Power dispatching method and device of power system, electronic equipment and storage medium

Abstract

The invention discloses a power dispatching method and device of a power system, electronic equipment and a storage medium, and relates to the field of power. The method comprises the steps of obtaining operation state data of a power system and meteorological information of an area where the power system is located; on the basis of the operation state data and the meteorological information, predicting a power load needing to be provided by the power system through a time sequence prediction algorithm to obtain a predicted power load; according to the predicted power load, a power dispatching scheme is determined through a game theory model, the game theory model is used for simulating interaction between the multiple power generation nodes and the multiple energy storage nodes, and the power dispatching scheme is used for representing the power generation amount of the multiple power generation nodes and the power storage amount of the multiple energy storage nodes; and performing power dispatching on the power system according to the power dispatching scheme. According to the invention, the technical problem of low accuracy of power dispatching is solved.

Description

Technical Field

[0001] The present invention relates to the field of electric power, and in particular to a method, device, electronic equipment and storage medium for dispatching electric power in an electric power system. Background Art

[0002] Distributed photovoltaic energy is a solar power generation system that installs photovoltaic panels in different locations. Compared with centralized photovoltaic power stations, distributed photovoltaic energy is smaller in scale, more flexible, can reduce losses during power transmission, and improve energy efficiency.

[0003] As for the power dispatching methods for distributed photovoltaic energy, the existing dispatching schemes fail to fully consider the characteristics of distributed energy, resulting in low accuracy of dispatching decisions.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] Embodiments of the present invention provide a power dispatching method, device, electronic device and storage medium for a power system, so as to at least solve the technical problem of low accuracy of power dispatching.

[0006] According to one aspect of an embodiment of the present invention, there is provided a power dispatching method for a power system, wherein the power system includes a plurality of power generation nodes and a plurality of energy storage nodes, and comprises: obtaining operating status data of the power system and meteorological information of an area where the power system is located; based on the operating status data and the meteorological information, predicting the power load that the power system needs to provide by a time series prediction algorithm to obtain a predicted power load; according to the predicted power load, determining a power dispatching scheme by a game theory model, wherein the game theory model is used to simulate the interaction between a plurality of power generation nodes and a plurality of energy storage nodes, and the power dispatching scheme is used to characterize the power generation of a plurality of power generation nodes and the storage power of a plurality of energy storage nodes; and performing power dispatching on the power system according to the power dispatching scheme.

[0007] Furthermore, according to the predicted power load, a power dispatching plan is determined through a game theory model, including: according to the operating status data and the predicted power load, a preliminary power dispatching plan is determined through a reinforcement learning dispatching algorithm, wherein the preliminary power dispatching plan is used to characterize the power generation conditions of multiple power generation nodes and the power supply conditions of multiple energy storage nodes; the preliminary power dispatching plan is adjusted through the game theory model to obtain a power dispatching plan.

[0008] Furthermore, the preliminary power dispatching plan is adjusted through the game theory model to obtain the power dispatching plan, including: constructing a game theory model based on the preliminary power dispatching plan; solving the game theory model to obtain the power dispatching plan, wherein the power dispatching plan satisfies the Nash equilibrium condition.

[0009] Furthermore, based on the preliminary power dispatching plan, a game theory model is constructed, including: determining the utility function of the game theory model based on the interactive utility between multiple intelligent agents, wherein different intelligent agents are obtained by combining different power generation nodes and different energy storage nodes; determining the strategy space of the game theory model according to the preliminary power dispatching plan, the strategy space containing multiple values ​​of power generation of different power generation nodes and multiple values ​​of stored power of different energy storage nodes; and constructing the game theory model based on the utility function and the strategy space.

[0010] Furthermore, the method also includes: based on the operating status data, performing fault detection on the power system through an abnormality detection algorithm to obtain fault information; generating a fault recovery plan based on the fault information; and recovering the fault in the power system according to the fault recovery plan.

[0011] Furthermore, generating a fault recovery plan based on the fault information includes: locating the fault based on the fault information to obtain the fault location; evaluating the impact range of the fault based on the fault location to obtain a fault evaluation result; and determining a fault recovery plan based on the fault evaluation result.

[0012] Furthermore, the operation status data of the power system is obtained, including: collecting the original operation data of the power system based on the long-distance wide area network protocol; performing data cleaning operations on the original operation data to obtain cleaned data; using a sliding window method to perform feature extraction operations on the cleaned data to obtain the operation status data of the power system.

[0013] According to another aspect of an embodiment of the present invention, there is also provided a power dispatching device of a power system, comprising: an acquisition module, used to acquire operating status data of the power system, wherein the operating status data of the power system comprises historical operating status data and current operating status data; a prediction module, used to predict the power load that the power system needs to provide based on the current operating status data and meteorological forecast information through a time series prediction algorithm to obtain a predicted power load, wherein the time series prediction algorithm is obtained by training with historical operating status data; a determination module, used to determine a power dispatching plan according to the predicted power load through a game theory model, wherein the game theory model is used to simulate the interaction between power generation nodes and energy storage nodes, and the power dispatching plan is used to characterize the power generation of multiple power generation nodes and the storage power of multiple energy storage nodes in the power system; a power generation module, used to perform power dispatching according to the power dispatching plan.

[0014] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the above-mentioned power dispatching method of the power system is executed when the program is running.

[0015] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein when the executable program runs, the device where the storage medium is located is controlled to execute the above-mentioned power dispatching method of the power system.

[0016] In an embodiment of the present invention, the operation status data of the power system and the meteorological information of the area where the power system is located are obtained; based on the operation status data and the meteorological information, the power load that the power system needs to provide is predicted by a time series prediction algorithm to obtain a predicted power load; according to the predicted power load, a power dispatching scheme is determined by a game theory model, wherein the game theory model is used to simulate the interaction between multiple power generation nodes and multiple energy storage nodes, and the power dispatching scheme is used to characterize the power generation of multiple power generation nodes and the storage power of multiple energy storage nodes; the power system is dispatched according to the power dispatching scheme. It is easy to notice that by introducing meteorological information, the accuracy of load forecasting is further enhanced, and the game theory model is used to determine the power dispatching scheme based on the predicted power load, which fully considers the mutual influence between the power generation nodes and the energy storage nodes in the power system, improves the rationality of the power dispatching scheme, and dispatches the power system according to the power dispatching scheme, thereby achieving the technical effect of improving the accuracy of power dispatching, thereby solving the technical problem of low accuracy of power dispatching. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1 is a flow chart of a power dispatching method of a power system according to an embodiment of the present invention;

[0019] Figure 2 is a schematic diagram of an optional power dispatching system according to an embodiment of the present invention;

[0020] Figure 3 is a schematic diagram of a power dispatching device of a power system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] Example 1

[0024] According to an embodiment of the present invention, an embodiment of a power dispatching method of a power system is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] In the embodiments of the present application, the application of the power dispatching method of the power system to a virtual power plant and a distributed photovoltaic system is taken as an example for explanation.

[0026] Figure 1 is a flow chart of a power dispatching method for a power system according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0027] Step S102, obtaining the operating status data of the power system and the meteorological information of the area where the power system is located.

[0028] The operating status data in the above steps may include but are not limited to power output, voltage, current, temperature, device status, etc.

[0029] The meteorological information in the above steps is data related to the weather, which may include but is not limited to weather conditions, light intensity, etc.

[0030] In an optional embodiment, the operation status data can be obtained through IoT (Internet of Things) devices in the perception layer of the power dispatching system. IoT devices may include but are not limited to smart meters, temperature and humidity sensors, voltage sensors, current sensors, etc. IoT devices are deployed in power generation nodes, energy storage nodes, transmission lines and other parts of the power system, and the output power of the power generation node is collected through smart meters, the voltage and current of the power line are detected through voltage sensors and current sensors respectively, and the temperature information of the power equipment during operation is collected through temperature sensors to obtain the operation status data.

[0031] When obtaining meteorological information of the area where the power system is located, the meteorological information of the area where the power system is located can be obtained through weather forecast.

[0032] Step S104, based on the operation status data and meteorological information, the power load that the power system needs to provide is predicted by a time series prediction algorithm to obtain a predicted power load.

[0033] The time series prediction algorithm in the above steps is used to predict the power load in the future based on the operating status data and meteorological information. The time series prediction algorithm can be an LSTM (Long Short-Term Memory) model, a moving average model, or an autoregressive integrated moving average model, but is not limited thereto.

[0034] The predicted power load in the above steps is used to characterize the power demand of the power system in the future period.

[0035] In an optional embodiment, the power load can be predicted in the cloud layer of the power dispatching system, that is, the cloud computing platform. The LSTM model can be selected as the time series prediction algorithm. The LSTM model is a recursive neural network suitable for processing time series data. The formula for updating the LSTM unit state is as follows:

[0036] i t =σ(W i ·[h t-1 ,x t ]+b i );

[0037] Among them, i t is the activation value of the input gate at time t; σ is the activation function, which is used to compress the activation value between 0 and 1 to play an activation role; W i is the weight matrix of the input gate; h t-1 is the hidden state at time t-1; x t is the input at time t; b i is the bias vector of the input gate.

[0038] f t =σ(W f ·[h t-1 ,x t ]+b f );

[0039] Among them, f t is the activation value of the forget gate at time t; σ is the activation function, which is used to compress the activation value between 0 and 1 to play an activation role; W f is the weight matrix of the forget gate; h t-1 is the hidden state at time t-1; x t is the input at time t; b f is the bias vector of the forget gate.

[0040] o t =σ(W o ·[h t-1 ,x t ]+b o );

[0041] Among them, t is the activation value of the output gate at time t; σ is the activation function, which is used to compress the activation value between 0 and 1 to play an activation role; W o is the weight matrix of the output gate; h t-1 is the hidden state at time t-1; x t is the input at time t; b o is the bias vector of the output gate.

[0042]

[0043] in, is the candidate unit state at time t; tanh is the hyperbolic tangent function; W C is the weight matrix of the candidate unit state; h t-1 is the hidden state at time t-1; x t is the input at time t; b C is the bias vector of the candidate unit state.

[0044]

[0045] Among them, C t is the cell state at time t; f t is the activation value of the forget gate at time t; C t-1 is the unit state at time t-1; i t is the activation value of the input gate at time t; is the candidate unit state at time t.

[0046] h t =o t tanh(C t );

[0047] Among them, t is the activation value of the output gate at time t; tanh is the hyperbolic tangent function; C t is the cell state at time t.

[0048] The operating status data and meteorological information are sorted by time to obtain the operating status data and meteorological information in the form of time series. The operating status data and meteorological information in the form of time series are input into the LSTM model, and the power load that the power system needs to provide in the future period is predicted by the LSTM model to obtain the predicted power load.

[0049] Step S106, determining a power dispatching plan through a game theory model according to the predicted power load, wherein the game theory model is used to simulate the interaction between multiple power generation nodes and multiple energy storage nodes, and the power dispatching plan is used to characterize the power generation of multiple power generation nodes and the storage power of multiple energy storage nodes.

[0050] The power generation node in the above steps is a device used to generate electricity in the power system, which may be a wind power station, a small gas turbine, etc., but is not limited thereto.

[0051] The energy storage node in the above steps is a device or system in the power system that can store and release electrical energy, and can be a battery energy storage system, etc., but is not limited thereto.

[0052] The game theory model in the above steps is used to analyze the strategic interactions between participants. In power system scheduling, the game theory model can determine the optimal power scheduling plan by finding Nash equilibrium based on the competition and cooperation between multiple power generation nodes and energy storage nodes and the interests of each power generation node and energy storage node.

[0053] In an optional embodiment, a game theory model can be first constructed based on the interaction between multiple power generation nodes and multiple energy storage nodes. Specifically, under the game theory framework, each power generation node and energy storage node can be regarded as a participant, a cost function can be constructed for each participant, and then game rules can be set to define how the participants interact with each other. Then, the constructed game theory model is solved with the predicted power load as a constraint to obtain a power dispatching plan with the lowest total cost.

[0054] Step S108, dispatching power to the power system according to the power dispatching plan.

[0055] In an optional embodiment, the power dispatching scheme is parsed to obtain the expected power generation of each power generation node and the expected storage capacity of each energy storage node. The dispatching center of the power dispatching system then generates specific operation instructions based on the parsing results. The operation instructions can be used for, but not limited to, adjusting the output power of the power generation equipment, controlling the charging and discharging operations of the energy storage equipment, etc. The dispatching center then sends the instructions to the corresponding power generation nodes and energy storage nodes through the control network of the power system. The power generation node adjusts the output of its generator set according to the received instructions to achieve the predetermined power generation; the energy storage node performs charging and discharging operations according to the dispatching scheme to maintain the specified storage capacity, thereby completing the power dispatching of the power system.

[0056] In an embodiment of the present invention, the operation status data of the power system and the meteorological information of the area where the power system is located are obtained; based on the operation status data and the meteorological information, the power load that the power system needs to provide is predicted by a time series prediction algorithm to obtain a predicted power load; according to the predicted power load, a power dispatching scheme is determined by a game theory model, wherein the game theory model is used to simulate the interaction between multiple power generation nodes and multiple energy storage nodes, and the power dispatching scheme is used to characterize the power generation of multiple power generation nodes and the storage power of multiple energy storage nodes; the power system is dispatched according to the power dispatching scheme. It is easy to notice that by introducing meteorological information, the accuracy of load forecasting is further enhanced, and the game theory model is used to determine the power dispatching scheme based on the predicted power load, which fully considers the mutual influence between the power generation nodes and the energy storage nodes in the power system, improves the rationality of the power dispatching scheme, and dispatches the power system according to the power dispatching scheme, thereby achieving the technical effect of improving the accuracy of power dispatching, thereby solving the technical problem of low accuracy of power dispatching.

[0057] In one embodiment of the present application, a power dispatching plan is determined through a game theory model based on predicted power load, including: determining a preliminary power dispatching plan through a reinforcement learning scheduling algorithm based on operating status data and predicted power load, wherein the preliminary power dispatching plan is used to characterize the power generation conditions of multiple power generation nodes and the power supply conditions of multiple energy storage nodes; and adjusting the preliminary power dispatching plan through a game theory model to obtain a power dispatching plan.

[0058] The reinforcement learning scheduling algorithm in the above steps can determine a preliminary power scheduling plan according to the predicted power load. The reinforcement learning scheduling algorithm can be a policy gradient method (Policy Gradient), but is not limited thereto.

[0059] The preliminary power dispatching plan in the above steps is used to characterize the overall power generation of the power generation nodes and the overall power supply of the energy storage nodes in the power system, that is, in order to meet the predicted power load, the overall power generation of multiple power generation nodes and the overall power supply or storage of multiple energy storage nodes. The difference between the preliminary power dispatching plan and the power dispatching plan is that the power dispatching plan is more detailed and can characterize the power generation that each power generation node should generate, and the power that each energy storage node should provide or store.

[0060] In an optional embodiment, the policy gradient method can be selected as the reinforcement learning scheduling algorithm, and the predicted power load can be used as the input of the policy gradient method. The action is selected according to the current state of the power system. For example, the action can be to increase photovoltaic output, mobilize energy storage battery power supply, etc., and the strategy for selecting the action is adjusted according to the feedback given by the policy gradient method until a better preliminary power scheduling plan is obtained.

[0061] The core update formula of the policy gradient method is as follows:

[0062]

[0063] Among them, J(θ) is the objective function; E is the expected value; is the gradient operator; π θ (a t ∣s t ) is the policy function, which can characterize the t Next, the agent chooses action a t The probability of θ Control of R t is the return starting from time t.

[0064] In one implementation of the present application, a preliminary power dispatching plan is adjusted through a game theory model to obtain a power dispatching plan, including: constructing a game theory model based on the preliminary power dispatching plan; solving the game theory model to obtain a power dispatching plan, wherein the power dispatching plan satisfies Nash equilibrium conditions.

[0065] The Nash equilibrium condition in the above steps means that each power generation node and energy storage node has selected the optimal strategy that can meet the predicted power load, and this strategy combination is stable for all participants, that is, no participant can obtain additional benefits or reduce costs by unilaterally changing the decision.

[0066] In an optional embodiment, a game theory model is constructed with the overall power generation of the power generation nodes in the preliminary power dispatching plan and the overall power that the energy storage nodes should provide or store as constraints to be met. Then the game theory model is solved. During the solution process, each power generation node and each energy storage node are regarded as participants, and the initial strategy of each power generation node and each energy storage node is set, and then it is iterated. During the iteration process, each participant updates its own strategy according to the current strategy and benefit function, while keeping the strategies of other participants unchanged, and then checks whether the current strategy combination meets the Nash equilibrium condition, that is, no participant can obtain higher benefits by unilaterally changing the strategy. Repeat the iterative process multiple times until the strategy combination meets the Nash equilibrium condition, and then the strategy combination that meets the Nash equilibrium condition is used as the power dispatching plan.

[0067] In one embodiment of the present application, a game theory model is constructed based on a preliminary power dispatching plan, including: determining a utility function of the game theory model based on the interactive utility between multiple intelligent agents, wherein different intelligent agents are obtained by combining different power generation nodes and different energy storage nodes; determining a strategy space of the game theory model according to the preliminary power dispatching plan, the strategy space containing multiple values ​​of power generation of different power generation nodes and multiple values ​​of stored power of different energy storage nodes; and constructing a game theory model based on the utility function and the strategy space.

[0068] The utility function in the above steps is an indicator used to measure the benefits of the agent under different strategy combinations.

[0069] The strategy space in the above steps is the set of all possible choices for each agent, including the possible values ​​of the power generation node and the possible values ​​of the energy storage node. The possible values ​​of the power generation node are used to characterize the power generation of the power generation node, and the possible values ​​of the energy storage node are used to characterize the charging and discharging state and power level of the energy storage node.

[0070] In an optional embodiment, different power generation nodes and different energy storage nodes are combined to obtain multiple different intelligent agents. Then, the impact of the strategy selection between multiple intelligent agents on each other's utility is considered to construct the utility function of the game theory model. The utility function of the game theory model is as follows:

[0071] U i =∑ j≠i u ij ;

[0072] Among them, U i is the total utility of agent i, u ij is the interaction utility between agent i and agent j.

[0073] Then, the overall power generation of the power generation nodes in the preliminary power dispatch plan and the overall power that the energy storage nodes should provide or store are used as constraints to be met. Under the premise of meeting the constraints, the various power generation values ​​that the power generation nodes can take and the various charging and discharging states and power level values ​​that the energy storage nodes can take are determined to obtain the strategy space. Then, a game theory model is constructed based on the utility function and strategy space.

[0074] In one embodiment of the present application, the method further includes: based on the operating status data, performing fault detection on the power system through an abnormality detection algorithm to obtain fault information; generating a fault recovery plan based on the fault information; and recovering the fault in the power system according to the fault recovery plan.

[0075] The anomaly detection algorithm in the above steps is used to identify anomalies in the operating status data that do not conform to the expected pattern or historical behavior, and can be a PCA (Principal Component Analysis) algorithm or an AutoEncoder algorithm, but is not limited thereto.

[0076] The fault information in the above steps is the specific information of the fault in the power system identified by the anomaly detection algorithm, including the fault type, fault severity and fault occurrence time, but not limited thereto.

[0077] The fault recovery plan in the above steps refers to a series of operation strategies formulated to restore normal operation after a power system fault occurs, which may include but is not limited to restarting equipment, transferring loads, dispatching backup power supplies, adjusting network topology, etc.

[0078] In an optional embodiment, the power system can be detected for faults through the edge layer of the power dispatching system. Specifically, the PCA algorithm can be selected as the anomaly detection algorithm, and the operating status data can be input into the PCA algorithm, which outputs the fault information of the power system. Then, based on the fault severity in the fault information, a fault recovery plan is determined. The fault recovery plan may include but is not limited to local restart, reporting to the cloud, dispatching backup power, adjusting load distribution, etc. Then, the measures in the fault recovery plan are executed to recover the fault in the power system.

[0079] In one implementation of the present application, a fault recovery plan is generated based on fault information, including: locating the fault based on the fault information to obtain the fault location; evaluating the impact range of the fault based on the fault location to obtain a fault evaluation result; and determining a fault recovery plan based on the fault evaluation result.

[0080] In an optional embodiment, the fault location can be determined based on the fault characteristics in the fault information and the topological structure of the power system. Then, the number of lines, equipment, and users affected by the fault is determined based on the fault location to determine the maximum damage that the fault may cause and obtain a fault assessment result. Then, combined with the fault assessment results and the current state and backup resources of the power system, the available recovery means are determined to determine the fault recovery plan. The recovery plan may include but is not limited to quickly isolating the fault area to avoid the spread of the fault; dispatching backup power supplies or energy storage systems to maintain a balance between power supply and demand; adjusting the power flow path to bypass the faulty line, etc.

[0081] In one embodiment of the present application, obtaining the operating status data of the power system includes: collecting the original operating data of the power system based on a long distance wide area network protocol; performing a data cleaning operation on the original operating data to obtain cleaned data; using a sliding window method to perform a feature extraction operation on the cleaned data to obtain the operating status data of the power system.

[0082] The Long-range Wide Area Network Protocol (LoRaWAN) in the above steps can be used in low-power, long-distance IoT communication scenarios and can transmit data over long distances.

[0083] The raw operating data in the above steps are unprocessed data collected directly from the power system through IoT devices.

[0084] The data cleaning operations in the above steps may include but are not limited to: outlier detection and removal, missing value filling, data format standardization, etc.

[0085] The sliding window method in the above steps is a data processing technology used to extract continuous data segments of a specific length from the data, which helps to capture the dynamic changes in the operating status of the power system.

[0086] The feature extraction operation in the above steps is used to extract feature information from the data after data cleaning, so as to simplify the data and reduce the amount of calculation.

[0087] In an optional embodiment, a LoRaWAN network is established at the perception layer of the power dispatching system, the IoT devices at the perception layer are connected to the network center node, and original operation data is obtained from multiple IoT devices deployed in the power system based on the LoRaWAN network.

[0088] In the LoRaWAN network, the propagation loss formula of the signal from the transmitting point to the receiving point is as follows:

[0089]

[0090] Among them, L is the propagation loss of the signal from the transmitting point to the receiving point, L0 is the loss of the reference distance, n is the environmental attenuation factor, d is the propagation distance, and d0 is the reference distance.

[0091] The formula for how the LoRaWAN gateway's transmission power changes with distance is as follows:

[0092]

[0093] Among them, P tx is the transmission power of the LoRaWAN gateway, is the reference transmission power, α is the environmental attenuation factor, and d is the distance.

[0094] The collected raw operation data can then be transmitted from the perception layer to the edge layer of the power dispatching system, that is, the edge computing node, where the raw operation data is cleaned. During the data cleaning process, the DBSCAN algorithm can be used to detect and mark outliers in the raw operation data. The DBSCAN algorithm is a density-based clustering algorithm. The formula for defining the neighborhood of the DBSCAN algorithm is as follows:

[0095] N ∈ (p)={q∈D|dist(p,q)≤∈};

[0096] Among them, N∈(p) is the neighborhood of point p, that is, the set of all points within a given radius and whose distance from point p is within a certain range; ∈ is the given radius; dist is the distance function, which can be Euclidean distance or Manhattan distance, but not limited to this; D is the set of all data points, and q is any point in D.

[0097] After detecting and marking the outliers in the original operation data, the outliers can be removed. Then, whether there are missing values ​​in the original operation data is detected, and the missing values ​​are processed using a filling method, which may include but is not limited to forward filling, backward filling, mean filling or prediction filling. The original operation data is then standardized to ensure that the format and unit of the data are consistent, and the cleaned data is obtained.

[0098] Then, the sliding window method is used to extract features from the cleaned data. When performing feature extraction, a sliding window is defined, and the size and step size of the window are determined. Then, the sliding window is moved on the cleaned data one time step at a time, and the cleaned data in the window is sliced ​​to form a series of data segments. Feature calculation is performed on the data in each data segment. The feature calculation may include: calculating the mean of the power output, voltage, and current in the data segment, calculating the standard deviation of the power output, voltage, and current in the data segment, and calculating the maximum and minimum values ​​of the power output, voltage, and current in the data segment, but is not limited to this. Then, according to the needs of the power system dispatch, the results of the feature calculation are combined to obtain the operating status data.

[0099] A preferred embodiment is described below. Figure 2 is a schematic diagram of an optional power dispatching system according to an embodiment of the present invention, such as Figure 2 As shown, the power dispatching system includes photovoltaic Internet of Things devices, an edge layer and a cloud layer. Among them, the photovoltaic Internet of Things devices can send data to the edge layer through the LoRaWAN protocol, and the edge layer can send data to the cloud layer through the HTTPS (Hypertext Transfer Protocol Secure) protocol. The photovoltaic Internet of Things device is used to collect the operating status data of the power system, the edge layer is used to perform data cleaning operations and feature extraction operations on the operating status data, and the cloud layer is used to determine the power dispatching plan based on the data processed by the data cleaning operations and feature extraction operations. When determining the power dispatching plan, the power load that the power system needs to provide can be predicted according to the LSTM model to obtain the predicted power load, and then the predicted power load is processed using the policy gradient method to obtain a preliminary power dispatching plan, and then the preliminary power dispatching plan is adjusted according to the game theory model to obtain a power dispatching plan.

[0100] Example 2

[0101] According to an embodiment of the present invention, an embodiment of a power dispatching device of a power system is provided, and the device can execute the power dispatching method of the power system provided in the above-mentioned embodiment 1. The specific implementation method and preferred application scenario are the same as those of the above-mentioned embodiment 1, and will not be repeated here.

[0102] Figure 3 is a schematic diagram of a power dispatching device of a power system according to an embodiment of the present invention, such as Figure 3 As shown:

[0103] The acquisition module 30 is used to acquire the operating status data of the power system, wherein the operating status data of the power system includes historical operating status data and current operating status data.

[0104] The prediction module 32 is used to predict the power load that the power system needs to provide based on the current operating status data and meteorological forecast information through a time series prediction algorithm to obtain a predicted power load, wherein the time series prediction algorithm is trained through historical operating status data.

[0105] The determination module 34 is used to determine the power dispatching plan through the game theory model according to the predicted power load, wherein the game theory model is used to simulate the interaction between the power generation node and the energy storage node, and the power dispatching plan is used to characterize the power generation of multiple power generation nodes and the storage power of multiple energy storage nodes in the power system.

[0106] The power generation module 36 is used to perform power dispatch according to the power dispatch plan.

[0107] The determination module includes: a determination unit, which is used to determine a preliminary power dispatching plan through a reinforcement learning scheduling algorithm based on operating status data and predicted power load, wherein the preliminary power dispatching plan is used to characterize the power generation conditions of multiple power generation nodes and the power supply conditions of multiple energy storage nodes; an adjustment unit, which is used to adjust the preliminary power dispatching plan through a game theory model to obtain a power dispatching plan.

[0108] The adjustment unit is also used to construct a game theory model based on the preliminary power dispatching plan; solve the game theory model to obtain the power dispatching plan, wherein the power dispatching plan satisfies the Nash equilibrium condition.

[0109] The adjustment unit is also used to determine the utility function of the game theory model based on the interactive utility between multiple intelligent agents, wherein different intelligent agents are obtained by combining different power generation nodes and different energy storage nodes; according to the preliminary power dispatch plan, determine the strategy space of the game theory model, the strategy space contains multiple values ​​of the power generation of different power generation nodes and multiple values ​​of the storage power of different energy storage nodes; construct the game theory model based on the utility function and the strategy space.

[0110] The acquisition module includes: a detection unit, which is used to perform fault detection on the power system based on the operating status data through an abnormal detection algorithm to obtain fault information; a generation unit, which is used to generate a fault recovery plan based on the fault information; and a recovery unit, which is used to recover the fault in the power system according to the fault recovery plan.

[0111] The generating unit is further used to locate the fault based on the fault information to obtain the fault location; evaluate the impact range of the fault based on the fault location to obtain a fault evaluation result; and determine a fault recovery plan according to the fault evaluation result.

[0112] The acquisition module also includes: a collection unit, which is used to collect the original operation data of the power system based on the long-distance wide area network protocol; a data cleaning unit, which is used to perform data cleaning operations on the original operation data to obtain data after data cleaning; a feature extraction unit, which is used to perform feature extraction operations on the data after data cleaning using a sliding window method to obtain the operation status data of the power system.

[0113] Example 3

[0114] According to an embodiment of the present invention, there is also provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the power dispatching method of the power system in Embodiment 1 is executed when the program is running.

[0115] Example 4

[0116] An embodiment of the present application also provides a computer-readable storage medium, which includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the power dispatching method of the power system in each embodiment of the present invention.

[0117] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0118] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0119] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0120] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0121] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0122] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0123] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A power dispatching method for a power system, characterized in that: The power system includes a plurality of power generation nodes and a plurality of energy storage nodes, and the method includes: Acquiring operating status data of the power system and meteorological information of the area where the power system is located; Based on the operating status data and the meteorological information, predicting the power load that the power system needs to provide by using a time series prediction algorithm to obtain a predicted power load; According to the predicted power load, a power dispatching scheme is determined by a game theory model, wherein the game theory model is used to simulate the interaction between the multiple power generation nodes and the multiple energy storage nodes, and the power dispatching scheme is used to characterize the power generation of the multiple power generation nodes and the storage power of the multiple energy storage nodes; The power system is dispatched according to the power dispatch plan.

2. The power dispatching method of the power system according to claim 1, characterized in that: According to the predicted power load, a power dispatching scheme is determined through a game theory model, including: Determine a preliminary power dispatching scheme by a reinforcement learning scheduling algorithm according to the operating status data and the predicted power load, wherein the preliminary power dispatching scheme is used to characterize the power generation conditions of the multiple power generation nodes and the power supply conditions of the multiple energy storage nodes; The preliminary power dispatching plan is adjusted through a game theory model to obtain the power dispatching plan.

3. The power dispatching method of the power system according to claim 2, characterized in that: The preliminary power dispatching plan is adjusted by a game theory model to obtain the power dispatching plan, including: Based on the preliminary power dispatching plan, constructing the game theory model; The game theory model is solved to obtain the power dispatching scheme, wherein the power dispatching scheme satisfies the Nash equilibrium condition.

4. The power dispatching method of the power system according to claim 3, characterized in that: Based on the preliminary power dispatching plan, the game theory model is constructed, including: Determining a utility function of the game theory model based on the interactive utility between multiple agents, wherein different agents are obtained by combining different power generation nodes and different energy storage nodes; According to the preliminary power dispatching plan, determining the strategy space of the game theory model, wherein the strategy space includes multiple values ​​of power generation of different power generation nodes and multiple values ​​of stored power of different energy storage nodes; The game theory model is constructed based on the utility function and the strategy space.

5. The power dispatching method of the power system according to claim 1, characterized in that: The method further comprises: Based on the operating status data, performing fault detection on the power system through an abnormality detection algorithm to obtain fault information; generating a fault recovery plan based on the fault information; The fault in the power system is restored according to the fault recovery plan.

6. The power dispatching method of the power system according to claim 5, characterized in that: Generating a fault recovery plan based on the fault information includes: Locating the fault based on the fault information to obtain the fault location; Evaluate the impact range of the fault based on the fault location to obtain a fault evaluation result; The fault recovery plan is determined according to the fault assessment result.

7. The power dispatching method of the power system according to claim 1, characterized in that: Obtain the operating status data of the power system, including: Collecting original operation data of the power system based on a long-distance wide area network protocol; Performing a data cleaning operation on the original operation data to obtain cleaned data; A sliding window method is used to perform feature extraction operations on the cleaned data to obtain operating status data of the power system.

8. A power dispatching device for a power system, characterized in that: include: An acquisition module, used for acquiring operation status data of the power system, wherein the operation status data of the power system includes historical operation status data and current operation status data; A prediction module, configured to predict the power load that the power system needs to provide based on the current operating status data and the meteorological forecast information by using a time series prediction algorithm to obtain a predicted power load, wherein the time series prediction algorithm is trained by using the historical operating status data; A determination module, configured to determine a power dispatching scheme through a game theory model according to the predicted power load, wherein the game theory model is used to simulate the interaction between power generation nodes and energy storage nodes, and the power dispatching scheme is used to characterize the power generation of multiple power generation nodes and the storage power of multiple energy storage nodes in the power system; The power generation module is used to perform power dispatch according to the power dispatch plan.

9. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program, when running, executes the power dispatching method of the power system described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the storage medium is located is controlled to execute the power dispatching method of the power system according to any one of claims 1 to 7.

Images