Reservoir level regulation and control method and device

Abstract

The invention provides a reservoir level regulation and control method and device which can be applied to the technical field of water conservancy. The method comprises the following steps: establishing a dam simulation model according to monitoring data of a dam; according to the dam simulation model, calculating safety coefficients of the dam at different water levels; according to an optimization target, the current state data of the reservoir and the safety coefficients of the dam at different water levels, the water level variation of the reservoir is determined, and the optimization target comprises at least one of ecological flow meeting, flood control and power generation. According to the reservoir water level regulation and control method and device provided by the embodiment of the invention, the working state of the dam is simulated and analyzed, the reservoir water level regulation and control scheme can be gradually refined under the target guidance of flood control, power generation and ecology under the condition of ensuring the safety of the working state, and the water level regulation and control scheme of the next water storage period is finally obtained; and safe, economic and ecological multi-target benefit improvement is realized.

Description

Technical Field

[0001] The present application relates to the field of water conservancy technology, and in particular to a reservoir water level control method and device. Background Art

[0002] During the dam impoundment process, the working state of the dam should be monitored to ensure the safety of operation. In addition, the development laws of the dam body temperature field, deformation field, and seepage field should be analyzed based on simulation and monitoring data, and the upstream water level control plan for dam operation should be adjusted. As the water level rises, the temperature of the upstream surface concrete within the elevation range of the water level change is affected by the water temperature, and the deformation of the dam has a certain impact on the working state and long-term safety.

[0003] The regulation of water levels is also closely related to flood control. In fact, the dynamic adjustment of water levels is to make full use of the reservoir's ability to regulate small and medium-sized floods, alleviate the downstream flood control pressure and improve other benefits. According to the water and rainfall information, combined with the law of seasonal changes, reasonable water level regulation can effectively solve the problem of flood resource utilization. Not only that, scientific water level regulation schemes can also improve power generation efficiency and benefit the sustainable development of energy. In conjunction with other clean energy sources such as wind power and photovoltaics, the efficient utilization of natural resources has been greatly improved. Finally, on the basis of meeting a series of social and economic goals such as reservoir flood control and power generation, the impact on the water ecology of the basin should also be considered, the ecological needs should be fully coordinated, and the water level regulation scheme should be further optimized in detail.

[0004] When considering the optimization of multi-objective schemes for reservoir water level regulation, current research generally takes flood dispatching as the main consideration, optimizes storage capacity allocation between cascade reservoirs, and lacks a detailed water level dynamic adjustment scheme for a single reservoir, or takes ecological flow as the core element, and simply describes other goals such as flood control and power generation with a few objective function formulas, lacking further scientific description. In other words, in the existing technology, the methods used are relatively biased, and the results achieved are not comprehensive. Some reservoir regulation schemes are based on monitoring data and simulation deformation analysis, and the water level regulation strategies obtained are very rough, and cannot be adjusted and analyzed in combination with the actual changes in water conditions and economic benefits. While other reservoir regulation schemes are based on traditional operations research or control equations, although optimization algorithms are used, they are often limited to a certain professional category, taking one of the multi-objective factors as the main line, and other objectives are just foils. Therefore, the multi-objective optimization of water level regulation schemes can actually only partially take into account other objectives. Summary of the invention

[0005] In response to the problems in the prior art, the embodiments of the present application provide a reservoir water level control method and device, which can at least partially solve the problems in the prior art.

[0006] On the one hand, the present application proposes a reservoir water level control method, comprising:

[0007] Establish a dam simulation model based on the dam monitoring data;

[0008] According to the dam simulation model, calculate the safety factor of the dam at different water levels;

[0009] Determine the water level change of the reservoir according to the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation;

[0010] The water level of the reservoir is regulated according to the change in the water level of the reservoir.

[0011] In some embodiments, the monitoring data includes at least one of the following: reservoir inflow, reservoir outflow, upstream water level data, downstream water level data, rainfall data, temperature data, water consumption data of downstream residents and enterprises, and minimum ecological flow.

[0012] In some embodiments, the safety factor of the dam at different water levels is calculated according to the dam simulation model, including:

[0013] According to the safety standards and mechanical models, the allowable stress and displacement values ​​of all grid points of the dam are calculated to obtain the safety threshold of each grid point;

[0014] The dam simulation model is used to perform simulation to obtain a water level-displacement relationship and a water level-stress relationship of each grid point of the dam;

[0015] The water level-safety factor relationship of the dam is determined based on the water level-displacement relationship, the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point.

[0016] In some embodiments, determining the water level-safety factor relationship of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point includes:

[0017] Determine a water level-safety score for each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point;

[0018] Based on the water level-safety score of each grid point, the water level-safety factor relationship of the dam is determined.

[0019] In some embodiments, the water level-stress relationship of each grid point of the dam includes at least one of the following: a water level-cross-river stress relationship, a water level-down-river stress relationship, and a water level-vertical stress relationship; the water level-displacement relationship of each grid point of the dam includes at least one of the following: a water level-cross-river displacement relationship, a water level-down-river displacement relationship, and a water level-vertical displacement relationship;

[0020] The safety threshold of each grid point includes a safety stress threshold and a safety displacement threshold. The safety stress threshold includes at least one of the following: a cross-river stress threshold, a downstream stress threshold, and a vertical stress threshold; the safety displacement threshold includes at least one of the following: a cross-river displacement threshold, a downstream displacement threshold, and a vertical displacement threshold.

[0021] In some embodiments, determining the water level-safety score of each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point includes:

[0022] For each water level, according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam, the displacement and stress of each grid point of the dam at the water level are calculated;

[0023] Determining the safety category of each grid point according to the displacement and stress of each grid point of the dam at the water level and the safety threshold of the grid point;

[0024] Determining the safety score of the grid point of the dam at the water level according to the safety category of each grid point of the dam and the safety score corresponding to each safety category;

[0025] According to the safety score of each grid point of the dam at each water level, a water level-safety score of each grid point of the dam is determined.

[0026] In some embodiments, determining the water level-safety factor relationship of the dam according to the water level-safety score of each grid point includes:

[0027] The water level-safety factor relationship of the dam is determined based on the water level-safety score of each grid point and the weight of the dam zone to which each grid point belongs; wherein the weight of each dam zone is determined based on the construction and water storage stage of the dam.

[0028] In some embodiments, determining the water level change of the reservoir according to the optimization target, the current state data of the reservoir, and the safety factor of the dam at different water levels includes:

[0029] The current state data of the reservoir is input into a pre-trained prediction model to obtain a recommended water level change output by the prediction model, wherein the prediction model is trained based on the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels.

[0030] In some embodiments, before inputting the current state data of the reservoir into a pre-trained prediction model to obtain the recommended water level change output by the prediction model, the method further includes:

[0031] Establish model training instances, each instance includes the initial water level of the reservoir, the current water level and the inflow flow, wherein the inflow flow is randomly generated to simulate the inflow conditions on different days;

[0032] According to the initial water level, current water level and inflow of the reservoir in each instance, the deep Q network model is trained to obtain a trained prediction model. During the training process, for each instance, the deep Q network model selects and executes a water level change according to a greedy strategy, and calculates the new water level, power generation and reservoir safety factor according to the water level change, then calculates the reward value according to the new water level, power generation and reservoir safety factor, and finally updates the parameters of the network Q according to the new water level and reward value.

[0033] On the other hand, the present application proposes a reservoir water level control device, comprising:

[0034] The first building module is used to build a dam simulation model based on the monitoring data of the dam;

[0035] A calculation module is used to calculate the safety factor of the dam at different water levels according to the dam simulation model;

[0036] A determination module, used to determine the water level change of the reservoir according to the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation;

[0037] The control module is used to control the water level of the reservoir according to the change of the water level of the reservoir.

[0038] In some embodiments, the monitoring data includes at least one of the following: reservoir inflow, reservoir outflow, upstream water level data, downstream water level data, rainfall data, temperature data, water consumption data of downstream residents and enterprises, and minimum ecological flow.

[0039] In some embodiments, the computing module is specifically used for:

[0040] According to the safety standards and mechanical models, the allowable stress and displacement values ​​of all grid points of the dam are calculated to obtain the safety threshold of each grid point;

[0041] The dam simulation model is used to perform simulation to obtain a water level-displacement relationship and a water level-stress relationship of each grid point of the dam;

[0042] The water level-safety factor relationship of the dam is determined based on the water level-displacement relationship, the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point.

[0043] In some embodiments, the calculation module determines the water level-safety factor relationship of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, including:

[0044] Determine a water level-safety score for each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point;

[0045] Based on the water level-safety score of each grid point, the water level-safety factor relationship of the dam is determined.

[0046] In some embodiments, the water level-stress relationship of each grid point of the dam includes at least one of the following: a water level-cross-river stress relationship, a water level-down-river stress relationship, and a water level-vertical stress relationship; the water level-displacement relationship of each grid point of the dam includes at least one of the following: a water level-cross-river displacement relationship, a water level-down-river displacement relationship, and a water level-vertical displacement relationship;

[0047] The safety threshold of each grid point includes a safety stress threshold and a safety displacement threshold. The safety stress threshold includes at least one of the following: a cross-river stress threshold, a downstream stress threshold, and a vertical stress threshold; the safety displacement threshold includes at least one of the following: a cross-river displacement threshold, a downstream displacement threshold, and a vertical displacement threshold.

[0048] In some embodiments, the calculation module determines the water level-safety score of each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, including:

[0049] For each water level, according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam, the displacement and stress of each grid point of the dam at the water level are calculated;

[0050] Determining the safety category of each grid point according to the displacement and stress of each grid point of the dam at the water level and the safety threshold of the grid point;

[0051] Determining the safety score of the grid point of the dam at the water level according to the safety category of each grid point of the dam and the safety score corresponding to each safety category;

[0052] According to the safety score of each grid point of the dam at each water level, a water level-safety score of each grid point of the dam is determined.

[0053] In some embodiments, the calculation module determines the water level-safety factor relationship of the dam according to the water level-safety score of each grid point, including:

[0054] The water level-safety factor relationship of the dam is determined based on the water level-safety score of each grid point and the weight of the dam zone to which each grid point belongs; wherein the weight of each dam zone is determined based on the construction and water storage stage of the dam.

[0055] In some embodiments, the determining module is specifically used to:

[0056] The current state data of the reservoir is input into a pre-trained prediction model to obtain a recommended water level change output by the prediction model, wherein the prediction model is trained based on the state data of the reservoir and the safety factor of the dam at different water levels.

[0057] In some embodiments, the apparatus further comprises:

[0058] The second establishment module is used to establish model training instances, each instance includes the initial water level of the reservoir, the current water level and the inflow flow, wherein the inflow flow is randomly generated to simulate the inflow conditions on different days;

[0059] A training module is used to train the deep Q network model according to the initial water level, current water level and inflow of the reservoir in each instance to obtain a trained prediction model, wherein during the training process, for each instance, the deep Q network model selects and executes a water level change according to a greedy strategy, and calculates the new water level, power generation and reservoir safety factor according to the water level change, then calculates the reward value according to the new water level, power generation and reservoir safety factor, and finally updates the parameters of the network Q according to the new water level and reward value.

[0060] An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the reservoir water level control method described in any of the above embodiments are implemented.

[0061] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the reservoir water level control method described in any of the above embodiments are implemented.

[0062] An embodiment of the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments.

[0063] The reservoir water level control method and device provided in the embodiment of the present application establish a dam simulation model based on the monitoring data of the dam; calculate the safety factor of the dam at different water levels based on the dam simulation model; determine the water level change of the reservoir based on the optimization target, the current state data of the reservoir, and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation; and control the water level of the reservoir based on the water level change of the reservoir. In this way, by simulating and analyzing the working state of the dam, the reservoir water level control plan can be gradually refined under the guidance of the goals of flood control, power generation, and ecology while ensuring the safety of the working state, and finally obtain the water level control plan for the next water storage period, thereby achieving multi-objective benefit improvement in safety, economy, and ecology. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0065] Figure 1 It is a flow chart of a reservoir water level control method provided in one embodiment of the present application.

[0066] Figure 2 It is a partial flow chart of a reservoir water level control method provided in one embodiment of the present application.

[0067] Figure 3 It is a partial flow chart of a reservoir water level control method provided in one embodiment of the present application.

[0068] Figure 4 It is a structural schematic diagram of a reservoir water level control device provided in one embodiment of the present application.

[0069] Figure 5It is a partial structural schematic diagram of a reservoir water level control device provided in one embodiment of the present application.

[0070] Figure 6 It is a flow chart of a reservoir water level control method provided in one embodiment of the present application.

[0071] Figure 7 It is a structural diagram of a deep Q network model provided in an embodiment of the present application.

[0072] Figure 8 It is a schematic diagram of a finite element model of a dam provided in an embodiment of the present application.

[0073] Fig. 9 It is a schematic diagram of stress simulation results for the 785m and 825m working conditions provided in one embodiment of the present application.

[0074] Fig.10 It is a schematic diagram of displacement simulation results of 785m and 825m working conditions provided in one embodiment of the present application.

[0075] Fig.11 This is a graph of the upstream water level changes of a hydropower station in the next year provided by an embodiment of the present application.

[0076] Fig.12 This is a diagram of upstream water level changes in a hydropower station in a flood year provided by an embodiment of the present application.

[0077] Fig.13 This is a diagram of upstream water level changes in a hydropower station in a dry year provided by an embodiment of the present application.

[0078] Fig.14 It is a structural schematic diagram of a reservoir water level control device provided in one embodiment of the present application.

[0079] Fig.15 It is a schematic diagram of the physical structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. Here, the illustrative embodiments of the present application and their descriptions are used to explain the present application, but are not intended to limit the present application. It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments can be arbitrarily arranged with each other.

[0081] The terms “first”, “second”, etc. used in this document do not specifically refer to an order or sequence, nor are they used to limit this application. They are only used to distinguish elements or operations described with the same technical terms.

[0082] The words “include,” “including,” “have,” “contain,” etc. used in this document are open-ended terms, meaning including but not limited to.

[0083] As used herein, "and / or" includes any and all permutations of the items described.

[0084] The execution subject of the reservoir water level control method provided in this application includes but is not limited to a computer.

[0085] Figure 1 is a flow chart of a reservoir water level control method provided in an embodiment of the present application, such as Figure 1 As shown, the reservoir water level control method provided in the embodiment of the present application includes:

[0086] S101. Establish a dam simulation model based on the monitoring data of the dam;

[0087] In step S101, specifically, the structural design data of the dam (such as height, width, material type, etc.) can be collected, real-time monitoring data (such as stress, displacement, seepage, temperature, etc.) can be obtained, and historical water level and flow data can be collected. The collected data is used to build a three-dimensional mathematical model of the dam in the software, and boundary conditions and load conditions (such as water pressure, earth pressure, etc.) are defined. The simulation results can also be compared and verified with the actual monitoring data to ensure the accuracy of the model.

[0088] S102. Calculate the safety factor of the dam at different water levels according to the dam simulation model;

[0089] In step S102, a simulation is performed based on the dam simulation model to obtain data on the temperature, stress, and displacement of each grid point on the dam as the water level changes. The original data of the simulation is processed to calculate the safety factor of the dam at different water levels.

[0090] S103. Determine the water level change of the reservoir according to the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation.

[0091] In step S103, the current state data of the reservoir is collected, including the current water level, inflow and outflow, rainfall forecast, etc. The safety factor calculated in step S102 is applied, combined with the current state data, to analyze the possible water level changes to meet the optimization goals of ecological flow, flood control, and power generation. Water level management suggestions are put forward to ensure that the reservoir operates within a safe range and prevent risks such as dam breach.

[0092] S104: regulating the water level of the reservoir according to the change in the water level of the reservoir.

[0093] In step S104, the water level of the reservoir is regulated according to the water level change determined in step S103 to meet the needs of ecological flow, flood control, and power generation.

[0094] The reservoir water level control method provided in the embodiment of the present application establishes a dam simulation model based on the monitoring data of the dam; calculates the safety factor of the dam at different water levels based on the dam simulation model; determines the water level change of the reservoir based on the optimization target, the current state data of the reservoir, and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation; and controls the water level of the reservoir based on the water level change of the reservoir. In this way, by simulating and analyzing the working state of the dam, the reservoir water level control plan can be gradually refined under the guidance of the goals of flood control, power generation, and ecology while ensuring the safety of the working state, and finally obtains the water level control plan for the next water storage period, thereby achieving multi-objective benefit improvement in safety, economy, and ecology.

[0095] In some embodiments, the monitoring data includes at least one of the following: reservoir inflow, reservoir outflow, upstream water level data, downstream water level data, rainfall data, temperature data, water consumption data of downstream residents and enterprises, and minimum ecological flow.

[0096] like Figure 2 As shown, in some embodiments, according to the dam simulation model, the safety factor of the dam at different water levels is calculated, including:

[0097] S1021. Calculate the allowable stress and displacement values ​​of all grid points of the dam according to the safety standards and mechanical models to obtain a safety threshold value of each grid point;

[0098] In step S1021, according to the relevant engineering safety standards, it is first necessary to define the safety standards of the dam under different working conditions, including the allowable stress and displacement values. These standards will be determined based on the design specifications and historical data of the dam. Grid point analysis: The dam model is divided into multiple grid points. For each grid point, the stress and displacement values ​​are calculated using a mechanical model (such as finite element analysis), and compared with the allowable stress and displacement values ​​to obtain the safety threshold of each grid point.

[0099] S1022, using the dam simulation model to perform simulation to obtain a water level-displacement relationship and a water level-stress relationship of each grid point of the dam;

[0100] In step S1022, the water level parameters are adjusted, and the water level-displacement relationship and water level-stress relationship are simulated and analyzed for each grid point to obtain corresponding data, indicating how the displacement and stress change when the water level changes.

[0101] S1023. Determine the water level-safety factor relationship of the dam based on the water level-displacement relationship, the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point.

[0102] In step S1023, the water level-displacement relationship and water level-stress relationship obtained in step S1022 are comprehensively calculated with the safety threshold obtained in step S1021 to obtain the overall safety factor of the dam at different water levels, providing a basis for relevant engineering decisions.

[0103] like Figure 3 As shown, in some embodiments, the water level-safety factor relationship of the dam is determined according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, including:

[0104] S10231. Determine a water level-safety score for each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point;

[0105] In step S10231, for each grid point, the safety score of the grid point is determined based on the water level-displacement relationship, the water level-stress relationship and the safety threshold of the grid point.

[0106] S10232. Determine the water level-safety factor relationship of the dam based on the water level-safety score of each grid point.

[0107] In step S10232, for each water level, the safety factor of the dam at the water level is determined according to the safety scores of each grid point of the dam at the water level. According to the safety factor of the dam at each water level, the water level-safety factor relationship of the dam is established.

[0108] In some embodiments, the water level-stress relationship of each grid point of the dam includes at least one of the following: water level-cross-river stress relationship, water level-down-river stress relationship, and water level-vertical stress relationship; the water level-displacement relationship of each grid point of the dam includes at least one of the following: water level-cross-river displacement relationship, water level-down-river displacement relationship, and water level-vertical displacement relationship; the safety threshold of each grid point includes a safety stress threshold and a safety displacement threshold, and the safety stress threshold includes at least one of the following: cross-river stress threshold, down-river stress threshold, and vertical stress threshold; the safety displacement threshold includes at least one of the following: cross-river displacement threshold, down-river displacement threshold, and vertical displacement threshold.

[0109] like Figure 4As shown, in some embodiments, according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, the water level-safety score of each grid point of the dam is determined, including:

[0110] S102311. For each water level, according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam, calculate the displacement and stress of each grid point of the dam at the water level;

[0111] S102312, determining the safety category of each grid point according to the displacement and stress of each grid point of the dam at the water level and the safety threshold of the grid point;

[0112] S102313. Determine the safety score of the grid point of the dam at the water level according to the safety category of each grid point of the dam and the safety score corresponding to each safety category;

[0113] S102314. Determine the water level-safety score of each grid point of the dam according to the safety score of each grid point of the dam at each water level.

[0114] In some embodiments, the water level-safety factor relationship of the dam is determined based on the water level-safety score of each grid point, including: determining the water level-safety factor relationship of the dam based on the water level-safety score of each grid point and the weight of the dam zone to which each grid point belongs; wherein the weight of each dam zone is determined based on the construction and water storage stage of the dam.

[0115] In some embodiments, the water level change of the reservoir is determined based on the optimization objective, the current status data of the reservoir, and the safety factor of the dam at different water levels, including: inputting the current status data of the reservoir into a pre-trained prediction model to obtain the recommended water level change output by the prediction model, wherein the prediction model is trained based on the optimization objective, the current status data of the reservoir, and the safety factor of the dam at different water levels.

[0116] like Figure 5 As shown, in some embodiments, before inputting the current state data of the reservoir into a pre-trained prediction model to obtain the recommended water level change output by the prediction model, the method further includes:

[0117] S104, establishing model training instances, each instance including the initial water level of the reservoir, the current water level and the inflow flow, wherein the inflow flow is randomly generated to simulate the inflow conditions on different days;

[0118] S105. According to the initial water level, current water level and inflow of the reservoir in each instance, the deep Q network model is trained to obtain a trained prediction model, wherein, during the training process, for each instance, the deep Q network model selects and executes a water level change according to a greedy strategy, and calculates the new water level, power generation and reservoir safety factor according to the water level change, then calculates the reward value according to the new water level, power generation and reservoir safety factor, and finally updates the parameters of the network Q according to the new water level and reward value.

[0119] In order to better understand the present application, the reservoir water level control method provided by the present application is described in detail below through a specific embodiment.

[0120] like Figure 6 As shown, the reservoir water level control method provided in this embodiment includes:

[0121] Step 1: Obtain data. Collect hydrological and meteorological data, mainly including reservoir inflow, outflow and upstream and downstream water level data; rainfall data; temperature data; downstream residential and enterprise water consumption data; minimum ecological flow, etc. These data should be referenceable, or can be converted into reference values ​​for the time period to be planned through existing technical means. Corresponding historical data are needed as references for wet years, normal years, and dry years to ensure rationality.

[0122] Step 2: Simulate and determine the safety factor. The stress field calculation model and the seepage field calculation model are obtained by the finite element method, and the temperature, stress, and displacement data of each grid point on the dam are obtained by simulation as the water level changes. The original data of the simulation is processed, and the classification algorithm of machine learning is used to determine whether the state of a certain point is safe or unsafe. By analyzing the key and important parts of the dam, the safety factor weights of each point are obtained and the overall safety factor of the dam is calculated.

[0123] Step 3: Get water level planning based on the deep Q network. Establish a reinforcement learning model for reservoir water level regulation, perform simulation data input, adjust the model logic structure and hyperparameters, and then use actual data to plan water level regulation for a certain period of time to meet the needs of flood control, power generation, and safety. Finally, output the saved data to obtain the reservoir water level regulation strategy, which gives recommended values ​​for the upstream water level of the reservoir every day.

[0124] Specific implementation of step 2:

[0125] 2.1: Ensure the rationality of simulation: Before planning the water level plan, first consider the construction process and water storage conditions of the dam as comprehensively as possible, include the deadweight, temperature, and water pressure load into the model, and clarify the mechanical, thermal, and seepage boundary conditions. This process requires monitoring data of the dam under certain working conditions. After the simulation, the dam body temperature, dam body deformation, and dam body seepage simulation values ​​should be compared with the actual monitoring values. If they do not match, the model should be corrected. If they are relatively consistent, it means that the dam simulation can reasonably predict the changes in the future working properties of the dam.

[0126] 2.2: Calculate the safety threshold. According to the safety standards and mechanical model (i.e., dam simulation model), the allowable stress and displacement values ​​of all grid points are calculated to obtain the limit for judging whether the point is safe or not. The safety stress threshold of point x is Yx, and the safety displacement threshold is Zx.

[0127] 2.3: Perform simulation. Use finite element software to perform simulation and obtain the stress and displacement of the data points that change with the simulation time. Through the time-water level relationship, the water level-stress (f) and water level-displacement relationship (g) of the data points are reflected. Generally, there is more than one category of stress and displacement, such as cross-river stress, downstream stress, and vertical stress, which are recorded as water level-stress relationship A (f1), water level-stress relationship B (f2), water level-stress relationship C (f3), etc. Select a part of the data points in particular, and pay attention to the partition of the data points. The key parts such as the dam heel, dam axis, dam toe, dam rear bridge, corridor, etc. require more detailed data. The partitions of the dam are recorded as Area1, Area2, Area3,...

[0128] 2.4: Obtain the water level - safety factor relationship. Each data point x has a safety score x_goal. For a certain water level H, all points that do not exceed the threshold, that is, f1,2,3 <= Yx, g1,2,3 <= Zx, such points are in the X1 set (safe set); there are points where the threshold is exceeded but the sum of all f1,2,3 and g1,2,3 exceeds only once, such points are in the X2 set (unsafe set); those that exceed the safety threshold two or more times are in the X3 set (dangerous set). The safety score of points in the safe set is 1, the safety score of points in the unsafe set is 0.5, and the score of points in the dangerous set is 0. According to the importance of the positions of different dam sections, the expert scoring method is used to assign corresponding safety weights q_Area_1, q_Area_2, q_Area_3,.... The safety factor of the dam is finally Σxi_goal * q_Area_i * 1 for points in X1 + Σxj_goal * q_Area_j * 0.5 for points in X2 + 0. Thus, the water level - safety factor relationship is obtained, H(s) = S, when the water level is s, the corresponding safety factor is S, 0 < S <= 1. It can be seen that the water level - safety factor is not fixed, so targeted assessments are required at different stages of dam construction and water storage.

[0129] Specific implementation of step 3:

[0130] 3.1: Define functions and basic parameters. Key parameters include the maximum output of the hydropower station, the daily water level limit variation range, etc.; functional relationships include the relationship between reservoir capacity and water level, the relationship between the discharge flow (the discharge flow is subject to the minimum ecological flow constraint) and the downstream water level, the safety factor relationship function (given in step 2), the relationship between the minimum water consumption and the inflow (obtained by fitting through the hydrological data in step 1), the power generation output function P = KQH (the output coefficient K depends on the reservoir, 8.0 - 8.6), etc.

[0131] 3.2: Environment definition. Write an environment definition class: Each instance contains the initial water level (initial_level), the current water level (current_level), and the inflow (q_in). The inflow is randomly generated to simulate the inflow situation on different days. According to the season in which the date is located, the generation range of the inflow is different. The reset method resets the environment to the initial state for starting a new Episode. The step method executes an action (the change in water level), calculates the new water level, power generation, and safety factor, and calculates the reward based on these values.

[0132] 3.3 Actions and Rewards. Actions are based on the change in the current water level, subject to the maximum change limit (change_max) and the water level range limit (not lower than the dead water level, not higher than the maximum reservoir capacity regulation water level). The reward calculation takes into account the power generation output and safety factor, and dynamically adjusts the weight of the two in the reward based on the inflow and current water level. When the inflow and water level exceed a certain threshold, the safety factor accounts for a larger proportion of the reward.

[0133] 3.4 Model definition. The deep Q network model is used for training, such as Figure 7 As shown in the figure, the model uses a neural network to replace the Q value table of traditional Q-learning. It mainly consists of three parts: Q network, Target network and Experience Replay component. The Q neural network is used to train the best state-action value, the Target neural network is used to calculate the Q value corresponding to the action taken in the next state, and Experience Replay is used to interact with the environment and generate data for training the neural network. You can quickly build a DQN model using pytorch.

[0134] The training process of DQN mainly includes the following steps:

[0135] Define state space and action space: First, we need to define the state space of the problem and the available action space. The state space is the reasonable range of the water level, and the action space is based on the change in the current water level. Since DQN can only be used in discrete action space, the water level change is discretized into discrete data of 0.1m.

[0136] Building a neural network: DQN uses a deep neural network to estimate the state-action-value function. This network is usually a convolutional neural network (CNN), which accepts the state as input and outputs the Q value corresponding to each action. Depending on the training results, the network is adjusted appropriately. In this case, a two-layer network is used.

[0137] Define the experience replay buffer: In order to reduce the correlation between samples and improve training efficiency, DQN uses an experience replay buffer to store experience tuples of the agent's interaction with the environment, and randomly extracts samples from previous experience for training. Easy to define in the code.

[0138] Define the update rule of the Q-value function: DQN uses the Q-learning algorithm to update the Q-value function. The update of the Q value is based on the Bellman equation, which updates Q(s,a) by observing the new state and reward. It is easy to define in the code.

[0139] Training process: During the training process, the input of the neural network is the state s, and the output is the score of all actions a. The training of the neural network is an optimization problem. The parameters of the neural network are updated through back propagation and gradient descent. The goal is to minimize the difference between the network output and the target value. See the next step for the detailed process.

[0140] Exploration and Exploitation: During training, DQN uses the ε-greedy strategy to strike a balance between exploration and exploitation. This means that the agent has a certain probability of randomly selecting an action for exploration, and also has a certain probability of selecting the best action at the moment for exploitation. In this case, the default ε-greedy is used and then adjusted.

[0141] 3.5 Training model

[0142] The pseudo code of the DQN algorithm is as follows:

[0143]

[0144] Use default parameters for training. Each epoch will generate loss during training. Use Python code to draw loss curves, average reward, etc. to observe the performance of the model. Adjust batch size, experience playback set size, neural network structure, GAMMA value, learning rate, e_greedy parameter, etc. to finally get a better model performance and take its training result. For the same reservoir, the model adjusted with one year's input data should be able to be used to predict other years.

[0145] 3.6 Application. Input the inflow value and rainfall value for the next forecast period (e.g. one year), and the final output and saved data is the reservoir water level control strategy, which gives the recommended value for the upstream water level of the reservoir every day.

[0146] The pseudo code for step 3 is as follows:

[0147] 1. Define functions and basic parameters, and define the environment

[0148]

[0149]

[0150] 2. Actions, Rewards, and Model Definition

[0151]

[0152]

[0153] 3. Training Model

[0154] train_model(model)

[0155] Read inflow and rainfall data

[0156] initialization

[0157] for day in range(365):

[0158] …

[0159] Output the predicted water level and save the result to a CSV file. The following specific example is used to illustrate the technical solution.

[0160] Taking a large hydropower station as an example, the normal water storage level of the hydropower station is 825m, the flood control limit water level is 785m, the dead water level is 765m, the flood control storage capacity is 7.5 billion m3, the regulation storage capacity is 10.436 billion m3, and the reservoir has annual regulation performance; the installed capacity of the power station is 14,000MW, and the average power generation over many years is 64.095 billion kW·h. It is a giant hydropower project of 10 million kW.

[0161] Investigate hydrological and meteorological data, such as data from a weather station:

[0162]

[0163] The stress field calculation model is the overall model of the entire dam section, and the calculation grid is as follows: Figure 8 The simulation range of the entire dam section is centered on the dam axis, with a total simulation range of 1427m×800m×700m, a total number of computational grid elements of 303958, and a number of nodes of 262528. The dam coordinates are set as the X direction across the river, the Y direction along the river, and the Z direction vertically. The dam concrete design divisions include A, B, and C zones, which are respectively C 180 40. C 180 35. C 180 30Concrete.

[0164] The reservoir was subjected to mechanical simulation at multiple water levels to obtain time-stress and time-displacement data. The data points of the dam were divided into: 7# dam heel, 7# dam axis, 7# dam toe, 8# dam heel, 8# dam axis, 8# dam toe, etc. The weights were determined to be 0.15, 0.12, etc., and the safety factor was expressed as S = a*math.log(s)+b.

[0165] Fig. 9 , Fig.10 The following are simulation examples for the 785m and 825m conditions.

[0166] The long-term calculation is carried out on a daily basis, starting from June 1 of this year and ending on June 1 of the following year.

[0167] Some of the input data is shown in the following table:

[0168] 2021-6-28 6000 2021-6-29 800 2021-6-30 1200 2021-7-1 1200 2021-7-2 1200 2021-7-3 4200 2021-7-4 2600 2021-7-5 2400 2021-7-6 2490 2021-7-7 2100 2021-7-8 3510 2021-7-9 5000 2021-7-10 5000 2021-7-11 2900 2021-7-12 1650

[0169] During model training, the inflow of the day is randomly generated based on the average distribution (800, 10000). The action (water level change) is within the safety limit of 1.5m, and the water level after the action is not lower than the water level corresponding to the dead storage capacity (760m), and the water level does not exceed the mandatory limit of 825m. The power output is calculated using the initial upstream water level and the downstream water level difference, and the output coefficient is 8.3. Calculate the reward reward = ((power_output / c_max) + (safety_coefficient / 2)), where the maximum output c_max = 14000. Adjust the weights of the safety factor and power output in the reward according to the inflow. When the inflow is higher than 5000 and the current water level is higher than 823, the safety factor is more important; when the inflow is higher than 5000 but the current water level is not higher than 823, the power output and the safety factor are equally important; in other cases, the power output is more important.

[0170] The network of QNetwork is as follows:

[0171] def __init__(self):

[0172] super(QNetwork,self).__init__()

[0173] Assume that each episode (round) has a maximum of 30 steps and train 1800 episodes.

[0174] Get as Fig.11 The optimized upstream water level changes (normal water year) are shown.

[0175] In reality, the predicted year should be a normal year. In order to show the usability of the model, the results can also be obtained by using the data input of the wet year and the dry year (see Fig.12 and Fig.13 Obviously, in a wet year, there is enough inflow, and the outflow is also large, but the water level is low to cope with floods; in a dry year, the water level is raised to store water, and power generation must also be taken into account, but the main purpose is to ensure water use (including the minimum ecological flow).

[0176] It can be seen that the reservoir water level control method provided by this application is an organic combination of working performance simulation analysis and multi-objective reservoir scheduling; it takes into account the safety, economy, and ecology of water storage. It can provide guidance for water level control during the dam water storage period and give full play to the economic and social benefits of the reservoir.

[0177] Based on the same inventive concept, the present application also provides a reservoir water level control device, such as Fig.14 As shown, the reservoir water level control device provided in the embodiment of the present application includes:

[0178] The first building module 21 is used to build a dam simulation model according to the monitoring data of the dam;

[0179] A calculation module 22, used for calculating the safety factor of the dam at different water levels according to the dam simulation model;

[0180] The determination module 23 is used to determine the water level change of the reservoir according to the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation;

[0181] The control module 24 controls the water level of the reservoir according to the change in the water level of the reservoir.

[0182] The reservoir water level control device provided in the embodiment of the present application establishes a dam simulation model based on the monitoring data of the dam; calculates the safety factor of the dam at different water levels based on the dam simulation model; determines the water level change of the reservoir based on the optimization target, the current state data of the reservoir, and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation. In this way, by simulating and analyzing the working state of the dam, the reservoir water level control plan can be gradually refined under the guidance of the goals of flood control, power generation, and ecology while ensuring the safety of the working state, and finally obtains the water level control plan for the next water storage period, thereby achieving multi-objective benefit improvement in safety, economy, and ecology.

[0183] In some embodiments, the monitoring data includes at least one of the following: reservoir inflow, reservoir outflow, upstream water level data, downstream water level data, rainfall data, temperature data, water consumption data of downstream residents and enterprises, and minimum ecological flow.

[0184] In some embodiments, the computing module is specifically used for:

[0185] According to the safety standards and mechanical models, the allowable stress and displacement values ​​of all grid points of the dam are calculated to obtain the safety threshold of each grid point;

[0186] The dam simulation model is used to perform simulation to obtain a water level-displacement relationship and a water level-stress relationship of each grid point of the dam;

[0187] The water level-safety factor relationship of the dam is determined based on the water level-displacement relationship, the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point.

[0188] In some embodiments, the calculation module determines the water level-safety factor relationship of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, including:

[0189] Determine a water level-safety score for each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point;

[0190] Based on the water level-safety score of each grid point, the water level-safety factor relationship of the dam is determined.

[0191] In some embodiments, the water level-stress relationship of each grid point of the dam includes at least one of the following: a water level-cross-river stress relationship, a water level-down-river stress relationship, and a water level-vertical stress relationship; the water level-displacement relationship of each grid point of the dam includes at least one of the following: a water level-cross-river displacement relationship, a water level-down-river displacement relationship, and a water level-vertical displacement relationship;

[0192] The safety threshold of each grid point includes a safety stress threshold and a safety displacement threshold. The safety stress threshold includes at least one of the following: a cross-river stress threshold, a downstream stress threshold, and a vertical stress threshold; the safety displacement threshold includes at least one of the following: a cross-river displacement threshold, a downstream displacement threshold, and a vertical displacement threshold.

[0193] In some embodiments, the calculation module determines the water level-safety score of each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, including:

[0194] For each water level, according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam, the displacement and stress of each grid point of the dam at the water level are calculated;

[0195] Determining the safety category of each grid point according to the displacement and stress of each grid point of the dam at the water level and the safety threshold of the grid point;

[0196] Determining the safety score of the grid point of the dam at the water level according to the safety category of each grid point of the dam and the safety score corresponding to each safety category;

[0197] According to the safety score of each grid point of the dam at each water level, a water level-safety score of each grid point of the dam is determined.

[0198] In some embodiments, the calculation module determines the water level-safety factor relationship of the dam according to the water level-safety score of each grid point, including:

[0199] The water level-safety factor relationship of the dam is determined based on the water level-safety score of each grid point and the weight of the dam zone to which each grid point belongs; wherein the weight of each dam zone is determined based on the construction and water storage stage of the dam.

[0200] In some embodiments, the determining module is specifically used to:

[0201] The current state data of the reservoir is input into a pre-trained prediction model to obtain a recommended water level change output by the prediction model, wherein the prediction model is trained based on the state data of the reservoir and the safety factor of the dam at different water levels.

[0202] In some embodiments, the apparatus further comprises:

[0203] The second establishment module is used to establish model training instances, each instance includes the initial water level of the reservoir, the current water level and the inflow flow, wherein the inflow flow is randomly generated to simulate the inflow conditions on different days;

[0204] A training module is used to train the deep Q network model according to the initial water level, current water level and inflow of the reservoir in each instance to obtain a trained prediction model, wherein during the training process, for each instance, the deep Q network model selects and executes a water level change according to a greedy strategy, and calculates the new water level, power generation and reservoir safety factor according to the water level change, then calculates the reward value according to the new water level, power generation and reservoir safety factor, and finally updates the parameters of the network Q according to the new water level and reward value.

[0205] The embodiments of the device provided in the embodiments of the present application can be specifically used to execute the processing flow applied to the above-mentioned various method embodiments. Its functions are not repeated here, and reference can be made to the detailed description of the above-mentioned method embodiments.

[0206] Fig.15 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application is shown in FIG. Fig.15 As shown, the electronic device may include: a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302 and the memory 303 communicate with each other through the communication bus 304. The processor 301 may call the logic instructions in the memory 303 to execute the method described in any of the above embodiments.

[0207] In addition, the logic instructions in the above-mentioned memory 303 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0208] This embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments.

[0209] This embodiment provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program enables the computer to execute the methods provided by the above-mentioned method embodiments.

[0210] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0211] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0212] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0213] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0214] In the description of this specification, the description with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0215] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A reservoir water level control method, It is characterized in that include: Establish a dam simulation model based on the dam monitoring data; According to the dam simulation model, calculate the safety factor of the dam at different water levels; Determine the water level change of the reservoir according to the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation; The water level of the reservoir is regulated according to the change in the water level of the reservoir.

2. The method according to claim 1, It is characterized in that The monitoring data includes at least one of the following: reservoir inflow, reservoir outflow, upstream water level data, downstream water level data, rainfall data, temperature data, water consumption data of downstream residents and enterprises, and minimum ecological flow.

3. The method according to claim 1, It is characterized in that According to the dam simulation model, the safety factor of the dam at different water levels is calculated, including: According to the safety standards and mechanical models, the allowable stress and displacement values ​​of all grid points of the dam are calculated to obtain the safety threshold of each grid point; The dam simulation model is used to perform simulation to obtain a water level-displacement relationship and a water level-stress relationship of each grid point of the dam; The water level-safety factor relationship of the dam is determined based on the water level-displacement relationship, the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point.

4. The method according to claim 3, It is characterized in that Determining the water level-safety factor relationship of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point includes: Determine a water level-safety score for each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point; Based on the water level-safety score of each grid point, the water level-safety factor relationship of the dam is determined.

5. The method according to claim 4, It is characterized in that The water level-stress relationship of each grid point of the dam includes at least one of the following: water level-cross-river stress relationship, water level-down-river stress relationship, and water level-vertical stress relationship; the water level-displacement relationship of each grid point of the dam includes at least one of the following: water level-cross-river displacement relationship, water level-down-river displacement relationship, and water level-vertical displacement relationship; The safety threshold of each grid point includes a safety stress threshold and a safety displacement threshold. The safety stress threshold includes at least one of the following: a cross-river stress threshold, a downstream stress threshold, and a vertical stress threshold; the safety displacement threshold includes at least one of the following: a cross-river displacement threshold, a downstream displacement threshold, and a vertical displacement threshold.

6. The method according to claim 5, It is characterized in that Determining the water level-safety score of each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point includes: For each water level, according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam, the displacement and stress of each grid point of the dam at the water level are calculated; Determining the safety category of each grid point according to the displacement and stress of each grid point of the dam at the water level and the safety threshold of the grid point; Determining the safety score of the grid point of the dam at the water level according to the safety category of each grid point of the dam and the safety score corresponding to each safety category; According to the safety score of each grid point of the dam at each water level, a water level-safety score of each grid point of the dam is determined.

7. The method according to claim 6, It is characterized in that Based on the water level-safety score of each grid point, the water level-safety factor relationship of the dam is determined, including: The water level-safety factor relationship of the dam is determined based on the water level-safety score of each grid point and the weight of the dam zone to which each grid point belongs; wherein the weight of each dam zone is determined based on the construction and water storage stage of the dam.

8. The method according to claim 1, It is characterized in that According to the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels, the water level change of the reservoir is determined, including: The current state data of the reservoir is input into a pre-trained prediction model to obtain a recommended water level change output by the prediction model, wherein the prediction model is trained based on the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels.

9. The method according to claim 8, It is characterized in that Before inputting the current state data of the reservoir into a pre-trained prediction model to obtain the recommended water level change output by the prediction model, the method further includes: Establish model training instances, each instance includes the initial water level of the reservoir, the current water level and the inflow, wherein the inflow is randomly generated to simulate the inflow conditions on different days; According to the initial water level, current water level and inflow of the reservoir in each instance, the deep Q network model is trained to obtain a trained prediction model. During the training process, for each instance, the deep Q network model selects and executes a water level change according to a greedy strategy, and calculates the new water level, power generation and reservoir safety factor according to the water level change, then calculates the reward value according to the new water level, power generation and reservoir safety factor, and finally updates the parameters of the network Q according to the new water level and reward value.

10. A reservoir water level control device, It is characterized in that include: The first building module is used to build a dam simulation model based on the monitoring data of the dam; A calculation module is used to calculate the safety factor of the dam at different water levels according to the dam simulation model; A determination module, used to determine the water level change of the reservoir according to the optimization target, the current state data of the reservoir and the safety factor of the dam at different water levels, wherein the optimization target includes at least one of the following: satisfying ecological flow, flood control, and power generation; The control module is used to control the water level of the reservoir according to the change of the water level of the reservoir.

11. The device according to claim 10, It is characterized in that The monitoring data includes at least one of the following: reservoir inflow, reservoir outflow, upstream water level data, downstream water level data, rainfall data, temperature data, water consumption data of downstream residents and enterprises, and minimum ecological flow.

12. The device according to claim 10, It is characterized in that The calculation module is specifically used for: According to the safety standards and mechanical models, the allowable stress and displacement values ​​of all grid points of the dam are calculated to obtain the safety threshold of each grid point; The dam simulation model is used to perform simulation to obtain a water level-displacement relationship and a water level-stress relationship of each grid point of the dam; The water level-safety factor relationship of the dam is determined based on the water level-displacement relationship, the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point.

13. The device according to claim 12, It is characterized in that The calculation module determines the water level-safety factor relationship of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, including: Determine a water level-safety score for each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point; Based on the water level-safety score of each grid point, the water level-safety factor relationship of the dam is determined.

14. The device according to claim 13, It is characterized in that The water level-stress relationship of each grid point of the dam includes at least one of the following: water level-cross-river stress relationship, water level-down-river stress relationship, and water level-vertical stress relationship; the water level-displacement relationship of each grid point of the dam includes at least one of the following: water level-cross-river displacement relationship, water level-down-river displacement relationship, and water level-vertical displacement relationship; The safety threshold of each grid point includes a safety stress threshold and a safety displacement threshold. The safety stress threshold includes at least one of the following: a cross-river stress threshold, a downstream stress threshold, and a vertical stress threshold; the safety displacement threshold includes at least one of the following: a cross-river displacement threshold, a downstream displacement threshold, and a vertical displacement threshold.

15. The device according to claim 14, It is characterized in that The calculation module determines the water level-safety score of each grid point of the dam according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam and the safety threshold of each grid point, including: For each water level, according to the water level-displacement relationship and the water level-stress relationship of each grid point of the dam, the displacement and stress of each grid point of the dam at the water level are calculated; Determining the safety category of each grid point according to the displacement and stress of each grid point of the dam at the water level and the safety threshold of the grid point; Determining the safety score of the grid point of the dam at the water level according to the safety category of each grid point of the dam and the safety score corresponding to each safety category; According to the safety score of each grid point of the dam at each water level, a water level-safety score of each grid point of the dam is determined.

16. The device according to claim 15, It is characterized in that The calculation module determines the water level-safety factor relationship of the dam according to the water level-safety score of each grid point, including: The water level-safety factor relationship of the dam is determined based on the water level-safety score of each grid point and the weight of the dam zone to which each grid point belongs; wherein the weight of each dam zone is determined based on the construction and water storage stage of the dam.

17. The device according to claim 10, It is characterized in that The determination module is specifically used for: The current state data of the reservoir is input into a pre-trained prediction model to obtain a recommended water level change output by the prediction model, wherein the prediction model is trained based on the state data of the reservoir and the safety factor of the dam at different water levels.

18. The device according to claim 17, It is characterized in that The device also includes: The second establishment module is used to establish model training instances, each instance includes the initial water level of the reservoir, the current water level and the inflow flow, wherein the inflow flow is randomly generated to simulate the inflow conditions on different days; A training module is used to train the deep Q network model according to the initial water level, current water level and inflow of the reservoir in each instance to obtain a trained prediction model, wherein during the training process, for each instance, the deep Q network model selects and executes a water level change according to a greedy strategy, and calculates the new water level, power generation and reservoir safety factor according to the water level change, then calculates the reward value according to the new water level, power generation and reservoir safety factor, and finally updates the parameters of the network Q according to the new water level and reward value.

19. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

20. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

21. A computer program product comprising a computer program stored on a non-transitory computer readable storage medium, It is characterized in that The computer program includes program instructions, and when the program instructions are executed by a computer, the steps of the method according to any one of claims 1 to 9 can be implemented.

Images