failure boundary, i.e., within the safe range, the offset distance is positive, and the larger the value, the higher the safety redundancy. When the parameter value is outside the boundary, i.e., within the danger range, the offset distance is negative, and the larger the absolute value, the more severe the deviation. Based on the offset distance calculation results for the entire time period, construct an offset distance spectrum, which includes three quantization indicators. The parameters are: minimum offset distance, duration of offset danger, and offset trend integral. The minimum offset distance is the minimum offset distance across all time periods, reflecting the extreme state where the parameter is closest to or furthest from the boundary during the protocol response. The duration of offset danger is the cumulative time the offset distance is less than a preset safety threshold, reflecting the sustained characteristics of the dangerous state when the protocol responds to disturbances. The preset safety threshold is usually set to 0, meaning the parameter is outside the boundary. The offset trend integral is the integral value of the offset distance over time across all time periods, reflecting the overall offset level and trend of the parameter during the protocol response. A larger integral value indicates better overall security performance. The calculation of the offset trend integral requires an integral formula. The derivation logic of the formula is as follows: the offset trend integral needs to comprehensively reflect the magnitude and duration of the offset distance. By integrating the offset distance over time across all time periods, the overall offset level of the parameter can be intuitively quantified. The integral result can be directly used for horizontal comparison of the performance of different protocols. Its calculation formula is:

[0104] ;

[0105] Where S is the offset trend integral; This marks the start time of the stress test. The end time of the stress test is t; D(t) is the offset distance between the ecological parameter evolution trajectory and the dynamic elastic failure boundary at time t. By comprehensively calculating the above three indicators, a complete offset distance spectrum is formed to fully quantify the disturbance response performance of each candidate protocol and combination.

[0106] In some embodiments of the present invention, step 4 is further included: tracking the actual execution process and results of the protocol combination that has obtained the right to execute, calculating the elastic capital consumed in the response and the resulting ecological technology debt, and generating an accounting result that includes immediate benefits, elastic consumption, and technology debt. The specific operations are as follows:

[0107] After determining the optimal protocol combination and initiating its actual implementation in step 3, a dedicated tracking mechanism needs to be established to record the entire execution process of the protocol combination in real time, including the initiation time, execution intensity, operational steps, and key control parameters of each protocol. Simultaneously, data on the state of the river and lake ecosystems during and after the protocol execution should be collected to ensure the completeness and timeliness of the execution process and results data. The tracking scope should cover the entire execution cycle of the protocol combination, including both dynamic process data during execution and ecosystem recovery and evolution data after execution. The data collection frequency should be consistent with the previous monitoring frequency to ensure data continuity and comparability. Based on the process and results data obtained through tracking, two key accounting tasks should be carried out: accounting for the elastic capital consumed in the response and accounting for the ecological technology debt generated. Elastic capital consumption focuses on the inherent elastic resources of the ecosystem invested during the execution of the protocol combination to maintain the ecological parameter equilibrium and resist disturbances, such as the water body's own elasticity. The ecosystem possesses naturally occurring resilience resources such as net capacity, self-regulating capacity of biological communities, and shoreline ecological buffering function. Ecological technology debt accounting focuses on the potential lagged effects and restoration deficits that may arise in the long-term stable operation of the ecosystem after the implementation of the agreement combination. Economic input costs, such as equipment operation and maintenance fees and reagent costs, are non-elastic capital direct inputs and are only deducted from costs when calculating immediate benefits, and are not included in the scope of core resource consumption. The accounting process needs to combine the elastic failure boundary model parameters, ecological parameter equilibrium state standards, and disturbance memory energy field data from the previous steps to ensure the objectivity and accuracy of the accounting results. Finally, based on the elastic capital consumption and ecological technology debt data obtained from the accounting, combined with the improvement effect of the agreement implementation on the trajectory of challenged ecological parameters, the immediate benefits brought by the implementation of the agreement combination are calculated. The three are integrated to generate standardized accounting results, which need to clearly reflect the short-term benefits, core resource consumption, and long-term potential impacts of the agreement combination implementation.

[0108] The accounting for the elastic capital consumed and the resulting eco-technology debt includes:

[0109] Step 41: During and after the implementation of the protocol combination, the actual observed trajectories of ecological parameters are recorded synchronously, and the predicted trajectory of natural fluctuations for the same period is generated. The net effect of the intervention is obtained by comparison. The specific operation is as follows:

[0110] After the protocol suite is initiated, two trajectory recording tasks need to be carried out simultaneously. First, the actual observation trajectory of ecological parameters needs to be recorded. This involves continuously collecting values ​​of core and related ecological parameters affected by resilience challenge events using deployed real-time monitoring equipment, and organizing these values ​​into a time series to form the actual observation trajectory. The trajectory data must fully cover the pre-set observation period during and after the protocol execution. The pre-set observation period is determined based on the recovery characteristics of the ecological parameters and the duration of the protocol's impact, ensuring a complete capture of the parameter evolution trend after intervention. Second, the generation of a concurrent natural fluctuation baseline prediction trajectory needs to be performed. This trajectory represents the trend of natural fluctuation and recovery of ecological parameters over time without protocol suite intervention. The generation process needs to be based on historical monitoring data of ecological parameters before protocol execution, the characteristics of the perturbation memory energy field, and the parameters of the elastic failure boundary model. The prediction algorithm, which is derived from the previous virtual river and lake model, is used to ensure the consistency between the predicted trajectory and the natural evolution law of the actual ecosystem. After obtaining the actual observed trajectory and the predicted trajectory of the natural fluctuation baseline, the parameter difference between the two trajectories at the same time is calculated by comparing them point by point in time. This difference is the net intervention effect of the protocol combination. When the net intervention effect is positive, it indicates that the implementation of the protocol combination has a promoting effect on the recovery of ecological parameters. When it is negative, it indicates that the implementation of the protocol combination may have an adverse effect on parameter evolution. When it approaches zero, it indicates that the implementation of the protocol combination has no significant intervention effect on parameter evolution. The calculation of the net intervention effect needs to exclude the influence of new disturbances during the same period. If new disturbance events occur during the period, the two trajectories need to be corrected based on the disturbance memory energy field data to ensure that the net effect only reflects the intervention effect of the protocol combination.

[0111] Step 42: Based on the net effect of the intervention, calculate the state maintenance energy and the process acceleration energy, and sum them to obtain the elastic capital consumption. The specific operation is as follows:

[0112] Based on the net effect of intervention, the elastic capital consumed in the execution of the protocol combination is quantitatively calculated, i.e., the core resource consumption, clarifying the input of the ecosystem's inherent elastic resources. Further explanation is needed here: Elastic capital refers to the ecosystem's inherent or long-term evolved capacity for resistance to disturbance, self-repair, and state maintenance; it belongs to the ecosystem's inherent resources. Human economic expenditures, such as engineering construction costs and monitoring equipment costs, belong to non-elastic capital inputs. The calculation of elastic capital consumption uses the net effect of intervention as input. By decomposing the energy input type corresponding to the net effect of intervention, it is divided into two parts: state maintenance energy and process acceleration energy. The sum of these two parts is the total elastic capital consumption. State maintenance energy refers to the elastic capital consumed during the execution of the protocol combination to prevent ecological parameters from further deviating from equilibrium and to maintain parameters within a safe range inside the elastic failure boundary. For example, dissolved oxygen consumed by the water body itself to offset pollutant disturbances, i.e., the core indicator of self-purification capacity; and the nutrients consumed by aquatic plants to purify water quality. Its calculation is based on the component of the net effect of intervention used to offset disturbances and stabilize parameter states, obtained by multiplying this component by the elastic capital conversion coefficient. The elastic capital conversion coefficient... The elastic capital consumed per unit of intervention effect is determined through calibration using historical data. Process acceleration energy refers to the elastic capital consumed during the execution of the protocol combination to accelerate the recovery of ecological parameters from a deviated state to an equilibrium state and shorten the recovery cycle. For example, to accelerate water quality recovery, the self-reproductive energy consumed by the microbial community and the inherent purification substances released by the sediment are used. Its calculation is based on the component of the net intervention effect used to improve the parameter recovery rate. It is also obtained by multiplying this component by the elastic capital conversion coefficient, and the conversion coefficient is consistent with the coefficient used in the state maintenance energy calculation to ensure the uniformity of the calculation standard. After calculating the state maintenance energy and process acceleration energy separately, the energy loss of both during the execution process needs to be considered. The loss coefficient is determined according to the operation type of the protocol execution and the characteristics of the ecosystem. After deducting the loss, the two are added together to obtain the final accounting value of elastic capital consumption. The calculation of elastic capital consumption needs to be combined with the dynamic parameters of the preceding elastic failure boundary model. When the elastic failure boundary elastic bandwidth is narrow, the net intervention effect corresponding to the unit state maintenance energy consumption is stronger. The calculation process needs to be adapted to this dynamic characteristic to ensure that the accounting result matches the actual state of the system.

[0113] Step 43: After the protocol execution is completed, compare the actual recovery trajectory of the parameters with the expected natural recovery trajectory to identify hysteresis distortions and calculate the eco-technical debt value. The specific operations are as follows:

[0114] After the protocol suite is fully implemented, it is necessary to continuously track the recovery and evolution of the affected ecological parameters to form the actual recovery trajectory of the parameters. This trajectory must cover the entire period from the end of protocol implementation to the point where the parameters tend to stabilize, ensuring that the long-term trend of parameter recovery can be captured. Simultaneously, an expected natural recovery trajectory of the parameters should be constructed. This trajectory represents the ideal evolutionary trend of parameters recovering from a disturbed deviation state to an equilibrium state under conditions of no protocol suite intervention and relying solely on the ecosystem's self-regulation capacity. The generation of the expected natural recovery trajectory should be based on the parameter state before protocol implementation, the decay law of the disturbance memory energy field, the equilibrium state standard of the elastic failure boundary model, and the natural recovery data of ecological parameters after similar historical disturbance events, ensuring the rationality and representativeness of the trajectory. The actual recovery trajectory and the expected natural recovery trajectory should be compared throughout the entire period to identify deviations between the two. In this case study, the focus is on identifying characteristic segments where the actual recovery trajectory lags behind the expected natural recovery trajectory, or where the recovery state exhibits distortion. These characteristic segments are termed "hysteresis distortion." Identifying hysteresis distortion requires setting quantitative judgment criteria, including a lag time threshold and a distortion amplitude threshold. When the difference between the parameter value of the actual recovery trajectory at a certain moment and the expected trajectory exceeds the distortion amplitude threshold, and the duration exceeds the lag time threshold, it is determined that hysteresis distortion exists. Based on the identified hysteresis distortion characteristics, the eco-technical debt value is calculated. The calculation logic is a weighted product of the amplitude and duration of the hysteresis distortion. The weighting coefficients are determined based on the importance of the affected ecological parameters and their impact on the long-term stability of the system. Calibration is performed through historical data regression analysis. The final quantitative value obtained is the eco-technical debt value; a larger value indicates a more significant long-term potential impact from the implementation of the protocol.

[0115] Step 44: Generate a three-dimensional accounting table. The three-dimensional accounting table records the immediate benefits, resilient capital consumption, and ecological technology debt value calculated based on the improvement of the ecological parameter trajectory challenged in the resilience challenge event relative to its resilient failure boundary after the protocol is implemented. The specific operations are as follows:

[0116] By integrating accounting data, a standardized three-dimensional accounting table is generated, comprehensively presenting the benefits of the protocol combination execution, the consumption of core resources (i.e., elastic capital consumption), and the potential impact (i.e., ecological technology debt), while clarifying the accounting positioning of economic input costs. The construction of the three-dimensional accounting table is based on the improvement of the ecological parameter trajectory challenged in the resilience challenge event. First, the improvement of the core ecological parameter trajectory relative to its elastic failure boundary after protocol execution is calculated. The improvement is calculated by the difference between the actual parameter value at a certain moment after protocol execution and the parameter value at the same disturbance stage before execution. At the same time, the dynamic position of the elastic failure boundary is combined for correction to ensure that the improvement can truly reflect the effect of protocol execution on the parameter stability state. Based on this improvement, the immediate benefits of protocol combination execution are calculated. Immediate benefits are positively correlated with the improvement; the larger the improvement, the higher the immediate benefits. The economic costs of human input, i.e., the direct input cost of non-elastic capital, need to be deducted during the calculation. Costs such as equipment procurement and manual maintenance are excluded, rather than core resource consumption, i.e., flexible capital consumption, to ensure the accuracy of revenue accounting. Subsequently, the core resource consumption (i.e., flexible capital consumption) calculated in step 42 and the ecological technology debt value calculated in step 43 are integrated with the calculated immediate revenue to construct a three-dimensional accounting table. The three dimensions of this accounting table are immediate revenue, flexible capital consumption, and ecological technology debt. Each dimension corresponds to specific quantitative values, calculation basis, and time attributes. The time attributes must clearly indicate the stage of the corresponding agreement execution or the time node of the accounting. The three-dimensional accounting table must label the core parameters and algorithm basis used in the calculation process for each quantitative value, including elastic failure boundary model parameters, intervention net effect data, hysteresis distortion characteristic parameters, etc., to ensure the traceability and verifiability of the accounting table. The generated three-dimensional accounting table must clearly present the relationship between the three and clearly distinguish the difference between core resource consumption and economic input expenditure.

[0117] In some embodiments of the present invention, step 5 is further included: dynamically correcting the elastic failure boundary model based on the calculation results. The specific operation is as follows:

[0118] Based on the results of the three-dimensional accounting table generated in step 4, the resilient failure boundary model is dynamically corrected to adapt it to the actual evolutionary characteristics of the river and lake ecosystem and the feedback from protocol execution, thereby improving the accuracy of subsequent resilience challenge event judgment and response decisions. While the initial parameter settings of the resilient failure boundary model, which serves as the basis for resilience challenge event judgment and protocol selection in the preceding steps, incorporate historical data and theoretical analysis, deviations from the actual system state may occur after actual protocol execution due to factors such as ecosystem resilience resource consumption and exposure of long-term potential impacts. By analyzing the protocol execution benefits, core resource consumption, and long-term potential impact data contained in the three-dimensional accounting table, the matching differences between model parameters and actual execution effects are explored, and the direction for model parameter optimization is identified. The correction process focuses on the adjustable parameters of the resilient failure boundary model, using quantitative extraction to identify key factors affecting model adaptability, establishing the correlation logic between factors and model parameters, and then making targeted adjustments to the model parameters. After correction, the resilient failure boundary model can more accurately reflect the current ecosystem's resistance to interference, parameter stability range, and the long-term impact of historical disturbances, providing more reliable model support for subsequent rounds of protocol triggering, selection, and execution.

[0119] The dynamic correction of the elastic failure boundary model based on the calculation results also includes the following steps:

[0120] Step 51: Analyze the three-dimensional accounting table, establish the association mapping between protocol types and vulnerabilities, and determine the association between vulnerabilities and elastic failure boundaries. The specific operations are as follows:

[0121] First, a comprehensive review of the data in the three-dimensional accounting table is conducted, including the immediate benefits, elastic capital consumption, and ecological technology debt values ​​corresponding to different protocol types or combinations of protocol types, as well as information on the trajectory improvement and lag distortion characteristics of challenged ecological parameters after the implementation of each protocol type. Based on this data, a comparative analysis is performed on the performance of different protocol types in responding to similar resilience challenges, identifying protocol types with excessively high elastic capital consumption, significant accumulation of ecological technology debt, or immediate benefits that fall short of expectations. The ecological parameter regulation scenarios corresponding to these protocol types often have weak links in the ecosystem that are susceptible to disturbance, i.e., ecosystem vulnerabilities. Vulnerabilities are specifically manifested in the tendency for one or more ecological parameters to deviate from equilibrium, insufficient elastic capital reserves, or poor response to disturbances. For lagging system nodes, such as when a certain water purification protocol is implemented, although the ammonia nitrogen parameter meets the standard in the short term, the elastic capital consumption is too large and it is easy to deviate again in the future. The equilibrium node corresponding to the ammonia nitrogen parameter is the vulnerability. After identifying the vulnerability, we further analyze its correlation with the elastic failure boundary to determine whether the formation of the vulnerability is due to the unreasonable setting of the dynamic boundary base value of the elastic failure boundary, the elastic bandwidth being too wide or too narrow, or the boundary modulation logic not matching the actual system's anti-interference capability. For example, the parameter equilibrium state corresponding to the vulnerability may have an elastic failure boundary bandwidth that is too narrow, leading to frequent triggering of resilience challenge events, or a bandwidth that is too wide, leading to untimely disturbance response. Finally, we establish the corresponding correlation between the vulnerability and the core parameters of the elastic failure boundary to form a complete correlation mapping system.

[0122] Step 52: Based on the correlation mapping, extract the cost-efficiency ratio, debt lag index, and sensitivity changes of correlation parameters from the three-dimensional accounting table as quantitative influencing factors. The specific operations are as follows:

[0123] Based on the correlation mapping established in step 51, core factors capable of quantifying the impact of vulnerabilities are extracted from the three-dimensional accounting table. This provides quantifiable input for model correction. The extracted quantitative impact factors include the cost-efficiency ratio, the debt lag index, and the sensitivity change of correlation parameters. All three factors need to be quantified by combining the data from the three-dimensional accounting table with the characteristics of vulnerabilities to ensure that the factors can accurately reflect the matching deviation between the model parameters and the actual system. The cost-efficiency ratio is used to quantify the elastic capital consumed per unit of immediate benefit. Its derivation logic is that the ratio of benefit to consumption intuitively reflects the resource utilization efficiency of the agreement execution, thereby determining whether the elastic failure boundary is exposed due to unreasonable resource consumption estimates. The calculation formula is as follows: ,in For the cost-efficiency ratio, The values ​​of flexible capital consumption recorded in the three-dimensional accounting table. The immediate return value is used as the benchmark. The formula transforms resource utilization efficiency into a quantifiable indicator through the linear ratio of elastic capital consumption to immediate return. A higher value indicates that more elastic capital is consumed per unit of return, and the lower the resource utilization efficiency. The debt lag index is used to quantify the long-term impact of ecotechnology debt. Its derivation logic combines the ecotechnology debt value with the duration of debt impact to characterize the long-term potential risk of vulnerabilities. The calculation formula is as follows: ,in The debt lag index This refers to the value of eco-technology debt recorded in the three-dimensional accounting table. The duration of the ecosystem's recovery to a stable state after the agreement's implementation is considered. A larger value indicates a more significant long-term impact of ecological technology debt. The sensitivity change of the associated parameters is used to quantify the change in the response sensitivity between the ecological parameters corresponding to the vulnerability and the associated ecological parameters. Its extraction requires comparing the slope difference of the associated ecological parameters with the core parameters before and after the agreement's implementation. The magnitude of the slope change is the sensitivity change value of the associated parameters. If the magnitude of the change exceeds the preset range, it indicates that the weight setting of the multi-factor coupling rules in the resilient failure boundary model may be unreasonable, leading to the exposure of the vulnerability. The extraction process of the three factors must be based on the three-dimensional accounting table data, combined with the correlation mapping between the vulnerability and the resilient failure boundary, to ensure that the factors can accurately reflect the needs of model correction.

[0124] Step 53: Based on the quantified influence factor, correct the modulation coefficients used to generate the dynamic elastic bandwidth, the decay time constant of the perturbation memory energy field, and the weights of the multi-factor coupling rules in the elastic failure boundary model. The specific operations are as follows:

[0125] The model parameters that need to be corrected include the modulation coefficient used to generate the dynamic elastic bandwidth, the decay time constant of the perturbation memory energy field, and the weights of the multi-factor coupling rules. These three parameters correspond to the bandwidth control of the elastic failure boundary, the weight of historical perturbation influence, and the logic of multi-factor synergy, respectively, and all affect the model's adaptability. The correction of the modulation coefficient for the dynamic elastic bandwidth is based on the cost-efficiency ratio. If the cost-efficiency ratio is too high, it indicates excessive elastic capital consumption, often due to excessively narrow elastic bandwidth leading to frequent protocol triggering and excessive resource consumption. In this case, the modulation coefficient needs to be increased to increase the dynamic elastic bandwidth and reserve more room for parameter fluctuations. If the cost-efficiency ratio is too low and the immediate return is stable, it indicates that the elastic bandwidth may be too wide, leading to a lag in the perturbation response. In this case, the modulation coefficient needs to be decreased to narrow the elastic bandwidth and improve response accuracy. The correction of the decay time constant of the perturbation memory energy field is based on the debt lag index. If the debt lag index is too high, it indicates that the long-term impact of historical perturbations has not been fully considered, often due to an excessively small decay time constant causing rapid decay of historical perturbation energy. In the process of reducing the sensitivity of the model, the decay time constant needs to be increased to extend the duration of the influence of historical disturbances, so that the model can fully take into account the long-term effects of historical disturbances. If the debt hysteresis index is low and the system recovers and stabilizes, the decay time constant can be appropriately reduced to improve the model's response sensitivity to recent disturbances. The weight correction of the multi-factor coupling rule is based on the change in the sensitivity of the associated parameters. If the change in the sensitivity of the associated parameters exceeds the preset range, it indicates that the weight of the associated parameters in the current coupling rule does not match the actual system response. In this case, the weight needs to be adjusted according to the direction of the change in sensitivity. For example, if the sensitivity of the associated parameters increases significantly, its weight in the multi-factor coupling operation needs to be increased, and vice versa, to ensure that the result of the multi-factor coupling operation can accurately reflect the dynamic characteristics of the actual system. During the correction process, multiple iterations of calibration are required. The corrected parameters are substituted into the elastic failure boundary model, and the boundary range output by the model is compared with the actual system state data until the model can accurately adapt to the current ecosystem's anti-interference ability and parameter stability requirements, thus completing the dynamic correction of the model.

[0126] In some embodiments of the present invention, step 6 is further included: periodically integrating the execution records of steps 1 to 5 to generate a resilience evolution map; based on the resilience evolution map, generating a new response protocol and adding it to the ecological response protocol cluster, the specific operations of which are as follows:

[0127] Fixed periodic execution nodes are set, with the cycle length determined based on the phased goals of river and lake ecological restoration, the frequency of disturbances, and the system evolution rate, typically ranging from 3 to 6 months. This ensures both sufficient execution data accumulation and timely response to system changes. At each cycle node, all execution records from steps 1 to 5 are comprehensively extracted, including ecological parameter equilibrium state deconstruction data, elastic failure boundary model parameters and correction records, resilience challenge event judgment records, protocol screening and execution trajectories, and three-dimensional accounting results, ensuring data coverage of the entire chain of decision-making, execution, accounting, and correction. Based on these... By recording data, integrating patterns, mining patterns, and extracting rules, a resilience evolution map is generated. This map visually presents the evolutionary trend, turning points, and influencing factors of river and lake ecosystem resilience over time. Subsequently, based on the system resilience shortcomings and evolutionary problems identified in the map, new response protocols are generated and added to the ecological response protocol cluster, realizing the dynamic expansion and optimization of the protocol library. Through this periodic optimization process, the entire intelligent decision-making method can continuously adapt to the long-term evolutionary characteristics of river and lake ecosystems, continuously improve its ability to respond to various disturbances, and ensure the sustainable progress of ecological restoration work.

[0128] Step 6 further includes the following steps:

[0129] Step 61: Periodically extract the execution records from steps 1 to 5, perform timeline alignment and state space reconstruction, identify turning points, extract patterns from the trajectory segments between turning points, and generate a structured resilience evolution knowledge network. The specific operations are as follows:

[0130] First, according to a preset cycle, all execution records from steps 1 to 5 are extracted in batches from the system database. The extraction scope includes, but is not limited to: the ecological parameter equilibrium state deconstruction results and protocol orchestration records of step 1; the construction and correction parameters of the elastic failure boundary model of step 2; the determination basis and results of resilience challenge events; the disturbance time series synthetic data of step 3; the protocol stress test results and execution right attribution records; the execution tracking data of step 4; the complete information of the three-dimensional accounting table; and the model correction parameters and process records of step 5. After extraction, the data needs to be preprocessed to unify the data format and timestamp. Missing data is filled in using interpolation based on historical similar scenarios, and abnormal data is corrected based on the characteristics of the disturbance memory energy field and the elastic failure boundary parameters to ensure the integrity and consistency of the data. Subsequently, a timeline alignment operation is performed. Using a unified time scale as a benchmark, the execution records of each step are linked and integrated in chronological order, enabling data from different stages to match each other in the time dimension. Based on this, state space reconstruction is carried out. Key variables such as core ecological parameter equilibrium indicators, elastic capital consumption values, ecological technology debt values, and dynamic parameters of elastic failure boundaries are selected as state dimensions to construct a multi-dimensional state space. The integrated data is mapped into this space to form a continuous evolutionary trajectory. Inflection points are identified through trajectory analysis. The determination of inflection points is based on abrupt changes in state variables, switching of protocol execution types, occurrence of resilience challenge events, or correction of model parameters. When the change in state variables at a certain moment exceeds a preset threshold, or when the aforementioned key events occur, it is determined to be a turning point. Pattern extraction is performed on the trajectory segments between adjacent turning points to extract the evolutionary characteristics of each trajectory segment, such as parameter recovery rate, elastic capital consumption trend, and boundary fluctuation pattern. Trajectory segments with similar characteristics are classified into standardized pattern units. Finally, all pattern units, turning point information, and the correlation relationships of data in each stage are integrated to construct a structured resilience evolutionary knowledge network. Each node in the network corresponds to a pattern unit or key event, and the connections between nodes represent evolutionary logic and correlation relationships, realizing the structured and knowledge-based transformation of execution records.

[0131] Step 62: Based on the structured resilience evolution knowledge network, perform pattern diagnosis and root cause correlation analysis to obtain deterioration trajectory patterns and root cause hypotheses. The specific operations are as follows:

[0132] First, a systematic pattern diagnosis is conducted on all pattern units in the structured resilience evolution knowledge network. This diagnosis process requires consideration of the ideal resilience evolution trajectory for river and lake ecological restoration. This ideal trajectory is pre-constructed based on ecological parameter equilibrium standards, the safety range of the elastic failure boundary, and the optimal resource consumption pattern. Each trajectory segment pattern unit is then compared and analyzed with the corresponding pattern on the ideal trajectory. Through this comparison, patterns deviating from the ideal trajectory and exhibiting a negative evolutionary trend are identified—these are deteriorating trajectory patterns. Typical characteristics of these patterns include lagging ecological parameter recovery, continuously increasing elastic capital consumption, accelerated accumulation of ecological technology debt, frequent resilience challenge events, or diminishing response effectiveness. For each identified deteriorating trajectory pattern, root cause correlation analysis is conducted. This analysis relies on the relationships between nodes in the knowledge network to trace the formation process of the pattern. Relevant factors include the type and effect of protocol execution in the preceding steps, the parameter settings of the elastic failure boundary model, the characteristics of the disturbance mode, and the initial state of ecological parameters. Root causes are located through correlation analysis and causal inference. For example, if a certain deterioration trajectory pattern is characterized by increasing elastic capital consumption and slow parameter recovery, correlation analysis reveals that this pattern corresponds to the execution of a specific type of protocol, and the excessively narrow elastic failure boundary bandwidth leads to frequent protocol triggering. Therefore, it can be preliminarily inferred that the root cause is insufficient resource utilization efficiency of this type of protocol and unreasonable boundary parameter settings. All located root causes are compiled into root cause hypotheses. Root cause hypotheses must be verifiable, clearly defining the causal logic between the cause and the deterioration trajectory pattern. Each hypothesis must be supported by specific data in the knowledge network to avoid unfounded speculation and ensure the objectivity and accuracy of root cause analysis.

[0133] Step 63: Based on the deterioration trajectory pattern and root cause hypothesis, generate strategic-level protocols through reverse design and input them into the ecological response protocol cluster. The specific operations are as follows:

[0134] The reverse design process is root cause hypothesis-driven, deriving intervention measures and implementation plans that can avoid or resolve the root causes and improve the trajectory of deterioration. The resulting strategic-level protocols differ from existing protocols in the preceding protocol suite, focusing more on long-term prevention and systemic improvement targeting the root causes, rather than simply responding to disturbances. In the specific design process, for each root cause hypothesis, the core objectives that the protocol must achieve are clearly defined, such as reducing elastic capital consumption, reducing the accumulation of ecological technology debt, optimizing parameter recovery efficiency, or adjusting the adaptability of elastic failure boundaries. Combining ecological parameter equilibrium standards, the characteristics of elastic failure boundary models, and the laws of disturbance memory energy fields, the execution logic, triggering conditions, control parameters, and operational procedures of the protocol are designed. For example, for a certain type of protocol with insufficient resource utilization efficiency and excessively narrow boundary bandwidth as a root cause hypothesis, the designed strategic-level protocol may include optimized control parameters, such as increasing unit elastic capital. The intervention effect; the dynamic triggering conditions are to adjust the triggering timing in combination with the boundary fluctuation law, as well as the supporting boundary adaptation rules, such as synchronously optimizing the elastic bandwidth modulation logic. After the protocol design is completed, it needs to be verified through a virtual river and lake model. The new protocol is substituted into the scenario corresponding to the historical deterioration trajectory pattern to simulate its execution effect and verify whether the protocol can effectively improve the deterioration trend and solve the root cause. If the verification fails, the protocol design parameters are adjusted until the effect requirements are met. After the strategic-level protocol is verified, it needs to be classified and coded according to the classification standard and indexing rules of the ecological response protocol cluster and added to the protocol cluster. At the same time, the retrieval and calling mechanism of the protocol cluster is updated to ensure that the new protocol can be accurately screened and called in the subsequent protocol activation and orchestration process in step 1, so as to realize the dynamic optimization and expansion of the ecological response protocol cluster and continuously improve the adaptability and effectiveness of the entire intelligent decision-making method.

[0135] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and its improved concepts, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, characterized in that, include: Step 1: Deconstruct the target steady state in the river and lake ecosystem into multiple ecological parameter equilibrium states, and pre-set a corresponding response protocol for each ecological parameter equilibrium state. The response protocol clarifies the coping measures, execution process, triggering conditions and expected effects when the parameter equilibrium state deviates. The collection of all response protocols constitutes an ecological response protocol cluster. Based on the real-time status of the river and lake ecosystem, activate and arrange the protocol execution sequence from the ecological response protocol cluster. Step 2: Construct a resilient failure boundary model for each ecological parameter equilibrium state. The resilient failure boundary model is a dynamic state envelope surface, which is dynamically adjusted according to the changes in the state of the river and lake ecosystem and the influence of historical disturbances. Monitor the position and trajectory of the ecological parameter in the state envelope surface. When the trajectory touches and attempts to cross the current resilient failure boundary with a preset angle and acceleration, it is determined as a resilience challenge event, triggering the protocol sub-cluster in the ecological response protocol cluster corresponding to the parameter equilibrium state and disturbance type, and starting the activation process of the protocol sub-cluster. Step 3: After multiple response protocols are triggered, based on the perturbation pattern represented by the perturbation memory energy site, synthesize multiple simulated perturbation time sequences of ecological parameter equilibrium states within a preset future time period; enable all triggered candidate protocols and their combinations to simultaneously perform stress tests on the perturbation time sequences in a parallel virtual river and lake model, and evaluate the offset distance between the ecological parameter trajectory and the failure boundary of each candidate protocol or combination under each perturbation time sequence; based on the evaluation results, select the protocol or protocol combination with the best overall performance and grant it actual execution rights; The simulated perturbation time series of multiple ecological parameter equilibrium states over a future preset period are synthesized, including: Historical disturbance data is split into independent disturbance event segments according to the time dimension. For each segment of disturbance event, pattern primitive extraction is performed to extract the feature dimensions of each disturbance event, including three dimensions: intensity envelope, spectral features, and covariance relationship. The intensity envelope characterizes the boundary of the change in the intensity of the disturbance event over time, the spectral features characterize the frequency distribution characteristics of the disturbance event, and the covariance relationship characterizes the mutual influence of the disturbance event with other related ecological parameter disturbances. The three-dimensional features extracted from each disturbance event segment are integrated to form standardized disturbance model primitives, and a disturbance model primitive library is constructed. Based on the current perturbation memory energy field, elastic failure boundary, and resilience challenge event pattern, generate primitive selection and combination rules; Based on the combination rules, multiple future perturbation time series are synthesized using the path-dependent Monte Carlo method.

2. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... Also includes: Step 4: Track the actual execution process and results of the combination of agreements that have obtained the right to execute, calculate the elastic capital consumed in the response and the ecological technology debt generated, and generate the accounting results including immediate benefits, elastic consumption and technology debt; Step 5: Dynamically revise the elastic failure boundary model based on the calculation results; Step 6: Periodically integrate the execution records of steps 1 to 5 to generate a resilience evolution map; based on the resilience evolution map, generate new response protocols and add them to the ecological response protocol cluster.

3. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... The target steady state in river and lake ecosystems is deconstructed into multiple ecological parameter equilibrium states, including: Step 11: Define the target steady state as a complex consisting of structural steady state, process steady state, and functional steady state; Step 12: Based on structural steady state, process steady state and functional steady state, perform control factor mapping for each dimension to identify the corresponding ecological parameters for each dimension; Step 13: Define the conditional domains of the ecological parameters under multiple different baseline scenarios.

4. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... Construct an elastic failure boundary model for the equilibrium state of each ecological parameter, including: Step 21: Extract perturbation imprints from the time series of each ecological parameter, identify deviation events that exceed their conditional state domain, quantify the deviation magnitude, duration, and recovery rate of each deviation event; convert each deviation event into a perturbation energy unit with decay characteristics, and superimpose all historical perturbation energy units to form a perturbation memory energy field. Step 22: Based on the real-time status of the perturbation memory energy field, seasonal phase signal, and associated ecological parameters, perform multi-factor coupling operation to calculate the dynamic boundary base value; Step 23: Based on the dynamic boundary baseline value, elastic bandwidth modulation is performed according to the state of the perturbation memory energy field and associated ecological parameters to generate an elastic failure boundary jointly defined by the dynamic boundary baseline value and the dynamic elastic bandwidth.

5. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... The determination of resilience challenge events includes: Step 24: Track the movement of ecological parameters in real time within the state envelope defined by the elastic failure boundary, and continuously calculate the trajectory dynamic vector, which includes the pointing component, the approaching component, and the inertial component. Step 25: Perform joint analysis of the trajectory dynamic vector and the dynamic characteristics of the current elastic failure boundary, perform crossing risk prediction, and calculate the instantaneous crossing risk index; Step 26: Perform pattern recognition on the time series of instantaneous traversal risk index. When a pulse impact pattern or continuous pressure pattern that matches the predefined pattern is identified, it is determined to be a resilience challenge event, and the corresponding protocol sub-cluster is activated from the ecological response protocol cluster to start the competition process.

6. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... Perform stress tests on the perturbation timings and evaluate the offset distance between the ecological parameter trajectories and the failure boundary for each candidate protocol or combination under each perturbation timing, including: Step 34: Create an independent virtual river and lake model copy for each triggered candidate protocol, using the state including real-time values ​​of ecological parameters, elastic failure boundary and disturbance memory energy field as a baseline snapshot, and input the protocol preloaded influence field according to the mechanism of action of the candidate protocol. Step 35: Apply future perturbation timings in parallel to each virtual river and lake model replica, so that the protocol responds dynamically according to conditions and action modes, and records the evolution trajectory of ecological parameters. Step 36: Calculate the offset distance spectrum based on the evolution trajectory and the dynamic elastic failure boundary at the corresponding time. The offset distance spectrum includes the minimum offset distance, the duration of offset danger, and the offset trend integral.

7. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... Track the actual execution process and results of the portfolio of agreements for which enforcement rights have been obtained, and calculate the elastic capital consumed and the resulting eco-technology debt, including: Step 41: During and after the implementation of the protocol combination, the actual observation trajectory of ecological parameters is recorded synchronously, and the baseline prediction trajectory of natural fluctuations during the same period is generated. The net effect of intervention is obtained by comparison. Step 42: Based on the net effect of the intervention, calculate the state maintenance energy and the process acceleration energy, and sum them up to obtain the elastic capital consumption; Step 43: After the agreement is executed, compare the actual recovery trajectory of the parameters with the expected natural recovery trajectory to identify hysteresis distortions and calculate the value of eco-technical debt. Step 44: Generate a three-dimensional accounting table. The three-dimensional accounting table records the immediate benefits, resilient capital consumption, and ecological technology debt value calculated based on the improvement of the ecological parameter trajectory challenged in the resilience challenge event relative to its resilient failure boundary after the protocol is implemented.

8. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... Based on the calculation results, the elastic failure boundary model is dynamically revised, including: Step 51: Analyze the three-dimensional accounting table, establish the association mapping between protocol type and vulnerability, and determine the association between vulnerability and elastic failure boundary; Step 52: Based on the correlation mapping, extract the cost efficiency ratio, debt lag index and sensitivity changes of correlation parameters from the three-dimensional accounting table as quantitative influencing factors. Step 53: Based on the quantified influence factor, correct the modulation coefficient used to generate the dynamic elastic bandwidth, the decay time constant of the perturbation memory energy field, and the weights of the multi-factor coupling rule in the elastic failure boundary model.

9. The intelligent decision-making method for river and lake health restoration schemes based on the theory of gradual ecological restoration, as described in claim 1, is characterized in that... Generate a resilience evolution map, and based on the resilience evolution map, generate new response protocols and add them to the ecological response protocol cluster, including: Step 61: Periodically extract the execution records from Step 1 to Step 5, perform time axis alignment and state space reconstruction, identify turning points and extract patterns from the trajectory segments between turning points to generate a structured resilience evolution knowledge network. Step 62: Based on the structured resilience evolution knowledge network, perform pattern diagnosis and root cause correlation analysis to obtain the deterioration trajectory pattern and root cause hypothesis; Step 63: Based on the deterioration trajectory pattern and root cause hypothesis, generate strategic-level protocols through reverse design and input them into the ecological response protocol cluster.

Images