Multi-objective optimization method for fish habitat restoration project based on reservoir operation constraints

Abstract

This invention discloses a multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints. It relates to the field of habitat restoration technology and addresses the technical problems of existing technologies, such as poor climate adaptability, reliance on experience, unsustainability in mountainous rivers, conflicts with reservoir scheduling, insufficient connectivity, difficulty in ensuring long-term restoration effects, and difficulty in significantly improving the accuracy of fish habitat restoration while ensuring the safe and efficient operation of reservoirs. The method employs CMIP6 climate downscaling hydrological prediction, embedding rigid constraints of reservoir scheduling, quantitative modeling of habitat suitability, collaborative evaluation of vertical and horizontal connectivity, dynamic coupling of sediment-landform habitats, and the NSGA-III algorithm for solving multi-objective optimization and closed-loop monitoring and evaluation. Combining the ecological habits of target fish species, such as temperature orientation, migration, reproduction, and foraging, a habitat hydraulic index-habitat suitability response model is constructed to assess the impact of key factors such as flow velocity, water depth, water level, and water temperature on the habitat.

Description

Technical Field

[0001] This invention belongs to the field of habitat restoration, specifically a multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints. Background Technology

[0002] A multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints is a method that comprehensively considers the impact of reservoir scheduling on fish habitats. By optimizing scheduling strategies, it aims to achieve synergistic optimization of socio-economic goals such as flood control, power generation, and water supply, as well as multiple objectives including fish habitat restoration and ecological protection. Reservoir scheduling constraints refer to a series of conditions that a reservoir must meet during operation, including physical constraints such as water level limits, flow requirements, discharge capacity, and reservoir capacity curves, as well as socio-economic functional requirements such as flood control, power generation, and water supply. These constraints directly affect reservoir scheduling decisions, thereby impacting the ecological environment of downstream fish habitats.

[0003] Existing fish habitat restoration projects rely solely on historical hydrological data, failing to account for future climate change and extreme weather. This results in extremely poor adaptability to scenarios such as drought, floods, sudden water level changes, and increased flow velocities, leading to project failure and short lifespan. Ecological restoration is disconnected from reservoir operation; restoration plans often exceed the reservoir's actual operational capacity, making implementation impossible. Furthermore, they fail to provide stable habitats for fish in areas with large water level differences and unstable flow. Relying on experience-based judgments and lacking quantitative assessments of hydraulic indicators such as flow velocity, water depth, and water temperature, they cannot accurately match the actual needs of fish for temperature regulation, migration, spawning, and juvenile rearing, resulting in poor targeting and limited effectiveness. Optimizing only local habitat areas while ignoring overall river connectivity prevents fish from completing their full migration and floodplain utilization, hindering the true restoration of ecosystem structure and function. Existing schemes also cannot predict riverbed erosion and deposition, particularly in mountainous rivers, where restoration projects are easily eroded, deposited, or structurally unstable, leading to rapid attenuation of restoration effects and even secondary damage, resulting in extremely poor sustainability. Summary of the Invention

[0004] This invention aims to solve at least one of the technical problems existing in the prior art. To this end, this invention proposes a multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints. This method addresses the technical problems of existing technologies, such as poor climate adaptability, reliance on experience, unsustainable mountain rivers, conflicts with reservoir scheduling, insufficient connectivity, difficulty in ensuring long-term restoration effects, and difficulty in significantly improving the accuracy of fish habitat restoration while ensuring the safe and efficient operation of reservoirs.

[0005] To address the aforementioned problems, the first aspect of this invention provides a multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints, comprising the following steps: Weather forecasts are made by combining the CMIP6 climate model with regional meteorological observation data, and the data is then downscaled and converted into watershed hydrological parameters. The rigid constraints of reservoir scheduling rules are clarified, and regional hydrological situation simulation is conducted based on the watershed hydrological parameters and rigid constraints to obtain the hydrological situation simulation results. Based on the ecological habits of the target fish species and combined with the rigid constraints of reservoir scheduling rules, a habitat hydraulic index habitat suitability response model is constructed to analyze habitat hydraulic suitability and establish connectivity evaluation indicators between longitudinal migration channels and transverse floodplain wetlands. The rigid constraints of reservoir scheduling rules are used as input boundary conditions to construct a coupled model of sediment transport-riverbed geomorphology-fish habitat suitability. Based on the coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability, connectivity evaluation index, and habitat hydraulic suitability, combined with the characteristics of mountain rivers, a multi-objective optimization model is constructed. The multi-objective optimization model is solved by the NSGA-III improved non-dominated sorting genetic algorithm to obtain the Pareto optimal solution set and determine the optimal restoration scheme. Establish a long-term dynamic monitoring system for the restoration area, compare the collected hydrological data with the simulation results of the coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, calculate the deviation, and analyze the comprehensive evaluation index to assess the restoration effect.

[0006] Optionally, in one example of the above aspects, weather forecasts are made by combining the CMIP6 climate model with regional meteorological observation data, and then converted into watershed hydrological parameters through downscaling, including the following steps: Using the CMIP6 climate model and combined with regional meteorological observation data, we predict extreme weather data for a predetermined time period in the future. By downscaling, we can analyze the extreme precipitation intensity of extreme weather events and predict runoff during drought periods. The extreme precipitation intensity and drought runoff forecasts obtained from extreme weather events are converted into watershed hydrological parameters.

[0007] Optionally, in one example of the above aspects, the rigid constraints of the reservoir scheduling rules are explicitly defined. Based on the watershed hydrological parameters and the rigid constraints, a regional hydrological situation simulation is performed to obtain the hydrological situation simulation results, including the following steps: Clearly define the rigid constraints of reservoir scheduling rules, including: water level constraints, discharge constraints, and water supply constraints; The water depth H(x,t) and flow velocity U(x,t) in region x at time t are obtained as the results of the hydrological situation simulation.

[0008] Optionally, in one example of the above aspects, based on the ecological habits of the target fish species and combined with the rigid constraints of reservoir scheduling rules, a habitat hydraulic index habitat suitability response model is constructed to analyze habitat hydraulic suitability, including the following steps: Based on the ecological habits of the target fish species, combined with the rigid constraints of reservoir scheduling rules, and the results of hydrological situation simulation, including water depth H(x,t) in region x at time t and flow velocity U(x,t) in region x at time t, a habitat suitability response model based on habitat hydraulic indicators is constructed. For the velocity suitability function, depth suitability function, and temperature suitability function, velocity, depth, and temperature are used as hydrological parameters, and the same type of Logistic function is set. Perform fitting.

[0009] Optionally, in one example of the above aspects, establishing a connectivity evaluation index between longitudinal migration channels and transverse floodplain wetlands includes the following steps: Based on the habitat suitability response model of habitat hydraulic indicators, the hydraulic suitability of habitat is analyzed. Combining the connectivity between longitudinal migration channels and transverse floodplain wetlands, an evaluation index for the connectivity between longitudinal migration channels and transverse floodplain wetlands is constructed. The connectivity co-optimization objective is set as follows: maximizing the vertical connectivity index (LCI) and the horizontal connectivity index (TCI).

[0010] Optionally, in one example of the above aspects, the rigid constraints of the reservoir scheduling rules are used as input boundary conditions to construct a coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, including the following steps: Using the rigid constraints of reservoir scheduling rules as input boundary conditions, a coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability is constructed to predict changes in habitat quality.

[0011] Optionally, in one example of the above aspects, based on a coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, connectivity evaluation indicators, and habitat hydraulic suitability, combined with the characteristics of mountain rivers, a multi-objective optimization model is constructed, including the following steps: Based on a coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability, connectivity evaluation indicators, and habitat hydraulic suitability, a multi-objective optimization model is constructed by embedding rigid and flexible constraints, including: To maximize ecological benefits, calculate the ecological benefit analysis values; The objective is to minimize project costs; therefore, the project cost analysis values ​​are calculated. To achieve the goal of maximizing restoration sustainability, calculate restoration sustainability analysis values.

[0012] Optionally, in one example of the above aspects, rigid constraints and flexible constraints include: Rigid constraints are set as rigid constraints for reservoir operation; flexible constraints include gradient constraints, meandering constraints, engineering construction constraints, and ecological constraints.

[0013] Optionally, in one example of the above aspects, the multi-objective optimization model is solved using the NSGA-III improved non-dominated sorting genetic algorithm to obtain the Pareto optimal solution set and determine the optimal repair scheme, including the following steps: An improved non-dominated sorting genetic algorithm using NSGA-III, incorporating both rigid and flexible constraints, is used to solve a multi-objective optimization model, including: Initialize the population, generate the initial population for the repair plan, and determine the population size; A fitness assessment is conducted, and the analytical values ​​of the ecological benefit maximization objective, engineering cost minimization objective, and restoration sustainability maximization objective for each individual are calculated. Combining rigid and flexible constraints, infeasible solutions are eliminated, and feasible solutions are retained. The feasible solutions are sorted in a non-dominated manner and divided into Pareto levels. The convergence speed and diversity of the algorithm are balanced by a dynamic adjustment mechanism of crossover probability and mutation probability. When the number of iterations reaches the preset number or the optimal solution tends to stabilize, the iteration is terminated, the Pareto optimal solution set is output, and the repair scheme corresponding to the Pareto optimal solution set is sent to experts, who will select the optimal repair scheme based on actual needs.

[0014] Optionally, in one example of the above aspects, based on the hydrological situation simulation results, the adaptive adjustment coefficient is analyzed according to the dynamic changes in water level, flow velocity, and water depth, and the comprehensive evaluation index is analyzed to evaluate the restoration effect, including the following steps: Based on the results of hydrological situation simulation, adaptive adjustment coefficients are analyzed according to the dynamic changes in water level, flow velocity, and water depth. An annual assessment of the restoration effect is conducted, and a comprehensive evaluation index is analyzed based on ecological benefits, engineering feasibility, and sustainability. If the adaptive adjustment coefficient is less than 0.6 or the comprehensive evaluation index is less than 0.6, a warning for the repair plan will be sent to the experts; otherwise, the original repair plan will continue to be implemented.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention utilizes the CMIP6 climate model, regional meteorological observations, and statistical downscaling to transform future climate change into reliable watershed hydrological parameters. It addresses the root cause of traditional restoration methods that rely solely on current hydrological data and are unable to cope with future droughts, torrential rains, and drastic changes in water level and flow velocity. This ensures that habitat restoration projects can continue to function stably under future conditions, preventing projects from becoming ineffective immediately after construction and avoiding resource waste. It establishes rigid constraints and uses them as mandatory boundary conditions for hydrological simulation, habitat assessment, and project optimization, ensuring that ecological restoration does not compromise the safety of reservoir operations. This resolves the conflict between traditional ecological restoration and reservoir management, and the resulting impracticality of such projects.

[0016] This invention combines the ecological habits of target fish species, such as temperature orientation, migration, reproduction, and foraging, to construct a habitat hydraulic index-habitat suitability response model. This model enables a quantitative and refined assessment of the impact of key factors such as flow velocity, water depth, water level, and water temperature on habitats, eliminating reliance on qualitative experience and significantly improving the accuracy of restoration plans. Simultaneously, it constructs evaluation indicators for vertical and horizontal connectivity, overcoming the bottleneck of traditional restoration methods that only focus on local habitat area while neglecting the overall connectivity of rivers, truly meeting the needs of fish throughout their entire life cycle.

[0017] This invention dynamically couples sediment transport, riverbed erosion and deposition, geomorphological evolution, and habitat suitability, enabling long-term prediction of the impact of river topography changes on habitats. It is particularly well-suited to the characteristics of mountain rivers with steep gradients, high meandering, and intense water and sediment movement, ensuring the long-term stability of restoration projects and preventing erosion or siltation damage. Coupled with multi-dimensional objectives such as habitat suitability, engineering costs, and reservoir benefit losses, it employs the NSGA-III improved non-dominated sorting genetic algorithm to solve the multi-objective optimization model for global optimization, obtaining a Pareto optimal solution set. Based on decision preferences, it can output multiple optimal restoration schemes, such as ecological priority, economic priority, and balanced development, demonstrating strong engineering practicality. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 The first aspect of this invention provides a multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints, comprising the following steps: Weather forecasts are made by combining the CMIP6 climate model with regional meteorological observation data, and the data is then downscaled and converted into watershed hydrological parameters. The rigid constraints of reservoir scheduling rules are clarified, and regional hydrological situation simulation is conducted based on the watershed hydrological parameters and rigid constraints to obtain the hydrological situation simulation results. Based on the ecological habits of the target fish species and combined with the rigid constraints of reservoir scheduling rules, a habitat hydraulic index habitat suitability response model is constructed to analyze the habitat hydraulic suitability and establish connectivity evaluation indicators between longitudinal migration channels and transverse floodplain wetlands. The rigid constraints of reservoir scheduling rules are used as input boundary conditions to construct a coupled model of sediment transport-riverbed geomorphology-fish habitat suitability. Based on the coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability, connectivity evaluation index, and habitat hydraulic suitability, combined with the characteristics of mountain rivers, a multi-objective optimization model is constructed. The multi-objective optimization model is solved by an improved NSGA-III non-dominated sorting genetic algorithm to obtain the Pareto optimal solution set and determine the optimal restoration scheme. Establish a long-term dynamic monitoring system for the restoration area, compare the collected hydrological data with the simulation results of the coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, calculate the deviation, and analyze the comprehensive evaluation index to assess the restoration effect.

[0022] Specifically, in this embodiment, the future climate change is transformed into reliable watershed hydrological parameters by using the CMIP6 climate model, regional meteorological observation, and statistical downscaling. This addresses the problem that traditional restoration methods rely solely on current hydrology and cannot cope with future droughts, rainstorms, and drastic changes in water level and flow velocity. This ensures that habitat restoration projects can still function stably in the future environment and avoids the project becoming ineffective and resources being wasted.

[0023] Clearly define rigid constraints and use them as mandatory boundary conditions for hydrological simulation, habitat assessment, and engineering optimization to ensure that ecological restoration does not exceed the bottom line of reservoir safety operation, and to solve the problems of conflict between traditional ecological restoration and reservoir scheduling, and the inability of projects to be implemented.

[0024] By combining the ecological habits of target fish species, such as temperature orientation, migration, reproduction, and foraging, a habitat hydraulic index-habitat suitability response model is constructed. This model enables quantitative and refined assessment of the impact of key factors such as flow velocity, water depth, water level, and water temperature on habitats, eliminating reliance on qualitative experience and significantly improving the accuracy of restoration plans.

[0025] Simultaneously construct evaluation indicators for vertical and horizontal connectivity to break through the bottleneck of traditional restoration that only focuses on the area of ​​local habitats and ignores the overall connectivity function of rivers, and truly meet the needs of fish throughout their entire life cycle.

[0026] By dynamically coupling sediment transport, riverbed erosion and deposition, geomorphological evolution, and habitat suitability, this method enables long-term prediction of the impact of river topography changes on habitats. It is particularly well-suited to the characteristics of mountain rivers—large gradients, high meandering, and intense water and sediment transport—ensuring the long-term stability of restoration projects and preventing erosion or siltation damage. Coupled with multi-dimensional objectives such as habitat suitability, engineering costs, and reservoir benefit losses, the improved NSGA-III non-dominated sorting genetic algorithm is used to solve the multi-objective optimization model for global optimization, obtaining a Pareto optimal solution set. Based on decision preferences, it can output multiple optimal restoration schemes, such as ecological priority, economic priority, and balanced development, demonstrating strong engineering practicality.

[0027] In one embodiment of the present invention, weather forecasting is performed by combining the CMIP6 climate model with regional meteorological observation data, and then converted into watershed hydrological parameters through downscaling, including the following steps: Using the CMIP6 climate model and combined with regional meteorological observation data, we predict extreme weather data for a predetermined time period in the future. By downscaling, we can analyze the extreme precipitation intensity of extreme weather events and predict runoff during drought periods. The formula for predicting extreme precipitation intensity is as follows: in, Let t be the extreme precipitation intensity. The historical baseline precipitation intensity is given by λ, which is the temperature sensitivity coefficient. In this embodiment, the temperature sensitivity coefficient λ = 0.025. ΔT(t) is the change in temperature at time t, and η(t) is the extreme weather probability coefficient. In this embodiment, 0 ≤ η ≤ 1.2, which is output by the CMIP6 model. The formula for predicting runoff during dry periods is as follows: in, For runoff at time t during the dry season, The average runoff over many years, kd is the drought attenuation coefficient. In this embodiment, the drought attenuation coefficient kd = 0.03, D(t) is the number of days of drought, and γ is the reservoir scheduling runoff regulation coefficient. In this embodiment, 0.6 ≤ γ ≤ 1.0, which is determined by the reservoir scheduling rules. The extreme precipitation intensity and drought runoff forecasts obtained from extreme weather events are converted into watershed hydrological parameters.

[0028] In one embodiment of the present invention, the rigid constraints of the reservoir scheduling rules are clearly defined. Based on the watershed hydrological parameters and the rigid constraints, a regional hydrological situation simulation is performed to obtain the hydrological situation simulation results, including the following steps: Clearly define the rigid constraints of reservoir scheduling rules, including: water level constraints, discharge constraints, and water supply constraints; The water level constraint is: Zmin≤Z(t)≤Zmax, where Z(t) is the actual water level of the reservoir at time t, Zmin is the minimum water level allowed by the reservoir, which may cause the water intake facilities to fail (such as pumps running dry), water quality to deteriorate (sediment disturbance) or affect power generation efficiency (insufficient head); Zmax is the maximum water level set by the reservoir during the flood season to reserve flood control capacity, which may cause the risk of dam failure or flooding of upstream areas. The discharge constraint is: Qmin≤Qds(t)≤Qmax, where Qmin is the ecological base flow, Qmax is the maximum discharge, and Qds(t) is the discharge flow of the reservoir at time t. The water supply constraint is: Qg(t)≥Qxu(t), where Qg(t) is the actual water supply provided by the reservoir to the coastal cities, agriculture or industry at time t, and Qxu(t) is the water demand for production and life along the coast at time t. Based on the predicted watershed hydrological parameters, the regional hydrological situation is simulated using the following formula: Where H(x,t) is the water depth in region x at time t. Let x be the elevation difference in region x, U(x,t) be the flow velocity in region x at time t, n be the Manning coefficient, A(x,t) be the cross-sectional area of ​​the water passage, and R(x,t) be the hydraulic radius. The water depth H(x,t) and flow velocity U(x,t) in region x at time t are obtained as the results of the hydrological situation simulation.

[0029] In one embodiment of the present invention, based on the ecological habits of the target fish species and combined with the rigid constraints of reservoir scheduling rules, a habitat hydraulic index habitat suitability response model is constructed to analyze the habitat hydraulic suitability, including the following steps: Based on the ecological habits of the target fish species, combined with the rigid constraints of reservoir scheduling rules, and the hydrological situation simulation results including water depth H(x,t) and flow velocity U(x,t) in region x at time t, a habitat suitability response model based on habitat hydraulic indicators is constructed, with the following formula: in, Let x be the hydraulic suitability index of the habitat at time t. Let U(x,t) be the flow velocity suitability function, representing the suitability of the flow velocity U(x,t) in region x at time t for the target fish species. Let H(x,t) be the water depth suitability function, representing the suitability of the water depth H(x,t) in region x at time t for the target fish species. Let T(x,t) be the water temperature suitability function, representing the suitability of the water temperature T(x,t) in region x at time t for the target fish species. Let C(x,t) be the dissolved oxygen suitability function, representing the suitability of dissolved oxygen C(x,t) in region x at time t for the target fish. , , and In this embodiment, the weighting coefficient is used. =0.35, =0.3, =0.2, =0.15, corresponding to flow velocity, water depth, water temperature, and dissolved oxygen, respectively; , , , Fit using the Logistic function; For the velocity suitability function, depth suitability function, and temperature suitability function, velocity, depth, and temperature are used as hydrological parameters, and the same type of Logistic function is set. The fitting process is performed using the following formula: in, Let Gmin be the velocity suitability function of region x at time t, the depth suitability function of region x, or the temperature suitability function of region x at time t, and Gmin be the target fish species and... The minimum suitable hydrological parameter corresponding to the parameter type, Gmax is the target fish species and The maximum suitable hydrological parameter corresponding to the parameter type, Gopt, is the target fish species and The optimal hydrological parameter corresponding to the parameter type, k is the fitting coefficient of the Logistic function. In this embodiment, k=0.8, which is used to adjust the steepness of the Logistic function to ensure that the evaluation results are more consistent with the actual situation; The dissolved oxygen suitability function is set as follows: Wherein, Cmin is the minimum suitable dissolved oxygen concentration for the target fish; Copt is the optimal dissolved oxygen concentration for the target fish.

[0030] In one embodiment of the present invention, establishing a connectivity evaluation index between longitudinal migration channels and transverse floodplain wetlands includes the following steps: Based on the habitat suitability response model using habitat hydraulic indicators, habitat hydraulic suitability is analyzed. Combining the connectivity between longitudinal migration channels and transverse floodplain wetlands, a connectivity evaluation index for longitudinal migration channels and transverse floodplain wetlands is constructed. The formulas include: Formula for calculating the vertical connectivity index: Wherein, LCI is the vertical connectivity index. Let be the habitat hydraulic suitability index for the i-th migratory channel. Let be the length of the i-th migratory channel. Let be the obstruction degree of the i-th migratory channel, i∈(1,2,…,ns), and ns be the total number of migratory channels; The resistance of the i-th migratory channel The calculation formula is: In the formula, n represents the number of physical obstacles in the i-th migratory channel. Let ik be the actual height of the i-th obstacle. Fi represents the maximum possible height at the location of the ikth obstacle, determined based on the height of similar obstacles in historical data; m represents the number of fish that successfully pass through the i-th migratory channel within a specific time period; M represents the total number of fish that attempt to pass through the i-th migratory channel within the same time period; by calculating the proportion of fish that successfully pass through, and then subtracting this proportion from 1, we obtain the fish passage factor. This factor directly reflects the actual difficulty for fish to pass through this segment of the channel. The lower the proportion, the greater the obstacle, and the closer the Fi value is to 1. and For the corresponding weighting coefficients, in this embodiment, =0.75, =0.25.

[0031] Formula for calculating lateral connectivity index: Wherein, TCI is the lateral connectivity index. Let be the area of ​​the j-th floodplain wetland. This represents the habitat hydraulic suitability index for the corresponding j-th floodplain wetland. Let be the total floodplain area of ​​the restoration area, j∈(1,2,…,ms), where ms is the total number of floodplain wetlands; The connectivity co-optimization objective is set as follows: maximizing the vertical connectivity index (LCI) and the horizontal connectivity index (TCI).

[0032] In one embodiment of the present invention, the rigid constraints of the reservoir scheduling rules are used as input boundary conditions to construct a coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, including the following steps: Using the rigid constraints of reservoir operation rules as input boundary conditions, a coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability is constructed to predict habitat quality changes. The formula is as follows: in, The habitat suitability index after T years. Let Z(t) be the increase in riverbed elevation within year T, and Z(t) be the actual water level at time t. This represents the increase in sediment concentration within year T. Let be the sediment concentration at time t. The influence coefficient of landform. In this embodiment, the sediment influence coefficient is used. =0.6, =0.4, enabling accurate prediction of long-term habitat stability.

[0033] In one embodiment of the present invention, a multi-objective optimization model is constructed based on a coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, connectivity evaluation indicators, and habitat hydraulic suitability, combined with the characteristics of mountain rivers, including the following steps: Based on a coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability, connectivity evaluation indicators, and habitat hydraulic suitability, a multi-objective optimization model is constructed by embedding rigid and flexible constraints, including: To maximize ecological benefits, calculate the ecological benefit analysis values: in, This represents the value of the ecological benefit analysis. Let A be the hydraulic suitability index of the habitat in region x at time t, and let A be the total area of ​​the area to be restored. The objective is to minimize project costs; therefore, the project cost analysis values ​​are calculated as follows: in, This represents the cost analysis value of the project. , , , The unit costs are respectively for ecological bank protection, fishway construction, dredging, and wetland restoration. For the area of ​​the revetment construction, The length of the fishway. To dredge the silt volume, The area of ​​wetland restoration; To maximize the goal of restoration sustainability, calculate the restoration sustainability analysis value: in, This indicates the value of the restoration sustainability analysis. The habitat suitability index is set as T years from now, and the predicted restoration period is set to T years.

[0034] In one embodiment of the present invention, the rigid constraint and the flexible constraint include: Rigid constraints are set as rigid constraints for reservoir operation; flexible constraints include gradient constraints, meandering constraints, engineering construction constraints, and ecological constraints. Among them, the gradient constraint is: Jfmin≤Jf≤Jfmax, where Jf is the riverbed gradient of the area to be repaired, Jfmin is the minimum riverbed gradient of the area to be repaired, and Jfmax is the maximum riverbed gradient of the area to be repaired. Mooring constraint: W≥Wmin, where W is the river's mooring degree and Wmin is the preset minimum river mooring degree; Construction constraints: Afx≤Aab, Tsg≤Tlm, where Aab is the area available for restoration, Tlm is the construction period limit, Afx is the planned area for ecological restoration, and Tsg is the actual construction period; Ecological constraints: HSI(x,t)≥0.5 (core habitat), Qr(t)≥Qlm, where Qr(t) is the ecological flow at time t, and Qlm is the minimum flow required to maintain the basic ecological functions of the river.

[0035] In one embodiment of the present invention, a multi-objective optimization model is solved using the NSGA-III improved non-dominated sorting genetic algorithm to obtain a Pareto optimal solution set and determine the optimal repair scheme, including the following steps: This paper improves the non-dominated sorting genetic algorithm using NSGA-III by introducing adaptive crossover and mutation operators to enhance solution efficiency and optimal solution quality. It also embeds rigid and flexible constraints to solve multi-objective optimization models, including: Initialize the population, generate the initial population for the restoration plan, such as the combination of parameters like the location of the bank protection, the length of the fishway, and the scope of dredging, and determine the population size; A fitness assessment is conducted, and the analytical values ​​of the ecological benefit maximization objective, engineering cost minimization objective, and restoration sustainability maximization objective for each individual are calculated. Combining rigid and flexible constraints, infeasible solutions are eliminated, and feasible solutions are retained. Perform non-dominated sorting on feasible solutions, classify Pareto levels, and calculate crowding degree; The crossover probability Pc is calculated as follows: Pc = 0.7 + 0.2 * ranki / rankmax, and the mutation probability Pm = 0.05 + 0.05 * 1 / crowdi, where ranki is the individual Pareto rank, rankmax is the individual maximum Pareto rank, and crowdi is the crowding level. When the number of iterations reaches the preset number or the optimal solution tends to stabilize, the iteration is terminated, the Pareto optimal solution set is output, and the repair scheme corresponding to the Pareto optimal solution set is sent to experts, who will select the optimal repair scheme based on actual needs.

[0036] In this embodiment, the parameters in the optimal repair scheme include: Riverbank protection construction area ( This parameter determines the scale of the ecological revetment project. Different solutions may correspond to different revetment construction areas. The size of the area will affect the project cost and ecological benefits. A larger revetment area may provide better ecological protection, but it will also increase the project cost.

[0037] Fishway length ( Fishway length is a key factor influencing fish migration. Longer fishways may be more conducive to fish migration, improve habitat connectivity, and thus enhance ecological benefits, but they also increase engineering costs. In the Pareto optimal solution set, the appropriate fishway length is determined based on different objective balances.

[0038] Dredging volume ( The dredging volume reflects the scale of the dredging project. A reasonable dredging volume can improve the riverbed gradient and hydraulic conditions, significantly impacting ecological benefits and the sustainability of restoration. Different solutions will yield different dredging volumes, seeking an optimal balance among multiple objectives.

[0039] wetland restoration area ( The area of ​​wetland restoration directly affects the ecological function of wetlands. A larger wetland restoration area helps improve ecological benefits, such as improving water quality and providing habitats for organisms, but it also increases engineering costs. The solution set determines the appropriate wetland restoration area based on the trade-offs of the objective function.

[0040] Riverbed gradient J(x): Under the premise of satisfying the gradient constraint (Jmin≤J(x)≤Jmax), different solutions may correspond to different riverbed gradient distributions. A suitable riverbed gradient can adapt to the large gradient characteristics of mountain rivers and ensure the natural flow characteristics and ecological functions of the river.

[0041] River meandering degree W: To satisfy the meandering degree constraint (W≥Wmin), the solution set will include restoration schemes with different meandering degrees. Maintaining a certain river meandering degree helps to maintain the natural form of the river, provides a suitable living environment for fish and other organisms, and has a positive effect on ecological benefits and the sustainability of restoration.

[0042] The planned ecological restoration area, Afx, is constrained by engineering construction constraints (Afx ≤ Aab). Different solutions will determine different planned restoration areas to strike a balance between engineering construction feasibility and multiple objectives.

[0043] The actual construction period Tsg must meet the construction period limit (Tsg≤Tlm). The solution set will determine a suitable construction period based on the construction conditions and target requirements to ensure that the project can be completed on time without affecting the achievement of other objectives.

[0044] In one embodiment of the present invention, based on the results of hydrological situation simulation, an adaptive adjustment coefficient is analyzed according to the dynamic changes in water level, flow velocity, and water depth, and a comprehensive evaluation index is analyzed to evaluate the restoration effect, including the following steps: Based on the hydrological situation simulation results, and according to the dynamic changes in water level, flow velocity, and water depth, the adaptive adjustment coefficient is analyzed, and the formula is: in, The adaptive adjustment coefficient for the repair scheme of region x at time t (0≤ ≤1), The optimal flow velocity for the target fish species in region x. Let U(x,t) be the optimal water depth for the target fish species in region x, U(x,t) be the flow velocity in region x at time t, and H(x,t) be the water depth in region x to be restored at time t. Annual assessments of restoration effectiveness are conducted, analyzing a comprehensive evaluation index based on ecological benefits, engineering feasibility, and sustainability. Where E is the comprehensive evaluation index. To maximize project cost; like If E < 0.6 or E < 0.6, send a warning about the repair plan to the experts; otherwise, continue with the original repair plan.

[0045] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints, characterized in that, Includes the following steps: Weather forecasts are made by combining the CMIP6 climate model with regional meteorological observation data, and the data is then downscaled and converted into watershed hydrological parameters. The rigid constraints of reservoir scheduling rules are defined, and regional hydrological situation simulations are conducted based on the watershed hydrological parameters and rigid constraints to obtain the hydrological situation simulation results. Based on the ecological habits of the target fish species and combined with the rigid constraints of reservoir scheduling rules, a habitat hydraulic index habitat suitability response model is constructed to analyze the habitat hydraulic suitability and establish connectivity evaluation indicators between longitudinal migration channels and transverse floodplain wetlands. The rigid constraints of reservoir scheduling rules are used as input boundary conditions to construct a coupled model of sediment transport-riverbed geomorphology-fish habitat suitability. Based on the coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, connectivity evaluation index, and habitat hydraulic suitability, a multi-objective optimization model is constructed in combination with the characteristics of mountain rivers. The multi-objective optimization model was solved by using the NSGA-III improved non-dominated sorting genetic algorithm to obtain the Pareto optimal solution set and determine the optimal repair scheme. Establish a long-term dynamic monitoring system for the restoration area, compare the collected hydrological data with the simulation results of the sediment transport-riverbed geomorphology-fish habitat suitability coupling model, calculate the deviation, and analyze the comprehensive evaluation index to assess the restoration effect.

2. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 1, characterized in that, Weather forecasts are made using the CMIP6 climate model combined with regional meteorological observation data, and then downscaled to convert the data into watershed hydrological parameters. The process includes the following steps: Using the CMIP6 climate model and combined with regional meteorological observation data, we predict extreme weather data for a predetermined time period in the future. By downscaling, we can analyze the extreme precipitation intensity of extreme weather events and predict runoff during drought periods. The formula for predicting extreme precipitation intensity is as follows: in, Let t be the extreme precipitation intensity. λ is the historical baseline precipitation intensity, λ is the temperature sensitivity coefficient, ΔT(t) is the temperature change at time t, and η(t) is the extreme weather probability coefficient. The formula for predicting runoff during dry periods is as follows: in, This represents the runoff at time t during the dry season. γ is the multi-year average runoff, kd is the drought attenuation coefficient, D(t) is the number of drought days, and γ is the reservoir scheduling runoff regulation coefficient. The extreme precipitation intensity and drought runoff forecasts obtained from extreme weather events are converted into watershed hydrological parameters.

3. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 1, characterized in that, Define the rigid constraints of the reservoir scheduling rules, and based on the watershed hydrological parameters and rigid constraints, conduct regional hydrological situation simulation to obtain the hydrological situation simulation results, including the following steps: Clearly define the rigid constraints of reservoir scheduling rules, including: water level constraints, discharge constraints, and water supply constraints; The water level constraint is: Zmin≤Z(t)≤Zmax, where Z(t) is the actual water level of the reservoir at time t, Zmin is the minimum water level allowed by the reservoir, and Zmax is the maximum water level set by the reservoir during the flood season to reserve flood control capacity. The discharge constraint is: Qmin≤Qds(t)≤Qmax, where Qmin is the ecological base flow, Qmax is the maximum discharge, and Qds(t) is the discharge flow of the reservoir at time t. The water supply constraint is: Qg(t)≥Qxu(t), where Qg(t) is the actual water supply provided by the reservoir to the coastal cities, agriculture or industry at time t, and Qxu(t) is the water demand for production and life along the coast at time t. Based on the predicted watershed hydrological parameters, the regional hydrological situation is simulated using the following formula: Where H(x,t) is the water depth in region x at time t. Let x be the elevation difference in region x, U(x,t) be the flow velocity in region x at time t, n be the Manning coefficient, A(x,t) be the cross-sectional area of ​​the water passage, and R(x,t) be the hydraulic radius. The water depth H(x,t) and flow velocity U(x,t) in region x at time t are obtained as the results of the hydrological situation simulation.

4. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 1, characterized in that, Based on the ecological habits of the target fish species and the rigid constraints of reservoir scheduling rules, a habitat hydraulic index-based habitat suitability response model is constructed to analyze habitat hydraulic suitability, including the following steps: Based on the ecological habits of the target fish species, combined with the rigid constraints of reservoir scheduling rules, and the hydrological situation simulation results including water depth H(x,t) and flow velocity U(x,t) in region x at time t, a habitat suitability response model based on habitat hydraulic indicators is constructed, with the following formula: in, Let x be the hydraulic suitability index of the habitat at time t. Let U(x,t) be the flow velocity suitability function, representing the suitability of the flow velocity U(x,t) in region x at time t for the target fish species. Let H(x,t) be the water depth suitability function, representing the suitability of the water depth H(x,t) in region x at time t for the target fish species. Let T(x,t) be the water temperature suitability function, representing the suitability of the water temperature T(x,t) in region x at time t for the target fish species. Let C(x,t) be the dissolved oxygen suitability function, representing the suitability of dissolved oxygen C(x,t) in region x at time t for the target fish. , , and These are the weighting coefficients; For the velocity suitability function, depth suitability function, and temperature suitability function, velocity, depth, and temperature are used as hydrological parameters, and the same type of Logistic function is set. The fitting process is performed using the following formula: in, Let Gmin be the velocity suitability function of region x at time t, the depth suitability function of region x, or the temperature suitability function of region x at time t, and Gmin be the target fish species and... The minimum suitable hydrological parameter corresponding to the parameter type, Gmax is the target fish species and The maximum suitable hydrological parameter corresponding to the parameter type, Gopt, is the target fish species and The optimal hydrological parameters corresponding to the parameter type, where k is the fitting coefficient of the Logistic function; The dissolved oxygen suitability function is set as follows: Wherein, Cmin is the minimum suitable dissolved oxygen concentration for the target fish; Copt is the optimal dissolved oxygen concentration for the target fish.

5. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 1, characterized in that, Establish connectivity evaluation indicators between longitudinal migration channels and transverse floodplain wetlands, including the following steps: Based on the habitat suitability response model using habitat hydraulic indicators, habitat hydraulic suitability is analyzed. Combining the connectivity between longitudinal migration channels and transverse floodplain wetlands, a connectivity evaluation index for longitudinal migration channels and transverse floodplain wetlands is constructed. The formulas include: Formula for calculating the vertical connectivity index: Wherein, LCI is the vertical connectivity index. Let be the habitat hydraulic suitability index for the i-th migratory passage. Let be the length of the i-th migratory channel. Let be the obstruction degree of the i-th migratory channel, i∈(1,2,…,ns), and ns be the total number of migratory channels; The resistance of the i-th migratory channel The calculation formula is: In the formula, n represents the number of physical obstacles in the i-th migratory channel. Let be the actual height of the ikth obstacle. is the maximum height that the location of the ikth obstacle can reach, which is determined based on the height of the location of similar obstacles in historical data; m is the number of fish that successfully pass through the i-th migratory channel in a specific time period, and M is the total number of fish that attempt to pass through the i-th migratory channel in the same time period. and These are the corresponding weighting coefficients; Formula for calculating lateral connectivity index: Wherein, TCI is the lateral connectivity index. Let be the area of ​​the j-th floodplain wetland. This represents the habitat hydraulic suitability index for the corresponding j-th floodplain wetland. Let be the total floodplain area of ​​the restoration area, j∈(1,2,…,ms), where ms is the total number of floodplain wetlands; The connectivity co-optimization objective is set as follows: maximizing the vertical connectivity index (LCI) and the horizontal connectivity index (TCI).

6. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 1, characterized in that, Using the rigid constraints of reservoir operation rules as input boundary conditions, a coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability is constructed, including the following steps: Using the rigid constraints of reservoir operation rules as input boundary conditions, a coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability is constructed to predict habitat quality changes. The formula is as follows: in, The habitat suitability index after T years. Let Z(t) be the increase in riverbed elevation within year T, and Z(t) be the actual water level at time t. This represents the increase in sediment concentration within year T. Let be the sediment concentration at time t. The influence coefficient of landform. The sediment impact coefficient.

7. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 1, characterized in that, Based on the coupled model of sediment transport-riverbed geomorphology-fish habitat suitability, connectivity evaluation indicators, and habitat hydraulic suitability, combined with the characteristics of mountain rivers, a multi-objective optimization model is constructed, including the following steps: Based on a coupled model of sediment transport, riverbed geomorphology, and fish habitat suitability, connectivity evaluation indicators, and habitat hydraulic suitability, a multi-objective optimization model is constructed by embedding rigid and flexible constraints, including: To maximize ecological benefits, calculate the ecological benefit analysis values: in, This represents the value of the ecological benefit analysis. Let A be the hydraulic suitability index of the habitat in region x at time t, and let A be the total area of ​​the area to be restored. The objective is to minimize project costs; therefore, the project cost analysis values ​​are calculated as follows: in, This represents the cost analysis value of the project. , , , The unit costs are respectively for ecological bank protection, fishway construction, dredging, and wetland restoration. For the area of ​​the revetment construction, The length of the fishway. To dredge the silt volume, The area of ​​wetland restoration; To maximize the goal of restoration sustainability, calculate the restoration sustainability analysis value: in, This indicates the value of the restoration sustainability analysis. The habitat suitability index is set as T years from now, and the predicted restoration period is set to T years.

8. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 7, characterized in that, Rigid constraints and flexible constraints include: Rigid constraints are set as rigid constraints for reservoir operation; flexible constraints include gradient constraints, meandering constraints, engineering construction constraints, and ecological constraints. Among them, the gradient constraint is: Jfmin≤Jf≤Jfmax, where Jf is the riverbed gradient of the area to be repaired, Jfmin is the minimum riverbed gradient of the area to be repaired, and Jfmax is the maximum riverbed gradient of the area to be repaired. Mooring constraint: W≥Wmin, where W is the river's mooring degree and Wmin is the preset minimum river mooring degree; Construction constraints: Afx≤Aab, Tsg≤Tlm, where Aab is the area available for restoration, Tlm is the construction period limit, Afx is the planned area for ecological restoration, and Tsg is the actual construction period; Ecological constraints: HSI(x,t)≥0.5, Qr(t)≥Qlm, where Qr(t) is the ecological flow at time t, and Qlm is the minimum flow required to maintain the basic ecological functions of the river.

9. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 1, characterized in that, The multi-objective optimization model is solved using the NSGA-III improved non-dominated sorting genetic algorithm to obtain the Pareto optimal solution set and determine the optimal repair scheme, including the following steps: An improved non-dominated sorting genetic algorithm using NSGA-III, incorporating both rigid and flexible constraints, is used to solve a multi-objective optimization model, including: Initialize the population, generate the initial population for the repair plan, and determine the population size; A fitness assessment is conducted, and the analytical values ​​of the ecological benefit maximization objective, engineering cost minimization objective, and restoration sustainability maximization objective for each individual are calculated. Combining rigid and flexible constraints, infeasible solutions are eliminated, and feasible solutions are retained. The feasible solutions are sorted in a non-dominated manner and divided into Pareto levels. The convergence speed and diversity of the algorithm are balanced by a dynamic adjustment mechanism of crossover probability and mutation probability. When the number of iterations reaches the preset number or the optimal solution tends to stabilize, the iteration is terminated, the Pareto optimal solution set is output, and the repair scheme corresponding to the Pareto optimal solution set is sent to experts, who will select the optimal repair scheme based on actual needs.

10. The multi-objective optimization method for fish habitat restoration projects based on reservoir scheduling constraints according to claim 7, characterized in that, Based on the hydrological situation simulation results, and considering the dynamic changes in water level, flow velocity, and water depth, adaptive adjustment coefficients are analyzed, and a comprehensive evaluation index is used to assess the restoration effect. This includes the following steps: Based on the hydrological situation simulation results, and considering the dynamic changes in water level, flow velocity, and water depth, an adaptive adjustment coefficient is analyzed. The formula is as follows: in, The adaptive adjustment coefficient for the repair scheme of region x at time t (0≤ ≤1), The optimal flow velocity for the target fish species in region x. Let U(x,t) be the optimal water depth for the target fish species in region x, U(x,t) be the flow velocity in region x at time t, and H(x,t) be the water depth in region x to be restored at time t. Annual assessments of restoration effectiveness are conducted, analyzing a comprehensive evaluation index based on ecological benefits, engineering feasibility, and sustainability. Where E is the comprehensive evaluation index. To maximize project cost; like If E < 0.6 or E < 0.6, send a warning about the repair plan to the experts; otherwise, continue with the original repair plan.

Images