Machine learning based pumped storage power plant water energy parameter calculation method and device

Abstract

The present application relates to the technical field of pumped storage, in particular to a pumped storage power station water energy parameter calculation method and device based on machine learning. The method determines the reservoir storage curve coefficient and the corresponding water level data as input and output samples to train a preset machine learning model. The water level calculation model trained in this way can quickly calculate the water energy parameters of the pumped storage power station, avoiding the problem of low calculation efficiency and long time consumption caused by iterative calculation of various optimization models one by one in related technologies. At the same time, when determining the reservoir storage curve coefficient, the reservoir storage curve constraint is considered, which can reduce the calculation amount and avoid the inaccuracy caused by random determination of the coefficient. In addition, the penalty constraint determined by the water level data is added to the loss function used in the model training, which can quickly correct the prediction direction in the training process, and finally output the water level parameters that meet the actual engineering requirements.

Description

Technical Field

[0001] This invention relates to the field of pumped storage technology, specifically to a method and apparatus for calculating hydropower parameters of pumped storage power stations based on machine learning. Background Technology

[0002] Calculating the hydroelectric parameters of a pumped storage power station is a complex technical task. It requires comprehensive consideration of the reservoir capacity of the upper and lower reservoirs, the parameter requirements for stable operation of the generating units, and other factors to find the optimal set of parameters as the basis for the planning and design of the power station.

[0003] Existing methods for optimizing pumped storage hydropower parameters, which seek the optimal solution through iterative calculations, have drawbacks: in the early planning and site selection stage, there is often a large number of pumped storage power stations that need to have their hydropower parameters calculated. Traditional methods can only perform iterative calculations one by one, resulting in low overall calculation efficiency and a long time consumption, which often affects the efficiency of planning and site selection work. Summary of the Invention

[0004] In view of this, the present invention provides a method and apparatus for calculating hydropower parameters of pumped storage power stations based on machine learning, so as to solve the problem that the calculation of hydropower parameters of pumped storage power stations is time-consuming in the prior art.

[0005] In a first aspect, the present invention provides a method for calculating hydropower parameters of a pumped storage power station based on machine learning. The method includes: determining the coefficients of a reservoir capacity curve simulation model using the Monte Carlo method based on reservoir capacity curve constraints and reservoir bottom elevation distribution range; inputting multiple sets of reservoir capacity curves determined by the reservoir capacity curve simulation model into a preset optimization model to obtain hydropower parameters corresponding to each set of reservoir capacity curves; adding penalty constraints to a preset loss function based on water level data in the hydropower parameters to obtain a loss function for the preset machine learning model; training the preset machine learning model with coefficients and reservoir bottom elevation as inputs and water level data in the hydropower parameters as outputs to obtain a water level calculation model; determining the water level data of the pumped storage power station to be built based on the reservoir capacity curve coefficients and the water level calculation model, and calculating the hydropower parameters.

[0006] The machine learning-based method for calculating hydropower parameters of pumped storage power stations provided in this invention trains a pre-set machine learning model by determining the reservoir capacity curve coefficient and the corresponding water level data as input and output samples. The resulting water level calculation model can quickly calculate hydropower parameters for pumped storage power stations, avoiding the problems of low computational efficiency and long processing times caused by iterative calculations using various optimization models in related technologies. Furthermore, the reservoir capacity curve coefficient is considered when determining the coefficient, reducing computational load and avoiding inaccuracies that may result from randomly determined coefficients. Additionally, a penalty constraint determined by water level data is added to the loss function used during model training, enabling the model to quickly correct its prediction direction during training and ultimately output water level parameters that meet the requirements of actual engineering projects.

[0007] In one optional implementation, the coefficients of the reservoir capacity curve simulation model are determined using the Monte Carlo method based on the reservoir capacity curve constraints and the reservoir bottom elevation distribution range. This includes: determining the reservoir bottom elevation distribution range according to the target area digital elevation model; randomly generating the reservoir bottom elevation within the reservoir bottom elevation distribution range and randomly generating coefficients within a preset range; substituting the reservoir bottom elevation and coefficients into the reservoir capacity curve model and determining whether the reservoir capacity curve simulation model satisfies the reservoir capacity curve constraints; and when the reservoir capacity curve constraints are satisfied, using the corresponding coefficients as the coefficients of the reservoir capacity curve simulation model.

[0008] In this invention, randomly generated coefficients are validated based on the reservoir bottom elevation. This ensures that the coefficients accurately adapt to the terrain and engineering requirements, making the reservoir capacity curve simulation model more realistic. This lays a solid foundation for subsequent hydropower parameter calculations and improves the accuracy and reliability of pumped storage power station parameter calculations.

[0009] In one alternative implementation, the storage capacity curve constraint includes storage capacity curve monotonicity constraint, storage capacity curve convexity constraint, and storage bottom volume constraint.

[0010] In this invention, by setting monotonicity constraints, convexity constraints, and bottom volume constraints of the reservoir capacity curve, the determined model can satisfy the topographic and physical laws.

[0011] In one optional implementation, the reservoir bottom elevation distribution range is determined based on the target area digital elevation model, including: using the preset area of ​​the pumped storage power station to be built as the target area, constructing a digital elevation model of the target area; determining the maximum and minimum elevations of the terrain based on the digital elevation model; using the maximum and minimum elevations of the terrain as the distribution range of the reservoir bottom elevation, and within the distribution range, following a uniform probability distribution.

[0012] In this invention, by generating a target region based on geographical scope, the final trained model can better fit the characteristics of the target region.

[0013] In one optional implementation, the reservoir bottom elevation distribution range includes a lower reservoir bottom elevation distribution range and an upper reservoir bottom elevation distribution range. Randomly generating the reservoir bottom elevation within the reservoir bottom elevation distribution range includes: randomly generating the lower reservoir bottom elevation with uniform probability within the lower reservoir bottom elevation distribution range, and randomly generating the upper reservoir bottom elevation with uniform probability within the upper reservoir bottom elevation distribution range; determining the relationship between the generated lower and upper reservoir bottom elevations; when the lower reservoir bottom elevation is greater than the upper reservoir bottom elevation, exchanging the lower and upper reservoir bottom elevations; determining whether the difference between the exchanged upper and lower reservoir bottom elevations meets the applicable head for the pumped storage power station to be built; when it does, using the exchanged upper and lower reservoir bottom elevations as the randomly generated reservoir bottom elevations.

[0014] In this invention, by judging and verifying the generated reservoir bottom elevation, the generated reservoir bottom elevation is made to better fit the actual terrain and engineering needs, ensuring that the upper reservoir bottom elevation is higher than the lower reservoir bottom elevation and meets the applicable head constraint. This provides reasonable and logical basic data for subsequent reservoir capacity curve construction and hydropower parameter calculation, thereby improving the scientificity and reliability of pumped storage power station parameter calculation.

[0015] In one optional implementation, the water level data includes normal storage water level and dead water level. Based on the water level data in the hydropower parameters, a penalty constraint is added to a preset loss function to obtain the loss function of a preset machine learning model. This includes: determining a first penalty constraint based on the relationship between the normal storage water level and the dead water level; determining a second penalty constraint based on the relationship between the dead water level and the reservoir bottom elevation, where the reservoir bottom elevation is determined based on the pumped storage power station to be built; and determining the loss function of the preset machine learning model based on the preset loss function and the first and second penalty constraints.

[0016] In one alternative implementation, the first penalty constraint is expressed by the following formula:

[0017]

[0018] In the formula, This indicates the normal water level of the upper reservoir. This indicates the dead water level of the upper reservoir. This indicates the normal water level of the lower reservoir. Indicates the dead water level of the lower reservoir;

[0019] The second penalty constraint is expressed by the following formula:

[0020]

[0021] In the formula, Z u Indicates the elevation of the bottom of the upper reservoir. This indicates the dead water level of the upper reservoir. Indicates the elevation of the reservoir bottom. This indicates the dead water level of the reservoir.

[0022] In this invention, the penalty constraint is determined based on the relationship between water level data and elevation data in the formula, which can ensure the rationality of water level operation and the matching of water level with terrain.

[0023] Secondly, this invention provides a machine learning-based hydropower parameter calculation device for pumped storage power stations. The device includes: a coefficient determination module, used to determine the coefficients of a reservoir capacity curve simulation model using the Monte Carlo method based on reservoir capacity curve constraints and reservoir bottom elevation distribution intervals; a hydropower parameter optimization module, used to input multiple sets of reservoir capacity curves determined by the reservoir capacity curve simulation model into a preset optimization model to obtain the hydropower parameters corresponding to each set of reservoir capacity curves; a loss function construction module, used to add penalty constraints to a preset loss function based on water level data in the hydropower parameters to obtain the loss function of a preset machine learning model; a training module, used to train the preset machine learning model with coefficients and reservoir bottom elevation as inputs and water level data in the hydropower parameters as outputs to obtain a water level calculation model; and a hydropower parameter calculation module, used to determine the water level data of the pumped storage power station to be built based on the reservoir capacity curve coefficients and the water level calculation model, and to calculate the hydropower parameters.

[0024] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the machine learning-based method for calculating hydropower parameters of pumped storage power stations as described in the first aspect or any corresponding embodiment.

[0025] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the machine learning-based method for calculating hydroelectric parameters of a pumped storage power station as described in the first aspect or any corresponding embodiment.

[0026] Fifthly, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the machine learning-based method for calculating hydroelectric parameters of a pumped storage power station as described in the first aspect or any of its corresponding embodiments. Attached Figure Description

[0027] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0028] Figure 1 This is a flowchart illustrating a machine learning-based method for calculating hydroelectric parameters of a pumped storage power station according to an embodiment of the present invention.

[0029] Figure 2 This is a structural block diagram of a machine learning-based hydropower station hydropower parameter calculation device according to an embodiment of the present invention.

[0030] Figure 3 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] According to an embodiment of the present invention, a method for calculating hydroelectric parameters of a pumped storage power station based on machine learning is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0033] This embodiment provides a machine learning-based method for calculating hydroelectric parameters of pumped storage power stations, which can be used in electronic devices such as computers, mobile phones, and tablets. Figure 1 This is a flowchart of a machine learning-based method for calculating hydroelectric parameters of a pumped storage power station according to an embodiment of the present invention, as shown below. Figure 1 As shown, the process includes the following steps:

[0034] Step S101: Based on the reservoir capacity curve constraints and the reservoir bottom elevation distribution range, the coefficients of the reservoir capacity curve simulation model are determined using the Monte Carlo method. The reservoir capacity curve simulation model is a model representing the relationship between the reservoir's water level and capacity, or a model describing the reservoir capacity curve. This model can be expressed using the following formula:

[0035]

[0036] In the formula, , , as well as The coefficients represent the model, Z represents the water level, and V represents the reservoir capacity. It should be noted that this formula is only an illustrative example of the model; in other embodiments, other formulas can be used as the reservoir capacity curve simulation model according to actual conditions.

[0037] Specifically, in this embodiment, by determining the coefficients of the reservoir capacity curve simulation model, multiple reservoir capacity curves can be simulated using this model. To reduce computational load and avoid inaccuracies that might result from randomly determining the coefficients, reservoir capacity curve constraints and reservoir bottom elevation distribution intervals are set. These reservoir capacity curve constraints can be understood as the physical constraints that the reservoir capacity curve simulation model must satisfy. That is, when determining the model coefficients using the Monte Carlo method and combining them with the reservoir bottom elevation within the distribution interval, the model must satisfy these reservoir capacity curve constraints.

[0038] It should be noted that the Monte Carlo method is a numerical statistical method for random sampling and statistical analysis. In this embodiment, when determining the model coefficients, the reservoir bottom elevation distribution range is used as the sampling boundary for the Monte Carlo method, and the reservoir capacity curve constraint is used as the screening criterion for the Monte Carlo method. This enables the determination of model coefficients that satisfy the reservoir capacity curve constraint within the reservoir bottom elevation distribution range.

[0039] Furthermore, pumped storage power stations consist of an upper reservoir and a lower reservoir, utilizing the head difference between the two reservoirs to generate hydroelectric power. Therefore, the reservoir capacity curve simulation model includes both an upper reservoir capacity curve simulation model and a lower reservoir capacity curve simulation model. This step is necessary to determine the coefficients of both models when determining the model coefficients.

[0040] Step S102 involves inputting multiple sets of reservoir capacity curves determined by the reservoir capacity curve simulation model into a preset optimization model to obtain the hydropower parameters corresponding to each set of reservoir capacity curves. Specifically, after determining the coefficients of the upper reservoir capacity curve simulation model and the lower reservoir capacity curve simulation model, the two models can be used to determine the upper reservoir capacity curve and the lower reservoir capacity curve, respectively. The coefficients of the two models determined using the Monte Carlo method according to the above steps may each include multiple values, thus yielding multiple sets of reservoir capacity curves. Each set of reservoir capacity curves includes one upper reservoir capacity curve (corresponding to one upper reservoir capacity curve simulation model) and one lower reservoir capacity curve (corresponding to one lower reservoir capacity curve simulation model).

[0041] Specifically, this preset optimization model is used to determine the optimal hydropower parameters corresponding to each set of reservoir capacity curves, thereby providing samples for subsequent machine learning training. These hydropower parameters include the normal water level, regulating capacity, and dead water level of the upper and lower reservoirs of the power station; the installed capacity, maximum head, minimum head, rated head, maximum lift, and minimum lift of the power station. For this preset optimization model, calculation models for hydropower parameters in related technologies can be adopted, such as the pumped storage power station hydropower parameter optimization mathematical model that includes the objective function and constraints of the maximum scale criterion, or the pumped storage power station hydropower parameter optimization mathematical model that considers the reserve reservoir capacity, etc. Therefore, the calculation of hydropower parameters can be implemented with reference to related technologies. This embodiment does not limit the specific calculation process of the hydropower parameters.

[0042] Step S103: Based on the water level data in the hydropower parameters, a penalty constraint is added to the preset loss function to obtain the loss function of the preset machine learning model. Specifically, in the machine learning models mentioned in related technologies, the loss function mostly adopts absolute error loss, mean squared error loss, or cross-entropy, etc., to achieve model optimization by measuring the relationship between the true value and the predicted value. In this embodiment, based on the preset loss function (for example, the loss function used in related technologies), the physical constraints of the pumped storage power station are considered, thereby constructing a loss function that includes the preset loss function and the penalty constraint.

[0043] Meanwhile, since the water level data in the hydropower parameters is used as the output when training the preset machine learning model, this embodiment uses the water level data to determine the penalty constraint so that the model can quickly correct the prediction direction during the training process and finally output water level parameters that meet the actual engineering requirements.

[0044] Step S104: Using coefficients and reservoir bottom elevation as inputs and water level data from the hydropower parameters as outputs, a preset machine learning model is trained to obtain a water level calculation model. This preset machine learning model can be a deep learning model such as a neural network model; this embodiment does not specifically limit the type of machine learning model used. Furthermore, since the calculation basis for the hydropower parameters of a pumped storage power station is water level data, this embodiment extracts water level data from the calculated hydropower parameters as output samples for model training. It should also be noted that when calculating hydropower parameters based on multiple sets of reservoir capacity curves in the aforementioned steps, only water level data can be calculated to determine the output samples for model training.

[0045] Specifically, based on the above analysis, a pumped storage power station includes an upper reservoir and a lower reservoir. Therefore, the coefficients here include the coefficients of the upper reservoir capacity curve simulation model, the coefficients of the lower reservoir capacity curve simulation model, the upper reservoir bottom elevation, and the lower reservoir bottom elevation. These upper and lower reservoir bottom elevations are determined from the reservoir bottom elevation distribution range when determining the model coefficients. Water level data includes the normal water level of the upper reservoir, the dead water level of the upper reservoir, the normal water level of the lower reservoir, and the dead water level of the lower reservoir. When training the model, the training process can be referenced from related technologies. Input samples are input into the model to obtain model predictions. The model predictions and output samples (i.e., true values) are substituted into the model's loss function, and the model parameters are adjusted according to the calculated loss function. By repeatedly performing this process—that is, inputting different samples—the trained model, i.e., the water level calculation model, is finally obtained.

[0046] Step S105: Determine the water level data of the pumped storage power station to be built based on the reservoir capacity curve coefficient and water level calculation model, and calculate the hydropower parameters. Specifically, for the pumped storage power station to be built, the corresponding reservoir capacity curve can be determined based on topographic information, thereby fitting the reservoir capacity curve coefficient. This topographic information refers to the topographic information of the pumped storage power station's construction location, which can be obtained through topographic maps, etc. Then, based on the obtained topographic information (including reservoir bottom elevation, etc.), determine the water surface area corresponding to different water levels. Next, use volume algorithms such as the frustum volume method to calculate the volume corresponding to different water levels, thus obtaining the reservoir capacity. This yields multiple water level-reservoir capacity data (which can further generate the reservoir capacity curve). Finally, based on these data, perform polynomial fitting using a reservoir capacity curve simulation model to obtain the reservoir capacity curve coefficient corresponding to the pumped storage power station.

[0047] The determined reservoir capacity curve coefficients and reservoir bottom elevation are then input into the trained water level calculation model to obtain the water level data corresponding to the pumped storage power station. Further, water level parameters are determined based on this data. For example, by combining the normal and dead water levels of the upper and lower reservoirs with the reservoir capacity curve, the reservoir capacity corresponding to each water level can be obtained. Then, based on the water level data (normal and dead water levels of the upper and lower reservoirs), the corresponding reservoir capacities for these two water levels can be looked up on the reservoir capacity curve, and the difference can be used to obtain the regulating capacity. Based on the water level and reservoir capacity, the head and energy storage capacity are calculated, and then the installed capacity is determined, etc. It should be noted that the process of calculating hydropower parameters from water level data can be implemented using relevant technologies, and will not be elaborated upon here.

[0048] The machine learning-based method for calculating hydropower parameters of pumped storage power stations provided in this invention trains a pre-set machine learning model by determining the reservoir capacity curve coefficient and the corresponding water level data as input and output samples. The resulting water level calculation model can quickly calculate hydropower parameters for pumped storage power stations, avoiding the problems of low computational efficiency and long processing times caused by iterative calculations using various optimization models in related technologies. Furthermore, the reservoir capacity curve coefficient is considered when determining the coefficient, reducing computational load and avoiding inaccuracies that may result from randomly determined coefficients. Additionally, a penalty constraint determined by water level data is added to the loss function used during model training, enabling the model to quickly correct its prediction direction during training and ultimately output water level parameters that meet the requirements of actual engineering projects.

[0049] This embodiment provides a method for calculating hydroelectric parameters of a pumped storage power station based on machine learning. The process includes the following steps:

[0050] Step S201: Based on the reservoir capacity curve constraints and the reservoir bottom elevation distribution range, the coefficients of the reservoir capacity curve simulation model are determined using the Monte Carlo method.

[0051] Specifically, step S201 includes:

[0052] Step S2011: Determine the reservoir bottom elevation distribution range based on the target area digital elevation model. Specifically, a digital elevation model refers to a digital model that stores the topographic elevation information of the target area in the form of a regular grid or an irregular triangular network, etc., through digital means. Therefore, the elevation information can be determined from this model to obtain the reservoir bottom elevation distribution range.

[0053] In one optional implementation, step S2011 includes:

[0054] Step a1: Using the preset area of ​​the pumped-storage power station to be built as the target area, a digital elevation model of the target area is constructed. Specifically, the target area can be a large region, such as a province or a region with similar terrain, so that the machine learning model trained subsequently can better fit the characteristics of the target area. In this embodiment, a certain range within the area where the pumped-storage power station to be built is located is used as the target area, such as a range of several hundred kilometers or a thousand kilometers. When a pumped-storage power station is subsequently built within this target area, the water level calculation model trained in this embodiment can be used to determine the water level. However, if the pumped-storage power station is built outside the target area, it is necessary to re-acquire data and train the model according to the method of this embodiment to ensure the accuracy of the calculated hydropower parameters.

[0055] After determining the target area, topographic data of that area can be acquired to construct a digital elevation model. Topographic data can be obtained through publicly available geographic data platforms, aerial remote sensing, or ground-based measurements; this embodiment does not impose specific limitations on this method.

[0056] Step a2: Determine the maximum and minimum terrain elevations based on the digital elevation model. Specifically, the digital elevation model stores the terrain elevation information of the target area, from which the maximum terrain elevation Zmax and minimum terrain elevation Zmin can be extracted.

[0057] Step a3: The maximum and minimum topographic elevations are used as the distribution intervals for the reservoir bottom elevation, and within these intervals, the elevation follows a uniform probability distribution. Specifically, a pumped storage power station includes an upper reservoir and a lower reservoir; therefore, the distribution intervals here include the distribution intervals for the upper and lower reservoir bottom elevations, and within these corresponding intervals, the reservoir bottom elevation is uniformly distributed. That is:

[0058] Lower reservoir bottom elevation Z d : Follows a uniform probability distribution: Z d ~ U(Zmin, Zmax);

[0059] Upper reservoir bottom elevation Z u : Follows a uniform probability distribution: Z u ~ U(Zmin, Zmax).

[0060] Step S2012 involves randomly generating the reservoir bottom elevation within the distribution range and randomly generating coefficients within a preset range. Specifically, since the reservoir bottom elevation follows a uniform probability distribution within the distribution range, values ​​are taken from the distribution range with the same probability when generating the reservoir bottom elevation, resulting in a randomly generated reservoir bottom elevation. For the coefficients of the reservoir capacity curve simulation model, to speed up calculations, the range of coefficient values ​​(i.e., the preset range) can be predetermined, and then the coefficients are generated within this range. For example, A3 can take values ​​between -0.01 and 0.01, A2 between -1 and 1, A1 between -1000 and 1000, and A0 can be randomly selected.

[0061] In an optional implementation, step S222 includes: randomly generating the lower reservoir bottom elevation with uniform probability within the lower reservoir bottom elevation distribution range, and randomly generating the upper reservoir bottom elevation with uniform probability within the upper reservoir bottom elevation distribution range; determining the relationship between the generated lower and upper reservoir bottom elevations; when the lower reservoir bottom elevation is greater than the upper reservoir bottom elevation, exchanging the lower and upper reservoir bottom elevations; determining whether the difference between the exchanged upper and lower reservoir bottom elevations meets the applicable head for the pumped storage power station to be built; when it does, using the exchanged upper and lower reservoir bottom elevations as the randomly generated reservoir bottom elevations.

[0062] Specifically, for reservoir bottom elevations randomly generated within a distribution range, it is necessary to first determine whether they meet relevant requirements. Firstly, since the upper and lower reservoir bottom elevations are randomly generated within the same range, it is possible for the lower bottom elevation to be greater than the upper bottom elevation. In this case, the upper and lower reservoir bottom elevations can be interchanged; that is, the randomly generated lower bottom elevation can be used as the upper bottom elevation, and vice versa. If the upper bottom elevation is greater than the lower bottom elevation, it is necessary to further determine whether the difference between the upper and lower bottom elevations (i.e., the elevation difference) meets the applicable head requirement. In any pumped storage power station, a minimum applicable head Zhmin and a maximum applicable head Zhmax are designed during construction, both of which are greater than 0. For example, in a certain region, Zhmin = 50m and Zhmax = 800m. After calculating the elevation difference, it is necessary to determine whether it is between the minimum applicable head and the maximum applicable head. If it exceeds this range, the elevation of the upper reservoir and the lower reservoir need to be regenerated until the above requirements are met. The upper reservoir elevation and the lower reservoir elevation that meet the above requirements are then used as a set of reservoir bottom elevations.

[0063] Step S2013: Substitute the reservoir bottom elevation and coefficients into the reservoir capacity curve model, and determine whether the reservoir capacity curve simulation model meets the reservoir capacity curve constraints.

[0064] Step S2014: When the storage capacity curve constraint is satisfied, the corresponding coefficient is used as the coefficient of the storage capacity curve simulation model.

[0065] The storage capacity curve constraints include monotonicity constraints, convexity constraints, and bottom volume constraints. Specifically, the monotonicity constraint is expressed as dV / dZ > 0, which can be understood as the storage capacity increasing with Z, meaning the storage capacity curve simulation model formed by the constraint satisfies physical laws. The convexity constraint is expressed as... The constraint can be understood as the increase in reservoir capacity as Z increases, meaning that the reservoir capacity curve simulation model formed by the constraint satisfies the topographic law; the reservoir bottom volume constraint is expressed as V(Z0) = 0, which can be understood as the bottom elevation of each reservoir must be 0.

[0066] Specifically, in this embodiment, a rejection sampling method is used to determine the model coefficients. That is, if a generated coefficient makes the model satisfy the reservoir capacity curve constraint, then the coefficient is accepted; otherwise, the coefficient is rejected. Specifically, when determining the model coefficients, it is necessary to determine the coefficients of the upper reservoir capacity curve simulation model and the lower reservoir capacity curve simulation model separately. When determining the coefficients of the upper reservoir capacity curve simulation model, the randomly generated upper reservoir bottom elevation (as water level) and coefficients are substituted into the model to calculate the reservoir capacity. Then, it is determined whether the relationship between the reservoir capacity and water level satisfies the aforementioned reservoir capacity curve constraint. If it does, the coefficient is accepted; otherwise, it is rejected. Then, the next upper reservoir bottom elevation and coefficient are substituted to continue the judgment, ultimately obtaining multiple sets of upper reservoir capacity curve simulation model coefficients that satisfy the reservoir capacity curve constraint. Furthermore, one set of coefficients can yield one upper reservoir capacity curve, and multiple sets of coefficients can yield multiple upper reservoir capacity curves. Similarly, by substituting the lower reservoir bottom elevation and coefficients into the model, the coefficients of multiple sets of lower reservoir capacity curve simulation models can be determined through the rejection sampling method, and multiple lower reservoir capacity curves can be obtained.

[0067] It should be noted that in the above-mentioned random generation of reservoir bottom elevations, multiple sets of reservoir bottom elevations are randomly generated. In this step, the upper and lower reservoir bottom elevations from a set of elevations can be substituted into the corresponding models to obtain the corresponding coefficients and reservoir capacity curves. Therefore, each set of reservoir bottom elevations can correspond to a set of reservoir capacity curves, and thus multiple sets of reservoir capacity curves can be determined in the end.

[0068] Step S202 involves inputting multiple sets of reservoir capacity curves determined by the reservoir capacity curve simulation model into a preset optimization model to obtain the hydropower parameters corresponding to each set of reservoir capacity curves; for details, please refer to [link to relevant documentation]. Figure 1 Step S102 of the illustrated embodiment will not be described again here.

[0069] Step S203: Add penalty constraints to the preset loss function based on the water level data in the hydropower parameters to obtain the loss function of the preset machine learning model.

[0070] Specifically, step S203 includes:

[0071] Step S2031: Determine the first penalty constraint based on the relationship between the normal water level and the dead water level. Specifically, the first penalty constraint is expressed by the following formula:

[0072]

[0073] In the formula, This indicates the normal water level of the upper reservoir. This indicates the dead water level of the upper reservoir. This indicates the normal water level of the lower reservoir. This indicates the dead water level of the reservoir.

[0074] Step S2032: Determine the second penalty constraint based on the relationship between the dead water level and the reservoir bottom elevation. The reservoir bottom elevation is determined based on the pumped storage power station to be built. Specifically, the second penalty constraint is expressed by the following formula:

[0075]

[0076] In the formula, Z u Indicates the elevation of the bottom of the upper reservoir. This indicates the dead water level of the upper reservoir. Indicates the elevation of the reservoir bottom. This indicates the dead water level of the lower reservoir. It should be noted that when using this penalty constraint during the training of the preset machine learning model, the elevations of the upper and lower reservoir bottoms can be substituted into the topographic information corresponding to the pumped storage power station to be built, which determines the upper and lower reservoir bottom elevations.

[0077] Step S2033: Determine the loss function of the preset machine learning model based on the preset loss function and the first and second penalty constraints. Specifically, when the preset loss function uses Mean Squared Error (MSE), the loss function of the preset machine learning model is expressed by the following formula:

[0078]

[0079]

[0080] In the formula, λ represents the penalty coefficient. By adding a first penalty constraint and a second penalty constraint to the preset loss function, the logical inversion of water level can be avoided, ensuring the rationality of reservoir operation.

[0081] Step S204: Using coefficients and reservoir bottom elevation as inputs, and water level data from the hydropower parameters as outputs, train the preset machine learning model to obtain the water level calculation model; for details, please refer to [link to relevant documentation]. Figure 1 Step S104 of the illustrated embodiment will not be described again here.

[0082] Step S205: Determine the water level data of the pumped storage power station to be built based on the reservoir capacity curve coefficient and water level calculation model, and calculate the hydropower parameters. For details, please refer to [link to relevant documentation]. Figure 1 Step S105 of the illustrated embodiment will not be described again here.

[0083] As a specific application embodiment of the present invention, the method for calculating hydropower parameters of a pumped storage power station based on machine learning is implemented using the following process:

[0084] S1. Determine the target area and acquire its DEM (Digital Elevation Model) data. The target area should be determined in conjunction with terrain features. For example, mountainous terrain and hilly / plain terrain should be treated as different target areas for training and applying machine learning models separately. This ensures that the resulting machine learning models are more closely aligned with the characteristics of the target area. In this way, a generalized model is generated across the entire target area, meaning the machine learning model trained in this embodiment can be used throughout the target area.

[0085] S2. Construct a model for generating upper and lower reservoir capacity curves that considers physical constraints, and use the Monte Carlo method to generate a large number of upper and lower reservoir capacity curve samples. Each set of upper and lower reservoir capacity curves includes the relationship curve between the water level and capacity of the upper reservoir and the relationship curve between the water level and capacity of the lower reservoir. Each set of curves is generated based on the following key parameters: the bottom elevation of the upper reservoir and the bottom elevation of the lower reservoir.

[0086] Specifically, step S2 above includes the following steps:

[0087] S21. Taking into account the terrain features and the characteristics of pumped storage units, the distribution range of reservoir bottom elevation is determined.

[0088] First, terrain feature parameters are extracted based on the digital elevation model (DEM) of the target area, mainly the maximum elevation Zmax and minimum elevation Zmin of the target area.

[0089] Then, determine the distribution range of reservoir bottom elevation:

[0090] Lower reservoir bottom elevation Z d : Follows a uniform probability distribution: Z d ~ U(Zmin, Zmax);

[0091] Upper reservoir bottom elevation Z u : Follows a uniform probability distribution: Z u ~ U(Zmin, Zmax).

[0092] S22, Establish a reservoir capacity curve simulation model.

[0093] The model is represented by the following formula:

[0094]

[0095] The model meets the following constraints:

[0096] 1) Monotonicity constraint of storage capacity curve: dV / dZ>0, which means that the storage capacity curve increases as Z increases.

[0097] 2) Convexity constraint of storage capacity curve: In other words, as Z increases, the increase in storage capacity also increases.

[0098] 3) Reservoir bottom volume constraint: V(Z0) = 0, for each reservoir, the bottom elevation must be 0.

[0099] S23, generate N sets of upper and lower reservoir capacity curve data, where N is a large number, for example, N=100000. Specifically, randomly simulate the upper and lower reservoir capacity curves respectively to obtain the power station's capacity curve. The specific steps for each simulation include:

[0100] S231, Z is randomly generated within the distribution interval. d Z u If Z d >Z u Then swap Z d and Z u ; Calculate Z h = Z u - Z d If Z h If the requirements are not met, regenerate until they are met. Here, the elevation difference Z... h It should satisfy: Zhmin ≤ Z h ≤Zhmax, where Zhmin and Zhmax are the minimum and maximum applicable head under the design and manufacturing level of pumped storage units, respectively, and are both greater than 0. For example, in a certain region, Zhmin = 50m and Zhmax = 800m.

[0101] S232, based on Z d and Z u Four coefficients, A3, A2, A1, and A0, are randomly generated using the rejection sampling method (ensuring that the generated coefficients meet all constraints of the reservoir capacity curve model). The model is then used to generate the reservoir capacity curves for the lower and upper reservoirs, respectively. When simulating the reservoir capacity curves, the four coefficients A3, A2, A1, and A0 can be positive or negative. In practice, a range can be given for these four numbers to speed up the calculation. For example, based on experience, A3 can take values ​​between -0.01 and 0.01, A2 between -1 and 1, A1 between -1000 and 1000, and A0 can be randomly selected.

[0102] S3. For each pumped storage sample in S2, its hydropower parameters are calculated independently using an optimization model based on the reservoir capacity curves of its upper and lower reservoirs.

[0103] S4. Obtain N input and output sample data of pumped storage power stations. Using the input and output data samples, consider physical constraints to improve the artificial neural network, and then train it to obtain a machine learning model for calculating the hydropower parameters of the target area.

[0104] Specifically, step S4 above includes:

[0105] S41, obtain all input and output data. The input samples include multiple sets of samples, each set containing coefficients from the upper reservoir capacity curve simulation model, coefficients from the lower reservoir capacity curve simulation model, and the elevations of the upper and lower reservoir bottoms. For the output samples, the normal water level of the upper reservoir, the dead water level, and the normal water level of the lower reservoir for each power station are obtained from the hydropower parameter calculation results.

[0106] S42, using input and output data samples, the artificial neural network is improved considering physical constraints, and then retrained to obtain a machine learning model for calculating hydropower parameters. During the improvement process, the physical constraints of pumped storage power stations are considered, and the loss function is constructed as shown in the following formula:

[0107]

[0108]

[0109] In the formula, It is the penalty coefficient, which can be set according to the actual situation, for example, it can be set to 1000.

[0110] When using S5, first obtain the coefficients of the polynomials of the upper and lower reservoir capacity curves of the power station to be calculated, input them into the artificial neural network, and obtain the normal water level and dead water level of the upper reservoir, and the normal water level and dead water level of the lower reservoir.

[0111] Specifically, in actual design, the reservoir capacity curve of pumped storage is often calculated based on topographic maps. Therefore, the reservoir capacity curves of the upper and lower reservoirs are first calculated, which are the data of water level and reservoir capacity. Then, polynomials are used to fit the data to obtain several parameters A of the polynomials. 3， A 2， A 1， A 0， B 3， B 2， B 1， B0, and Z d Z u Then, it is input into the machine learning model.

[0112] S6. Based on the above characteristic water levels, calculate the hydropower parameters (installed capacity, head, etc.) of the power station.

[0113] This embodiment also provides a machine learning-based hydroelectric power station hydroelectric parameter calculation device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0114] This embodiment provides a machine learning-based device for calculating hydroelectric parameters of a pumped storage power station, such as... Figure 2 As shown, it includes:

[0115] The coefficient determination module 21 is used to determine the coefficients of the reservoir capacity curve simulation model based on the reservoir capacity curve constraint and the reservoir bottom elevation distribution range using the Monte Carlo method.

[0116] The hydropower parameter optimization module 22 is used to input multiple sets of reservoir capacity curves determined by the reservoir capacity curve simulation model into the preset optimization model to obtain the hydropower parameters corresponding to each set of reservoir capacity curves.

[0117] The loss function construction module 23 is used to add penalty constraints to the preset loss function based on the water level data in the hydropower parameters to obtain the loss function of the preset machine learning model.

[0118] Training module 24 is used to train a preset machine learning model with coefficients and reservoir bottom elevation as inputs and water level data in hydropower parameters as outputs to obtain a water level calculation model.

[0119] The hydropower parameter calculation module 25 is used to determine the water level data of the pumped storage power station to be built based on the reservoir capacity curve coefficient and water level calculation model, and to calculate the hydropower parameters.

[0120] In one optional implementation, the coefficient determination module includes: an interval determination module, used to determine the reservoir bottom elevation distribution interval based on the target area digital elevation model; a random generation module, used to randomly generate the reservoir bottom elevation within the reservoir bottom elevation distribution interval and randomly generate coefficients within a preset interval; a judgment module, used to substitute the reservoir bottom elevation and coefficients into the reservoir capacity curve model and judge whether the reservoir capacity curve simulation model meets the reservoir capacity curve constraints; and a coefficient determination submodule, used to use the corresponding coefficients as the coefficients of the reservoir capacity curve simulation model when the reservoir capacity curve constraints are met.

[0121] In one alternative implementation, the storage capacity curve constraint includes storage capacity curve monotonicity constraint, storage capacity curve convexity constraint, and storage bottom volume constraint.

[0122] In one optional implementation, the interval determination module is specifically used to construct a digital elevation model of the target area, taking the preset area range of the pumped storage power station to be built as the target area; determine the maximum elevation and minimum elevation of the terrain based on the digital elevation model; and use the maximum elevation and minimum elevation of the terrain as the distribution interval of the reservoir bottom elevation, and within the distribution interval, it follows a uniform probability distribution.

[0123] In one optional implementation, the reservoir bottom elevation distribution range includes a lower reservoir bottom elevation distribution range and an upper reservoir bottom elevation distribution range. A random generation module is specifically used to randomly generate the lower reservoir bottom elevation with uniform probability within the lower reservoir bottom elevation distribution range, and to randomly generate the upper reservoir bottom elevation with uniform probability within the upper reservoir bottom elevation distribution range; determine the relationship between the generated lower and upper reservoir bottom elevations; when the lower reservoir bottom elevation is greater than the upper reservoir bottom elevation, swap the lower and upper reservoir bottom elevations; determine whether the difference between the swapped upper and lower reservoir bottom elevations meets the applicable head for the pumped storage power station to be built; when it does, use the swapped upper and lower reservoir bottom elevations as the randomly generated reservoir bottom elevations.

[0124] In one optional implementation, the water level data includes normal storage water level and dead water level. The loss function construction module is specifically used to: determine a first penalty constraint based on the relationship between the normal storage water level and the dead water level; determine a second penalty constraint based on the relationship between the dead water level and the reservoir bottom elevation, wherein the reservoir bottom elevation is determined based on the pumped storage power station to be built; and determine the loss function of a preset machine learning model based on a preset loss function and the first and second penalty constraints.

[0125] In one alternative implementation, the first penalty constraint is expressed by the following formula:

[0126]

[0127] In the formula, This indicates the normal water level of the upper reservoir. This indicates the dead water level of the upper reservoir. This indicates the normal water level of the lower reservoir. Indicates the dead water level of the lower reservoir;

[0128] The second penalty constraint is expressed by the following formula:

[0129]

[0130] In the formula, Z u Indicates the elevation of the bottom of the upper reservoir. This indicates the dead water level of the upper reservoir. Indicates the elevation of the reservoir bottom. This indicates the dead water level of the reservoir.

[0131] Further functional descriptions of the above modules are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0132] This invention also provides a computer device having the above-described features. Figure 2 The device shown is a machine learning-based calculation device for hydroelectric parameters of a pumped storage power station.

[0133] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 3 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 3 Take a processor 10 as an example.

[0134] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0135] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0136] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0137] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0138] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0139] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0140] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0141] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for calculating hydroelectric parameters of a pumped storage power station based on machine learning, characterized in that, The method includes: Based on the constraints of the reservoir capacity curve and the distribution range of reservoir bottom elevation, the coefficients of the reservoir capacity curve simulation model are determined using the Monte Carlo method. Multiple sets of reservoir capacity curves determined by the reservoir capacity curve simulation model are input into the preset optimization model to obtain the hydropower parameters corresponding to each set of reservoir capacity curves. Based on the water level data in the hydropower parameters, a penalty constraint is added to the preset loss function to obtain the loss function of the preset machine learning model; Using the coefficients and reservoir bottom elevation as inputs, and the water level data in the hydropower parameters as outputs, the preset machine learning model is trained to obtain a water level calculation model. The water level data of the pumped storage power station to be built is determined based on the reservoir capacity curve coefficient and the water level calculation model, and the hydropower parameters are calculated.

2. The method according to claim 1, characterized in that, Based on the constraints of the reservoir capacity curve and the distribution range of reservoir bottom elevation, the coefficients of the reservoir capacity curve simulation model are determined using the Monte Carlo method, including: Determine the reservoir bottom elevation distribution range based on the digital elevation model of the target area; Randomly generate the reservoir bottom elevation within the reservoir bottom elevation distribution range, and randomly generate coefficients within a preset range; Substitute the reservoir bottom elevation and coefficients into the reservoir capacity curve model, and determine whether the reservoir capacity curve simulation model meets the reservoir capacity curve constraints. When the storage capacity curve constraint is satisfied, the corresponding coefficients are used as coefficients in the storage capacity curve simulation model.

3. The method according to claim 1, characterized in that, The storage capacity curve constraints include storage capacity curve monotonicity constraints, storage capacity curve convexity constraints, and storage bottom volume constraints.

4. The method according to claim 2, characterized in that, The reservoir bottom elevation distribution range is determined based on the digital elevation model of the target area, including: Using the pre-defined area of ​​the pumped storage power station to be built as the target area, a digital elevation model of the target area is constructed. The maximum and minimum terrain elevations are determined based on the digital elevation model. The maximum and minimum topographic elevations are used as the distribution range of the reservoir bottom elevation, and within the distribution range, it follows a uniform probability distribution.

5. The method according to claim 2, characterized in that, The reservoir bottom elevation distribution range includes a lower reservoir bottom elevation distribution range and an upper reservoir bottom elevation distribution range. Reservoir bottom elevations are randomly generated within this range, including: Within the lower reservoir bottom elevation distribution range, the lower reservoir bottom elevation is randomly generated with uniform probability, and within the upper reservoir bottom elevation distribution range, the upper reservoir bottom elevation is randomly generated with uniform probability. Determine the relationship between the generated lower reservoir bottom elevation and the upper reservoir bottom elevation; When the lower reservoir bottom elevation is greater than the upper reservoir bottom elevation, the lower reservoir bottom elevation and the upper reservoir bottom elevation are swapped. Determine whether the difference between the upper reservoir bottom elevation and the lower reservoir bottom elevation after the exchange meets the applicable head for the pumped storage power station to be built; When the conditions are met, the swapped upper reservoir bottom elevation and lower reservoir bottom elevation are used as the randomly generated reservoir bottom elevation.

6. The method according to claim 1, characterized in that, Water level data includes normal storage water level and dead water level. Based on the water level data in the hydropower parameters, a penalty constraint is added to a preset loss function to obtain the loss function of the preset machine learning model, which includes: The first penalty constraint is determined based on the relationship between the normal water level and the dead water level. The second penalty constraint is determined based on the relationship between the dead water level and the reservoir bottom elevation, wherein the reservoir bottom elevation is determined based on the pumped storage power station to be built. The loss function of the preset machine learning model is determined based on the preset loss function and the first and second penalty constraints.

7. The method according to claim 6, characterized in that, The first penalty constraint is expressed by the following formula: In the formula, This indicates the normal water level of the upper reservoir. Indicates the dead water level of the upper reservoir. This indicates the normal water level of the lower reservoir. Indicates the dead water level of the lower reservoir; The second penalty constraint is expressed by the following formula: In the formula, Z u Indicates the elevation of the bottom of the upper reservoir. Indicates the dead water level of the upper reservoir. Indicates the elevation of the reservoir bottom. This indicates the dead water level of the reservoir.

8. A machine learning-based hydroelectric power station hydroelectric parameter calculation device, characterized in that, The device includes: The coefficient determination module is used to determine the coefficients of the reservoir capacity curve simulation model based on the reservoir capacity curve constraints and the reservoir bottom elevation distribution range, using the Monte Carlo method. The hydropower parameter optimization module is used to input multiple sets of reservoir capacity curves determined by the reservoir capacity curve simulation model into a preset optimization model to obtain the hydropower parameters corresponding to each set of reservoir capacity curves. The loss function construction module is used to add penalty constraints to the preset loss function based on the water level data in the hydropower parameters to obtain the loss function of the preset machine learning model. The training module is used to train the preset machine learning model with the coefficients and reservoir bottom elevation as inputs and the water level data in the hydropower parameters as outputs, so as to obtain the water level calculation model. The hydropower parameter calculation module is used to determine the water level data of the pumped storage power station to be built based on the reservoir capacity curve coefficient and the water level calculation model, and to calculate the hydropower parameters.

9. A computer device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the machine learning-based method for calculating hydropower parameters of a pumped storage power station as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the machine learning-based method for calculating hydroelectric parameters of a pumped storage power station as described in any one of claims 1 to 7.

Images