Reservoir optimization scheduling system and scheduling method for generating multiple targets based on AI

Abstract

The invention relates to a reservoir optimization scheduling system and scheduling method for generating multiple objectives based on AI, the reservoir optimization scheduling system comprises a data input module, a multi-objective optimization module, a constraint processing module, a result output module and a visualization module, the data input module is configured to obtain basic data of a reservoir; the multi-objective optimization module is configured to construct an optimization model based on a particle swarm optimization algorithm, minimize the maximum discharged flow, minimize the flood exceeding storage capacity and minimize the discharged flow change rate as objective functions, and search the optimal discharged flow process through particle iteration; the constraint processing module is configured to perform constraint check on the discharged flow and the water level of the reservoir and apply punishment to solutions violating constraints so as to ensure the feasibility of the scheduling scheme; the result output module is configured to output the optimized discharge flow process, the optimized water level process and the optimized flood limit exceeding storage capacity process; the visualization module is configured to draw a water level-flow process chart and optimize a target convergence curve.

Description

Technical Field

[0001] This application relates to water resource optimization and scheduling, specifically to a reservoir optimization and scheduling system and method based on AI-generated multi-objectives. Background Technology

[0002] Reservoir scheduling is a core component of water resource management, aiming to achieve synergistic optimization of multiple objectives, including flood control, power generation, and water supply, while ensuring project safety. Traditional reservoir scheduling methods often rely on empirical rules or single-objective optimization, making it difficult to address multi-objective balance problems under complex hydrological conditions. For example, solely pursuing the minimization of maximum outflow may lead to reservoirs exceeding flood control limits, while excessive control of outflow capacity may increase downstream flood control pressure. Existing PSO-based reservoir scheduling models often suffer from incomplete objective considerations, such as neglecting the impact of outflow stability on downstream river channels and coarse constraint handling, thus limiting the practicality of scheduling schemes. Summary of the Invention

[0003] The first aspect of this application provides an AI-based multi-objective reservoir optimization scheduling system, which can effectively balance reservoir flood control safety and downstream river stability, thereby improving the comprehensive utilization efficiency of water resources.

[0004] The first aspect of this application provides a reservoir optimization scheduling system based on AI-generated multi-objectives, which includes: a data input module configured to acquire basic data of the reservoir; The multi-objective optimization module is configured to construct an optimization model based on the particle swarm optimization algorithm, with the objective functions being minimizing the maximum outflow, minimizing the reservoir capacity exceeding the flood limit, and minimizing the rate of change of outflow. The optimal outflow process is searched through particle iteration. The objective functions can be transformed from semantics into mathematical functions through an AI model and support dynamic changes in the combination of objectives. The constraint processing module is configured to perform constraint checks on the outflow and reservoir water level, and to impose penalties on solutions that violate the constraints in order to ensure the feasibility of the scheduling scheme. The result output module is configured to output the optimized discharge flow process, water level process, and reservoir capacity exceeding the flood limit process. A visualization module is configured to plot the water level-flow rate process diagram and the optimization target convergence curve.

[0005] In addition, the AI-based multi-objective reservoir optimization scheduling system provided in this application may also have the following additional technical features: In one optional scheme, the objective function of the multi-objective optimization module is in the form of a weighted summation, and the expression of the objective function is: fitness=wq⋅fq+wv⋅fv+wdq⋅fdq+penalty; In the formula, fq is the normalized maximum outflow, fv is the normalized total reservoir capacity exceeding the flood limit, fdq is the normalized total rate of change of outflow, wq, wv, and wdq are the weights of each objective, and penalty is the penalty for constraint violation.

[0006] In one optional embodiment, the constraint conditions used by the constraint processing module during constraint checking include discharge flow constraints, water level constraints, and water balance constraints; wherein the discharge flow constraints satisfy: 0 ≤ q(t) ≤ q max The water level constraint satisfies: z min ≤z(t)≤z max The water balance constraint satisfies: V(t+1)=V(t)+(I(t)−q(t))⋅Δt; In the formula, q max To maximize the allowable discharge flow, z min For the lowest water level, z max V represents the highest water level, V represents the water storage capacity, I represents the inflow rate, and Δt represents the duration of the time period.

[0007] In one optional scheme, the basic data acquired by the data input module includes the water level-reservoir capacity curve, inflow process line, initial water level, constraint parameters, and algorithm parameters; the constraint parameters include the highest water level, the lowest water level, the flood limit water level, and the maximum outflow; and the algorithm parameters include the number of particles and the number of iterations.

[0008] In one alternative approach, the particle swarm optimization algorithm in the multi-objective optimization module includes the following steps: Initialize the particle swarm and velocity: Set the number of particles, maximum number of iterations, inertia weight w, learning factors c1 and c2, and randomly generate the initial particle discharge flow process and velocity; Calculate particle fitness: For each particle, calculate the fitness according to the objective function, and simultaneously record the maximum discharge flow, the reservoir capacity exceeding the flood limit, and the rate of change of flow. Update individual optimality and global optimality: Compare the current fitness of a particle with the historical optimality to update the individual optimality; compare the individual optimalities of all particles to update the global optimality; Particle iteration: Adjust particle velocity and position according to the particle swarm optimization velocity-position update formula, and limit the velocity range to avoid divergence; Constraint handling: Penalize particles that violate constraints, forcibly correcting the discharge flow or storage volume to a feasible range; Convergence judgment: After reaching the maximum number of iterations, output the global optimal solution, that is, the optimal discharge flow process.

[0009] The second aspect of this application provides a reservoir optimization scheduling method based on AI-generated multi-objectives, which mainly includes the following steps: S1: Input basic data and parameters for the reservoir; S2: Construct a multi-objective optimization model, and define the objective function and constraints; S3: The particle swarm optimization algorithm is used to solve the model and generate the optimal discharge flow process; S4: Output and visualize the optimization results.

[0010] The beneficial effects of this application are as follows: This application's AI-based multi-objective reservoir optimization scheduling system and method is the first to incorporate the rate of change in outflow into the optimization objective, balancing flood control safety with downstream river stability and avoiding the impact of sudden flow changes on the river channel. A penalty mechanism is used to handle water level and flow constraints, ensuring that the scheduling scheme complies with engineering safety standards. Based on the rapid convergence characteristics of the particle swarm optimization algorithm, it can quickly generate the optimal scheme under complex hydrological conditions, improving the efficiency of scheduling decisions. Furthermore, a results visualization module intuitively displays the scheduling process, facilitating understanding and application by decision-makers. Overall, it is conducive to achieving multi-objective collaborative optimization, ensuring the safety and feasibility of the scheduling scheme, improving optimization efficiency, effectively balancing reservoir flood control safety with downstream river stability, and enhancing the comprehensive utilization benefits of water resources.

[0011] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this application. Attached Figure Description

[0012] Figure 1 A schematic diagram of the particle swarm optimization algorithm provided in this application during runtime.

[0013] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. Detailed Implementation

[0014] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0015] It should be understood that the described embodiments are merely some embodiments of this application, and not all embodiments. All other technical solutions obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0016] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0017] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0018] like Figure 1 As shown, the first aspect of this application provides an AI-based multi-objective reservoir optimization scheduling system, which is mainly applied to comprehensive scheduling scenarios such as reservoir flood control and water conservancy. It constructs an optimization model containing multi-objective functions and strict constraints, and uses the particle swarm optimization (PSO) algorithm to intelligently generate the optimal scheduling scheme, achieving a balance between reservoir safety and benefits.

[0019] Specifically, this AI-based multi-objective reservoir optimization scheduling system mainly includes a data input module, a multi-objective optimization module, a constraint processing module, a result output module, and a visualization module. These modules are interconnected to achieve data exchange and transmission. The data input module is configured to acquire basic reservoir data; the multi-objective optimization module is configured to construct an optimization model based on the particle swarm optimization algorithm, with the objective functions being minimizing the maximum outflow, minimizing the flood limit capacity, and minimizing the rate of change of outflow. The optimal outflow process is searched through particle iteration; the objective function can be semantically transformed into a mathematical function using the AI ​​model, and supports dynamic changes in the combination of objectives; the constraint processing module is configured to perform constraint checks on the outflow and reservoir water level (storage capacity), and impose penalties on solutions that violate constraints to ensure the feasibility of the scheduling scheme; the result output module is configured to output key results such as the optimized outflow process, water level process, and flood limit capacity process; and the visualization module is configured to draw water level-flow process diagrams and optimization objective convergence curves, intuitively displaying the scheduling effect.

[0020] This embodiment of the AI-based multi-objective reservoir optimization scheduling system incorporates the rate of change in outflow into the optimization objective for the first time, balancing flood control safety with downstream river stability and avoiding the impact of sudden flow changes on the river channel. A penalty mechanism is used to handle water level and flow constraints, ensuring that the scheduling plan complies with engineering safety standards. Based on the rapid convergence characteristics of the particle swarm optimization algorithm, it can quickly generate the optimal plan under complex hydrological conditions, improving the efficiency of scheduling decisions. Furthermore, a results visualization module intuitively displays the scheduling process, facilitating understanding and application by decision-makers. Overall, it is beneficial for achieving multi-objective collaborative optimization, ensuring the safety and feasibility of the scheduling plan, improving optimization efficiency, effectively balancing reservoir flood control safety with downstream river stability, and enhancing the comprehensive utilization benefits of water resources.

[0021] In one specific embodiment, the objective function of the multi-objective optimization module is in the form of a weighted summation, meaning the multi-objective optimization module can optimize three objectives simultaneously. The comprehensive fitness function is constructed by weighted summation, and the expression of the objective function is: fitness=wq⋅fq+wv⋅fv+wdq⋅fdq+penalty; In the formula, fq is the normalized maximum outflow (objective 1: minimize), fv is the normalized total excess reservoir capacity (objective 2: minimize), fdq is the normalized total rate of change of outflow (objective 3: minimize), wq, wv, and wdq are the weights of each objective (summing up to 1), and penalty is the penalty for constraint violation.

[0022] In one specific embodiment, the constraint conditions for the constraint processing module during constraint checking include discharge flow constraint, water level constraint, and water balance constraint; wherein, the discharge flow constraint satisfies: 0≤q(t)≤q max The water level constraint satisfies: z min ≤z(t)≤z max The water balance constraint satisfies: V(t+1)=V(t)+(I(t)−q(t))⋅Δt; In the formula, q max To maximize the allowable discharge flow, z min For the lowest water level, z max V represents the highest water level, V represents the water storage capacity, I represents the inflow rate, and Δt represents the duration of the time period.

[0023] In one specific embodiment, the basic data acquired by the data input module includes the water level-reservoir capacity curve, inflow process line, initial water level, constraint parameters, and algorithm parameters; the constraint parameters include the highest water level, the lowest water level, the flood limit water level, and the maximum outflow, and the algorithm parameters include the number of particles and the number of iterations.

[0024] In one specific embodiment, the particle swarm optimization algorithm in the multi-objective optimization module includes the following steps: Initialize the particle swarm and velocity: Set the number of particles, maximum number of iterations, inertia weight w, learning factors c1 and c2, and randomly generate the initial particle discharge flow process and velocity; Calculate particle fitness: For each particle, calculate the fitness according to the objective function, and simultaneously record the maximum discharge flow, the reservoir capacity exceeding the flood limit, and the rate of change of flow. Update individual optimality and global optimality: Compare the current fitness of a particle with the historical optimality to update the individual optimality; compare the individual optimalities of all particles to update the global optimality; Particle Iteration: Adjust particle velocity and position according to the Particle Swarm Optimization (PSO) velocity-position update formula, and limit the velocity range to avoid divergence; Constraint handling: Penalize particles that violate constraints, forcibly correcting the discharge flow or storage volume to a feasible range; Convergence judgment: After reaching the maximum number of iterations, output the global optimal solution, that is, the optimal discharge flow process.

[0025] The second aspect of this application provides a reservoir optimization scheduling method based on AI-generated multi-objectives, which mainly includes the following steps: S1: Input basic data and parameters for the reservoir; S2: Construct a multi-objective optimization model, and define the objective function and constraints; S3: The particle swarm optimization algorithm is used to solve the model and generate the optimal discharge flow process; S4: Output and visualize the optimization results.

[0026] Since the reservoir optimization scheduling system in the first embodiment facilitates multi-objective collaborative optimization, ensures the safety and feasibility of the scheduling scheme, improves optimization efficiency, and effectively balances reservoir flood control safety with downstream river stability, thereby enhancing the comprehensive utilization benefits of water resources, the reservoir optimization scheduling method based on the operating logic of this system also has similar effects. Therefore, this paper will not elaborate further on this aspect.

[0027] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An AI-based multi-objective reservoir optimal scheduling system, characterized in that, The application relates to a reservoir discharge optimization system, which comprises the following modules: a data input module configured to obtain basic data of a reservoir; a multi-objective optimization module configured to construct an optimization model based on a particle swarm optimization algorithm, to minimize the maximum discharge, to minimize the over-flood limited storage capacity, and to minimize the discharge change rate as an objective function, and to search for an optimal discharge process through particle iteration; the objective function can convert semantics into a mathematical function through an AI model and support dynamic changes of target combinations; a constraint processing module configured to perform constraint checking on the discharge and the water level of the reservoir, and to impose a penalty on solutions violating the constraints, so as to ensure the feasibility of the scheduling scheme; a result output module configured to output the optimized discharge process, the water level process and the over-flood limited storage capacity process; a visualization module configured to draw a water level-flow process graph and an optimization target convergence curve.

2. The AI-based multi-objective reservoir optimal scheduling system of claim 1, wherein, The objective function of the multi-objective optimization module is in a weighted summation form, and the expression of the objective function is as follows: fitness = 1 - penalty wq ⋅ fq + wv ⋅ fv + wdq ⋅ fdq +penalty; wherein fq is the normalized maximum outflow, fv is the normalized total amount of flood control storage, fdq is the normalized total amount of outflow rate change, wq, wv, wdq are the weights of each objective, and penalty is the penalty term for constraint violation.

3. The AI-based multi-objective reservoir optimal scheduling system of claim 1 or 2, wherein, The constraint conditions of the constraint processing module when performing constraint checking include a downstream flow constraint, a water level constraint and a water balance constraint; wherein the downstream flow constraint satisfies: 0≤ q ( t )≤ q max ; the water level constraint satisfies: z min ≤ z ( t )≤ z max ; and the water balance constraint satisfies: V ( t +1)= V ( t )+( I ( t )− q ( t ))⋅Δ t ; wherein q max Qmax is the maximum allowable outflow, z min Qmin is the minimum water level, z max Qmax is the maximum water level, V Q is the storage volume, I Q is the inflow, Δ t T is the time period length.

4. The AI-based multi-objective reservoir optimal scheduling system of claim 3, wherein, The basic data obtained by the data input module include a water level-storage capacity curve, an inflow process line, an initial water level, constraint parameters and algorithm parameters; the constraint parameters include a maximum water level, a minimum water level, a flood limit water level and a maximum discharge, and the algorithm parameters include the number of particles and the number of iterations.

5. The AI-based multi-objective reservoir optimal scheduling system of claim 1 or 2 or 4, wherein, The particle swarm optimization algorithm in the multi-objective optimization module comprises the following steps: initializing the particle swarm and the velocity: setting the number of particles, the maximum number of iterations, the inertia weight w, the learning factors c1 and c2, randomly generating the discharge process and the velocity of the initial particles; calculating the fitness of the particles: for each particle, the fitness is calculated according to the objective function, and the maximum discharge, the over-flood limited storage capacity and the flow change rate are recorded synchronously; updating the individual optimum and the global optimum: comparing the current fitness of the particles with the historical optimum, and updating the individual optimum; comparing the individual optimum of all the particles, and updating the global optimum; particle iteration: adjusting the particle velocity and position according to the particle swarm optimization velocity-position update formula, and limiting the velocity range to avoid divergence; constraint processing: imposing a penalty on the particles violating the constraints, and forcibly correcting the discharge or the storage capacity to a feasible range; convergence judgment: after the maximum number of iterations is reached, the global optimal solution, i.e. the optimal discharge process, is output.

6. An AI-based method for generating multi-objective reservoir optimal scheduling, characterized in that, The application further relates to a reservoir discharge optimization method, which comprises the following steps: S1: inputting the basic data and parameters of the reservoir; S2: constructing a multi-objective optimization model, and defining the objective function and the constraint conditions; S3: solving the model by using the particle swarm optimization algorithm, and generating the optimal discharge process; S4: outputting and visualizing the optimization results.

Images