Short-term optimal scheduling method and device for hydropower station coupled with hydraulic transient characteristics

Abstract

The present application relates to the technical field of power system optimal operation, in particular to a hydropower station short-term optimal scheduling method and device coupled with hydraulic transient characteristics, and the method comprises the following steps: constructing a hydropower station short-term optimal scheduling model with the minimum water consumption as the target in the hydropower station; determining a hydraulic transient characteristic index system based on a pre-constructed power priority simulation control model of the hydropower unit, thereby constructing a surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network; decoupling the hydropower station short-term optimal scheduling model and the surrogate model based on a heuristic algorithm with dynamic constraint update, so as to perform short-term optimal scheduling on the hydropower station and obtain a hydropower station short-term optimal scheduling result coupled with hydraulic transient characteristics. Thus, the problem that the hydropower station short-term optimal scheduling method in the related art needs a large number of iterations to converge, resulting in a very high calculation cost and being unable to meet the actual operation requirements of the rapidly changing hydropower station, and reducing the efficiency and flexibility of the scheduling decision, etc. can be solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system optimal operation, in particular to a short-term optimal scheduling method and device for a hydropower station coupled with hydraulic transient characteristics. BACKGROUND

[0002] The global energy transition has accelerated the development and utilization of new energy such as wind and solar energy. However, the inherent volatility of these new energy sources poses a significant challenge to the stability of the power system. By utilizing the flexibility of hydropower in rapid power response and frequency regulation, the integration of multi-energy complementary systems of hydropower, wind power and solar power provides a promising solution to the above problems. These systems can improve the utilization rate of renewable energy, stabilize power supply and reduce dependence on fossil fuels. In multi-energy complementary systems relying on hydropower regulation, the performance of hydropower units is crucial to the operational efficiency and stability of the entire complementary system.

[0003] Important components of hydropower units, such as spiral cases and draft tubes, are directly affected during the hydraulic transient process triggered by power regulation. For example, excessive pressure in the spiral case during power regulation can cause severe unit vibration, and even lead to equipment damage or endanger personnel safety when exceeding mechanical limits. Ignoring hydraulic transient characteristics in short-term scheduling can lead to unstable operation, equipment wear and even failure. Therefore, incorporating hydraulic transient characteristics into short-term scheduling is key to balancing operational objectives and ensuring safe and reliable operation of hydropower stations.

[0004] In related technologies, the short-term optimal scheduling of hydropower stations coupled with hydraulic transient characteristics mainly has the following deficiencies: Logically, at each time in the optimization scheduling process, the change in unit power command can be input into a dynamic simulation model to calculate transient hydraulic performance indicators such as the maximum pressure in the spiral case. Then, these indicators can be fed back to the scheduling model to form a closed-loop feedback cycle. Although this method can integrate hydraulic transient characteristics into scheduling decisions, it requires explicit power commands at each time step: Explicit data-heuristic algorithms: These methods usually require a large number of iterations to converge, resulting in high computational costs (several years); Implicit data-mathematical solution (Gurobi, etc.): The steps in the solution process are implicit data, not actual mathematical explicit data, which cannot be directly input into the control model for solution.

[0005] Therefore, the short-term optimal scheduling method for hydropower stations in related technologies requires a large number of iterations to converge, resulting in very high computational costs and poor real-time response capability, which cannot meet the actual operational requirements of hydropower stations that change rapidly, reducing the efficiency and flexibility of scheduling decisions, and urgently needs to be solved. SUMMARY

[0006] The application provides a hydropower station short-term optimization scheduling method and device coupled with hydraulic transient characteristics to solve the problem of the high calculation cost and poor real-time response capability of the short-term optimization scheduling method of the hydropower station in the prior art, which cannot meet the actual operation requirements of the fast-changing hydropower station and reduces the efficiency and flexibility of scheduling decisions.

[0007] The first aspect of the application provides a hydropower station short-term optimization scheduling method coupled with hydraulic transient characteristics, comprising the following steps: constructing a hydropower station short-term optimization scheduling model with the minimum water consumption as the target in a target hydropower station; determining a hydraulic transient characteristic index system of a hydropower unit in the target hydropower station based on a pre-constructed power priority simulation control model of the hydropower unit, and constructing a surrogate model of a radial basis function neural network-based simulation control model of the hydropower unit based on the hydraulic transient characteristic index system; decoupling the hydropower station short-term optimization scheduling model and the surrogate model of the radial basis function neural network-based simulation control model of the hydropower unit based on a heuristic algorithm with dynamic constraint updating to perform short-term optimization scheduling on the target hydropower station and obtain a short-term optimization scheduling result of the target hydropower station coupled with hydraulic transient characteristics.

[0008] Optionally, in an embodiment of the application, the determination of the hydraulic transient characteristic index system of the hydropower unit in the target hydropower station based on the pre-constructed power priority simulation control model of the hydropower unit comprises: determining at least one of a maximum flow deviation index, a maximum pressure index of a spiral case, a minimum pressure index of a draft tube, a maximum water level fluctuation index of a surge tank, and a guide vane adjustment mileage index in the hydraulic transient characteristic index system based on the pre-constructed power priority simulation control model of the hydropower unit.

[0009] Optionally, in an embodiment of the application, the decoupling of the hydropower station short-term optimization scheduling model and the surrogate model of the radial basis function neural network-based simulation control model of the hydropower unit based on the heuristic algorithm with dynamic constraint updating comprises: obtaining a short-term scheduling plan of the target hydropower station according to the hydropower station short-term optimization scheduling model; calculating each hydraulic transient characteristic index of each time of the hydropower unit by using the surrogate model of the radial basis function neural network-based simulation control model of the hydropower unit; determining the hydraulic transient performance of the hydropower unit according to the each hydraulic transient characteristic index of the hydropower unit; updating the dynamic ramping rate constraint of the hydropower unit based on the hydraulic transient performance, the short-term scheduling plan, and a preset hydropower unit operation limiting factor to obtain an updated dynamic ramping rate constraint; and decoupling the hydropower station short-term optimization scheduling model and the surrogate model of the radial basis function neural network-based simulation control model of the hydropower unit based on the updated dynamic ramping rate constraint.

[0010] Optionally, in one embodiment of the present invention, the dynamic gradeability constraint is expressed as:

[0011]

[0012] in, for t +1 moment hydroelectric generator j power, for t Shike Hydropower Unit j power, For hydroelectric generator units j The initial gradient, These are the limiting factors for the operation of hydropower units. This represents the number of iterations.

[0013] Optionally, in one embodiment of the present invention, the short-term optimal scheduling model for the hydropower station is expressed as:

[0014]

[0015] in, For hydroelectric generator units j At any moment t Start-stop status; T and N These are the number of optimized scheduling periods and the number of generating units, respectively. For hydroelectric generator units j At any moment t Traffic; and These represent the number of times the unit was started and shut down, respectively. and These represent the flow consumption during unit startup and shutdown, respectively.

[0016] A second aspect of the present invention provides a short-term optimal scheduling device for a hydropower station coupled with hydraulic transient characteristics, comprising: a first construction module for constructing a short-term optimal scheduling model for a target hydropower station with the objective of minimizing water consumption; a second construction module for determining the hydraulic transient characteristic index system of the hydropower units in the target hydropower station based on a pre-constructed power-priority simulation control model of the hydropower units, and constructing a surrogate model of the hydropower unit simulation control model based on the hydraulic transient characteristic index system; and a scheduling module for decoupling the short-term optimal scheduling model of the hydropower station and the surrogate model of the hydropower unit simulation control model based on the radial basis function neural network based on a heuristic algorithm of dynamic constraint update, so as to perform short-term optimal scheduling of the target hydropower station and obtain the short-term optimal scheduling result of the target hydropower station coupled with hydraulic transient characteristics.

[0017] Optionally, in one embodiment of the present invention, the second construction module includes: a first determining unit, used to determine at least one of the following in the hydraulic transient characteristic index system: maximum flow deviation index, volute maximum pressure index, tailrace minimum pressure index, surge tank maximum water level fluctuation index, and guide vane adjustment mileage index, based on the pre-constructed hydropower unit power priority simulation control model.

[0018] Optionally, in one embodiment of the present invention, the scheduling module includes: an acquisition unit, configured to acquire a short-term scheduling plan for the target hydropower station based on the short-term optimal scheduling model of the hydropower station; a calculation unit, configured to calculate each hydraulic transient characteristic index at each moment in the hydropower unit using a surrogate model of the hydropower unit simulation control model based on the radial basis function neural network; a second determination unit, configured to determine the hydraulic transient performance of the hydropower unit based on each hydraulic transient characteristic index in the hydropower unit; an update unit, configured to update the dynamic ramp rate constraint of the hydropower unit based on the hydraulic transient performance, the short-term scheduling plan, and a preset hydropower unit operation constraint factor, to obtain the updated dynamic ramp rate constraint; and a processing unit, configured to decouple the short-term optimal scheduling model of the hydropower station and the surrogate model of the hydropower unit simulation control model based on the updated dynamic ramp rate constraint.

[0019] Optionally, in one embodiment of the present invention, the dynamic gradeability constraint is expressed as:

[0020]

[0021] in, for t +1 moment hydroelectric generator j power, for t Shike Hydropower Unit j power, For hydroelectric generator units j The initial gradient, These are the limiting factors for the operation of hydropower units. This represents the number of iterations.

[0022] Optionally, in one embodiment of the present invention, the short-term optimal scheduling model for the hydropower station is expressed as:

[0023]

[0024] in, For hydroelectric generator units j At any moment t Start-stop status; T and N These are the number of optimized scheduling periods and the number of generating units, respectively. for a hydroelectric unit j at time t flow rate; and respectively the number of start-ups and shut-downs of the unit; and respectively the flow rate consumed by the start-ups and shut-downs of the unit.

[0025] The third aspect of the present application provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the short-term optimal scheduling method of a hydropower station coupled with hydraulic transient characteristics as described in the above embodiments.

[0026] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, wherein the program is executed by a processor to implement the short-term optimal scheduling method of a hydropower station coupled with hydraulic transient characteristics as described above.

[0027] The fifth aspect of the present application provides a computer program product, comprising a computer program, wherein the computer program is executed to implement the short-term optimal scheduling method of a hydropower station coupled with hydraulic transient characteristics as described above.

[0028] The embodiments of the present application can construct a short-term optimal scheduling model of a hydropower station with the minimum water consumption as the target, then determine a hydraulic transient characteristic index system based on a power priority simulation control model of a hydroelectric unit, thereby constructing a surrogate model of the simulation control model of the hydroelectric unit based on a radial basis function neural network, and then decoupling the short-term optimal scheduling model of the hydropower station and the surrogate model to perform short-term optimal scheduling on the hydropower station, so as to obtain a short-term optimal scheduling result of the hydropower station coupled with hydraulic transient characteristics, effectively meeting the actual operation requirements of the rapidly changing hydropower station, and improving the efficiency and flexibility of scheduling decisions. Therefore, the problems in the related art that the short-term optimal scheduling method of the hydropower station needs a large number of iterations to converge, resulting in a very high calculation cost, failing to meet the actual operation requirements of the rapidly changing hydropower station, and reducing the efficiency and flexibility of scheduling decisions are solved.

[0029] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0030] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, taken in conjunction with the accompanying drawings, in which:

[0031] Figure 1A flow chart of a short-term optimal scheduling method of a hydropower station coupled with hydraulic transient characteristics according to an embodiment of the present application is provided.

[0032] Figure 2 A schematic diagram of an established RBF neural network for one specific embodiment of the present application is provided.

[0033] Figure 3 A schematic diagram of a training and verification scheme of an established RBF neural network for one specific embodiment of the present application is provided.

[0034] Figure 4 A load, wind power and photovoltaic output diagram for one specific embodiment of the present application is provided.

[0035] Figure 5 A dynamic simulation result of an established control model for one specific embodiment of the present application is provided.

[0036] Figure 6 A comparison result of a coupling model and a traditional model for one specific embodiment of the present application is provided.

[0037] Figure 7 An iteration result of a heuristic algorithm based on dynamic constraint updating for one specific embodiment of the present application is provided.

[0038] Figure 8 A short-term scheduling result of a hydropower station coupled with hydraulic transient characteristics for one specific embodiment of the present application is provided.

[0039] Figure 9 A structural schematic diagram of a short-term optimal scheduling device of a hydropower station coupled with hydraulic transient characteristics according to an embodiment of the present application is provided.

[0040] Figure 10 A structural schematic diagram of an electronic device according to an embodiment of the present application is provided. DETAILED DESCRIPTION

[0041] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.

[0042] A method and device for short-term optimal scheduling of a hydropower station coupling hydraulic transient characteristics are described below with reference to the accompanying drawings. The short-term optimal scheduling method of the hydropower station in the related art mentioned above requires a large number of iterations to converge, resulting in a very high calculation cost and poor real-time response capability, which cannot meet the actual operation requirements of the rapidly changing hydropower station and reduces the efficiency and flexibility of scheduling decisions. To solve the above problems, the present application provides a method for short-term optimal scheduling of a hydropower station coupling hydraulic transient characteristics. In this method, a short-term optimal scheduling model of the hydropower station with the minimum water consumption as the target can be constructed. Then, based on the power priority simulation control model of the hydropower unit, the hydraulic transient characteristic index system is determined to construct the surrogate model of the simulation control model of the hydropower unit based on the radial basis function neural network. Then, the short-term optimal scheduling model of the hydropower station and the surrogate model are decoupled to perform short-term optimal scheduling of the hydropower station and obtain the short-term optimal scheduling result of the hydropower station coupling hydraulic transient characteristics, which effectively meets the actual operation requirements of the rapidly changing hydropower station and improves the efficiency and flexibility of scheduling decisions.

[0043] Specifically, Figure 1 A flowchart of a method for short-term optimal scheduling of a hydropower station coupling hydraulic transient characteristics is provided.

[0044] As Figure 1 shown, the method for short-term optimal scheduling of a hydropower station coupling hydraulic transient characteristics includes the following steps:

[0045] In step S101, a short-term optimal scheduling model of the hydropower station with the minimum water consumption as the target is constructed.

[0046] In the present embodiment, the target hydropower station is the hydropower station that needs to be short-term optimized and scheduled.

[0047] It can be understood that the present embodiment can construct a short-term optimal scheduling model of the hydropower station with the minimum water consumption as the target, for example, first set the objective function of the short-term optimal scheduling model of the hydropower station, wherein the short-term scheduling of the hydropower station aims at economic scheduling:

[0048]

[0049] wherein, is the power of the hydropower unit j at time t t; and T and Nrespectively, are the number of optimized dispatching periods and the number of units; for the hydropower unit j at time t ; and respectively, are the number of start-up and shut-down times of the unit; and respectively, are the start-up and shut-down flow rates of the unit.

[0050] Further, the constraint conditions of the short-term optimization scheduling model of the hydropower station are set, which are as follows:

[0051] The power balance constraint is set. Due to the non-storability of electricity, the power supply and load demand should be balanced at all times. This constraint takes into account the influence of wind power and photovoltaic power, that is:

[0052]

[0053] wherein, and respectively, are the output of wind power and photovoltaic power; and respectively, are t the output power and load of the hydropower unit at time j .

[0054] The water balance constraint of the hydropower station is set. The change of reservoir capacity is related to the runoff, power generation flow and abandoned water flow, that is:

[0055]

[0056] wherein, and respectively, are the reservoir capacities at time t and t -1 of the hydropower station; and respectively, are the inflow and outflow of the reservoir at time t ; is the time interval.

[0057] The abandoned wind and light rate constraint is set. The abandoned electricity rate of wind power and photovoltaic power should be kept within a certain range, that is:

[0058]

[0059]

[0060] wherein, and respectively, are the predicted output of wind power and photovoltaic power (MW) at time t ; and respectively, are the maximum allowable wind curtailment rate and light curtailment rate.

[0061] The reservoir capacity constraint is set, i.e. the reservoir of the hydropower station should be within a certain reservoir capacity range, i.e.

[0062]

[0063] wherein, and are the minimum and maximum reservoir capacities, respectively.

[0064] The output limit constraint of the hydropower unit is set, i.e.

[0065]

[0066]

[0067]

[0068] wherein, P min and P max are the minimum and maximum hydropower unit outputs, respectively. is the efficiency of the hydropower unit at the time point t, j is the net water head of the hydropower unit at the time point t. t The head limit constraint of the hydropower unit is set, i.e. t j

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075] wherein, and are the reservoir water level and tailrace water level of the hydropower station at the time point t, t , and are the total water head loss, the along-the-way water head loss and the local water head loss of the hydropower unit at the time point t, is the functional relationship between the flow and the tailrace water level, t j ​​​​​​​The function relationship between the water level and the reservoir capacity.

[0076] Setting the minimum start-stop time constraint of the hydroelectric generating set:

[0077]

[0078] Wherein, and are the cumulative start-up and shut-down time (h) of the hydroelectric generating set, j respectively. and are the cumulative minimum start-up and shut-down time (h) of the hydroelectric generating set, j respectively.

[0079] In step S102, based on the pre-constructed power priority simulation control model of the hydroelectric generating set, the hydraulic transient characteristic index system of the hydroelectric generating set in the target hydropower station is determined, and based on the hydraulic transient characteristic index system, the surrogate model of the simulation control model of the hydroelectric generating set based on the radial basis function neural network is constructed.

[0080] In the embodiment of the present application, a power priority simulation control model of the hydroelectric generating set is first constructed, which includes three subsystems, namely a hydraulic subsystem, a mechanical subsystem and an electrical subsystem.

[0081] Among them, the simulation of the hydraulic subsystem is the key of the transition process simulation model based on power control, mainly including the modeling of the water turbine and the water channel. For the water turbine, the present application mainly focuses on the application of the Francis turbine which is more common, and its modeling can be divided into two aspects: (1) for small power fluctuation conditions, the water turbine can be linearized according to its characteristic curve. However, this method is not suitable for large power fluctuation conditions, because under large power fluctuation conditions, the water turbine will deviate from the initial operating point; (2) for large power regulation conditions, a nonlinear water turbine model is constructed by interpolation based on the characteristic curve of the water turbine, and the functional relationship of the basic unit parameters in the characteristic curve is as follows:

[0082]

[0083] Wherein, are the unit speed, unit torque and unit flow, respectively; are the speed, torque and flow, respectively; is the working water head of the water turbine; D 1 is the diameter of the water turbine.

[0084] For the modeling of the hydraulic pipeline of the hydropower station, the unsteady flow in the transition process can be represented by the continuity equation and the momentum equation, as shown in the following formula:

[0085]

[0086]

[0087] wherein, A is the pipe area; G is the gravitational constant; α is the angle between the pipe and the horizontal line; is the distance along the pipe axis; is the gravitational acceleration; is the water hammer wave speed in the pipe; is the pipe diameter; then, by the transfer function method, the hydraulic pipe under multiple pipe connections is modeled.

[0088] Secondly, the mechanical subsystem mainly refers to the speed governor, and the key is a power control mode, which applies a PI controller, in which the input is the actual output power and the target power, and the difference signal between the actual output power and the target power is controlled by the PI controller.

[0089] Thirdly, for the electrical subsystem, considering that the hydropower station is operated in a grid-connected mode, the rotor speed of the synchronous generator changes the least during power disturbance and can be ignored. Therefore, under the grid-connected condition, the model of the hydropower station in the power regulation control mode can be simplified by ignoring the rotor motion equation of the synchronous generator. The output power is calculated by the following formula:

[0090]

[0091] wherein, m and n are the torque and the speed, respectively; η is the generator efficiency.

[0092] It can be understood that the embodiment of the present application can determine the hydraulic transient characteristic index system of the hydropower unit in the hydropower station based on the above-mentioned constructed power priority simulation control model of the hydropower unit, and specifically, for the entire transition process of the hydropower unit, the performance of the pipe, the spiral case, the draft tube, the surge chamber and the guide vane can represent the performance of the unit transition process from multiple dimensions, such as safety, stability and wear. These characteristic indexes are obtained from the above-mentioned dynamic power priority simulation control model of the hydropower unit, so that based on the hydraulic transient characteristic index system, a proxy model of the simulation control model of the hydropower unit based on the radial basis function neural network can be constructed to improve the efficiency of the scheduling decision.

[0093] In one embodiment of the present application, based on the pre-constructed power priority simulation control model of the hydroelectric generating set, the hydraulic transient characteristic index system of the hydroelectric generating set in the target hydropower station is determined, including: based on the pre-constructed power priority simulation control model of the hydroelectric generating set, at least one of the maximum flow deviation index, the maximum spiral case pressure index, the minimum tailrace pipe pressure index, the maximum surge chamber water level fluctuation index and the guide vane adjustment mileage index in the hydraulic transient characteristic index system is determined.

[0094] In actual execution process, for the entire transition process of the hydroelectric generating set, the performance of the pipe, the spiral case, the tailrace pipe, the surge chamber and the guide vane can characterize the performance of the unit transition process from multiple dimensions, such as safety, stability and wear.

[0095] The maximum flow deviation index is the deviation between the transient extreme flow value and the stable flow value in the power regulation stage, which reflects the stability of the hydroelectric generating set, that is:

[0096]

[0097] The maximum flow deviation index is the deviation between the transient extreme flow value and the stable flow value in the power regulation stage, which reflects the stability of the hydroelectric generating set, that is: The maximum flow deviation index is the deviation between the transient extreme flow value and the stable flow value in the power regulation stage, which reflects the stability of the hydroelectric generating set, that is:

[0098] The maximum spiral case pressure index: the spiral case pressure change will affect the unit oscillation, and when the maximum pressure of the spiral case under the condition of large flow exceeds the allowed upper limit, it will cause damage to the unit, and even endanger the safety of the hydropower station personnel. Therefore, based on the spiral case water hammer pressure change data in the dynamic simulation process, the maximum pressure is taken as the index to reflect the safety and stability of the hydroelectric generating set, that is:

[0099]

[0100] The maximum spiral case pressure index: the spiral case pressure change will affect the unit oscillation, and when the maximum pressure of the spiral case under the condition of large flow exceeds the allowed upper limit, it will cause damage to the unit, and even endanger the safety of the hydropower station personnel. Therefore, based on the spiral case water hammer pressure change data in the dynamic simulation process, the maximum pressure is taken as the index to reflect the safety and stability of the hydroelectric generating set, that is:

[0101] The minimum tailrace pipe pressure index: when the minimum pressure of the tailrace pipe is low, the tailrace pipe may appear cavitation or even liquid column separation phenomenon, which will seriously cause the unit to lift. Therefore, in the dynamic simulation process, the minimum pressure index of the tailrace pipe is obtained by collecting the pressure data of the tailrace pipe, which reflects the safety and stability of the hydroelectric generating set, that is:

[0102]

[0103] The minimum tailrace pipe pressure index: when the minimum pressure of the tailrace pipe is low, the tailrace pipe may appear cavitation or even liquid column separation phenomenon, which will seriously cause the unit to lift. Therefore, in the dynamic simulation process, the minimum pressure index of the tailrace pipe is obtained by collecting the pressure data of the tailrace pipe, which reflects the safety and stability of the hydroelectric generating set, that is:

[0104] ​​​​​​The maximum water level fluctuation index of the surge chamber is the maximum range of water level fluctuation in the surge chamber caused by the change of flow or pressure under the transition process condition, and reflects the safety and stability of the pressure conduit of the hydroelectric generating set, that is:

[0105]

[0106] wherein, is the maximum water level fluctuation index of the surge chamber; and are the pressure extreme value and the stable value of the surge chamber respectively.

[0107] The guide vane adjustment mileage index: the guide vane adjustment mileage represents the movement of the water turbine in response to the power adjustment, and can represent the wear of the generating set in the adjustment process, that is:

[0108]

[0109] wherein, is the guide vane adjustment mileage index, GVO is the guide vane opening degree.

[0110] In addition, in the embodiment of the present application, the RBF (Radial Basis Function) neural network model is established as a proxy model of the numerical simulation process in the power adjustment control model to improve the efficiency and provide a basis for the research on the optimal scheduling strategy of the hydropower station considering the hydraulic transient characteristics. The RBF neural network is an artificial neural network characterized by its simple structure and strong approximation ability. In the hidden layer, the radial basis function is used as the activation function, so that the complex nonlinear relationship can be modeled efficiently. The RBF network has the advantages of fast training, universal approximation characteristics, and robustness to noisy data, and is widely used in function approximation, pattern recognition, and optimization and control systems. It is worth mentioning that the radial basis function neural network has the unique property of obtaining its specific analytical expression after training, and its interpretability is significantly enhanced compared with many other neural network models. Therefore, the present application uses this network as a proxy model of the simulation control model described above.

[0111] As shown in Figure 2 , the RBF neural network used in the present application is a three-layer feedforward network, including an input layer, a hidden layer and an output layer. The RBF neural network has the structural characteristics of multiple inputs and multiple outputs: (1) in the power adjustment process, the initial power, the target power, the upper reservoir water level and the downstream tail water level constitute a four-dimensional input vector of the network; (2) in the dynamic transition process triggered by the power adjustment, five indexes (the maximum water level fluctuation index of the surge chamber, the maximum pressure fluctuation index of the surge chamber, the maximum flow fluctuation index of the surge chamber, the guide vane adjustment mileage index and the guide vane adjustment mileage index) constitute the output vector of the network. L 1-5The five-dimensional output vector constitutes the RBF neural network; the number of nodes in the input layer and the output layer are equal to the dimensions of the input vector and the output vector, respectively.

[0112] In addition, the training strategy of RBF neural network mainly includes the following three key steps: (1) Center selection: Greedy selection method is used to determine RBF centers. This method can effectively reduce the number of centers and avoid the network from becoming too complex; (2) Output layer training: Linear least squares method is used to train the weights of the output layer. This is a supervised learning method. The output layer has a linear structure, and the goal is to minimize the mean square error between the output value and the target value; (3) Determination of expansion factor: Heuristic method is used to set the expansion factor of the network to control the size of the receptive field of each neuron.

[0113] Prediction flowchart as follows Figure 3 As shown, the dataset originates from the preprocessing of the hydropower unit power-priority simulation control model described above. Under different upstream reservoir water levels and downstream tailrace water levels, 182 operating conditions were simulated, including both power increases and decreases. A dataset of hydraulic transient characteristic indicators was extracted from each simulation. Subsequently, the dataset was divided into training and testing sets in an 8:2 ratio for training and testing the accuracy and effectiveness of the established network.

[0114] To evaluate the accuracy and effectiveness of the proposed RBF network, RMSE (Root Mean Square Error) and [other metrics] were used. (Coefficient of Determination) as an evaluation indicator:

[0115] Among them, RMSE reflects the absolute error of the model. The smaller the value, the lower the prediction error and the better the model performance.

[0116]

[0117] in, n For the number of data items; and These are the actual value and the predicted value, respectively.

[0118] for This metric reflects the relative performance and fit of the neural network. The closer its value is to 1, the stronger the model's explanatory power for the target variable; the closer the value is to 0 or smaller, the worse the model's performance.

[0119]

[0120] in, , and is a true value, a predicted value, and a mean of true values.

[0121] In step S103, based on the heuristic algorithm of dynamic constraint updating, the proxy model of the short-term optimal scheduling model of the hydropower station and the simulation control model of the hydropower unit based on the radial basis function neural network are decoupled to perform short-term optimal scheduling on the target hydropower station, and a short-term optimal scheduling result of the target hydropower station with coupled hydraulic transient characteristics is obtained.

[0122] In the embodiment of the present application, if the RBF neural network is directly coupled with the hydraulic transient characteristics to establish the coupled scheduling model of the hydropower station, the nonlinear and non-convex characteristics of the RBF neural network model bring significant challenges to direct solving. Although the RBF neural network has been pre-trained in the above steps, it still exists as a nonlinear function in the optimization process and needs to be calculated by forward propagation in each iteration. These calculations include nonlinear mapping of all radial basis kernels and weighted summation of outputs. Frequent calling of the neural network in the iterative optimization process significantly increases the overall computational complexity. Since the optimization process usually involves a large number of iterations, repeated calling of the RBF network becomes the main bottleneck of computational efficiency. In addition, the complexity of nonlinear calculation further reduces the efficiency of the optimization process.

[0123] Therefore, the embodiment of the present application decouples the RBF neural network from the optimization process based on the heuristic algorithm of dynamic constraint updating. The core idea of the heuristic algorithm of dynamic constraint updating is to realize "soft connection" between the optimization model and the neural network model by imposing a dynamic ramp rate constraint on the hydropower unit. In addition, since the key factor affecting the hydraulic transient characteristics of the hydropower unit in power regulation control is the amplitude of power step change. By extending this concept to the short-term scheduling model of the hydropower station, the complex calculation in the optimization process can be simplified by dynamically controlling the power ramp rate of the unit at each time.

[0124] It can be understood that the embodiment of the present application can decouple the short-term optimal scheduling model of the hydropower station and the proxy model of the simulation control model of the hydropower unit based on the radial basis function neural network based on the heuristic algorithm of dynamic constraint updating, for example, as Figure 4As shown, for the hydroelectric generating units mainly used for power generation and regulation, the ramping rate is usually very high, for example, the hydroelectric generating units can usually transition from the no-load state to the full-power generation within 20 seconds to several minutes (the same is true for the load rejection process from full-power generation to shutdown). This indicates that, on the short-term scheduling time scale of 15 minutes to 1 hour, if the hydraulic transient characteristics are not considered, the maximum ascending / descending rate of the hydroelectric generating unit can be close to its maximum output power. Based on this characteristic, the method of dynamically adjusting the ramping rate constraint of the hydroelectric generating unit can be used to avoid frequent calling of the computationally complex neural network model, thereby decoupling the optimization-based short-term scheduling model and the neural network-based hydraulic transient characteristic proxy model, so that the short-term optimization scheduling of the hydropower station can be performed, and then the short-term optimization scheduling result of the hydropower station coupled with the hydraulic transient characteristics can be obtained, effectively meeting the actual operation requirements of the hydropower station that change rapidly, and improving the efficiency and flexibility of the scheduling decision.

[0125] Optionally, in an embodiment of the present application, a heuristic algorithm based on dynamic constraint updating is used to decouple the short-term optimization scheduling model of the hydropower station and the proxy model of the radial basis function neural network-based simulation control model of the hydroelectric generating unit, and the method comprises the following steps: obtaining the short-term scheduling plan of the target hydropower station according to the short-term optimization scheduling model of the hydropower station; calculating each hydraulic transient characteristic index of each time of the hydroelectric generating unit by using the proxy model of the radial basis function neural network-based simulation control model of the hydroelectric generating unit; determining the hydraulic transient performance of the hydroelectric generating unit according to each hydraulic transient characteristic index of the hydroelectric generating unit; updating the dynamic ramping rate constraint of the hydroelectric generating unit based on the hydraulic transient performance, the short-term scheduling plan and the preset operation restriction factor of the hydroelectric generating unit, to obtain the updated dynamic ramping rate constraint; and decoupling the short-term optimization scheduling model of the hydropower station and the proxy model of the radial basis function neural network-based simulation control model of the hydroelectric generating unit based on the updated dynamic ramping rate constraint.

[0126] As a possible implementation manner, the embodiment of the present application can obtain the short-term scheduling plan of the hydropower station, wherein the short-term scheduling plan of the hydropower station is solved by using the short-term optimization scheduling model of the hydropower station, combining the dynamic constraint and limiting the hydroelectric generating unit ramping rate generated by the above step. In the first iteration, the hydroelectric generating unit ramping rate is not additionally limited.

[0127] Then, the hydraulic transient performance of the hydropower station can be evaluated, the hydraulic transient performance of the optimal scheduling plan is evaluated by using the pre-trained RBF network, and five hydraulic transient characteristic indexes are calculated. In order to comprehensively evaluate the hydraulic transient performance of the unit at each time, the following formula can be used for normalization processing, that is:

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] wherein, is the limit value of flow, is the limit pressure of spiral case inlet, is the limit value of draft tube pressure, is the limit value of maximum water level fluctuation of surge chamber, is the limit value of gate adjustment range, is the weight of five indexes, is the hydraulic transient performance.

[0135] For example, is the limit pressure of spiral case inlet, the five indexes after normalization can be coupled by weighting method, and the weight can be obtained by methods such as AHP (Analytic Hierarchy Process); the hydraulic transient performance (i.e. ) is a comprehensive score (using percentage system), and the smaller the value is, the better the hydraulic transient performance is; for different unit generation plans at multiple time points (such as 24 time points) within a day, parallel computing can be used to simultaneously calculate the hydraulic transient characteristics at each time point, thereby accelerating the calculation process.

[0136] Then, iteration termination judgment can be performed, i.e. judging whether the solution meets the designed stopping criterion: when the stopping criterion is met, the iteration process is terminated and the final solution is output; otherwise, the dynamic constraints of the hydroelectric unit are updated. Among them, the hydraulic transient performance is the primary standard for judging whether to stop iteration, because the present application aims to optimize the water consumption under the premise of improving the hydraulic transient performance, therefore, a comprehensive index can be proposed for evaluating the performance of the hydropower station within the entire optimization period, which includes two aspects: average performance and worst performance of hydraulic transient, which are specifically shown in the following formula:

[0137]

[0138]

[0139]

[0140]

[0141] wherein, Score is the comprehensive score of the hydropower station; is the firstj Overall score of the Taiwanese team; For the first j The average score of the Taiwanese machine group; For the first j The worst score for the Taiwanese team; w 1 and w 2 represents the weight; For the first j Taiwanese unit t The overall score at any given moment.

[0142] In engineering practice, the dispatch center can flexibly set the iteration stop score by comprehensively considering water consumption, hydropower unit status and other relevant factors.

[0143] Secondly, embodiments of the present invention can update the dynamic constraints of hydropower units, introducing and updating the boundaries of hydropower units related to operational constraints to limit unit operation and affect hydraulic transient performance. The relevant constraints will be dynamically updated and passed to the step of obtaining the short-term scheduling plan of the hydropower station, so as to be taken into consideration in the next short-term scheduling of the hydropower station. The present invention will generate new constraints to limit or adjust hydropower output, thereby improving the overall hydraulic transient performance of the hydropower station. After each iteration, the generated constraints will further tighten the operating range of the hydropower units. The constraints shown in the following formula will be designed to dynamically limit the maximum power ramp rate (including power increase / decrease), i.e.:

[0144]

[0145] in, for t +1 moment hydroelectric generator j power, for t Shike Hydropower Unit j power, For hydroelectric generator units j The initial gradient, These are the limiting factors for the operation of hydropower units. This represents the number of iterations.

[0146] This constraint will be applied in each iteration based on the previous round of short-term hydropower station scheduling plan and preset hydropower unit operation limitation factors. Generate and update. It is important to note that... The value of determines the range of dynamic constraints in each iteration; in this invention, it is set to 0.01.

[0147] In this embodiment of the invention, to verify the effectiveness of the invention, the following example is taken of an energy base on the Jinsha River (wind power installed capacity 600 MW, photovoltaic installed capacity 550 MW, hydropower installed capacity 2160 MW) to quantitatively analyze the applicability and effectiveness of the invention in short-term dispatching of hydropower stations:

[0148] (1) Surrogate model effectiveness analysis: First, the dataset required for pre-training the RBF network was obtained through the simulation control model of the hydroelectric generating unit power priority. By setting different power and water level changes, 910 groups of data were simulated, and an example of the dynamic simulation results of the simulation control model is shown in FIG. 1. Based on the pre-training dataset, two evaluation indicators, RMSE and R2, were used to comprehensively analyze the data fitting ability of the RBF network model. In addition, the present application compares the surrogate model method with the classic linear regression and GAM (Generalized Additive Models) method. Figure 5

[0149] Table 1 is the running result table of the RBF neural network-based surrogate model, which shows the results of the three methods, i.e., the RMSE and R2 of the three methods. Table 1 is as follows:

[0150] Table 1

[0151]

[0152] On average, the RMSE value of the RBF network is 83.49% and 76.69% lower than that of linear regression and GAM, respectively, especially in terms of index L4. Similarly, the R2 values of linear regression and GAM are 62.58% and 30.86% lower than that of the RBF network, with a significant gap in index L5. These results show that the RBF network model is superior to traditional statistical methods in terms of accuracy (lower error) and fitting degree (higher R2). L

[0153] In addition, the RMSE index further highlights the superior prediction performance of the RBF network. The absolute prediction error of the method proposed in the present application is always smaller in all indicators. Notably, the R2 value of the RBF network model is close to 1, indicating that its fitting effect is very good. Even in the challenging index L5 (R2=0.8779), the fitting effect of the RBF network is still much better than that of linear regression and GAM. L The performance difference in index L5 can be attributed to its relatively complex nonlinear characteristics, which make its modeling itself more difficult. However, the ability of the RBF network to handle such complexity still surpasses the comparison methods. L

[0154] ​​​​​​​​​In summary, the RBF network performs well in all indicators and evaluation criteria, establishing its robustness and reliability as a data fitting surrogate model, especially when dealing with nonlinear relationships.

[0155] (2) Effectiveness analysis of short-term scheduling method coupling hydraulic transient characteristics: The effectiveness of the proposed coupling model compared to the traditional model is evaluated. As shown in Figure 6 , the comprehensive score gradually decreases from the initial iteration value of 38.39 (corresponding to a ramp rate of 360 MW) to 29.78 at the 89th iteration, reaching the stopping criterion of a score lower than 30 (corresponding to a ramp rate of 147 MW). This iteration process shows that the proposed model can effectively improve the comprehensive performance of hydraulic transients.

[0156] As shown in Figure 7 , by comparing the results of the 1st iteration (representing the performance of the traditional model) and the 89th iteration (representing the performance of the proposed model), the following conclusions about the differences in hydraulic transient performance and water consumption can be drawn: the proposed model's hydraulic transient comprehensive score decreases by 8.61 points, i.e., the comprehensive hydraulic transient performance improves by 22.43% compared to the traditional model. However, this improvement comes at the cost of a slight increase in water consumption, specifically an additional 2.8545×10 5 of water consumption, an increase of 0.23%. This is because the reduction in ramp rate causes the units to operate at non-optimal conditions (such as partial load regions) to meet the scheduling requirements, thereby increasing the water consumption per unit of electricity generated.

[0157] As shown in Figure 8 , further analysis of the hydraulic transient performance of each unit is conducted. In terms of average hydraulic transient performance, except for Unit 2, the scores of the remaining units under the proposed model are lower than those of the traditional model, especially for Units 1 and 5, which have improved significantly. This indicates that the proposed model can effectively improve the hydraulic transient performance of most units. In terms of maximum hydraulic transient score (i.e., the most unfavorable performance during scheduling), the proposed model has a more obvious advantage. On average, the proposed model improves the most unfavorable hydraulic transient performance of the five online units by 21.90% compared to the traditional model. This improvement shows that the proposed model not only improves the average hydraulic transient performance of each unit but also effectively mitigates the impact of the most unfavorable conditions in short-term scheduling.

[0158] Therefore, the application significantly improves the performance and stability of short-term scheduling of the hydropower station by deeply fusing the traditional scheduling model and dynamic hydraulic transient characteristics through the surrogate modeling technology. Firstly, in the aspect of hydraulic transient characteristic extraction, the RBF network-based surrogate model is proved to be able to effectively replace the control model for simulating the hydraulic transient characteristic index, the average RMSE of which is 0.1882, the average 0.9684, and the fitting accuracy is excellent. Compared with the classic statistical methods (such as linear regression and generalized additive model), the error of the RBF network surrogate model is reduced by 83.49% and 76.69% on average, and the fitting accuracy is improved by 62.58% and 30.86% on average. Secondly, in the aspect of coupling effect of the hydraulic transient characteristics, the proposed coupling model shows significant advantages in the short-term scheduling process. Compared with the traditional scheduling model, the hydraulic transient comprehensive performance is improved by 22.43%, and the water consumption is only increased by 0.23%, which reflects the good balance between performance improvement and resource utilization. In summary, the coupling method in the embodiment of the application not only performs excellently in characteristic extraction and performance optimization, but also provides an efficient and reliable solution for short-term optimal scheduling of the hydropower station.

[0159] The short-term optimal scheduling method for the hydropower station coupling hydraulic transient characteristics proposed in the embodiment of the application can construct a short-term optimal scheduling model of the hydropower station taking the minimum water consumption as the target, then determine a hydraulic transient characteristic index system based on a power priority simulation control model of the hydropower generating set, thereby constructing a surrogate model of the simulation control model of the hydropower generating set based on a radial basis function neural network, and then decouple the short-term optimal scheduling model of the hydropower station and the surrogate model to perform short-term optimal scheduling on the hydropower station, so as to obtain a short-term optimal scheduling result of the hydropower station coupling hydraulic transient characteristics, effectively meeting the actual operation demand of the rapidly changing hydropower station, and improving the efficiency and flexibility of scheduling decision. Thus, the problems in the related art that the short-term optimal scheduling method for the hydropower station needs a large number of iterations to converge, resulting in very high calculation cost, failing to meet the actual operation demand of the rapidly changing hydropower station, and reducing the efficiency and flexibility of scheduling decision are solved.

[0160] Secondly, the short-term optimal scheduling device for the hydropower station coupling hydraulic transient characteristics proposed in the embodiment of the application is described with reference to the accompanying drawings.

[0161] Figure 9 is a block schematic diagram of the short-term optimal scheduling device for the hydropower station coupling hydraulic transient characteristics in the embodiment of the application.

[0162] As shown in Figure 9 , the short-term optimal scheduling device for the hydropower station coupling hydraulic transient characteristics 10 comprises a first construction module 100, a second construction module 200 and a scheduling module 300.

[0163] Specifically, the first construction module 100 is configured to construct a short-term optimal scheduling model of a target hydropower station with a minimum water consumption as an objective.

[0164] The second construction module 200 is configured to determine a hydraulics transient characteristic index system of a hydropower unit in the target hydropower station based on a pre-constructed power priority simulation control model of the hydropower unit, and construct a surrogate model of the simulation control model of the hydropower unit based on the radial basis function neural network based on the hydraulics transient characteristic index system.

[0165] The scheduling module 300 is configured to decouple the short-term optimal scheduling model of the hydropower station and the surrogate model of the simulation control model of the hydropower unit based on the radial basis function neural network based on a heuristic algorithm with dynamic constraint updating, so as to perform short-term optimal scheduling on the target hydropower station, and obtain a short-term optimal scheduling result of the target hydropower station with coupled hydraulics transient characteristics.

[0166] Optionally, in an embodiment of the present application, the second construction module 200 comprises a first determination unit.

[0167] The first determination unit is configured to determine at least one of a maximum flow deviation index, a maximum spiral case pressure index, a minimum draft tube pressure index, a maximum surge tank water level fluctuation index, and a guide vane adjustment mileage index in the hydraulics transient characteristic index system based on the pre-constructed power priority simulation control model of the hydropower unit.

[0168] Optionally, in an embodiment of the present application, the scheduling module comprises an acquisition unit, a calculation unit, a second determination unit, an updating unit, and a processing unit.

[0169] The acquisition unit is configured to acquire a short-term scheduling plan of the target hydropower station according to the short-term optimal scheduling model of the hydropower station.

[0170] The calculation unit is configured to calculate each hydraulics transient characteristic index of the hydropower unit at each time point by using the surrogate model of the simulation control model of the hydropower unit based on the radial basis function neural network.

[0171] The second determination unit is configured to determine a hydraulics transient performance of the hydropower unit according to each hydraulics transient characteristic index of the hydropower unit.

[0172] The updating unit is configured to update a dynamic ramping rate constraint of the hydropower unit based on the hydraulics transient performance, the short-term scheduling plan, and a preset hydropower unit operation limiting factor, so as to obtain an updated dynamic ramping rate constraint.

[0173] The processing unit is configured to decouple the short-term optimal scheduling model of the hydropower station and the surrogate model of the simulation control model of the hydropower unit based on the radial basis function neural network based on the updated dynamic ramping rate constraint.

[0174] Optionally, in one embodiment of the present invention, the dynamic gradeability constraint is expressed as:

[0175]

[0176] in, for t +1 moment hydroelectric generator j power, for t Shike Hydropower Unit j power, For hydroelectric generator units j The initial gradient, These are the limiting factors for the operation of hydropower units. This represents the number of iterations.

[0177] Optionally, in one embodiment of the present invention, the short-term optimal scheduling model for hydropower stations is expressed as:

[0178]

[0179] in, For hydroelectric generator units j At any moment t The start / stop status; T and N These are the number of optimized scheduling periods and the number of generating units, respectively. For hydroelectric generator units j At any moment t Traffic; and These represent the number of times the unit was started and shut down, respectively. and These represent the flow consumption during unit startup and shutdown, respectively.

[0180] It should be noted that the foregoing explanation of the embodiment of the short-term optimal scheduling method for hydropower stations with coupled hydraulic transient characteristics also applies to the short-term optimal scheduling device for hydropower stations with coupled hydraulic transient characteristics in this embodiment, and will not be repeated here.

[0181] The short-term optimal scheduling device of a hydropower station coupling with hydraulic transient characteristics provided by the embodiment of the present application can construct a short-term optimal scheduling model of the hydropower station with the minimum water consumption as the target, then determine a hydraulic transient characteristic index system based on a power priority simulation control model of a hydropower unit, thereby constructing a surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network, and then decouple the short-term optimal scheduling model of the hydropower station and the surrogate model to perform short-term optimal scheduling on the hydropower station, so as to obtain a short-term optimal scheduling result of the hydropower station coupling with hydraulic transient characteristics, effectively meet the actual operation requirements of the hydropower station changing rapidly, and improve the efficiency and flexibility of scheduling decisions. Thus, the problems in the prior art that the short-term optimal scheduling method of the hydropower station needs a large number of iterations to converge, results in a very high calculation cost, cannot meet the actual operation requirements of the hydropower station changing rapidly, and reduces the efficiency and flexibility of scheduling decisions are solved.

[0182] Figure 10 The structural schematic diagram of the electronic device provided by the embodiment of the present application is provided. The electronic device can include:

[0183] The memory 1001, the processor 1002 and the computer program stored in the memory 1001 and executable on the processor 1002.

[0184] The processor 1002 implements the short-term optimal scheduling method of the hydropower station coupling with hydraulic transient characteristics provided in the above embodiment when executing the program.

[0185] Further, the electronic device further includes:

[0186] The communication interface 1003 is used for communication between the memory 1001 and the processor 1002.

[0187] The memory 1001 is used for storing the computer program executable on the processor 1002.

[0188] The memory 1001 can include a high-speed RAM memory, and can also include a non-volatile memory, for example, at least one disk memory.

[0189] If the memory 1001, the processor 1002 and the communication interface 1003 are implemented independently, the communication interface 1003, the memory 1001 and the processor 1002 can be connected with each other through a bus and complete communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 10 Only one thick line is used to represent the bus in the figure, but it does not mean that there is only one bus or only one type of bus.

[0190] Optionally, in a specific implementation, if the memory 1001, the processor 1002 and the communication interface 1003 are integrated on a chip, the memory 1001, the processor 1002 and the communication interface 1003 can complete communication between each other through an internal interface.

[0191] The processor 1002 can be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present application.

[0192] The embodiment further provides a computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the method for short-term optimal scheduling of a hydropower station coupled with hydraulic transient characteristics as above.

[0193] The embodiment further provides a computer program product, comprising a computer program, which, when executed by a processor, is used to implement the method for short-term optimal scheduling of a hydropower station coupled with hydraulic transient characteristics as above.

[0194] In the description of the application, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the application. The illustrative description of the above terms in the specification does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in one or N embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples, without contradiction.

[0195] In addition, the terms "first", "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the application, the meaning of "N" is at least two, for example, two, three, etc., unless otherwise specifically limited.

[0196] Any process or method descriptions in flow charts or described elsewhere herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for performing specific logic functions or steps in the process, and the various embodiments of the application include additional implementations in which the order of steps can differ from those shown or described, including a step can occur at the same time as others or can be performed in reverse order or can be repeated, and additional steps can be performed, without departing from the scope of the application.

[0197] The logic and / or steps represented in flow diagrams or otherwise described herein, for example, can be considered as a sequence of executable instructions, and can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a product of the manufacturing and / or processing. The computer-readable medium can include, but is not limited to, the following: an electronic connection (an electronic device with one or N wires), a portable computer diskette (a magnetic device), a RAM (random access memory), a ROM (read-only memory), an EPROM (erasable programmable ROM) or a Flash memory, an optical fiber, and a portable CD ROM. In addition, the computer-readable medium can even be paper or another suitable medium upon which the program can be printed, as the program can be electronically captured, via the optically scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and stored in the computer memory.

[0198] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. As such, if implemented in hardware, and in another embodiment, any of the following technologies, alone or in any combination, can be used: discrete logic circuitry having logic gates for implementing logic functions upon an application of data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and the like.

[0199] Those of ordinary skill in the art can understand that all or part of the steps involved in the above-mentioned embodiment methods can be completed by programs instructing relevant hardware, and the programs can be stored in a computer-readable storage medium. When the programs are executed, one or a combination of the steps of the methods is included.

[0200] In addition, each function unit in each embodiment of the present application can be integrated in one processing module, or each unit can be physically present separately, or two or more units can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module. When the integrated module is realized in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0201] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A method for short-term optimal scheduling of a hydropower station coupled with hydraulic transient characteristics, characterized in that, The method comprises the following steps: constructing a short-term optimal scheduling model of a target hydropower station with the minimum water consumption as an objective; determining a hydraulic transient characteristic index system of a hydropower unit in the target hydropower station based on a pre-constructed power-priority simulation control model of the hydropower unit, and constructing a surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network based on the hydraulic transient characteristic index system; decoupling the short-term optimal scheduling model of the target hydropower station and the surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network based on a heuristic algorithm of dynamic constraint updating to perform short-term optimal scheduling on the target hydropower station and obtain a short-term optimal scheduling result of the target hydropower station coupled with hydraulic transient characteristics; the heuristic algorithm of dynamic constraint updating for decoupling the short-term optimal scheduling model of the target hydropower station and the surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network comprises: obtaining a short-term scheduling plan of the target hydropower station according to the short-term optimal scheduling model of the target hydropower station; calculating each hydraulic transient characteristic index of each time of the hydropower unit by using the surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network; determining a hydraulic transient performance of the hydropower unit according to the each hydraulic transient characteristic index of the hydropower unit; updating a dynamic ramping rate constraint of the hydropower unit based on the hydraulic transient performance, the short-term scheduling plan and a preset hydropower unit operation limiting factor to obtain an updated dynamic ramping rate constraint; decoupling the short-term optimal scheduling model of the target hydropower station and the surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network based on the updated dynamic ramping rate constraint; the dynamic ramping rate constraint is expressed as: in, for t +1 moment hydroelectric generator j power, for t Shike Hydropower Unit j power, For hydroelectric generator units j The initial gradient, These are the limiting factors for the operation of hydropower units. This represents the number of iterations. the short-term optimal scheduling model of the target hydropower station is expressed as: wherein, is the water turbine generator j is the flow rate of the water turbine generator at time t is the start-stop state of the water turbine generator at time T and N are the number of optimized scheduling periods and the number of generators, respectively; is the water turbine generator j is the flow rate of the water turbine generator at time t is the flow rate of the water turbine generator at time and are the number of start-ups and shut-downs of the generator, respectively; and are the flow consumption of the start-up and shut-down of the generator, respectively.

2. The short-term optimal scheduling method of hydropower stations coupling hydraulic transient characteristics according to claim 1, characterized in that, the hydraulic transient characteristic index system of the hydropower unit in the target hydropower station is determined based on the pre-constructed power-priority simulation control model of the hydropower unit, and the surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network is constructed based on the hydraulic transient characteristic index system. at least one of a maximum flow deviation index, a maximum pressure index of a spiral case, a minimum pressure index of a tailrace pipe, a maximum water level fluctuation index of a surge tank and a guide vane adjustment mileage index in the hydraulic transient characteristic index system is determined based on the pre-constructed power-priority simulation control model of the hydropower unit.

3. A device for short-term optimal scheduling of a hydropower station coupled with hydraulic transient characteristics, characterized in that, comprise: a first construction module configured to construct a short-term optimal scheduling model of a target hydropower station with the minimum water consumption as an objective; a second construction module configured to determine a hydraulic transient characteristic index system of a hydropower unit in the target hydropower station based on a pre-constructed power-priority simulation control model of the hydropower unit, and construct a surrogate model of the simulation control model of the hydropower unit based on a radial basis function neural network based on the hydraulic transient characteristic index system; The scheduling module is configured to decouple the short-term optimal scheduling model of the hydropower station and the surrogate model of the simulation control model of the hydropower unit based on a heuristic algorithm of dynamic constraint updating, to perform short-term optimal scheduling on the target hydropower station, and to obtain a short-term optimal scheduling result of the target hydropower station coupled with the hydraulic transient characteristics. The scheduling module comprises: The obtaining unit is configured to obtain a short-term scheduling plan of the target hydropower station according to the short-term optimal scheduling model of the hydropower station. The calculating unit is configured to calculate each item of the hydraulic transient characteristic index of the hydropower unit at each time point by using the surrogate model of the simulation control model of the hydropower unit based on the radial basis function neural network. The second determining unit is configured to determine the hydraulic transient performance of the hydropower unit according to the each item of the hydraulic transient characteristic index of the hydropower unit. The updating unit is configured to update the dynamic ramping rate constraint of the hydropower unit based on the hydraulic transient performance, the short-term scheduling plan, and a preset hydropower unit operation limiting factor, to obtain an updated dynamic ramping rate constraint. The processing unit is configured to decouple the short-term optimal scheduling model of the hydropower station and the surrogate model of the simulation control model of the hydropower unit based on the updated dynamic ramping rate constraint. The dynamic ramping rate constraint is expressed as: in, for t +1 moment hydroelectric generator j power, for t Shike Hydropower Unit j power, For hydroelectric generator units j The initial gradient, These are the limiting factors for the operation of hydropower units. This represents the number of iterations. The short-term optimal scheduling model of the hydropower station is expressed as: wherein, is the water turbine generator j is the start-up time of the water turbine generator t is the stop time of the water turbine generator T and N are respectively the number of optimal scheduling periods and the number of units; is the water turbine generator j is the flow rate of the water turbine generator at time t is the flow rate of the water turbine generator at time and are respectively the number of start-ups and the number of shut-downs of the unit; and are respectively the flow consumption of the start-up and the flow consumption of the shut-down of the unit.

4. The device for short-term optimal scheduling of a hydropower station coupling hydraulic transient characteristics according to claim 3, characterized in that, The second construction module comprises: The first determining unit is configured to determine at least one of the maximum flow deviation index, the maximum pressure index of the spiral case, the minimum pressure index of the draft tube, the maximum water level fluctuation index of the surge tank, and the guide vane adjustment mileage index in the hydraulic transient characteristic index system based on the pre-constructed simulation control model of the hydropower unit.

5. An electronic device, comprising: Comprise: A memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the method for short-term optimal scheduling of a hydropower station coupled with hydraulic transient characteristics according to any one of claims 1-2.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method for short-term optimal scheduling of a hydropower station coupled with hydraulic transient characteristics according to any one of claims 1-2.

7. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the method for short-term optimal scheduling of a hydropower station coupled with hydraulic transient characteristics according to any one of claims 1-2.

Images