Regulation and control method for reducing flooding risk of tail platform of hydropower station

Abstract

The invention discloses a regulation and control method for reducing flooding risks of a tail platform of a hydropower station, and belongs to the technical field of reservoir flood control dispatching. The method aims at solving the technical problems that in the prior art, the flood risk of the tail platform of the hydropower station is evaluated depending on historical flood data and an empirical formula, prediction precision is low, and complex and changeable flood conditions are difficult to adapt. According to the method, the hydropower station engineering characteristic basic data is comprehensively collected, the tail water level prediction model based on the multi-layer perceptron (MLP) is constructed, and the prediction accuracy is improved through a large amount of historical data training; according to different tail water platform elevations, the critical discharge flow is accurately calculated by using the model; according to a typical design flood process, amplifying to different frequencies, and carrying out dynamic flood regulation calculation; finally, a critical tail water elevation table corresponding to different frequencies of flood is drawn, and scientific guidance is provided for actual dispatching operation. The method can effectively reduce the flooding risk of the tailgate platform, improve the flood prevention capability of the hydropower station, and guarantee the safe and stable operation of the hydropower station.

Description

Technical Field

[0001] The present invention relates to the technical field of reservoir flood control and dispatching, and in particular to a control method for reducing the flooding risk of a tailwater platform of a hydropower station. Background Art

[0002] As important water conservancy projects, reservoirs play a vital role in power generation, flood control, irrigation, and water supply. During floods, reservoirs regulate outflow to reduce flood peaks, thereby alleviating flood control pressures on downstream rivers. However, during this process, the safety of hydropower station tailwater platforms has become increasingly prominent, becoming a key factor restricting the safe operation of reservoirs.

[0003] Currently, the inundation risk of hydropower station tailwater platforms is primarily assessed using historical flood data and empirical formulas. This approach relies heavily on past hydrological records and engineering experience. Its core framework is to establish an empirical prediction model by statistically analyzing the relationship between tailwater level changes and outflow rates during historical flood events. The basic principle is to use historical data to predict future tailwater level changes, assuming that future flood events will follow similar patterns to those in the past. However, this approach has significant limitations: its prediction accuracy is highly dependent on the completeness and representativeness of historical data, and it struggles to adapt to the impacts of climate change and human activities on the hydrological cycle.

[0004] While existing technologies can, to a certain extent, provide a reference for flood control and scheduling at hydropower stations, their major problems cannot be ignored. First, low prediction accuracy is a major weakness of existing technologies. Due to the limitations of historical flood data and the complexity and uncertainty of flood events themselves, empirical formulas often struggle to accurately predict future changes in tailwater levels. Second, poor adaptability is also a major challenge facing existing technologies. With climate change and accelerated urbanization, extreme weather events are occurring more frequently, leading to changes in flood characteristics. Empirical formulas based on historical data often struggle to adapt to these new changes.

[0005] Furthermore, existing technologies overlook the complex relationship between a hydropower station's tailrace system's discharge capacity and tailrace level. During floods, rising tailrace levels not only affect the generating efficiency of the turbines but can also weaken the plant's overall discharge capacity and even cause backflow within the tailrace channel and tailrace pipe, further limiting discharge flow. These factors are often overlooked or oversimplified in existing technologies, resulting in ineffective control methods in practice.

[0006] Crucially, due to the mismatch between the calibration standards for the dam and the powerhouse, the power station faces a risk of flooding when it encounters a flood that exceeds its own calibration standards. Currently, the inundation risk assessment method for the tailwater platform of a hydropower station is unable to effectively address this complex situation. A more efficient control method is urgently needed to reduce the inundation risk and ensure that the units can participate in flood discharge under the dam's certified flood conditions, thereby maintaining the original design conditions without changing the dam's discharge conditions or raising the reservoir water level.

[0007] In summary, existing technologies for assessing and controlling the flooding risk of tailwater platforms at hydropower stations are significantly deficient, making them inadequate for the safe operation of current water conservancy projects. Therefore, the present invention aims to provide a more scientific and precise control method to effectively reduce the flooding risk of tailwater platforms at hydropower stations and ensure the safe and stable operation of hydropower stations. Summary of the Invention

[0008] The technical problem to be solved by this invention is to provide a control method for reducing the flooding risk of tailwater platforms at hydropower stations, addressing the technical field of reservoir flood control and scheduling, particularly the flooding risk faced by tailwater platforms during floods. Specifically, existing technologies primarily rely on historical flood data and empirical formulas to assess tailwater platform flooding risk. However, this method suffers from low prediction accuracy and poor adaptability, making it difficult to effectively address the technical challenges of complex and changing flood situations.

[0009] In order to achieve the above objectives, the present invention adopts the following technical solutions: Specifically, the control method for reducing the flooding risk of the tailwater platform of a hydropower station includes the following steps: 1. Collect basic data on hydropower station engineering characteristics: Collect reservoir characteristic water level data to clarify key water level parameters such as flood control high water level, design flood level, and verification flood level; draw water level and storage capacity curves to establish the corresponding relationship between water level and storage capacity; collect historical flood flow process and dam site water level process data to provide a basis for subsequent analysis and prediction.

[0010] 2. Construct a tailwater level prediction model based on a multi-layer perceptron: First, preprocess the collected data, including outlier processing and data normalization, to ensure data quality; then design the network structure of the multi-layer perceptron, determine the number of neurons in the input layer, hidden layer, and output layer, and initialize the network parameters; use the backpropagation algorithm to train the model, and use the gradient descent method to update the network parameters to minimize the error between the predicted value and the actual value; finally, verify the model by dividing it into training and validation sets, and adjust the network structure and training parameters according to the verification results until a satisfactory prediction model is obtained.

[0011] 3. Determine the critical discharge flow rate for the tailwater platform to remain submerged at different elevations: Different tailwater platform sealing elevations are proposed, and the wind and wave extremes are calculated based on local meteorological conditions and reservoir characteristics to obtain the tailwater level after accounting for the wind and wave extremes. The tailwater level-discharge flow relationship curve from the hydropower station design phase is used to find the initial critical discharge flow rate. This flow rate is then verified and adjusted using the tailwater level prediction model until the threshold requirements are met, thereby determining the critical discharge flow rate at different tailwater platform elevations.

[0012] 4. Based on the typical design flood process, the typical design flood process is amplified to different frequencies according to the peak flow ratio amplification method: Select a representative typical design flood process, and according to different flood frequency requirements, the peak flow and flow in each time period of the typical design flood process are amplified according to the peak flow ratio amplification method to obtain flood process lines of different frequencies.

[0013] 5. For flood processes with different frequencies amplified, the starting water level is formulated, and flood control calculations are performed based on the relationship between the tailwater platform elevation and the critical downstream flow rate: the starting water level for flood processes with different frequencies is formulated based on the operation and dispatching rules of the hydropower station and the actual situation; when the reservoir water level is lower than the flood control high water level, flood control dispatching is carried out according to the downstream flood protection objects; when the reservoir water level exceeds the flood control high water level, flood control is first carried out according to open discharge dispatching, and the upstream water inflow and downstream flow are monitored in real time, and dynamic adjustments are made according to the actual situation. When the downstream flow rate is higher than the critical flow rate corresponding to the tailwater platform, controlled discharge dispatching is performed; the flood control calculation process is repeated until the flood process ends and the highest flood control level is determined.

[0014] 6. Based on the flood control calculation results, a table of critical tailwater elevations corresponding to floods of different frequencies is drawn to guide actual scheduling operations: the data such as the highest flood control level, tailwater platform elevation, and critical downstream flow corresponding to flood processes of different frequencies are organized into a table. In actual scheduling operations, the safe downstream flow is determined by querying the table based on the real-time flood process and tailwater platform elevation, thereby effectively reducing the risk of tailwater platform flooding.

[0015] The present invention provides a control method for reducing the flooding risk of the tailwater platform of a hydropower station, which has the following beneficial effects: 1. The present invention solves the problem of flooding risk faced by tailwater platforms of hydropower stations in the field of reservoir flood control and scheduling technology, especially during floods. It overcomes the limitations of the existing technology of relying on historical flood data and empirical formulas for evaluation, which leads to inaccurate predictions and poor adaptability. It can effectively reduce the flooding risk of tailwater platforms of hydropower stations, improve the overall flood control capacity of hydropower stations, and ensure the safe and stable operation of hydropower stations.

[0016] 2. By constructing a tailwater level prediction model based on a multi-layer perceptron (MLP), the present invention achieves accurate prediction of tailwater levels under different operating conditions, providing a scientific basis for flood control scheduling. The model can fully consider multiple influencing factors and accurately predict the tailwater level under different operating conditions, providing a reliable basis for determining the critical discharge flow. Compared with existing technologies, the prediction accuracy is higher.

[0017] 3. The introduction of dynamic flood control calculation and elevation optimization technology enables this invention to calculate flood control for different tailwater platform plugging elevations, assess changes in inundation risk, and determine the optimal elevation to meet flood control needs. This not only improves the safety and reliability of hydropower stations but also provides strong support for the further development of reservoir flood control and scheduling technology. It also addresses the lack of systematic elevation optimization methods in existing technologies.

[0018] 4. By comprehensively considering factors such as the tailwater platform elevation, wind and wave height, and tailwater level prediction results, the present invention can reasonably determine the critical discharge flow rate at which the tailwater platform is not submerged at different elevations. In actual scheduling and operation, the discharge flow rate can be adjusted in a timely manner according to the real-time flood process and the tailwater platform elevation to ensure the safety of the tailwater platform.

[0019] 5. The present invention provides an intuitive and convenient reference for actual dispatching operations by drawing a table of critical tailwater elevations corresponding to floods of different frequencies. Dispatchers can quickly query the table based on real-time flood conditions to determine the safe discharge flow rate, thereby achieving scientific dispatching of hydropower stations and effectively reducing the risk of tailwater platform flooding. At the same time, it solves the problem of lack of intuitive and convenient dispatching guidance methods in the existing technology, and improves the efficiency and scientific nature of dispatching operations.

[0020] 6. This invention utilizes a tailwater level prediction model based on a multi-layer perceptron. Compared to traditional methods, it can more comprehensively consider the various factors affecting tailwater level. This model, using machine learning algorithms, significantly improves the accuracy of tailwater level predictions, providing more reliable data support for scheduling decisions. By accurately predicting tailwater levels, rationally determining critical discharge flows, and scientifically guiding scheduling operations, the risk of flooding at hydropower station tailwater platforms is effectively reduced, ensuring the safe and stable operation of hydropower stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is the technical roadmap of the temporary plugging method of the present invention; Figure 2 This is a diagram showing the calculation process of the critical discharge flow corresponding to the tailwater level of the present invention; Figure 3This is a diagram showing the evaluation effect of the water level prediction model based on the MLP model for Reservoir A in Example 3 of the present invention; Figure 4 This is a diagram of the flood control process at different tailwater platform elevations of Reservoir A in Example 3 of the present invention; Figure 5 This is a diagram showing changes in reservoir water levels at different tailwater platform elevations in Reservoir A in Example 3 of the present invention; Figure 6 This is a diagram of the tailwater level changes at different tailwater platform elevations of Reservoir A in Example 3 of the present invention. DETAILED DESCRIPTION

[0022] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0023] Example 1 See also Figure 1 and Figure 2 A control method for reducing the flooding risk of a tailwater platform of a hydropower station comprises the following steps: This embodiment takes a specific hydropower station as an example to explain in detail the specific implementation process of the control method for reducing the flooding risk of the tailwater platform of the hydropower station according to the present invention.

[0024] Step 1: Collect basic data on hydropower station engineering characteristics Collect the reservoir characteristic water level data of the hydropower station, including the flood control high water level, design flood level, and verification flood level; the water level storage capacity curve to clarify the reservoir storage capacity corresponding to different water levels; historical flood flow process and corresponding dam site water level process data to provide a basis for subsequent model construction and flood control calculations.

[0025] Step 2: Build a tailwater level prediction model based on a multi-layer perceptron (MLP) 1. Data preprocessing: The collected historical flood flow process and dam site water level process data are cleaned, outliers are removed, and normalization is performed to adjust the data range to an appropriate range to facilitate model training. The specific formula is as follows: (1) Where, express t The normalized value of the discharge flow at the moment; for t Discharge flow at all times, is the maximum downstream flow rate, is the minimum downstream flow rate; (2) Where, express t The normalized value of the tail water level at time t; for tTail water level at the moment, is the maximum tailwater level, is the minimum tailwater level; Based on the long series of historical flood discharge and tailwater level data of the hydropower station, the and The mapping matrix: (3) Where, Sequence The normalized value of the downstream flow rate is Sequence The normalized value of the tail water level, n is the sequence length.

[0026] 2. Model Training: Divide the preprocessed data into a training set and a validation set in a ratio of 7:3 or 8:2. Use the multilayer perceptron algorithm to determine the appropriate network structure, including the number of neurons in the input, hidden, and output layers. Use the training set to train the model, adjusting model parameters through a backpropagation algorithm to enable the model to learn the complex nonlinear relationship between flood flow and tailwater level.

[0027] 3. Model Validation: Use the validation set to validate the trained model and calculate prediction error metrics, such as mean squared error (MSE) and mean absolute error (MAE), to evaluate the model's predictive performance. If the model performance does not meet the requirements, adjust the network structure or training parameters and retrain and validate until a satisfactory prediction model is obtained.

[0028] Step 3: Determine the critical discharge flow rate at different elevations to avoid flooding of the tailwater platform 1. Drafting the tailwater platform plugging elevation: Based on the design requirements and actual operation of the hydropower station, different tailwater platform plugging elevations are drafted, as follows: (4) Where, For different tailwater platform elevations, , ; 2. Considering wind and wave height: Based on local meteorological conditions and reservoir characteristics, calculate the wind and wave height values ​​under different wind speeds and wind directions, and add them to the tailwater platform plugging elevation to obtain the tailwater level after considering wind and wave height, as follows: (5) Where, They are different tailwater levels after considering the high wind and waves, ; 3. Use the relationship between tailwater level and flow to calibrate the critical discharge flow: Use the tailwater level and flow relationship curve in the hydropower station design phase to find the relationship between tailwater level and flow. The corresponding flow value is used as the initial critical discharge flow ; Verify the critical discharge flow rate based on the tailwater level prediction model: the initial critical discharge flow rate As input data, it is input into the tail water level prediction model based on multi-layer perceptron to calculate the tail water level . Determine the output water level based on the threshold and water level If the formula (6) is satisfied, the output water level is considered to be and water level The same, then the input discharge flow is the critical discharge flow under the tailwater level ; If the formula (6) is not satisfied, the output water level is considered and water level If they are different, the downstream flow rate is adjusted by trial algorithm. , the adjusted Input the tailwater level prediction model based on the multi-layer perceptron to obtain the new tailwater level , until formula (6) is satisfied, at which point the discharge flow is the critical discharge flow at the tailwater level. , the specific formula is as follows: (6) (7) Where, It is the tailwater level of the hydropower station after deducting the wind and wave height from the tailwater platform elevation; The tailwater level is calculated using the tailwater level prediction model based on the multi-layer perceptron; the initial critical discharge flow ; Critical discharge flow rate under tailwater level ; For judgment and whether the critical threshold is equal; They represent the critical discharge flow corresponding to different tailwater levels.

[0029] Step 4: Based on the typical design flood process, amplify the typical design flood process to different frequencies according to the flood peak flow multiplication method. A representative typical design flood process is selected, and according to different flood frequency requirements, the peak flow and flow in each period of the typical design flood process are amplified according to the peak flow ratio amplification method to obtain flood process lines of different frequencies (flood processes of different frequencies include flood processes of corresponding frequencies such as reservoir design flood standards, reservoir verification flood standards, hydropower station plant design flood standards, and hydropower station plant verification flood standards).

[0030] Step 5: According to the flood process after different frequencies are amplified, the water level is formulated and the flood control calculation is performed based on the relationship between the tailwater platform elevation and the critical discharge flow. 1. Drafting the water level: Drafting the water level for different flood frequencies according to the operation and dispatching rules of the hydropower station and the actual situation. ; 2. Flood control calculation: When the reservoir water level is lower than the flood control high water level, flood control scheduling is carried out according to the downstream flood protection objects. By adjusting the reservoir's flood discharge facilities and controlling the downstream flow, the reservoir water level is raised according to the predetermined scheduling method until the reservoir water level rises to the flood control high water level.

[0031] When the reservoir water level exceeds the flood control high water level, flood control is first implemented according to open discharge scheduling. This involves opening the reservoir's flood discharge facilities as much as possible to discharge flood water at maximum capacity. During this process, upstream water inflow and downstream flow are monitored in real time. If upstream water inflow exceeds the reservoir's discharge capacity, water is discharged according to the discharge capacity. If upstream water inflow is less than the discharge capacity, the downstream flow is adjusted to the upstream water inflow according to the principle of water balance. If the downstream flow exceeds the critical flow corresponding to the tailwater platform, water is discharged according to this critical flow, and the flood control mode shifts from open discharge to controlled discharge.

[0032] Repeat the above flood control calculation process until the flood process ends, and calculate the highest flood control level. The specific calculation formula is as follows: (8) Where, is the outbound flow; For water coming from upstream; is the discharge capacity; is the critical discharge flow, To prevent high water levels from flooding, It is a flood control dispatching method for downstream flood protection objects.

[0033] Step 6: Based on the flood control calculation results, draw a table of critical tailwater elevations corresponding to floods of different frequencies to guide actual dispatching operations. Data such as the maximum flood level, tailwater platform elevation, and critical discharge flow corresponding to flood processes of varying frequencies are tabulated and compiled into a table of critical tailwater elevations for each flood frequency. During actual dispatch operations, this table is consulted based on real-time flood processes and tailwater platform elevations to quickly determine the safe discharge flow rate, effectively mitigating the risk of tailwater platform inundation.

[0034] Example 2 In another preferred embodiment, based on embodiment 1, this embodiment takes into account the impact of climate change on flood processes and hydropower station operations, and dynamically adjusts the control method.

[0035] Step 1: Collect data related to climate change Collect local climate change data, including trends and forecasts of temperature, precipitation, and extreme weather events. Also, collect runoff change data for the basin where the hydropower station is located to analyze the impact of climate change on flood processes.

[0036] Step 2: Update the tailwater level prediction model The tailwater level prediction model based on a multi-layer perceptron was updated based on climate change data and runoff variation data. Historical flood flow and dam site water level data were recollected and organized, and the model was retrained and validated to improve its prediction accuracy under different climate conditions, taking into account the impact of climate change on the relationship between flood flow and tailwater level.

[0037] Step 3: Dynamically adjust critical downstream flow Regularly re-determine the critical discharge flow rate required to prevent tailwater platform inundation at different tailwater platform elevations based on climate change and model updates. Dynamically adjust the critical discharge flow rate to account for changes in flood frequency and intensity caused by climate change, ensuring the safety of tailwater platforms under varying climatic conditions.

[0038] Step 4: Real-time flood process monitoring and dynamic flood control calculation Establish a real-time flood monitoring system to obtain real-time information on reservoir inflows, water levels, and other information. Perform real-time flood control calculations based on the real-time flood process and dynamically adjusted critical discharge flows. When the reservoir water level exceeds the flood control high water level, dynamically adjust the flood control mode and discharge flow based on real-time monitoring data and model predictions to ensure the safe operation of the reservoir.

[0039] Step 5: Regularly evaluate and optimize the control plan Regularly evaluate the effectiveness of control plans and analyze changes in tailwater platform flooding risks under different climatic conditions. Based on the evaluation results, optimize and adjust the control plans to continuously improve the adaptability and effectiveness of the control methods.

[0040] Example 3 In another preferred embodiment, based on Examples 1 and 2, this embodiment describes in detail a specific implementation plan of a control method for reducing the flooding risk of the tailwater platform of a hydropower station.

[0041] The dam calibration of Reservoir A is inconsistent with the calibration standards of the downstream power plant. The technical design of a single project lowers the calibration standards of the plant building and the corresponding tailwater platform elevation. There is a risk of flooding when encountering floods that exceed its own calibration flood standards. Temporary water blocking and sealing measures are needed to ensure that the units can participate in flood discharge under the dam calibration flood conditions, so as to maintain the original design conditions without changing the dam discharge conditions and without raising the reservoir water level.

[0042] The reservoir situation is analyzed using a control method for reducing the flooding risk of the tailwater platform of a hydropower station provided by the present invention, and a temporary plugging elevation and control method for the tailwater platform are provided. The specific steps are as follows: Step 1: Collect basic data on hydropower station engineering characteristics: Table 1 Characteristics of Reservoir A

[0043] Step 2: Based on the historical flood flow process and the corresponding tailwater level process at the dam site, a tailwater level prediction model based on a multi-layer perceptron (MLP) is constructed; Step 2.1 Data preprocessing: Select the outflow flow data and tailwater level data of Reservoir A since its construction, and normalize the water level and flow data; Step 2.2 Model training: The normalized data is usually divided into a training set and a test set in a ratio of 4:1. The multi-layer perceptron model is trained using the training set. The input feature is the historical discharge flow of the hydropower station, and the output feature is the predicted tailwater level. The model was evaluated using the test data, with the mean square error (MSE) and the coefficient of determination (R) set as the model evaluation parameters. 2 (R-squared). If the evaluation result is not good, adjust the parameters and repeat steps 2.2 to 2.3 until the evaluation result reaches a better value. The model achieves good results when the number of nodes is 60, the learning rate is 0.01 and the number of training times is 1000, with an MSE of 0.0080 and R 2 is 0.8622; the evaluation results are as follows Figure 3 .

[0044] Step 3: For different tailwater platform plugging elevations, the tailwater level prediction model based on the multi-layer perceptron is used to calibrate the relationship between tailwater level and flow in the reservoir design phase to determine the critical discharge flow at which the tailwater platform is not submerged at different tailwater platform elevations.

[0045] Step 3.1 Draft the tailwater platform plugging elevation: Figure 4 As shown, the elevation of the tailwater platform at the power station is Based on the consideration The increase in the tailwater level is used to formulate different tailwater level platform elevations: [82, 82.5, 83, 83.5, 84, 84.5, 85] (9) Step 3.2 Considering high wind and wave height: Consider high wind and wave height in combination with the design of hydropower station ,like Figure 5 As shown in the figure, different tailwater levels of hydropower stations are obtained by subtracting the wind and wave height from the proposed different tailwater platform elevations: [81.5, 82, 82.5, 83, 83.5, 84, 84.5] (10) Step 3.3 Determine the critical discharge flow: the tailwater level after subtracting the wind and wave height from the tailwater platform elevation of the hydropower station , using the relationship between tailwater level and flow in the hydropower station design phase, calculate the critical discharge flow ; Then, the critical discharge flow As input data, the tail water level prediction model based on multi-layer perceptron is used to calculate the tail water level , judge the output water level based on the threshold and water level If the formula (6) is satisfied, the output water level is considered to be and water level The same, then the input discharge flow is the critical discharge flow under the tailwater level ; If the formula (6) is not satisfied, the output water level is considered and water level If different, the downstream flow rate is adjusted by trial algorithm , the adjusted Input the tailwater level prediction model based on the multi-layer perceptron to obtain the new tailwater level , until formula (6) is satisfied, at which point the discharge flow is the critical discharge flow at the tailwater level. ,like Figure 6 As shown in the figure, the critical discharge flow corresponding to different tailwater levels is calculated as follows: [92510, 94880, 97220, 99550, 101890, 104230, 106580] (11) Step 4: Based on the typical design flood process, amplify the typical design flood process to different frequencies using the flood peak flow multiplication method: (12) Where, 122475m 3 / s Check the flood peak flow for Reservoir A. 70000m 3 / s This is a typical flood peak flow.

[0046] Step 5: For the flood process after amplification at different frequencies, formulate the water level and perform flood control calculation based on the relationship between the tailwater platform plugging elevation and the critical discharge flow. Specifically: Step 5.1: Draft the water level: In view of the flood process, when the operating water level of Reservoir A reaches the high flood control level, the reservoir will be opened for discharge first. At this time, the tailwater platform of the reservoir needs to consider the risk of flooding, so the water level is drafted. , for flood control high water level.

[0047] Step 5.2 Flood control calculation: Carry out flood control calculation for the tailwater level after considering the high wind and wave corresponding to different tailwater platform elevations. Specifically, when the reservoir water level is lower than the flood control high water level, carry out flood control scheduling according to the downstream flood control protection objects until the reservoir water level rises to the flood control high water level; when the reservoir water level exceeds the flood control high water level, first carry out flood control according to open discharge scheduling. During the open discharge scheduling flood control process, if the upstream water flow is greater than the reservoir discharge capacity, discharge will be carried out according to the discharge capacity: if the upstream water flow is less than the discharge capacity, then according to the principle of water balance, the discharge flow will be carried out according to the discharge capacity. The upstream water flow is discharged; if the discharged flow is greater than the critical flow corresponding to the tailwater platform, the water flow is discharged according to the critical flow, and the flood control mode is switched from open discharge to controlled discharge. The maximum discharged flow and the maximum flood control level are calculated; if the maximum flood control level does not exceed the verified flood level, it indicates that the upstream dam of the reservoir is safe to retain water; if the maximum flood control level exceeds the verified flood level, it indicates that the upstream dam is at risk and the tailwater platform elevation does not meet the flood control requirements. In this case, the downstream tailwater platform elevation needs to be raised, and the next round of flood control calculations is performed based on the next level of tailwater platform elevation: (8) Where, is the outbound flow; For water coming from upstream; is the discharge capacity; is the critical discharge flow, To prevent high water levels from flooding, It is a flood control dispatching method for downstream flood protection objects.

[0048] Step 6: Based on the flood control calculation results, draw a table of critical tailwater elevations corresponding to floods of different frequencies to guide actual dispatching operations.

[0049] Table 2 Critical tailwater levels of floods with different frequencies

[0050] Through the implementation of the three aforementioned embodiments, the present invention provides a scientific and dynamic control method that effectively reduces the impact of flooding on downstream areas. By optimizing the tailwater platform elevation design, the overall flood control capacity of the hydropower station is improved, ensuring the safety and reliability of the project.

[0051] In the preferred solution, the basic data described in Step 1 include the reservoir power station's flood control high water level, design flood level, verification flood level characteristic water level, water level storage capacity curve, historical flood flow process and corresponding dam site water level process; the above settings are all core parameters to ensure the safe operation of the reservoir power station; at the same time, meteorological and hydrological data are further integrated to construct a flood forecasting model to achieve accurate flood prediction and timely scheduling.

[0052] In the preferred solution, Step 2 involves building a tailwater level prediction model based on a multi-layer perceptron (MLP), including data preprocessing, model training, and model validation. This setup aims to leverage the MLP's powerful nonlinear fitting capabilities to accurately capture the complex patterns of tailwater level changes. Data preprocessing ensures data quality, model training optimizes parameter configuration, and model validation verifies the model's generalization capabilities, collectively supporting the reliable construction of the prediction model.

[0053] In the preferred solution, the model training is to divide the normalized data into a training set and a test set according to a set ratio, and use the training set to train the multi-layer perceptron model. The input feature is the normalized historical discharge flow of the hydropower station, and the output feature is the predicted tailwater level. The above settings are intended to improve the model's prediction accuracy for changes in the tailwater level of the hydropower station, by continuously adjusting the model parameters to minimize the prediction error, and using the test set to verify the model performance; at the same time, regularization technology is introduced to prevent overfitting and ensure the generalization ability of the model; finally, the predicted tailwater level result is output to provide decision support for the operation of the hydropower station.

[0054] In the preferred approach, the typical design flood process described in Step 4 is a representative flood process selected based on historical flood data. This setting ensures the accuracy and practicality of the simulation analysis. By using this typical design flood process, the research team can more effectively evaluate the performance of flood control facilities and provide a scientific basis for subsequent engineering design and decision-making.

[0055] In the preferred solution, the flood processes of different frequencies described in Step 5 include flood processes of corresponding frequencies such as the reservoir design flood standard, the reservoir verification flood standard, the hydropower station plant design flood standard, and the hydropower station plant verification flood standard. The above settings can comprehensively cover extreme flood events from low to extremely high frequencies, ensure the robustness of flood control planning and design, effectively respond to flood risks under different probabilities, and protect the safety of people's lives and property and the stable operation of infrastructure.

[0056] In the preferred solution, during the flood control calculation process described in Step 5, when the reservoir water level exceeds the flood control high water level, flood control is first implemented according to open discharge scheduling. If the discharge flow exceeds the critical flow corresponding to the tailwater platform, the flood control mode switches from open discharge to controlled discharge. This setting is intended to ensure reservoir safety and prevent downstream flooding. In controlled discharge mode, the discharge flow is strictly controlled by fine-tuning the gate opening to ensure that it does not exceed the carrying capacity of the tailwater platform, effectively balancing flood control and flood discharge needs.

[0057] In the preferred solution, Step 6 draws a table of critical tailwater elevations corresponding to floods of different frequencies, which is used to quickly determine the safe downstream flow rate based on the real-time flood process and the tailwater platform elevation during actual scheduling operations. The above settings can significantly improve the efficiency and accuracy of reservoir flood control scheduling, ensure that the reservoir can respond quickly under extreme weather conditions, effectively reduce the risk of downstream flood disasters, and protect the lives and property of the people.

[0058] In the preferred scheme, the method also includes regularly updating the tailwater level prediction model to adapt to climate changes and changes in the operating conditions of the hydropower station; the control method is implemented through a computer program and integrated into the flood control and scheduling system of the hydropower station; the above settings can significantly improve the accuracy and flexibility of the flood control and scheduling of the hydropower station, ensure efficient and stable operation in a complex and changing environment, and effectively protect the safety of life and property of the people in the downstream area and the harmony and stability of the ecological environment.

[0059] In summary, this invention proposes a highly targeted control method for reducing the risk of tailwater platform inundation at hydropower stations. This method precisely addresses the critical issue of inundation risk faced by tailwater platforms during floods in the field of reservoir flood control and scheduling technology. Existing technologies for addressing this issue have significant shortcomings, primarily relying on historical flood data and empirical formulas to assess tailwater platform inundation risk. However, this method suffers from low prediction accuracy and poor adaptability, making it difficult to effectively address the technical challenges of complex and changing flood situations.

[0060] The present invention has achieved innovative breakthroughs in multiple key technical links. In terms of tailwater level prediction, the existing technologies mostly use traditional or simple models, while the present invention is the first to construct a tailwater level prediction model based on a multi-layer perceptron. This model can more comprehensively consider the various factors affecting the tailwater level, and greatly improve the accuracy of the prediction with the help of machine learning algorithms, which is significantly different from the existing technology; when determining the critical discharge flow at which the tailwater platform is not submerged, the existing technology does not consider comprehensive factors or the estimation method is rough, while the present invention comprehensively considers the tailwater platform elevation, wind and wave height, and the tailwater level predicted based on the multi-layer perceptron, and determines the critical discharge flow through multi-step calculation and verification. This comprehensive method has not been reported in the existing technology; in addition, the existing technology lacks an intuitive and convenient way to guide the scheduling and operation of hydropower stations under different flood frequencies. The present invention provides a clear reference basis for dispatchers by drawing a table of critical tailwater elevations corresponding to floods of different frequencies, which facilitates quick decision-making in actual operations. This innovative tabular form has not been disclosed in the existing technology.

[0061] From the perspective of innovative value, the present invention introduces the multi-layer perceptron (MLP) machine learning model into hydropower station tailwater level prediction, which is a cross-domain innovative application. Utilizing the powerful learning ability of the MLP, it can more accurately capture the complex relationship between tailwater level and various influencing factors, significantly improving prediction accuracy. This innovative model construction and application method is unprecedented in the existing technology. At the same time, the present invention is not an isolated technical point, but a systematic solution. It covers the complete process from data collection, model construction, critical discharge flow determination to scheduling and operation guidance. Each step is closely linked and together constitutes a technical system that can effectively reduce the risk of tailwater platform flooding. This systematic solution is relatively lacking in the existing technology. More importantly, by accurately predicting the tailwater level, reasonably determining the critical discharge flow, and scientifically guiding scheduling and operation, the present invention can significantly improve the safe and stable operation level of the hydropower station and reduce the risk of tailwater platform flooding. This improvement in practical application value is unmatched by the existing technology, reflecting the technological progress and innovation of the present invention. Moreover, the invention is highly practical and operable, and can bring substantial help to the operation and management of the hydropower station.

Claims

1. A control method for reducing the flooding risk of a tailwater platform of a hydropower station, characterized in that: The following steps are involved: Step 1: Collect basic data on hydropower station engineering characteristics; Step 2: Based on the collected basic data, a tailwater level prediction model based on the multi-layer perceptron MLP is constructed; Step 3: For different tailwater platform elevations, the tailwater level prediction model is used to calibrate the tailwater level-flow relationship in the reservoir design phase to determine the critical discharge flow at which the tailwater platform is not submerged at different elevations; Step 4: Based on the typical design flood process, amplify the typical design flood process to different frequencies according to the flood peak flow multiplication ratio method; Step 5: For the flood process after amplification at different frequencies, the starting water level is formulated, and flood control calculations are performed based on the relationship between the tailwater platform elevation and the critical discharge flow; Step 6: Based on the flood control calculation results, draw a table of critical tailwater elevations corresponding to floods of different frequencies to guide actual scheduling operations.

2. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 1, characterized in that: The basic data described in Step 1 include the flood control high water level of the reservoir power station, the design flood level, the verification flood level characteristic water level, the water level storage capacity curve, the historical flood flow process and the corresponding dam site water level process.

3. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 2, characterized in that: Step 2 describes the construction of a tailwater level prediction model based on a multi-layer perceptron (MLP), including data preprocessing, model training, and model validation steps.

4. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 3, characterized in that: The data preprocessing is to select a long series of hydropower station historical flood process discharge and tailwater level data, and perform normalization processing, as follows: （1）； Where, express t The normalized value of the discharge flow at the moment; for t Discharge flow at all times, is the maximum downstream flow rate, is the minimum downstream flow rate; （2）； Where, express t The normalized value of the tail water level at time t; for t Tail water level at the moment, is the maximum tailwater level, is the minimum tailwater level; Based on the long series of historical flood discharge and tailwater level data of the hydropower station, the and The mapping matrix: （3）； Where, Sequence The normalized value of the downstream flow rate is Sequence The normalized value of the tail water level, n is the sequence length.

5. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 4, characterized in that: The model training is to divide the normalized data into a training set and a test set according to a set ratio, and use the training set to train the multi-layer perceptron model. The input feature is the normalized historical discharge flow of the hydropower station, and the output feature is the predicted tailwater level.

6. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 5, characterized in that: The critical discharge flow rate for the tailwater platform to remain unsubmerged at different elevations as described in Step 3 specifically includes: Step 3.1: Draft the tailwater platform plugging elevation, based on the existing tailwater platform elevation of the power station. Based on the consideration The elevations of different tailwater platforms are proposed to meet the increase in the tailwater level, as follows: （4）； Where, For different tailwater platform elevations, , ; Step 3.2: Consider the high wind and wave heights in the design of hydropower stations , the tailwater level of the hydropower station is calculated by subtracting the wind and wave height from the proposed different tailwater platform elevations, as follows: （5）； Where, They are different tailwater levels after considering the high wind and waves, ; Step 3.3: Determine the critical discharge flow based on the tailwater platform elevation minus the wind and wave height. , using the relationship between tailwater level and flow in the design phase, calculate the initial critical discharge flow ;Will Input the tail water level prediction model based on multi-layer perceptron to get the tail water level ;judge and Does it satisfy the following formula: （6）； Where, It is the tailwater level of the hydropower station after deducting the wind and wave height from the tailwater platform elevation; To calculate the tailwater level using the tailwater level prediction model based on multi-layer perceptron; For judgment and The critical threshold for equality. If satisfied, then is the critical discharge flow at the tailwater level If not satisfied, adjust the discharge flow rate through trial algorithm , the adjusted Input the tailwater level prediction model based on the multi-layer perceptron to obtain the new tailwater level , until formula (6) is satisfied, the discharge flow at this time is the critical discharge flow under the tailwater level ; （7）； Where, They represent the critical discharge flow corresponding to different tailwater levels.

7. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 6, characterized in that: The typical design flood process described in Step 4 is a representative flood process selected based on historical floods.

8. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 7, characterized in that: The different frequency flood processes described in Step 5 include flood processes of corresponding frequencies such as the reservoir design flood standard, the reservoir verification flood standard, the hydropower station plant design flood standard, and the hydropower station plant verification flood standard.

9. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 8, characterized in that: During the flood control calculation process described in Step 5, when the reservoir water level exceeds the flood control high water level, flood control is first carried out according to the open discharge scheduling. If the discharge flow is greater than the critical flow corresponding to the tailwater platform, the flood control mode is switched from open discharge to controlled discharge.

10. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 9, characterized in that: The specific implementation steps of the flood control calculation include: Step 5.1: Draft the water level for the typical design flood after amplification of different frequencies. Step 5.2: When the reservoir water level is below the flood control high water level, flood control scheduling is implemented based on the downstream flood control objects until the reservoir water level rises to the flood control high water level. When the reservoir water level exceeds the flood control high water level, open discharge scheduling is first implemented. If the upstream water inflow exceeds the reservoir's discharge capacity during open discharge, flood discharge is carried out according to the discharge capacity. If the upstream water inflow is less than the discharge capacity, discharge is carried out according to the upstream water flow based on the water balance principle. If the discharge flow exceeds the critical flow of the tailwater platform, flood discharge is carried out according to the critical flow, and the open discharge is switched to controlled discharge. The maximum flood control water level is calculated. If the highest flood control level does not exceed the verified flood level, the upstream dam of the reservoir is safe. If it exceeds the verified flood level, the upstream dam is at risk and the current tailwater platform elevation does not meet flood control requirements. The downstream tailwater platform elevation needs to be raised and flood control calculations need to be re-performed based on the new tailwater platform elevation. The details are as follows: （8）； Where, is the outbound flow; For water coming from upstream; is the discharge capacity; is the critical discharge flow, To prevent high water levels from flooding, It is a flood control dispatching method for downstream flood protection objects.

11. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 10, characterized in that: Step 6 draws a table of critical tailwater elevations corresponding to floods of different frequencies, which is used to quickly determine the safe discharge flow rate according to the real-time flood process and the tailwater platform elevation during actual scheduling operation.

12. A control method for reducing the flooding risk of a tailwater platform of a hydropower station according to claim 11, characterized in that: The method also includes regularly updating the tailwater level prediction model to adapt to climate change and changes in hydropower station operating conditions; the control method is implemented through a computer program and integrated into the hydropower station flood control scheduling system.

Images