Energy planning method, system and equipment of pumped storage power station and medium

By obtaining toughness indicators in pumped storage power plants and building an energy planning model, combined with the dual-delay depth deterministic strategy gradient algorithm, the energy planning problem under extreme climate conditions is solved, the stability and recovery capacity of the power plants are improved, and energy is ensured Reliability of supply and consumption of new energy.

CN119090206APending Publication Date: 2024-12-06STATE GRID ZHEJIANG ELECTRIC POWER CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411133367.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Existing research focuses less on how to effectively plan pumped storage power plants in extreme climates to ensure the stable operation of the power grid and the reliability of energy supply.

Method used

By obtaining the resilience indicators of pumped storage power plants under extreme climate conditions, building an energy planning model, and using a dual-delay depth deterministic strategy gradient algorithm for solving, optimizing the energy planning scheme, and improving the stability and recovery ability of the power plants under extreme conditions. .

Benefits of technology

The operation strategy of pumped storage power stations has been optimized, its stability and recovery capabilities under extreme climate conditions have been improved, the reliability and flexibility of energy supply have been improved, and the absorption of new energy and the economics of the power system have been promoted. Promote green and low-carbon development.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090206A_ABST
    Figure CN119090206A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of energy planning, and discloses an energy planning method, system, equipment and medium for a pumped storage power station, and the method comprises the steps: obtaining a toughness index of the pumped storage power station under an extreme climate condition, and constructing an energy planning model of the pumped storage power station according to the toughness index; the toughness index is calculated according to the occurrence probability of a fault scene and the corresponding load curve missing area of the pumped storage power station under the extreme climate condition; solving the energy planning model through a double-delay depth deterministic strategy gradient algorithm to obtain an energy planning scheme of the pumped storage power station; constructing an energy planning model of the pumped storage power station through the toughness index so as to optimize an operation strategy of the pumped storage power station; a double-delay depth deterministic strategy gradient algorithm is introduced to solve the complex continuous control problem of pumped storage planning so as to improve the solving efficiency of the model.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of energy planning, and in particular to an energy planning method, system, equipment and medium for a pumped storage power station. Background Art

[0002] With the deepening development of global energy transformation and the large-scale grid connection of renewable energy, traditional power systems are facing unprecedented challenges. Among them, how to effectively integrate unstable renewable energy generation and ensure the stable operation and supply reliability of the power grid has become a key task in the current power system planning and operation. Pumped Hydro Storage (PHS), as a mature large-scale energy storage technology, plays a vital role in balancing the load of the power grid, providing system peak-shaving capabilities, and ensuring the stability of energy supply. Being able to effectively support the resilience of the power grid under extreme climatic conditions is the development direction of the new power system. Therefore, how to quantify the planning of various power sources and energy storage under extreme conditions is the key to current research. Existing research focuses on how to optimize the grid connection system of new energy (wind power, photovoltaics, etc.) through pumped storage power stations, and few articles consider the future planning of pumped storage power stations under extreme climatic conditions.

[0003] It can be seen that how to solve the energy planning problem of pumped storage power stations under extreme conditions has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention

[0004] The present invention provides an energy planning method, system, equipment and medium for a pumped storage power station, which solve the problem of energy planning of a pumped storage power station under extreme conditions.

[0005] In order to solve the above technical problems, the first aspect of the present invention provides an energy planning method for a pumped storage power station, comprising:

[0006] Obtaining a resilience index of a pumped-storage power station under extreme climatic conditions, and constructing an energy planning model of the pumped-storage power station based on the resilience index; the resilience index is calculated by the probability of occurrence of a fault scenario of the pumped-storage power station under the extreme climatic conditions and the corresponding load curve missing area;

[0007] The energy planning model is solved by a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme for the pumped-storage power station.

[0008] As one preferred solution, the toughness index is expressed by the following formula:

[0009]

[0010] Where f3 is the toughness index; n is the probability of occurrence of scenario n; N is the number of failure scenarios; I mn is the power supply shortage degree of scenario n; T0 is the duration of the urban power supply network being affected by extreme climate; L(t) is the real load curve of the pumped storage power station when a major fault occurs due to extreme climate; L1(t) is the target load curve when the pumped storage power station operates without fault; R is the integral parameter; S n is the missing area of ​​the load curve.

[0011] As one of the preferred solutions, the energy planning model aims to minimize active network loss, minimize energy storage equipment investment cost and maximize resilience index.

[0012] As one of the preferred solutions, the constraints of the energy planning model include node voltage constraints, power balance constraints, energy storage operation constraints and energy storage quantity constraints.

[0013] As one of the preferred solutions, the energy planning model is solved by a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme for the pumped storage power station, including:

[0014] Initializing a network model and assigning random weights, and initializing a playback buffer, wherein the network model includes a Q network, a target Q network, and a policy network, wherein the Q network includes a first Q network and a second Q network, and wherein the target Q network includes a first target Q network and a second target Q network;

[0015] Based on the basic parameters in the pumped storage power station and the energy planning model, state variables, action space and reward functions are defined to interact with the environment, and the observed conversion data is stored in the playback buffer; the basic parameters include the reservoir water level, power demand, charging and discharging status, energy storage device status, node voltage and output of each power source at the current moment;

[0016] Randomly extracting conversion data from the playback buffer to train the network model, and iterating the environment interaction and model training steps until a preset number of iterations is reached to obtain the trained network model;

[0017] The key parameters of the pumped-storage power station are input into the trained network model for processing to obtain the energy planning scheme; the key parameters include grid parameters, boundary conditions and cost coefficients, and the energy planning scheme includes the installation nodes and capacity of energy storage equipment.

[0018] As one of the preferred solutions, based on the basic parameters in the pumped storage power station and the energy planning model, defining state variables, action space and reward functions to interact with the environment, and storing the observed conversion data in the playback buffer, includes:

[0019] The reservoir water level, power demand, charging and discharging state, and energy storage device state of the pumped-storage power station at the current moment are defined as the state variables, the node voltage of the pumped-storage power station and the output of each power source are defined as the action space, and the energy planning model is used as the reward function;

[0020] Observe the current state of the pumped-storage power station at each time step, select an action through the policy network, and execute the selected action in the environment to obtain a new state and reward;

[0021] The current state, selected action, new state, and reward are stored as the transition data in the replay buffer.

[0022] As one of the preferred solutions, randomly extracting conversion data from the playback buffer to train the network model, and iterating the environment interaction and model training steps until a preset number of iterations is reached to obtain the trained network model, including:

[0023] Randomly extracting conversion data from the playback buffer, and calculating a target Q value of the conversion data through the network model;

[0024] Performing gradient descent calculation on the loss of the Q network according to the target Q value, and updating a first network parameter, where the first network parameter is a network parameter of the Q network;

[0025] When the number of updates of the Q network reaches a preset number of conditions, the policy network is evaluated by the latest Q network to update the second network parameters, where the second network parameters are the network parameters of the policy network;

[0026] The first network parameters are correspondingly copied to the target Q network, and the environment interaction and model training steps are iterated until a preset number of iterations is reached to obtain the network model after training.

[0027] A second aspect of the present invention provides an energy planning system for a pumped storage power station, comprising:

[0028] A model building module, used to obtain the resilience index of the pumped-storage power station under extreme climatic conditions, and to build an energy planning model of the pumped-storage power station according to the resilience index; the resilience index is calculated by the probability of occurrence of the fault scenario of the pumped-storage power station under the extreme climatic conditions and the corresponding load curve missing area;

[0029] The model solving module is used to solve the energy planning model through a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme for the pumped-storage power station.

[0030] A third aspect of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the energy planning method for a pumped-storage power station as described above when executing the computer program.

[0031] A fourth aspect of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the energy planning method for the pumped-storage power station as described above is implemented.

[0032] Compared with the prior art, the beneficial effects of the embodiments of the present invention are as follows:

[0033] The present invention discloses a method, system, device and medium for energy planning of a pumped-storage power station. The method comprises obtaining a resilience index of a pumped-storage power station under extreme climatic conditions, and constructing an energy planning model of the pumped-storage power station according to the resilience index; the resilience index is obtained by calculating the probability of occurrence of a fault scenario of the pumped-storage power station under the extreme climatic conditions and the corresponding load curve missing area; the energy planning model is solved by a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme of the pumped-storage power station; the problem of energy planning of a pumped-storage power station under extreme conditions is solved. Please use resilience indicators to construct an energy planning model for pumped-storage power stations, which can optimize the operation strategy of pumped-storage power stations, improve their stability and resilience under extreme climatic conditions, ensure the safe and stable operation of the power station, improve the reliability and flexibility of energy supply, promote the absorption of new energy and the economy of the power system, and promote green and low-carbon development; this application introduces a double-delay deep deterministic policy gradient algorithm to solve the complex continuous control problem of pumped-storage planning, which improves the solution efficiency of the energy planning model, helps to formulate a more scientific and reasonable planning scheme for pumped-storage power stations, and improves the overall performance and reliability of pumped-storage power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is a flow chart of an energy planning method for a pumped storage power station provided by a certain embodiment of the present invention;

[0036] Figure 2 It is a trapezoidal curve diagram of the elasticity of a pumped storage power station system under a typhoon provided by a certain embodiment of the present invention;

[0037] Figure 3 is a flowchart of step S2 provided by an embodiment of the present invention;

[0038] Figure 4 is a flowchart of step S22 provided by an embodiment of the present invention;

[0039] Figure 5 is a flowchart of step S23 provided by an embodiment of the present invention;

[0040] Figure 6 is a structural diagram of an IEEE33 node system provided by a certain embodiment of the present invention;

[0041] Figure 7 It is a planning result diagram of IEEE33 node example scenario 1 provided by an embodiment of the present invention;

[0042] Figure 8 It is a planning result diagram of IEEE33 node example scenario 2 provided by an embodiment of the present invention;

[0043] Fig. 9 It is a structural diagram of an energy planning system for a pumped storage power station provided by a certain embodiment of the present invention;

[0044] Fig.10 It is a structural diagram of an electronic device provided by a certain embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0047] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0048] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0049] In one embodiment, if Figure 1 As shown, the first aspect of the present invention provides an energy planning method for a pumped storage power station, comprising:

[0050] S1. Obtaining a resilience index of a pumped-storage power station under extreme climatic conditions, and constructing an energy planning model of the pumped-storage power station according to the resilience index; the resilience index is calculated by the probability of occurrence of a fault scenario of the pumped-storage power station under the extreme climatic conditions and the corresponding load curve missing area;

[0051] In one embodiment, the toughness index is expressed by the following formula:

[0052]

[0053] Where f3 is the toughness index; nis the probability of occurrence of scenario n; N is the number of failure scenarios; I mn is the power supply shortage degree of scenario n; T0 is the duration of the urban power supply network being affected by extreme climate; L(t) is the real load curve of the pumped storage power station when a major fault occurs due to extreme climate; L1(t) is the target load curve when the pumped storage power station operates without fault; R is the integral parameter, that is, the difference between the integral value of the actual operation curve of the pumped storage power station and the integral value of the ideal operation curve; S n is the missing area of ​​the load curve.

[0054] Specifically, this application constructs an energy planning model for a pumped storage power station based on resilience indicators under extreme climate conditions; among them, extreme climate conditions include drought, floods, typhoons, etc. This application takes typhoons as an example, such as Figure 2 As shown, Figure 2 This is a trapezoidal curve of the elasticity of the pumped storage power station system under a typhoon. Since T0 indicates the duration of the city's power supply network being affected by extreme climate, including the time it takes for the typhoon to pass through the power grid in the pumped storage power station and the duration of the power grid restoring power supply, and the time required for fault recovery in each scenario is different, T0 should be much greater than the corresponding t in each scenario. Figure 2 The area enclosed by the true curve and the target curve represents the missing area of ​​the load curve. This application reflects the resilience of the power grid through the missing area of ​​the pumped-storage power station system function under extreme climatic conditions, and uses the probability of occurrence of fault scenarios and the corresponding missing area of ​​the load curve to estimate the resilience of the power grid, that is, the resilience index, to reflect the stability and recovery ability of the pumped-storage power station in the face of extreme climate, thereby providing a better planning model.

[0055] In one embodiment, the energy planning model aims to minimize active network loss, minimize energy storage equipment investment cost, and maximize resilience index; wherein the energy planning model is represented by the following formula:

[0056] F=[minf1,minf2,maxf3]

[0057]

[0058] Where F is the energy planning model; f1 is the active network loss; N b is the number of branches; G k is the current output conductance of the branch between node i and node j; U i and U j are the voltage amplitudes of nodes i and j respectively; δ ij is the voltage phase angle difference between node i and node j; f2 is the investment cost of energy storage equipment; C e is the energy storage capacity cost; C pis the energy storage power cost; C op,i is the energy storage operation cost of node i; S 1,2,3 It is the sum of the compensation income from the energy storage equipment participating in primary frequency regulation, peak regulation compensation income and peak-valley arbitrage income.

[0059] Specifically, this application constructs an energy planning model with the maximum resilience index and the minimum active network loss and energy storage equipment investment cost, in order to set the comprehensive optimization goal of the energy storage optimization model, and solve the problem of energy planning of pumped storage power stations under extreme conditions. This application can optimize the operation strategy of pumped storage power stations and improve their stability and recovery capabilities under extreme climate conditions by constructing an energy quantification analysis model for pumped storage power stations under extreme climate conditions, namely an energy planning model, so as to ensure the safe and stable operation of the power station, improve the reliability and flexibility of energy supply, promote the absorption of new energy and the economy of the power system, and promote green and low-carbon development.

[0060] The constraints of the energy planning model include node voltage constraints, power balance constraints, energy storage operation constraints and energy storage quantity constraints; wherein the node voltage constraint is expressed by the following formula:

[0061] V min ≤V i (t)≤V max

[0062] Where V max 、V min is the node voltage V i The upper and lower limits of (t).

[0063] The power balance constraint is expressed as follows:

[0064]

[0065] Where P G (t) is the active output of the power supply; n g The number of energy storage devices installed; is the active power output of the i-th distributed energy storage at time t; P load (t) is the total load of the system at time t.

[0066] Energy storage operation constraints include pumped storage unit output constraints, pumped storage unit operating condition constraints, pumped storage power station water constraints and pumped storage unit reserve capacity constraints; among them, the pumped storage unit output constraints are expressed by the following formula:

[0067]

[0068] In the formula, are the generating power and pumping power of pumped storage unit j respectively; are the maximum and minimum power generation of pumped storage unit j, respectively; are the maximum and minimum pumping power of pumped storage unit j, respectively; are the state variables of power generation and pumping of the pumped storage unit in period t, respectively, which restricts the unit from operating in power generation and pumping modes at the same time.

[0069] Since a section of penstock is usually included between the upper reservoir and the pump-turbine of a pumped-storage power station, no matter how many pumped-storage units there are, they cannot operate in both power generation and pumping modes at the same time; that is, if at least one pumped-storage unit is operating in power generation mode, then no other unit can operate in pumping mode at the same time, and vice versa. The operating condition constraint of the pumped-storage unit can ensure that a single pumped-storage unit cannot operate in both power generation and pumping modes at the same time, which is expressed by the following formula:

[0070]

[0071] Where J is the number of pumped storage units.

[0072] The water quantity constraint of a pumped storage power station is expressed by the following formula:

[0073]

[0074] Where, X t is the storage capacity of the upper reservoir of the pumped storage power station at time t; are the conversion coefficients from electricity to hydropower when the pumped storage unit j is in pumping and generating conditions; X max , X min are the maximum and minimum water storage capacity of the upper reservoir respectively; X0 is the initial storage capacity of the upper reservoir; X T It is the storage capacity of the upper reservoir at the end of the scheduling period.

[0075] The reserve capacity constraint of pumped storage units is expressed by the following formula:

[0076]

[0077] In the formula, They are the upper and lower and spare capacities provided by pumped storage unit j during period t under power generation conditions; The upper and lower spare capacities provided by pumped-storage unit j during period t under pumping conditions respectively.

[0078] The energy storage quantity constraint is expressed as follows:

[0079]

[0080] In the formula, The maximum number of energy storage devices installed.

[0081] This application ensures the stable operation of the pumped-storage power station system through node voltage constraints to improve the power supply quality; ensures the balance between supply and demand through power balance constraints to optimize resource allocation; improves the flexibility of the pumped-storage power station system through energy storage operation constraints to promote the consumption of new energy and reduce demand response costs; reasonably controls investment costs through energy storage quantity constraints to optimize energy storage configuration; and the comprehensive application of the aforementioned constraints helps to achieve the optimal planning goal of the energy planning model.

[0082] S2. Solve the energy planning model through the double-delayed deep deterministic policy gradient algorithm to obtain the energy planning scheme of the pumped-storage power station; Specifically, this application adopts the double-delayed deep deterministic policy gradient (TwinDelayed Deep Deterministic Policy Gradient, TD3) algorithm, a model-free reinforcement learning algorithm, to solve the overestimation bias problem of the Q-value function in the deep deterministic policy gradient (DDPG) algorithm. The TD3 algorithm described in this application introduces two Q networks and target policy smoothing and delayed update strategies, making it particularly suitable for the complex continuous control problems of pumped-storage planning power stations to maximize efficiency and minimize costs.

[0083] In one embodiment, step S2 is as follows Figure 3 As shown, including:

[0084] S21, initializing a network model and assigning a random weight, and initializing a playback buffer, wherein the network model includes a Q network, a target Q network, and a policy network, wherein the Q network includes a first Q network and a second Q network, and wherein the target Q network includes a first target Q network and a second target Q network;

[0085] Specifically, the initialization process of the TD3 algorithm is as follows: first, the first Q network and the second Q network are initialized and assigned random weights, the first target Q network and the second target Q network are initialized so that their weights are the same as those of the Q network, and the policy network is initialized and assigned random weights. In addition, the network model also includes a target policy network, which may be initialized and assigned random weights; then an experience replay buffer is initialized to store the observed conversion data; finally, hyperparameters are initialized, including but not limited to the discount factor, batch size, network learning rate, and delay parameters for policy updates.

[0086] S22, based on the basic parameters in the pumped storage power station and the energy planning model, define state variables, action space and reward function to interact with the environment, and store the observed conversion data in the playback buffer; the basic parameters include the reservoir water level, power demand, charging and discharging status, energy storage device status, node voltage and output of each power source at the current moment;

[0087] In one embodiment, step S22 is as follows: Figure 4 As shown, including:

[0088] S221, defining the reservoir water level, power demand, charge and discharge status, and energy storage device status of the pumped-storage power station at the current moment as the state variables, defining the node voltage of the pumped-storage power station and the output of each power source as the action space, and using the energy planning model as the reward function;

[0089] S222, observing the current state of the pumped-storage power station at each time step, selecting an action through the strategy network, and executing the selected action in the environment to obtain a new state and reward;

[0090] S223. Store the current state, the selected action, the new state, and the reward as the conversion data in the playback buffer.

[0091] Specifically, in the environmental interaction part, the state variables, action space and reward function of the TD3 algorithm are first defined according to the basic parameters of the pumped storage power station and the energy planning model. The action constraints of the TD3 algorithm are the constraints of the energy planning model. Then the state observation is performed: at each time step t, the current state s of the pumped storage power station system is observed. t ; followed by action selection: using the policy network (with noise added for exploration) to select an action at, which represents a decision such as pumping water from the reservoir (charging) or releasing water to generate electricity (discharging); followed by execution: executing the selected action in the environment, resulting in a new state s t +1 and reward r t , the reward can be based on factors such as energy efficiency, cost savings, or supply-demand balance; finally, storage conversion: the observed conversion (s t ,at,r t ,s t+1) is stored in the replay buffer for experience replay. Through interaction with the environment, the TD3 algorithm can find a balance between exploration (exploring unknown states and actions) and utilization (using known information to select the best action). In the early stages of training, experience is accumulated through exploration, which is conducive to data collection and accumulation, evaluation of optimization strategies, delayed update of dual networks, and balance between exploration and utilization, and promotes efficient learning and stable performance of the TD3 algorithm in continuous control problems.

[0092] S23, randomly extracting conversion data from the playback buffer to train the network model, and iterating the environment interaction and model training steps until a preset number of iterations is reached to obtain the trained network model;

[0093] In one embodiment, step S23 is as follows: Figure 5 As shown, including:

[0094] S231, randomly extracting conversion data from the playback buffer, and calculating the target Q value of the conversion data through the network model; specifically, after randomly extracting a small batch of conversion data from the playback buffer, for each extracted conversion data, the target Q value is calculated through the target network, as shown in the following formula:

[0095] y=r+γmin(Q 1′ (s t+1 ,π′(s t+1 )),Q 2′ (s t+1 ,π′(s t+1 )))

[0096] Where y is the target Q value; r is the immediate reward; γ is the discount factor; Q1 is the first Q network; π′ is the target policy network; Q2 is the second Q network.

[0097] S232, performing gradient descent calculation on the loss of the Q network by using the target Q value, and updating a first network parameter, where the first network parameter is a network parameter of the Q network; specifically, performing gradient descent calculation on the losses of the two Q networks by using the following formula:

[0098]

[0099] In the formula, L(θ i ) is the loss of Q network; N is the total number of Q networks; θ i is the first network parameter.

[0100] S233. When the number of updates of the Q network reaches a preset number of conditions, the policy network is evaluated by the latest Q network to update the second network parameter, where the second network parameter is the network parameter of the policy network. Specifically, the present application adopts a delayed update strategy, and updates the policy network every several iterations to maximize the Q value. The update process is performed by the following formula:

[0101]

[0102] Where π is the policy network; θ π is the second network parameter.

[0103] S234, copying the first network parameters to the target Q network, iterating the environment interaction and model training steps until a preset number of iterations is reached, and obtaining the network model after training; specifically, the update process of the target Q network is as follows:

[0104] θ i ′←τθ i +(1-τ)θ i '

[0105] θ π ′←τθ π +(1-τ)θ π '

[0106] In the formula, θ i ' is the first network parameter after update; θ π ' is the updated second network parameter; τ∈(0,1).

[0107] Then repeat the steps of environment interaction, playback buffer update and model training until the algorithm converges or reaches a predefined number of iterations. Convergence is usually indicated by the stability of the policy and Q value. When updating the Q network, the present application adds random noise ε to smooth the target policy, which helps to increase the generalization ability and stability of the algorithm. The noise ε is usually sampled from a truncated normal distribution and added to the action output by the policy network: Then, this smoothed action is used to calculate the target Q value; a delayed update strategy is adopted, that is, the policy network is updated once after the Q network is updated multiple times, which helps the policy network to update the policy based on a more stable Q value, reducing the policy instability caused by Q value fluctuations, so that the TD3 algorithm can effectively learn stable policies in the continuous action space and reduce the problems of overestimation of Q values ​​and error accumulation.

[0108] S24. Input the key parameters of the pumped-storage power station into the trained network model for processing to obtain the energy planning scheme; the key parameters include grid parameters, boundary conditions and cost coefficients, and the energy planning scheme includes the installation nodes and capacity of energy storage equipment; specifically, input the grid parameters, boundary conditions and cost coefficients of the pumped-storage power station into the trained target Q network to obtain the energy planning scheme including the installation nodes and capacity of the energy storage equipment.

[0109] This application takes the improved IEEE 33-node electrical control system as an example and uses Matlab software to perform calculation analysis. The IEEE 33-node system structure is as follows: Figure 6 As shown in the figure, the system contains a total of 33 nodes, 32 branches and 1 transformer. The system base voltage is 12.66kV, the node voltage allowable range is 0.9~1.1 (pu), the initial grid active load is 5500kW, and the grid load during the fault is 0.5~0.8 (pu). In the example, up to 3 energy storage devices are connected, and the location nodes where the energy storage devices can be connected are {2, 3, ..., 33}. When the maximum rated power and rated capacity of the energy storage device are 350kW and 1000kW respectively, the corresponding cost coefficients are 7000 yuan / kW and 35 yuan / kWh respectively, and the annual operation and maintenance cost per unit power is 63 yuan / kWh. The discharge efficiency of the energy storage device is 0.9, and the charge state range is 0.1~0.9. By default, the important loads in the power grid have been connected to a transfer line. According to the different land costs in different cities, the energy storage sites are divided into three areas. The cost of each site is 910,000 yuan (nodes 1-9, 19-22), 700,000 yuan (nodes 23-25, 26-33) and 490,000 yuan (nodes 10-18). Taking a certain area in Guangdong as an example, the peak-to-valley ratio of energy storage equipment is 1.7:1:0.38, and the normal electricity price is 0.5283 yuan / kWh; the energy storage equipment is treated as a negative PQ node with a power factor of 0.9.

[0110] In order to verify the effectiveness of the scheme to which this application belongs, scenario 1 is set: no resilience protection measures are added, scenario 2: energy storage equipment is added for resilience protection, and the weighted load loss of the power grid is used as an intuitive expression of the system function loss. The unit load loss cost of non-important loads is 70 yuan / kWh, and the unit load loss cost of important loads is 7000 yuan / kWh; the extreme climate condition is that the typhoon transit time is selected as 15 minutes. The technical and economic comparison of different scenarios under typhoon disasters in the IEEE33 node example is as follows:

[0111]

[0112]

[0113] As can be seen from the table, the loss of important loads in scenario 1 is 35.7%, while the loss of important loads in scenario 2 is 0; the loss of non-important loads in scenario 1 is 58.17, and the loss of non-important loads in scenario 2 is 7.28% less than that in scenario 1. At the same time, the weighted load loss is greatly reduced, only 8.54% of that in scenario 1, and the resilience index is greatly improved. In addition, the network loss in scenario 2 is also 36.6% lower than that in scenario 1, which shows that the effect of reducing network loss has been achieved.

[0114] The planning results of IEEE33 node scenario 1 are as follows: Figure 7 As shown in Figure 2, the power supply lines of important load nodes 16 and 31 were damaged, and non-important nodes 12-18, 20-22, 25, 26-33 were also damaged, resulting in a large loss of load and huge economic losses. The planning results of IEEE 33-node example scenario 2 are shown in Figure 2. Figure 8 As shown, while ensuring that the fault location remains unchanged, after adding three energy storage devices, all important node loads are guaranteed to have power supply, and non-important load nodes 17, 18, and 32 are also guaranteed to have power supply. According to the location and capacity of the energy storage equipment calculated according to the example, the cost can be recovered in 7 to 8 years. It can be seen that the effectiveness of the scheme described in this application can effectively reduce the losses of pumped-storage power stations under extreme climatic conditions.

[0115] In the embodiment of the present application, based on how to solve the problem of energy planning of pumped-storage power stations under extreme conditions, a method for energy planning of pumped-storage power stations is designed, which realizes the acquisition of the resilience index of the pumped-storage power station under extreme climatic conditions, and constructs an energy planning model of the pumped-storage power station according to the resilience index; the resilience index is obtained by calculating the probability of occurrence of the fault scenario of the pumped-storage power station under the extreme climatic conditions and the corresponding load curve missing area; the energy planning model is solved by the double-delay deep deterministic policy gradient algorithm to obtain the technical solution of the energy planning scheme of the pumped-storage power station; the problem of pumped-storage power stations under extreme conditions is solved. In order to solve the problem of energy planning under extreme weather conditions, this application constructs an energy planning model for pumped-storage power stations through resilience indicators, which can optimize the operation strategy of pumped-storage power stations, improve their stability and recovery capabilities under extreme climatic conditions, ensure the safe and stable operation of power stations, improve the reliability and flexibility of energy supply, promote the absorption of new energy and the economy of power systems, and promote green and low-carbon development; this application introduces a double-delay deep deterministic strategy gradient algorithm to solve the complex continuous control problem of pumped-storage planning, which improves the solution efficiency of the energy planning model, helps to formulate a more scientific and reasonable planning scheme for pumped-storage power stations, and improves the overall performance and reliability of pumped-storage power stations.

[0116] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0117] In another embodiment, if Fig. 9 As shown, the second aspect of the present invention provides an energy planning system for a pumped storage power station, comprising:

[0118] The model building module 10 is used to obtain the resilience index of the pumped storage power station under extreme climatic conditions, and to build an energy planning model of the pumped storage power station according to the resilience index; the resilience index is calculated by the probability of occurrence of the fault scenario of the pumped storage power station under the extreme climatic conditions and the corresponding load curve missing area;

[0119] The model solving module 20 is used to solve the energy planning model through a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning solution for the pumped-storage power station.

[0120] It should be noted that each module in the above-mentioned energy planning system based on a pumped-storage power station can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of an energy planning system for a pumped-storage power station, please refer to the definition of an energy planning method for a pumped-storage power station above. The two have the same functions and effects, which will not be repeated here.

[0121] A third aspect of the present invention provides an electronic device, the electronic device comprising:

[0122] processor, memory, and bus;

[0123] The bus is used to connect the processor and the memory;

[0124] The memory is used to store operation instructions;

[0125] The processor is used to call the operation instruction, and the executable instruction enables the processor to perform operations corresponding to the energy planning method of a pumped-storage power station as shown in the first aspect of the present application.

[0126] In an alternative embodiment, an electronic device is provided, such as Fig.10 As shown, Fig.10The electronic device 5000 shown includes: a processor 5001 and a memory 5003. The processor 5001 and the memory 5003 are connected, such as through a bus 5002. Optionally, the electronic device 5000 may also include a transceiver 5004. It should be noted that in actual applications, the transceiver 5004 is not limited to one, and the structure of the electronic device 5000 does not constitute a limitation on the embodiments of the present application.

[0127] Processor 5001 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 5001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0128] The bus 5002 may include a path to transmit information between the above components. The bus 5002 may be a PCI bus or an EISA bus, etc. The bus 5002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.10 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0129] The memory 5003 may be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or an EEPROM, a CD-ROM or other optical disk storage, an optical disk storage (including a compressed optical disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0130] The memory 5003 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 5001. The processor 5001 is used to execute the application code stored in the memory 5003 to implement the content shown in any of the above method embodiments.

[0131] Among them, electronic devices include but are not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc.

[0132] The fourth aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the energy planning method for a pumped-storage power station shown in the first aspect of the present application is implemented.

[0133] Another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0134] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented.

[0135] In summary, the present invention relates to the field of energy planning technology, and discloses a method, system, device and medium for energy planning of a pumped-storage power station. The method comprises obtaining a resilience index of a pumped-storage power station under extreme climatic conditions, and constructing an energy planning model of the pumped-storage power station according to the resilience index; the resilience index is calculated by the probability of occurrence of a fault scenario of the pumped-storage power station under the extreme climatic conditions and the corresponding load curve missing area; the energy planning model is solved by a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme for the pumped-storage power station; and the energy planning of the pumped-storage power station under extreme conditions is solved. In order to solve the problem of pumped storage planning, this application constructs an energy planning model for pumped storage power stations through resilience indicators, which can optimize the operation strategy of pumped storage power stations, improve their stability and recovery capabilities under extreme climatic conditions, ensure the safe and stable operation of power stations, improve the reliability and flexibility of energy supply, promote the absorption of new energy and the economy of power systems, and promote green and low-carbon development; this application introduces a double-delay deep deterministic strategy gradient algorithm to solve the complex continuous control problem of pumped storage planning, which improves the solution efficiency of the energy planning model, helps to formulate a more scientific and reasonable planning scheme for pumped storage power stations, and improves the overall performance and reliability of pumped storage power stations.

[0136] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0137] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. A method for energy planning of a pumped storage power station, characterized in that: include: Obtaining a resilience index of a pumped-storage power station under extreme climatic conditions, and constructing an energy planning model of the pumped-storage power station according to the resilience index; The resilience index is calculated by the probability of occurrence of the failure scenario of the pumped storage power station under the extreme climatic conditions and the corresponding load curve missing area; The energy planning model is solved by a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme for the pumped-storage power station.

2. The energy planning method for a pumped storage power station according to claim 1, characterized in that: The toughness index is expressed by the following formula: Where f3 is the toughness index; n is the probability of occurrence of scenario n; N is the number of failure scenarios; I mn is the power supply shortage degree of scenario n; T0 is the duration of the urban power supply network being affected by extreme climate; L(t) is the real load curve of the pumped storage power station when a major fault occurs due to extreme climate; L1(t) is the target load curve when the pumped storage power station operates without fault; R is the integral parameter; S n is the missing area of ​​the load curve.

3. The energy planning method for a pumped storage power station according to claim 1, characterized in that: The energy planning model aims to minimize active network loss, minimize energy storage equipment investment cost and maximize resilience index.

4. The energy planning method for a pumped storage power station according to claim 1, characterized in that: The constraints of the energy planning model include node voltage constraints, power balance constraints, energy storage operation constraints and energy storage quantity constraints.

5. The energy planning method for a pumped storage power station according to claim 1, characterized in that: The energy planning model is solved by using a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme for the pumped-storage power station, including: Initializing a network model and assigning random weights, and initializing a playback buffer, wherein the network model includes a Q network, a target Q network, and a policy network, wherein the Q network includes a first Q network and a second Q network, and wherein the target Q network includes a first target Q network and a second target Q network; Based on the basic parameters in the pumped storage power station and the energy planning model, state variables, action space and reward functions are defined to interact with the environment, and the observed conversion data is stored in the playback buffer; the basic parameters include the reservoir water level, power demand, charging and discharging status, energy storage device status, node voltage and output of each power source at the current moment; Randomly extracting conversion data from the playback buffer to train the network model, and iterating the environment interaction and model training steps until a preset number of iterations is reached to obtain the trained network model; The key parameters of the pumped-storage power station are input into the trained network model for processing to obtain the energy planning scheme; the key parameters include grid parameters, boundary conditions and cost coefficients, and the energy planning scheme includes the installation nodes and capacity of energy storage equipment.

6. The energy planning method for a pumped storage power station according to claim 5, characterized in that: The method of defining state variables, action spaces, and reward functions based on basic parameters in the pumped storage power station and the energy planning model to interact with the environment and storing observed conversion data in the playback buffer comprises: The reservoir water level, power demand, charging and discharging state, and energy storage device state of the pumped-storage power station at the current moment are defined as the state variables, the node voltage of the pumped-storage power station and the output of each power source are defined as the action space, and the energy planning model is used as the reward function; Observe the current state of the pumped-storage power station at each time step, select an action through the policy network, and execute the selected action in the environment to obtain a new state and reward; The current state, selected action, new state, and reward are stored as the transition data in the replay buffer.

7. The energy planning method for a pumped storage power station according to claim 6, characterized in that: The randomly extracting conversion data from the playback buffer to train the network model, and iterating the environment interaction and model training steps until a preset number of iterations is reached to obtain the trained network model, including: Randomly extracting conversion data from the playback buffer, and calculating a target Q value of the conversion data through the network model; Performing gradient descent calculation on the loss of the Q network according to the target Q value, and updating a first network parameter, where the first network parameter is a network parameter of the Q network; When the number of updates of the Q network reaches a preset number of conditions, the policy network is evaluated by the latest Q network to update the second network parameters, where the second network parameters are the network parameters of the policy network; The first network parameters are correspondingly copied to the target Q network, and the environment interaction and model training steps are iterated until a preset number of iterations is reached to obtain the network model after training.

8. An energy planning system for a pumped storage power station, characterized in that: include: A model building module, used to obtain the resilience index of the pumped-storage power station under extreme climatic conditions, and to build an energy planning model of the pumped-storage power station according to the resilience index; the resilience index is calculated by the probability of occurrence of the fault scenario of the pumped-storage power station under the extreme climatic conditions and the corresponding load curve missing area; The model solving module is used to solve the energy planning model through a double-delayed deep deterministic policy gradient algorithm to obtain an energy planning scheme for the pumped-storage power station.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the energy planning method for a pumped-storage power station as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the energy planning method for the pumped-storage power station as described in any one of claims 1 to 7 is implemented.