A method for optimizing ecological flow of reservoir dry limited water level in stages

Abstract

The application discloses a reservoir dry-limit water level staged ecological flow optimization method, S1. and generates an ecological window set; S2. constructs a hydrological state set, and calculates a corresponding hydrological state transition probability matrix; S3. sets a physical relationship constraint; S4. constructs an improved particle swarm optimization model; S5. the final construction of the improved particle swarm optimization model is completed; S6. performing iterative search on the improved particle swarm optimization model, outputting the optimal particle solution composed of the reservoir water level control curve corresponding to the ecological window set and the ecological flow discharge sequence; S7. the optimal particle solution is converted into an executable control signal of the gate opening sequence, the diversion unit adjustment instruction and the real-time reservoir water level set value which can be recognized by the reservoir monitoring data acquisition and monitoring system, and the control signal is executed through the monitoring data acquisition and monitoring system to implement the staged reservoir scheduling. The application realizes the organic unification of the intelligentization and the engineering applicability of the multi-objective staged scheduling of the reservoir.

Description

Technical Field

[0001] This invention relates to the field of reservoir monitoring technology, and in particular to a method for optimizing ecological flow in stages according to the drought limit water level of a reservoir. Background Technology

[0002] With the continuous improvement of water resource regulation and watershed ecological protection concepts, multi-objective phased scheduling of reservoirs is playing an increasingly important role in the comprehensive management of water supply, power generation, flood control, and ecological protection. In recent years, the dual constraints of drought limit water level and ecological flow have become a hot research topic in ecohydrology and water resources engineering. How to achieve a dynamic balance between reservoir water level safety and ecological baseflow guarantee under the background of uncertain water inflow and variable ecological demands is a major challenge in current reservoir scheduling scientific research and engineering practice.

[0003] Existing methods for phased control of reservoir drought limits and ecological flows generally adopt "static linear" or "single threshold within the year" water level control lines. These methods often artificially set water level scheduling schemes based on historical experience or deterministic rules, lacking the ability to dynamically respond to nonlinear hydrological situations such as extreme droughts and sudden floods. Traditional models mainly achieve water level and flow scheduling through single-period linear programming or static threshold methods, which cannot differentiate the control of phased water demand and habitat protection goals for different ecological windows within the year. This leads to insufficient ecological flow discharge from reservoirs or increased risks to water supply security in the later stages during extreme drought years or climate change events.

[0004] On the other hand, while existing multi-objective swarm intelligence algorithms can optimize reservoir scheduling schemes to a certain extent, they are prone to premature convergence, local extrema, and parameter adjustment rigidity problems in complex scenarios such as high-dimensional phased scheduling, multiple constraints, and extreme climates. They are difficult to balance the dynamic equilibrium of ecological and safety objectives. In addition, the mainstream scheduling model is disconnected from the actual reservoir control system, and the calculation results are difficult to be converted into phased and executable gate adjustment commands in real time. This makes it difficult to implement the model results. Information lag and execution deviation further exacerbate the contradiction between ecological and water supply objectives.

[0005] Therefore, there is an urgent need for a new reservoir scheduling optimization method that can achieve phased scheduling adaptation, respond to extreme hydrological conditions, and enhance the coordinated guarantee capacity of ecology and water supply. Summary of the Invention

[0006] One objective of this invention is to propose a method for optimizing the ecological flow of reservoirs in stages according to drought limit water levels. This invention achieves an organic unity between the intelligent multi-objective staged scheduling of reservoirs and their engineering applicability.

[0007] A method for optimizing ecological flow in stages according to a reservoir drought limit water level, according to an embodiment of the present invention, includes the following steps:

[0008] S1. Construct an initial parameter set and preprocess it to form a standardized hydrological time series and generate an ecological window set;

[0009] S2. Using standardized hydrological time series, we divide the state into drought, normal water, abundant water and extreme drought, construct a set of hydrological states, and calculate the corresponding hydrological state transition probability matrix.

[0010] S3. Establish a multi-objective optimization problem with the objectives of minimizing ecological water shortage rate and minimizing reservoir water supply risk, and set physical relationship constraints;

[0011] S4. Construct an improved particle swarm optimization model, and use phased coding to represent particle positions as reservoir water level control variables and ecological flow control variables corresponding to the ecological window set;

[0012] S5. Embed the hydrological state transition probability matrix into the particle velocity update rule, assign the current hydrological state label to each particle, and completely improve the final construction of the particle swarm optimization model;

[0013] S6. Perform iterative search on the improved particle swarm optimization model, evaluate the particle fitness based on the multi-objective optimization problem, update the individual optimal solution and the global optimal solution of the population, adjust the particle position using the current hydrological status label of the particle, and output the optimal particle solution composed of the reservoir water level control curve and the ecological flow discharge sequence corresponding to the ecological window set.

[0014] S7. The optimal particle solution is converted into executable control signals that can be recognized by the reservoir monitoring data acquisition and monitoring system, including gate opening sequence, diversion unit adjustment command and real-time reservoir water level setpoint. The control signals are then executed through the monitoring data acquisition and monitoring system to implement phased reservoir scheduling.

[0015] Optionally, the initial parameter set includes historical water inflow data, real-time water inflow data, reservoir capacity-water level curve, drought limit water level control line parameters, ecological flow demand data, and annual scheduling period division data.

[0016] Optionally, S2 includes the following steps:

[0017] S21. Based on standardized hydrological time series Q={q1,q2,...,q T}, where q t This represents the standardized inflow rate at time step t, where T represents the total number of time steps. The hydrological time series is divided into intervals, and each q is assigned an interval based on the set hydrological boundary thresholds θ = {θ1, θ2, θ3}. t Mapped to the corresponding hydrological state s t∈S, we obtain the set of hydrological states S={D1,N1,W1,ED1}, where D1 represents drought state, N1 represents normal water state, W1 represents abundant water state, ED1 represents extreme drought state, and θ1,θ2,θ3 are standardized inflow thresholds used to delineate the boundaries of each hydrological state.

[0018] S22. Construct the hydrological state transition frequency matrix M = [m] based on the hydrological state set S. ij ] 4×4 , where m ij The hydrological state is represented by state s. i Transition to state s j The frequency of s, and s i ,s j ∈S;

[0019] S23. Calculate the hydrological state transition probability matrix P = [p] based on the hydrological state transition frequency matrix M. ij ] 4×4 , where element p ij This indicates that the hydrological state is s at time t. i Under the given conditions, the state transitions to state s at the next time step t+1. j The probability of.

[0020] Optionally, S3 includes the following steps:

[0021] S31. Calculate the ecological water shortage rate based on the ecological window set. The ecological water shortage rate is used to measure the degree to which the actual outflow within the ecological window does not reach the minimum ecological flow requirement of the ecological window.

[0022] S32. Calculate reservoir water supply risk based on annual scheduling time steps. Reservoir water supply risk is used to measure the degree of risk that the reservoir water level is lower than the drought limit water level control line at each time step during the scheduling year.

[0023] S33. Construct a bi-objective optimization function for the bi-objective optimization problem by weighted normalization of ecological water shortage rate and reservoir water supply risk. Normalize the ecological water shortage rate according to the preset upper and lower bounds, normalize the reservoir water supply risk according to the preset upper and lower bounds, multiply them by preset weight coefficients respectively, and add the two weighted results to obtain the value of the bi-objective optimization function.

[0024] S34. For all time steps of the scheduling year, set a constraint that the reservoir water level is not lower than the drought limit water level control line corresponding to that time step, so that the reservoir water level at each time step is greater than or equal to the drought limit water level control line of that time step.

[0025] S35. For all ecological windows in the ecological window set, set a constraint that the actual discharge flow is not lower than the minimum ecological flow requirement of the ecological window, so that the actual discharge flow of each ecological window is greater than or equal to the minimum ecological flow requirement of the ecological window. The actual discharge flow is the optimization variable, and the minimum ecological flow requirement of the ecological window is the ecological security benchmark.

[0026] S36. For all time steps of the scheduling year, set physical relationship constraints between reservoir capacity and discharge. Based on the reservoir capacity-water level curve, uniquely correspond the reservoir water level of each time step to the reservoir capacity of that time step. Based on the reservoir capacity-discharge relationship and the inflow of water at that time step, establish a one-to-one correspondence between the reservoir capacity and the reservoir water level and the discharge flow.

[0027] S37. Integrate the bi-objective optimization function and all constraints to form the modeling result of the multi-objective optimization problem.

[0028] Optionally, S4 includes the following steps:

[0029] S41. Initialize the particles using a phased-coupled coding method, and set the phased decision vector X for each particle. i Construct a set of ecological windows Ω = {ω1, ω2, ..., ω} for the whole year K}Water level control variable H of each window reservoir i,k and ecological flow control variable Q i,k And introduce a phased cross-coupling term C i,k Reflecting the dynamic connection between scheduling variables of adjacent ecological windows, the particle is encoded as follows:

[0030] X i ={[H i,1 Q i,1 C i,1 ],[H i,2 Q i,2 C i,2 ],…,[H i,K Q i,K C i,K ]};

[0031] Among them, H i,k Let Q represent the reservoir water level control variable for the i-th particle in the k-th ecological window. i,k Let C represent the ecological flow control variable for the i-th particle in the k-th ecological window. i,k This refers to phased cross-coupling terms;

[0032] S42. Introduce a phased adaptive inertia weighting mechanism, adjusting the particle inertia weight w according to the hydrological state of the current ecological window. i,k Real-time dynamic adjustment:

[0033] w i，k=w base ·(1+α·R k );

[0034] Among them, w base The basic inertia weight, α is the window sensitivity coefficient, and R k This represents the historical ecological water shortage risk of the k-th ecological window;

[0035] S43. Based on the current ecological flow control variable Q for each ecological window. i,k Minimum ecological flow demand and annual water supply safety margin for scheduling are considered in the multi-objective optimization function weight λ. eco,k ,λ sup,k Mapped to periodic variable coefficients:

[0036]

[0037] in, H represents the minimum ecological flow requirement for the k-th ecological window. lim,k This is the drought limit water level control line for the k-th window;

[0038] S44. Set the population structure as a global-stage two-layer collaborative architecture, that is, the global optimal solution of the particle swarm not only saves the optimal scheduling sequence for the whole year, but also feeds it back to each ecological window subgroup in real time, realizing the collaborative dynamic optimization of the two-level solutions for the whole year and the stage. The stage subgroup evolves independently under the guidance of the global particle swarm, optimizing the ability to respond to stage extreme climate.

[0039] S45. A phased-probability jump operator is introduced in the particle velocity update. The phased-probability jump operator combines the hydrological state transition probability matrix P with the current particle phased decision vector X. i The correlation triggers a probabilistic perturbation jump when the ecological window is in an extreme drought state, affecting the reservoir water level control variable H. i,k With ecological flow control variable Q i,k Introducing a probability perturbation term:

[0040]

[0041] in, For the new velocity of the i-th particle in the k-th ecological window, Let V be the old velocity of the i-th particle in the k-th ecological window. i,k Let the velocity of the i-th particle be the velocity of the k-th ecological window. For the individual particle's historical best, For the global historical optimum, γ is the extreme drought disturbance coefficient, and η is the global historical optimum. i,k It is a random perturbation that follows the hydrological state transition probability matrix P.

[0042] Optionally, S5 includes the following steps:

[0043] S51. The hydrological state transition probability matrix P = [p ij ] 4×4 Integrated into the particle velocity update mechanism, for each particle's k-th ecological window, based on the hydrological state set S and the current hydrological state label s i,k Using hydrological state transition probability The hydrological state transition of this particle in the next evolutionary step is determined by:

[0044]

[0045] in, This represents the updated hydrological status label of the i-th particle in the k-th ecological window, and Categorical(·) is a random sampling function based on the transition probability.

[0046] S52. Update the hydrological status labels Mapped to phased adaptive inertia weight w i,k Disturbance coefficient γ i,k and probability disturbance term η i,k The selection rule is based on the updated hydrological status label. At the same time, increase the phased perturbation coefficient and the amplitude of the probability perturbation term to increase the jumping ability of particles in the extreme drought window and maintain ecological and security constraints;

[0047] S53. The hydrological state transition probability is embedded into the particle velocity update mechanism. In each iteration, the current hydrological state label s of the particle is used as the basis for the update. i,k And its hydrological state transition probability, dynamically adjusting the particle's velocity and position;

[0048] S54. Form an improved particle swarm optimization model. The improved particle swarm optimization model adaptively adjusts the velocity and position of particles based on the hydrological state transition probability and the current hydrological state label of each ecological window.

[0049] Optionally, S6 includes the following steps:

[0050] S61. Iteratively search the improved particle swarm optimization model. In each iteration, for each particle i and ecological window k, the particle velocity is updated based on the hydrological state transition probability matrix P. and location

[0051] S62. In each iteration, the current stage decision vector for each particle is determined based on the multi-objective optimization function and constraints. Calculate the bi-objective optimization function value F i :

[0052]

[0053] Where, λ eco,k and λ sup,k The weights of the multi-objective optimization function for the k-th ecological window are... and These are the normalized ecological water shortage rate and water supply risk index of the i-th particle in the k-th ecological window, respectively.

[0054] S63. Based on the bi-objective optimization function value F for each particle i Update the individual historical optimal solution of the particle. Historical global optimal solution of particle swarm If the current solution is better than the historical best, then update to the latest value;

[0055] S64. At the end of each iteration, based on the particle's current hydrological state label s i,k The position and velocity of the particles are dynamically adjusted based on the hydrological state transition probability matrix P;

[0056] S65. Determine whether the improved particle swarm optimization model meets the preset convergence condition. The convergence condition can be set based on the global optimal solution change threshold, the maximum number of iterations, or the objective function convergence criterion.

[0057] S66. When the convergence condition is met, output the reservoir water level control curve corresponding to the ecological window set. With ecological flow discharge sequence The optimal particle solution is formed.

[0058] Optionally, step S7 further includes real-time acquisition of actual reservoir water level data and actual outflow data during the scheduling execution process, generating execution result data, calculating the execution deviation between the execution result data and the optimal particle solution, correcting the hydrological state transition probability matrix based on the execution deviation and adjusting the parameters of the improved particle swarm optimization model, and then returning to step S5 to perform rolling optimization based on the corrected hydrological state transition probability matrix and the adjusted particle swarm optimization model parameters, thereby realizing continuous closed-loop control of the phased coordinated scheduling of reservoir drought limit water level and ecological flow.

[0059] The beneficial effects of this invention are:

[0060] (1) This invention deeply integrates phased coupling coding and Markov chain hydrological state transition mechanism into particle swarm optimization model. The scheduling decision vector of each particle is constructed as reservoir water level, ecological flow and phased continuous variables. The particle inertia weight and velocity perturbation term are dynamically adjusted according to the actual hydrological state. When the ecological window is in extreme drought or sudden water flow, the particles can automatically jump perturb according to the hydrological state transition probability, thereby greatly improving the model's adaptability and global search capability in the face of uncertain water inflow and extreme climate events, reducing premature convergence and local optimum problems, and realizing the coordinated dynamic regulation of reservoir drought limit water level and ecological flow.

[0061] (2) This invention introduces a phased adaptive multi-objective weight mapping mechanism. Based on the actual discharge flow, minimum ecological flow demand and drought limit water level constraint of each ecological window, the weight coefficient of the dual-objective optimization function is dynamically adjusted in real time. It is combined with historical ecological water shortage risk to realize window sensitivity adjustment. Through phased normalization and variable weight, the algorithm can finely balance the dynamic conflict between ecological protection and reservoir safe water supply, realize the optimal matching of phased flow and reservoir water level within the ecological window, improve the ecological protection capacity and subsequent water supply security under multi-objective phased scheduling, and is significantly better than the existing static weight or global single-objective phased method.

[0062] (3) This invention directly transforms the optimal solution of the particle swarm-Markov chain hybrid model into phased executable gate opening commands and real-time reservoir water level settings. Through the linkage of the SCADA system and the reservoir automation control system, a closed-loop scheduling process from model optimization to engineering execution is formed. During the scheduling execution process, a Bayesian self-learning feedback mechanism is introduced to correct the hydrological state transition probability matrix and particle swarm parameters in real time based on the actual execution deviation. This continuous rolling optimization significantly reduces the risk of ecological flow and reservoir water level imbalance caused by information lag and execution error, improves the robustness of the model to sudden situations and consecutive extreme years, and realizes the organic unity of intelligent phased scheduling of reservoir multi-objectives and engineering applicability. Attached Figure Description

[0063] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0064] Figure 1 This is a flowchart of a method for optimizing ecological flow in stages according to the drought limit water level of a reservoir, as proposed in this invention. Detailed Implementation

[0065] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0066] refer to Figure 1 A method for optimizing ecological flow in stages at the drought limit water level of a reservoir includes the following steps:

[0067] S1. Construct an initial parameter set, perform wavelet denoising and time alignment processing based on the initial parameter set to form a standardized hydrological time series, and generate an ecological window set;

[0068] S2. Using standardized hydrological time series, we divide the state into drought, normal water, abundant water and extreme drought, construct a set of hydrological states, and calculate the corresponding hydrological state transition probability matrix.

[0069] S3. Establish a multi-objective optimization problem with the objectives of minimizing ecological water shortage rate and minimizing reservoir water supply risk, and set constraints such as the control line that the reservoir water level is not lower than the drought limit water level, the minimum flow constraint of the ecological window, and the physical relationship constraint of reservoir capacity-discharge.

[0070] S4. Construct an improved particle swarm optimization model, using phased coding to represent particle positions as reservoir water level control variables and ecological flow control variables corresponding to the ecological window set, and setting particle velocity, dynamic inertia weight, learning factor, population size and global optimal solution parameters;

[0071] S5. Embed the hydrological state transition probability matrix into the particle velocity update rule, assign the current hydrological state label to each particle, and completely improve the final construction of the particle swarm optimization model;

[0072] S6. Perform iterative search on the improved particle swarm optimization model, evaluate the particle fitness based on the multi-objective optimization problem in step S3, update the individual optimal solution and the global optimal solution of the population, adjust the particle position using the current hydrological state label of the particle, determine whether the improved particle swarm optimization model meets the convergence condition, and if it does, output the optimal particle solution composed of the reservoir water level control curve and the ecological flow discharge sequence corresponding to the ecological window set.

[0073] S7. The optimal particle solution is converted into executable control signals that can be recognized by the reservoir monitoring data acquisition and monitoring system, including gate opening sequence, diversion unit adjustment command and real-time reservoir water level setpoint. The phased reservoir scheduling is then implemented by executing the control signals through the monitoring data acquisition and monitoring system.

[0074] In this embodiment, the initial parameter set includes historical water inflow data, real-time water inflow data, reservoir capacity-water level curve, drought limit water level control line parameters, ecological flow demand data, and scheduling period division data.

[0075] In this embodiment, S2 includes the following steps:

[0076] S21. Based on standardized hydrological time series Q={q1,q2,...,qT}, where q t This represents the standardized inflow rate at time step t, where T represents the total number of time steps. The hydrological time series is divided into intervals, and each q is assigned an interval based on the set hydrological boundary thresholds θ = {θ1, θ2, θ3}. t Mapped to the corresponding hydrological state s t ∈S, thus obtaining the hydrological state set S:

[0077] S = {D1, N1, W1, ED1};

[0078] Wherein, D1 represents drought status, N1 represents normal water status, W1 represents abundant water status, ED1 represents extreme drought status, and θ1, θ2, and θ3 are standardized inflow thresholds used to delineate the boundaries of each hydrological status.

[0079] S22. Construct the hydrological state transition frequency matrix M = [m] based on the hydrological state set S. ij ] 4×4 , where m ij The hydrological state is represented by state s. i Transition to state s j The frequency of s, and s i ,s j ∈S;

[0080] S23. Calculate the hydrological state transition probability matrix P = [p] based on the hydrological state transition frequency matrix M. ij ] 4×4 , where element p ij This indicates that the hydrological state is s at time t. i Under the given conditions, the state transitions to state s at the next time step t+1. j The probability of.

[0081] In this embodiment, S3 includes the following steps:

[0082] S31. Calculate the ecological water shortage rate based on the ecological window set. The ecological water shortage rate is used to measure the degree to which the actual discharge flow within the ecological window does not reach the minimum ecological flow requirement of the ecological window. The ecological water shortage rate is calculated by traversing all ecological windows, taking the non-negative part of the difference between the minimum ecological flow requirement of the ecological window and the actual discharge flow of the ecological window for each ecological window, summing the above results of all ecological windows and taking the average value. The ecological window set is used to divide the ecological scheduling time period. The minimum ecological flow requirement of the ecological window is used to limit the water quantity guarantee standard of the ecosystem within each ecological window. The actual discharge flow is the water release result of the scheduling scheme output in the corresponding ecological window.

[0083] S32. Calculate reservoir water supply risk based on annual scheduling time steps. Reservoir water supply risk is used to measure the degree of risk that the reservoir water level is lower than the drought limit water level control line at each time step in the scheduling year. The calculation method of reservoir water supply risk is to traverse all time steps in the scheduling year, take the non-negative part of the difference between the drought limit water level control line and the reservoir water level at each time step, sum up the above results of all time steps and take the average value. The annual scheduling time step is all discrete time points of the annual scheduling, the drought limit water level control line is the minimum water level constraint to ensure safe water supply from the reservoir, and the reservoir water level is the actual reservoir water level output by the scheduling scheme.

[0084] S33. A bi-objective optimization function is constructed for the bi-objective optimization problem by weighted normalization of ecological water shortage rate and reservoir water supply risk. The calculation method is as follows: the ecological water shortage rate is normalized according to the preset upper and lower bounds, and the reservoir water supply risk is normalized according to the preset upper and lower bounds. Then, they are multiplied by preset weight coefficients respectively. Finally, the two weighted results are added together to obtain the value of the bi-objective optimization function. The value of the bi-objective optimization function is the basis for the fitness evaluation of the bi-objective optimization problem. The normalized upper and lower bounds of ecological water shortage rate and reservoir water supply risk are used to standardize the dimensions, and the weight coefficients are used to adjust the relative importance of the two objectives in the comprehensive evaluation.

[0085] S34. For all time steps of the scheduling year, set a constraint that the reservoir water level is not lower than the drought limit water level control line corresponding to that time step, so that the reservoir water level at each time step is greater than or equal to the drought limit water level control line for that time step.

[0086] Among them, the reservoir water level is the optimization variable, and the drought limit water level control line is the minimum safe water level;

[0087] S35. For all ecological windows in the ecological window set, set a constraint that the actual discharge flow is not lower than the minimum ecological flow requirement of the ecological window, so that the actual discharge flow of each ecological window is greater than or equal to the minimum ecological flow requirement of the ecological window. The actual discharge flow is the optimization variable, and the minimum ecological flow requirement of the ecological window is the ecological security benchmark.

[0088] S36. For all time steps of the scheduling year, set physical relationship constraints between reservoir capacity and discharge. Based on the reservoir capacity-water level curve, uniquely correspond the reservoir water level of each time step to the reservoir capacity of that time step. Then, based on the reservoir capacity-discharge relationship and the inflow rate of that time step, establish a one-to-one correspondence between the reservoir capacity, the reservoir water level and the discharge rate to ensure that the scheduling decisions of all time steps meet the physical constraints. The reservoir capacity-water level curve is the static function relationship between the reservoir capacity and the reservoir water level, the reservoir capacity-discharge relationship is the dynamic function relationship between the reservoir capacity and the discharge rate, and the inflow rate is the inflow rate from the upstream of the reservoir.

[0089] S37. Integrate the bi-objective optimization function and all constraints to form the modeling result of the multi-objective optimization problem.

[0090] In this embodiment, S4 includes the following steps:

[0091] S41. Initialize the particles using a phased-coupled coding method, and set the phased decision vector X for each particle. i Construct a set of ecological windows Ω = {ω1, ω2, ..., ω} for the whole year K}Water level control variable H of each window reservoir i,k and ecological flow control variable Q i,k And introduce a phased cross-coupling term C i,k Reflecting the dynamic connection between scheduling variables of adjacent ecological windows, the particle is encoded as follows:

[0092] X1={[H i，1 Q i，1 C i，1 ], [H i，2 Q i，2 C i，2 ],...,[H i，K Q i，K C i，K ]};

[0093] Among them, H i,k Let Q represent the reservoir water level control variable for the i-th particle in the k-th ecological window. i,k Let C represent the ecological flow control variable for the i-th particle in the k-th ecological window. i,k For phased cross-coupling terms, the continuity of water level and flow scheduling before and after the window is quantified;

[0094] The phased-coupled coding method proposed in S41 not only explicitly decomposes the decision variables of each particle into reservoir water level control variables and ecological flow control variables for each ecological window throughout the year, but also introduces cross-coupling terms for different phases to achieve dynamic connection of scheduling decision variables between different ecological stages. The design of the formula enables the scheduling particle coding itself to have both phased adaptive capability and flexible regulation of water level and flow changes between adjacent windows. This structure effectively overcomes the technical bottleneck of traditional particle swarm optimization, which can only perform global single-entity searches and is difficult to be compatible with phased ecological needs.

[0095] Furthermore, S41 directly maps ecological window risk assessment to particle search weights through an adaptive inertia weight mechanism. The dynamic weight calculation formula ensures that the model can adjust the search strategy in real time according to ecological risks. This allows the scheduling optimization to consider both the global aspects of phased scheduling and the flexible response to extreme situations, overcoming the limitations of existing technologies such as rigid inertia parameters and poor scheduling robustness, and greatly improving the flexibility and intelligence of phased scheduling. This formula system of phased-coupled coding and adaptive weights is an important manifestation of the structural innovation of particle swarm optimization in the field of watershed phased ecological scheduling.

[0096] S42. Introduce a phased adaptive inertia weighting mechanism, adjusting the particle inertia weight w according to the hydrological state of the current ecological window. i,k Real-time dynamic adjustment enables particle inertia weights to adaptively reflect window risk:

[0097] w i,k =w base ·(1+α·R k );

[0098] Among them, w base The basic inertia weight, α is the window sensitivity coefficient, and R k This represents the historical ecological water shortage risk of the k-th ecological window;

[0099] S43. Implement a multi-objective dynamic weight mapping mechanism for particles, based on the current ecological flow control variable Q of each ecological window. i,k Minimum ecological flow demand and annual water supply safety margin for scheduling are considered in the multi-objective optimization function weight λ. eco,k ,λ sup,k Mapped to periodic variable coefficients:

[0100]

[0101] in, H represents the minimum ecological flow requirement for the k-th ecological window. lim,k This is the drought limit water level control line for the k-th window;

[0102] S44. Set the population structure as a global-stage two-layer collaborative architecture, that is, the global optimal solution of the particle swarm not only saves the optimal scheduling sequence for the whole year, but also feeds it back to each ecological window subgroup in real time, realizing the collaborative dynamic optimization of the two-level solutions for the whole year and the stage. The stage subgroup evolves independently under the guidance of the global particle swarm, optimizing the ability to respond to stage extreme climate.

[0103] S45. A phased-probability jump operator is introduced in the particle velocity update. The phased-probability jump operator combines the hydrological state transition probability matrix P with the current particle phased decision vector X. i The correlation triggers a probabilistic perturbation jump when the ecological window is in an extreme drought state, affecting the reservoir water level control variable H. i,k With ecological flow control variable Q i,k Introducing a probability perturbation term:

[0104]

[0105] Among them, V i,k Let the velocity of the i-th particle be the velocity of the k-th ecological window. For the individual particle's historical best, For the global historical optimum, γ is the extreme drought disturbance coefficient, and η is the global historical optimum. i,k It is a random perturbation that follows the hydrological state transition probability matrix P.

[0106] In this embodiment, S5 includes the following steps:

[0107] S51. The hydrological state transition probability matrix P = [p ij ] 4×4 Integrated into the particle velocity update mechanism, for each particle's k-th ecological window, based on the hydrological state set S and the current hydrological state label s i,k Using hydrological state transition probability The hydrological state transition of this particle in the next evolutionary step is determined by:

[0108]

[0109] in, This represents the updated hydrological status label of the i-th particle in the k-th ecological window, and Categorical(·) is a random sampling function based on the transition probability.

[0110] S52. Update the hydrological status labels Mapped to phased adaptive inertia weight w i,k Disturbance coefficient γ i,k and probability disturbance term η i,k The selection rule is based on the updated hydrological status label. At the same time, increase the phased perturbation coefficient and the amplitude of the probability perturbation term to increase the jumping ability of particles in the extreme drought window and maintain ecological and security constraints;

[0111] S53. The hydrological state transition probability is embedded into the particle velocity update mechanism. In each iteration, the current hydrological state label s of the particle is used as the basis for the update. i,k Its hydrological state transition probability, dynamically adjusting the particle velocity and position:

[0112]

[0113] in, For the new velocity of the i-th particle in the k-th ecological window, w i,k Let c1 and c2 be the current periodic inertia weights, c1 and c2 be the learning factors, and r1 and r2 be uniformly random numbers in the interval [0,1]. Let be the individual historical optimal solution for the i-th particle in the k-th ecological window. For the global historical optimal solution of the k-th ecological window, γ i,k η is the periodic disturbance coefficient. i,k The disturbance is randomly generated based on the hydrological state transition probability matrix P;

[0114] Formula S53 defines the core mechanism for particle velocity update in particle swarm optimization during the phased scheduling optimization of ecological flow at reservoir drought-limited water levels, as described in this invention. The formula deeply couples the Markov chain hydrological state transition probability matrix P with the traditional particle swarm velocity update mechanism, specifically in the following aspects:

[0115] Inertia weight phased adaptive: w i,k It is not a fixed value, but rather dynamically adjusted according to the current ecological window and its corresponding historical ecological water shortage risk, phased hydrological status, etc., so that the particle can adaptively adjust its global search and local convergence capabilities in different ecological windows (such as extreme drought and abundant water).

[0116] Multi-objective phased optimization: Preserves the individual historical optimality of particles. and global historical best The approaching term ensures the global optimization property of phased decision-making.

[0117] Markov probability perturbation introduced: γ i,k ·η i,k Based on the hydrological state label of the ecological window in which the particle is located and its Markov state transition probability, the term dynamically generates a probability perturbation term. Under extreme hydrological conditions (such as extreme drought windows), the perturbation intensity is significantly amplified, enabling the particle's probability jump and multi-path escape, effectively overcoming the problems of premature convergence and local extrema.

[0118] Unlike existing PSO technologies that typically use static parameters and homogeneous perturbations, this invention drives the magnitude and frequency of the perturbation term through a hydrological state transition probability matrix. This directly couples the particle search dynamics with the uncertainties of real hydrology, achieving a fundamental leap from "empirical static optimization" to "probability-driven phased adaptive optimization," thus enhancing the model's global adaptability to extreme climate and sudden hydrological scenarios. Through the close coordination of particle phased encoding, phased option weighting, and phased probability perturbations, the S53 formula takes into account the spatiotemporal differences of the ecological window and changes in hydrological conditions, achieving phased collaborative optimization of ecological and water supply security objectives. It exhibits superior dynamic balance and anti-interference capabilities compared to traditional methods, especially under high-risk, low-inflow, or rapid state transition conditions. The probability perturbation term in the S53 formula not only theoretically increases the probability of obtaining the global optimal solution but also directly corresponds to the flexibility and real-time nature of scheduling instructions in practical engineering applications. This ensures that phased decisions can be adjusted in real time within the scheduling cycle for sudden climate events, significantly reducing the practical engineering risks of ecological dehydration and water supply interruption, and providing a solid technical foundation for reservoir ecological scheduling and safety assurance.

[0119] S54. An improved particle swarm optimization model is formed. The improved particle swarm optimization model adaptively adjusts the velocity and position of particles according to the hydrological state transition probability and the current hydrological state label of each ecological window, thereby improving the global optimization ability and ecological scheduling robustness of particles in extreme hydrological scenarios.

[0120] In this embodiment, S6 includes the following steps:

[0121] S61. Iteratively search the improved particle swarm optimization model. In each iteration, for each particle i and ecological window k, the particle velocity is updated based on the hydrological state transition probability matrix P. and location Update particle positions:

[0122]

[0123] Among them, X i,k Let H be the staged decision variable for the i-th particle in the k-th ecological window, including the reservoir water level control variable H. i,k With ecological flow control variable Q i,k ;

[0124] S62. In each iteration, the current stage decision vector for each particle is determined based on the multi-objective optimization function and constraints. Calculate the bi-objective optimization function value F i :

[0125]

[0126] Where, λ eco,k and λ sup,k The weights of the multi-objective optimization function for the k-th ecological window are... and These are the normalized ecological water shortage rate and water supply risk index of the i-th particle in the k-th ecological window, respectively.

[0127] The core of S62 lies in the construction of a bi-objective optimization for the phased scheduling problem. The formula normalizes the ecological water shortage rate and the reservoir water supply risk separately, and introduces a variable weight λ for each ecological window. eco,k and λ sup,k This approach enables flexible integration of phased multi-objectives and real-time priority adjustment. The dynamic setting of weights directly binds the particle's decision output in the current window to the achievement of ecological / water supply objectives, avoiding the phased scheduling imbalance caused by a single global weight in traditional methods.

[0128] The formula system breaks through the conventional group optimization method of single normalization objective or static weight configuration. It organically combines phased objective weights, phased normalization and dynamic mapping of historical risks, ensuring that the algorithm can take into account dynamic priority adjustment, weigh complex objective conflicts and achieve real-time optimal scheduling in different ecological stages.

[0129] S63. Based on the bi-objective optimization function value F for each particle i Update the individual historical optimal solution of the particle. Historical global optimal solution of particle swarm If the current solution is better than the historical best, then update to the latest value;

[0130] S64. At the end of each iteration, based on the particle's current hydrological state label s i,k The position and velocity of the particles are dynamically adjusted based on the hydrological state transition probability matrix P;

[0131] S65. Determine whether the improved particle swarm optimization model meets the preset convergence condition. The convergence condition can be set based on the global optimal solution change threshold, the maximum number of iterations, or the objective function convergence criterion.

[0132] S66. When the convergence condition is met, output the reservoir water level control curve corresponding to the ecological window set. With ecological flow discharge sequence The optimal particle solution. and These represent the optimal reservoir water level and optimal ecological flow control schemes for phased scheduling, respectively, serving as the optimization results of phased coordinated scheduling of reservoir drought limit water level and ecological flow.

[0133] In this embodiment, S7 further includes real-time acquisition of actual reservoir water level data and actual outflow data during the scheduling execution process, generating execution result data, calculating the execution deviation between the execution result data and the optimal particle solution, correcting the hydrological state transition probability matrix based on the execution deviation and adjusting the parameters of the improved particle swarm optimization model, and then returning to step S5 to perform rolling optimization based on the corrected hydrological state transition probability matrix and the adjusted particle swarm optimization model parameters, thereby realizing continuous closed-loop control of the phased coordinated scheduling of reservoir drought limit water level and ecological flow.

[0134] Example 1:

[0135] During a complete dispatch cycle, the dispatch center collected 8,760 hourly inflow data points for a certain reservoir from the previous year, ranging from 15 to 295 meters. 3 / s, Entering the new year, the dispatching system first uses wavelet noise reduction to process the water inflow time series data from today to the previous 60 days. The water inflow rates for the five consecutive days are 32, 28, 25, 23, and 19 m³ / s, respectively. 3 / s, after standardization and allocation to drought conditions, automatically generated standardized reservoir capacity-water level curves are as follows: when the reservoir capacity is 860 million cubic meters, the water level is 314.2m; when it is 1.03 billion cubic meters, the water level is 316.1m; and when it is 1.2 billion cubic meters, the water level is 318.3m. The phased division results are: January-March gestation period, April-May spawning period, June-October peak growth period, and November-December overwintering period.

[0136] Within a certain scheduling cycle, the minimum ecological flow requirements for the ecological window are set to 65, 90, 105, and 52 m³, respectively. 3 / s. The system automatically detected that the flow rate from day 5 to day 8 is less than 65m. 3 / s, triggering an ecological water shortage warning, the algorithm identifies the current window's extreme drought continuation probability as 0.68 through Markov chain state transition probabilities. Therefore, it automatically increases the inertia weight of the phased particles to 1.38 and injects probabilistic perturbations into the particle swarm optimization iteration to adjust particle positions. At this point, some particle decision outputs are adjusted to a water level of 314.8m and a discharge flow of 67m³ / s. 3 / s, the system automatically issues a water replenishment scheduling command, driving the gate opening range to increase by 2%, the real-time water level change in the reservoir area is 0.12m, and the water shortage rate is reduced from the original 8.1% to 1.7%.

[0137] As the ecological window shifts to the spawning season, the real-time water inflow increases to 89, 96, 101, 104, and 109 m. 3 / s, standardized mapping to a normal to high water level state. Based on phased-coupled coding, the system automatically adjusts the phased coupling term C to the interval 0.8-1.1 to ensure a smooth transition between water level and flow rate decisions. During particle swarm optimization, the multi-objective weights are mapped to an ecological weight of 0.62 and a water supply weight of 0.38. After normalization of the fitness function, the optimal particle solution is a water level of 316.3m and a discharge flow rate of 93m³. 3 / s, the system issues a gate opening adjustment command, and the monitoring station reports that the downstream ecological flow has reached 9599m³. 3 The flow rate fluctuates by 0 m / s, resulting in an ecological water shortage duration of 0 hours. Traditional dispatching methods show a flow rate fluctuation of 85-92 m / s during the same period. 3 / s, ecological water shortage duration 4 hours.

[0138] During the peak growth period, the inflow to the reservoir fluctuated twice, with the lowest flow rate at 42 m³ / h. 3 / s, maximum flow rate 273m 3 / s, rapid switching between extreme drought and abundant water conditions. The Markov chain hydrological state matrix is ​​automatically updated. In the particle swarm optimization algorithm, the probability perturbation coefficient for extreme drought increases to 1.8, and some particles actively increase the water level decision to 317.6m and decrease the discharge flow to 102m. 3 / s, quickly recovered to 109m after hydration. 3 / s. Throughout the entire cycle, the ecological flow compliance rate was 99.1%, the reservoir water level was below the drought limit for a cumulative period of only 3 hours, and the water supply scheduling error was 0.9%. Under traditional scheduling, the compliance rate was 94.7%, the water level was below the drought limit for 16 hours, and the water supply scheduling error was 3.4%.

[0139] During a certain extreme weather event, the water level suddenly dropped to 21-23 meters over two days. 3 / s, standardized as extreme drought. The system recognizes that phased particles require a rapid response under this condition, significantly increasing the perturbation term η. After the particle velocity jumps, the discharge flow rate is immediately adjusted to 54m³. 3 / s, the ecological water shortage alarm time is less than 20 minutes, and the flow interruption duration after water replenishment scheduling is 0 hours. If traditional scheduling is used, due to the lag in the response of the static scheduling curve, the actual flow interruption duration is 6 hours, and alarms are triggered at ecological monitoring points in some downstream river sections.

[0140] During the year-end scheduling summary phase, the model performs Bayesian self-learning corrections to address actual execution deviations. For example, during the peak growing season in October, the system detected a model-predicted discharge flow of 110 m³ / h. 3 / s, the actual average monitored value is 108.5m 3 The model automatically fine-tunes the Markov state transition matrix and particle swarm weight parameters. Full-cycle statistical data shows that, using the method of this invention, the annual ecological window discharge compliance rate reached 98.9%, the annual reservoir water level safety margin reached 98.4%, there were 0 ecological flow interruption events during extreme drought windows, and the average scheduling error was 1.3%. Under the traditional method, these indicators were 92.7%, 93.2%, 3 times, and 5.6%, respectively.

[0141] Table 1 shows a partial comparison of training samples (taking a portion of the scheduling cycles as an example):

[0142]

[0143] Compared to the whole year, the method of this invention increased the average downstream flow by 6.3%, reduced the ecological water shortage rate by 5.4 percentage points, increased the water supply security margin by 4.2 percentage points, reduced the risk of ecological flow interruption during extreme drought windows to 0, and achieved automatic closed-loop feedback for all scheduling operations with minimal execution deviation.

[0144] This embodiment 1 demonstrates the engineering application advantages of the method of the present invention in key aspects of extreme hydrological situation identification, phased scheduling adaptation, continuous regulation of ecological window, global-phased coordination of particle swarm and Markov chain dynamic disturbance mechanism. By controlling reservoir water level and ecological flow in phases, it not only significantly improves the ecological safety and water supply reliability of reservoir scheduling, but also comprehensively overcomes the core defects of traditional scheduling in extreme event response lag and model-execution gap.

[0145] This invention deeply integrates phased coupling coding and Markov chain hydrological state transition mechanism into a particle swarm optimization model. The scheduling decision vector of each particle is constructed as reservoir water level, ecological flow and phased continuous variables. The particle inertia weight and velocity perturbation term are dynamically adjusted according to the actual hydrological state. When the ecological window is in extreme drought or sudden water flow, the particles can automatically jump perturb according to the hydrological state transition probability, thereby greatly improving the model's adaptability and global search capability in the face of uncertain water inflow and extreme climate events, reducing premature convergence and local optima problems, and realizing the coordinated dynamic regulation of reservoir drought limit water level and ecological flow.

[0146] This invention introduces a phased adaptive multi-objective weight mapping mechanism. Based on the actual discharge flow, minimum ecological flow demand, and drought limit water level constraint of each ecological window, the weight coefficients of the bi-objective optimization function are dynamically adjusted in real time. The window sensitivity is adjusted by combining it with historical ecological water shortage risk. Through phased normalization and variable weights, the algorithm can finely balance the dynamic conflict between ecological protection and reservoir safe water supply, and achieve optimal matching between phased flow and reservoir water level within the ecological window. This improves the ecological security capability and subsequent water supply security under multi-objective phased scheduling, and is significantly better than existing static weight or global single-objective phased methods.

[0147] This invention directly transforms the optimal solution of the particle swarm optimization-Markov chain hybrid model into phased executable gate opening commands and real-time reservoir water level settings. Through linkage with the SCADA system and the reservoir automation control system, a closed-loop scheduling process from model optimization to engineering execution is formed. During the scheduling execution process, a Bayesian self-learning feedback mechanism is introduced to correct the hydrological state transition probability matrix and particle swarm parameters in real time based on actual execution deviations, continuously optimizing the process. This significantly reduces the risk of ecological flow and reservoir water level imbalance caused by information lag and execution errors, improves the model's robustness to sudden situations and consecutive extreme years, and achieves the organic unity of intelligent multi-objective phased scheduling of the reservoir and engineering applicability.

[0148] The above description is only a preferred embodiment of the present invention, but 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 inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for optimizing ecological flow in stages at the drought-limited water level of a reservoir, characterized in that, Includes the following steps: S1. Construct an initial parameter set and preprocess it to form a standardized hydrological time series and generate an ecological window set; S2. Using standardized hydrological time series, we divide the state into drought, normal water, abundant water and extreme drought, construct a set of hydrological states, and calculate the corresponding hydrological state transition probability matrix. S3. Establish a multi-objective optimization problem with the objectives of minimizing ecological water shortage rate and minimizing reservoir water supply risk, and set physical relationship constraints; S4. Construct an improved particle swarm optimization model, and use phased coding to represent particle positions as reservoir water level control variables and ecological flow control variables corresponding to the ecological window set; S5. Embed the hydrological state transition probability matrix into the particle velocity update rule, assign the current hydrological state label to each particle, and completely improve the final construction of the particle swarm optimization model; S6. Perform iterative search on the improved particle swarm optimization model, evaluate the particle fitness based on the multi-objective optimization problem, update the individual optimal solution and the global optimal solution of the population, adjust the particle position using the current hydrological status label of the particle, and output the optimal particle solution composed of the reservoir water level control curve and the ecological flow discharge sequence corresponding to the ecological window set. S7. The optimal particle solution is converted into executable control signals that can be recognized by the reservoir monitoring data acquisition and monitoring system, including gate opening sequence, diversion unit adjustment command and real-time reservoir water level setpoint. The phased reservoir scheduling is implemented by executing the control signals through the monitoring data acquisition and monitoring system. Specifically, S4 is: S41. Initialize the particles using a phased-coupled coding method, and set the phased decision vector for each particle. Constructed as a set of ecological windows throughout the year Water level control variables for each window reservoir and ecological flow control variables And introduce phased cross-coupling terms. Reflecting the dynamic connection between scheduling variables of adjacent ecological windows, the particle is encoded as follows: ； in, Indicates the first The particle in the first The reservoir water level control variable for each ecological window Indicates the first The particle in the first Ecological flow control variables for each ecological window This refers to phased cross-coupling terms; S42. Introduce a phased adaptive inertia weighting mechanism, adjusting particle inertia weights according to the hydrological state of the current ecological window. Real-time dynamic adjustment: ； in, Based on the inertia weight, The window sensitivity coefficient, Indicates the first The historical ecological water shortage risk of each ecological window; S43. Based on the current ecological flow control variables for each ecological window. Minimum ecological flow demand and annual water supply security margin for scheduling will be considered in the weighting of the multi-objective optimization function. Mapped to periodic variable coefficients: ； in, For the first Minimum ecological flow requirement for each ecological window For the first One window of drought limit water level control line; S44. Set the population structure as a global-stage two-layer collaborative architecture, that is, the global optimal solution of the particle swarm not only saves the optimal scheduling sequence for the whole year, but also feeds it back to each ecological window subgroup in real time, realizing the collaborative dynamic optimization of the two-level solutions for the whole year and the stage. The stage subgroup evolves independently under the guidance of the global particle swarm, optimizing the ability to respond to stage extreme climate. S45. A phased-probability jump operator is introduced in the particle velocity update. The phased-probability jump operator transforms the hydrological state transition probability matrix. With the current particle stage decision vector The correlation triggers a probabilistic perturbation jump when the ecological window is in a state of extreme drought, affecting the reservoir water level control variable. With ecological flow control variables Introducing a probability perturbation term: ； in, For the first The first particle A new speed for an ecological window For the first The first particle The old speed of an ecological window For the first The first particle An ecological window speed, For the individual particle's historical best, For the best overall historical performance, This represents the extreme drought disturbance coefficient. To obey the hydrological state transition probability matrix The random perturbation quantity.

2. The method for optimizing ecological flow in stages according to claim 1, characterized in that, The initial parameter set includes historical water inflow data, real-time water inflow data, reservoir capacity-water level curve, drought limit water level control line parameters, ecological flow demand data, and annual scheduling period division data.

3. The method for optimizing ecological flow in stages according to claim 2, characterized in that, S2 includes the following steps: S21. Based on standardized hydrological time series ,in Indicates the first Standardized inflow rate value at each time step This represents the total number of time steps, dividing the hydrological time series into intervals based on set hydrological boundary thresholds. Each Mapped to the corresponding hydrological state The hydrological state set is obtained. ,in, Indicates the drought situation. Indicates a level water state. Indicates a state of abundant water. Indicates an extreme drought condition. To standardize the inflow threshold, used to delineate the boundaries of various hydrological states; S22. Construct a hydrological state transition frequency matrix based on the hydrological state set S. ,in Indicates the hydrological state from state Transition to state The frequency of, and ; S23. Calculate the hydrological state transition probability matrix based on the hydrological state transition frequency matrix M. , of which elements Indicates at time Hydrological status is Under the conditions, the next moment Transition to state The probability of.

4. The method for optimizing ecological flow in stages according to claim 3, characterized in that, S3 includes the following steps: S31. Calculate the ecological water shortage rate based on the ecological window set. The ecological water shortage rate is used to measure the degree to which the actual outflow within the ecological window does not reach the minimum ecological flow requirement of the ecological window. S32. Calculate reservoir water supply risk based on annual scheduling time steps. Reservoir water supply risk is used to measure the degree of risk that the reservoir water level is lower than the drought limit water level control line at each time step during the scheduling year. S33. Construct a bi-objective optimization function for the bi-objective optimization problem by weighted normalization of ecological water shortage rate and reservoir water supply risk. Normalize the ecological water shortage rate according to the preset upper and lower bounds, normalize the reservoir water supply risk according to the preset upper and lower bounds, multiply them by preset weight coefficients respectively, and add the two weighted results to obtain the value of the bi-objective optimization function. S34. For all time steps of the scheduling year, set a constraint that the reservoir water level is not lower than the drought limit water level control line corresponding to that time step, so that the reservoir water level at each time step is greater than or equal to the drought limit water level control line of that time step. S35. For all ecological windows in the ecological window set, set a constraint that the actual discharge flow is not lower than the minimum ecological flow requirement of the ecological window, so that the actual discharge flow of each ecological window is greater than or equal to the minimum ecological flow requirement of the ecological window. The actual discharge flow is the optimization variable, and the minimum ecological flow requirement of the ecological window is the ecological security benchmark. S36. For all time steps of the scheduling year, set physical relationship constraints between reservoir capacity and discharge. Based on the reservoir capacity-water level curve, uniquely correspond the reservoir water level of each time step to the reservoir capacity of that time step. Based on the reservoir capacity-discharge relationship and the inflow of water at that time step, establish a one-to-one correspondence between the reservoir capacity and the reservoir water level and the discharge flow. S37. Integrate the bi-objective optimization function and all constraints to form the modeling result of the multi-objective optimization problem.

5. The method for optimizing ecological flow in stages according to claim 1, characterized in that, S5 includes the following steps: S51. Hydrological state transition probability matrix Integrated into the particle velocity update mechanism, for each particle's... An ecological window, based on a set of hydrological conditions. and current hydrological status labels Using hydrological state transition probability The hydrological state transition of this particle in the next evolutionary step is determined by: ； in, Indicates the first The particle in the first Updated hydrological status labels for each ecological window. It is a random sampling function based on the transition probability; S52. Update the hydrological status labels Mapping to phased adaptive inertia weights Disturbance coefficient and probability disturbance term The selection rule is based on the updated hydrological status label. At the same time, increase the phased perturbation coefficient and the amplitude of the probability perturbation term to increase the jumping ability of particles in the extreme drought window and maintain ecological and security constraints; S53. Embed the hydrological state transition probability into the particle velocity update mechanism, and update the velocity based on the particle's current hydrological state label in each iteration. And its hydrological state transition probability, dynamically adjusting the particle's velocity and position; S54. Form an improved particle swarm optimization model. The improved particle swarm optimization model adaptively adjusts the velocity and position of particles based on the hydrological state transition probability and the current hydrological state label of each ecological window.

6. The method for optimizing ecological flow in stages according to claim 5, characterized in that, S6 includes the following steps: S61. Perform an iterative search on the improved particle swarm optimization model. In each iteration, for each particle... and ecological window Using the hydrological state transition probability matrix Updated particle velocity and location ; S62. In each iteration, the current stage decision vector for each particle is determined based on the multi-objective optimization function and constraints. Calculate the value of the bi-objective optimization function : ； in, and For the first Weights of the multi-objective optimization function for each ecological window and The first The particle in the first Normalized ecological water shortage rate and water supply risk indicators for each ecological window; S63. Based on the bi-objective optimization function value for each particle. Update the individual historical optimal solution of the particle. Historical global optimal solution of particle swarm If the current solution is better than the historical best, then update it to the latest value; S64. At the end of each iteration, based on the particle's current hydrological state label. Combining the hydrological state transition probability matrix Dynamically adjust the position and velocity of particles; S65. Determine whether the improved particle swarm optimization model meets the preset convergence condition. The convergence condition can be set based on the global optimal solution change threshold, the maximum number of iterations, or the objective function convergence criterion. S66. When the convergence condition is met, output the reservoir water level control curve corresponding to the ecological window set. With ecological flow discharge sequence The optimal particle solution is formed.

7. The method for optimizing ecological flow in stages according to claim 6, characterized in that, S7 further includes real-time acquisition of actual reservoir water level data and actual outflow data during the scheduling execution process, generating execution result data, calculating the execution deviation between the execution result data and the optimal particle solution, correcting the hydrological state transition probability matrix based on the execution deviation and adjusting the parameters of the improved particle swarm optimization model, and then returning to step S5 to perform rolling optimization based on the corrected hydrological state transition probability matrix and the adjusted particle swarm optimization model parameters, thereby realizing continuous closed-loop control of the phased coordinated scheduling of reservoir drought limit water level and ecological flow.

Images