Distribution area distributed resource adjusting method, device, equipment and medium

Abstract

The invention relates to a transformer area distributed resource adjusting method and device, equipment and a medium, and the method comprises the steps: obtaining power grid data resources, and generating a power grid model based on the power grid data resources; the power grid model is converted into a graph model, the graph model is analyzed, topological structure information between transformer areas is obtained, and transformer areas with mutual assistance capability form a transformer area group; inputting the net load prediction characteristic factor data of each transformer area into a net load prediction model to obtain a net load prediction value of each transformer area, and superposing the net load prediction values belonging to the same transformer area group to obtain a net load prediction value of the transformer area group; and determining a transformer area group energy management strategy according to the net load prediction value of the transformer area group and the energy storage capacity of the transformer area group, and adopting the transformer area group energy management strategy to control distributed resources in the transformer area group. According to the invention, the reverse overload of the transformer area can be reduced while the photovoltaic absorption capability is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a method, device, equipment and medium for regulating distributed resources in a substation area. Background Art

[0002] The high proportion of "random, intermittent" distributed resource access, the large-scale application of medium and low voltage flexible DC and power electronics technology, has made the distribution network present an active, dynamic, and AC / DC hybrid development trend. The access of a large number of distributed photovoltaics has led to reverse power and even reverse overload problems in a large number of distribution stations. The configuration of energy storage can achieve dynamic capacity expansion in the substation area and effectively manage overload. However, the investment cost of energy storage is high, and the configuration capacity is often limited, which cannot completely solve the problem of reverse overload in the substation area.

[0003] The existing public patent document CN118889500A discloses a heavy overload management method based on the real-time load rate of the photovoltaic storage and charging area. This method only considers the photovoltaic storage and charging equipment, and does not consider the power mutual assistance capability generated by the flexible interconnection equipment connecting the areas to form an area group. Therefore, it is not suitable for the existing distribution network. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a distributed resource regulation method, device, equipment and medium in a substation area, which can improve the photovoltaic absorption capacity while reducing the reverse overload of the substation area.

[0005] The technical solution adopted by the present invention to solve the technical problem is: to provide a distributed resource adjustment method for a substation, comprising the following steps:

[0006] Acquire power grid data resources, and generate a power grid model based on the power grid data resources;

[0007] Convert the power grid model into a graph model, analyze the graph model, obtain topological structure information between substations, and group substations with mutual assistance capabilities into substation groups to obtain N substation groups, where N is a positive integer;

[0008] Input the net load prediction characteristic factor data of each substation into the net load prediction model to obtain the net load prediction value of each substation, and superimpose the net load prediction values ​​belonging to the same substation group to obtain the net load prediction values ​​of N substation groups;

[0009] The substation group energy management strategy of the nth substation group is determined according to the net load forecast value of the nth substation group and the energy storage capacity of the substation group, and the substation group energy management strategy of the nth substation group is used to control the distributed resources within the nth substation group, wherein n is a positive integer not greater than N.

[0010] The converting the power grid model into a graph model specifically includes: abstracting each device in the power grid model as a node in the graph model, and abstracting the connection relationship between each device as an edge in the graph model.

[0011] When analyzing the graph model, one or more of a breadth-first search algorithm, a depth-first search algorithm and a Dijkstra algorithm are used.

[0012] The net load forecasting model is obtained in the following way:

[0013] Based on the historical net load measurement data of all substations, the corresponding actual meteorological data and forecast data, the net load forecast characteristic factors of the substations are analyzed, and a forecast model based on long short-term memory neural network is established using a deep learning framework. The forecast model is optimized using a nonlinear weighted particle swarm optimization algorithm to obtain a net load forecast model.

[0014] The step of determining the energy management strategy of the nth substation group according to the net load prediction value of the nth substation group and the energy storage capacity of the substation group, and controlling the distributed resources in the nth substation group by using the energy management strategy of the nth substation group, specifically includes:

[0015] Determine the PV cumulative net reverse transmission capacity of the nth substation group according to the net load forecast value of the nth substation group;

[0016] Determine whether the accumulative net reverse transmission capacity of the nth grid area group exceeds the energy storage capacity of the nth grid area group;

[0017] If the accumulative net reverse transmission capacity of the nth group of substations does not exceed the energy storage capacity of the nth group of substations, the maximum reverse load reduction real-time energy regulation strategy is used to control the distributed resources in the nth group of substations;

[0018] If the accumulative net reverse transmission capacity of the nth grid area group exceeds the energy storage capacity of the nth grid area group, it is determined whether the maximum reverse transmission power of at least one grid area in the nth grid area group exceeds the maximum reverse load rate standard;

[0019] If the maximum reverse transmission power of at least one area in the nth area group exceeds the maximum reverse load rate standard, the uniform reverse load rate energy regulation strategy during the reverse transmission period is adopted to control the distributed resources in the nth area group;

[0020] If the maximum reverse power of no area in the nth area group exceeds the maximum reverse load rate standard, the energy regulation strategy of uniformly reducing the reverse load rate during the heavy load period is adopted to control the distributed resources in the nth area group.

[0021] The step of determining the photovoltaic cumulative net reverse transmission capacity of the nth group of substations according to the net load forecast value of the nth group of substations specifically includes:

[0022] Determine whether the net load forecast value of the nth substation group is less than zero;

[0023] If the net load forecast value of the nth group of grid stations is less than zero, the PV cumulative net reverse transmission capacity of the nth group of grid stations is determined as the absolute value of the net load forecast value of the nth group of grid stations;

[0024] If the net load forecast value of the nth substation group is greater than or equal to zero, the photovoltaic cumulative net reverse transmission capacity of the nth substation group is determined to be zero.

[0025] When the maximum reverse load rate real-time energy regulation strategy is used to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged according to the maximum charging power acceptable to the distributed resources, that is, ,in, is the charging power of the i-th distributed resource in the n-th substation group under the maximum reverse load reduction real-time energy regulation strategy, is the maximum charging power acceptable to the i-th distributed resource in the n-th substation group.

[0026] When the energy regulation strategy of uniform reverse load rate during the reverse transmission period is adopted to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged at a smaller value of a constant power or a maximum charging power acceptable to the device during the entire reverse transmission period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the n-th substation group under the uniform reverse load rate energy regulation strategy during the reverse period, is the maximum charging power acceptable to the i-th distributed resource in the n-th station group, S ES,i is the rated capacity of the ith distributed resource in the nth substation group, T over This is the entire reverse transmission period.

[0027] When the energy regulation strategy of uniformly decreasing the reverse load rate during the heavy load period is adopted to control the distributed resources in the substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged at a smaller value of a constant power or a maximum charging power acceptable to the device during the heavy load period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the n-th substation group under the energy regulation strategy of uniformly decreasing the reverse load rate during the heavy load period, is the maximum charging power acceptable to the i-th distributed resource in the n-th station group, S ES,i is the rated capacity of the ith distributed resource in the nth substation group, T LFDover This is the heavy load period.

[0028] The technical solution adopted by the present invention to solve the technical problem is: to provide a distributed resource adjustment device for a substation, comprising:

[0029] An acquisition module, used for acquiring power grid data resources and generating a power grid model based on the power grid data resources;

[0030] An analysis module, used to convert the power grid model into a graph model, analyze the graph model, obtain topological structure information between substations, and group substations with mutual assistance capabilities into substation groups to obtain N substation groups, where N is a positive integer;

[0031] A prediction module is used to input the net load prediction characteristic factor data of each substation into the net load prediction model to obtain the net load prediction value of each substation, and to superimpose the net load prediction values ​​belonging to the same substation group to obtain the net load prediction values ​​of N substation groups;

[0032] A control module is used to determine the substation group energy management strategy of the nth substation group based on the net load prediction value of the substation group and the energy storage capacity of the substation group, and use the substation group energy management strategy of the nth substation group to control the distributed resources within the nth substation group, wherein n is a positive integer not greater than N.

[0033] The technical solution adopted by the present invention to solve its technical problem is: to provide an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned distributed resource adjustment method when executing the computer program.

[0034] The technical solution adopted by the present invention to solve its technical problem is: providing a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned distributed resource adjustment method of the substation area are implemented.

[0035] Beneficial Effects

[0036] Due to the adoption of the above-mentioned technical scheme, the present invention has the following advantages and positive effects compared with the prior art: the present invention traverses and cyclically searches for the power mutual assistance capability between substations, organizes substations with power mutual assistance capability into substation groups, utilizes the power mutual assistance characteristics between substations, and adopts different substation group energy management strategies according to the results of substation net load prediction to adapt to the joint autonomy problem between substations and substations under the new distribution system. This energy management strategy has the advantages of predicting photovoltaic reverse overload substations to respond in advance, improving photovoltaic absorption, high overload control effect, fast speed and applicable real-time energy management. The present invention also changes the problems of incomplete distribution network equipment data, chaotic topology connectivity, inability to perceive model changes, and difficulty in distinguishing data authenticity by automatically acquiring, verifying and correcting substation data resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a flow chart of a method for adjusting distributed resources in a substation according to a first embodiment of the present invention;

[0038] Figure 2 It is a flow chart of determining the energy management strategy of the substation group in the first embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall within the scope limited by the appended claims of the application equally.

[0040] The first embodiment of the present invention relates to a distributed resource adjustment method for a station area, such as Figure 1 As shown, the following steps are included:

[0041] Step 1: Acquire power grid data resources and generate a power grid model based on the power grid data resources.

[0042] The power grid data resources obtained in this step include stock differentiated CIM model files, equipment parameters, svg graphics, operating data, and coordinate data. When used for the first time or when the substation equipment is changed, the stock differentiated CIM model files, equipment parameters, svg graphics, operating data, and coordinate data are obtained. For scenarios such as stock data errors, line reconnections, and new equipment, verification and correction can be performed.

[0043] The following description will be made using a photovoltaic device as an example.

[0044] First, connect with the local photovoltaic project filing system and equipment management system to obtain key information such as the geographical location (latitude and longitude), installed capacity, component type, and grid connection time of the photovoltaic power station. These static data provide the basis for the initial positioning and attribute annotation of the photovoltaic power station in the power grid topology. For example, it is necessary to clarify the specific geographical coordinates and size of a distributed photovoltaic power station in a specific area, so as to facilitate the subsequent analysis of its potential impact on the surrounding power grid nodes.

[0045] Next, the format of multi-source heterogeneous data is unified to ensure that photovoltaic static data reported by different manufacturers and different regions can be stored in the database in a standardized format to eliminate interference from data inconsistency.

[0046] Then, with the help of smart meters, communication gateways and Internet of Things technology, the operating parameters of distributed photovoltaic power generation, voltage, current, power factor, etc. are collected in real time, and the data is pushed to the resource pool at a frequency of seconds or minutes through high-speed communication networks such as power fiber optic private networks and 5G wireless private networks.

[0047] Afterwards, through the established data caching and preprocessing mechanism, the high-frequency dynamic measurement data is preliminarily screened and aggregated to eliminate outliers, reduce invalid data transmission and storage burdens, and ensure the accuracy of subsequent data association and analysis.

[0048] Finally, an integrated data aggregation framework is constructed to achieve standardized output of data models such as primary grid, secondary terminals, collection and measurement, and metering to obtain a power grid model.

[0049] This step is based on the commonly used system-side data model standard (IEC CIM) for power systems. It standardizes the description of various objects in the power system, such as generators, transformers, transmission lines, loads, etc., as well as their attributes and relationships, so that different systems can understand and process data in a consistent manner, breaking down data silos.

[0050] It can be seen that this implementation method changes the previous problems of incomplete distribution network equipment data, chaotic topology connectivity, imperceptible model changes, and difficulty in distinguishing data authenticity by automatically acquiring, verifying and correcting substation data resources.

[0051] Step 2: convert the power grid model into a graph model, analyze the graph model, obtain topological structure information between substations, and group substations with mutual assistance capabilities into substation groups to obtain N substation groups, where N is a positive integer.

[0052] This step reads the power grid model generated in step 1, abstracts each device in the power grid model (such as transformers, switches, lines, energy storage, photovoltaics, charging piles, etc.) as a node in the graph model, and abstracts the connection relationship between each device as an edge in the graph model, thereby converting the physical network of the substation into a mathematical graph model. The topological structure information of the substation is then obtained by searching and analyzing this graph model.

[0053] In this step, each device is abstracted as a node in the graph model, including:

[0054] Transformer node: When a transformer is considered as a node in a graph model, in addition to basic identification information, its node attributes can include static parameters such as transformer capacity, transformation ratio, short-circuit impedance, etc. These parameters are extracted from the corresponding device description part in the CIM model. For example, a 500kVA distribution transformer has a capacity attribute value of 500kVA. These attributes play a key role in the subsequent analysis of power transmission, voltage conversion, etc. in the substation area.

[0055] Switch node: As a node in the graph model, the switch has unique properties, such as the type of switch (knife switch, circuit breaker, etc.), rated voltage, rated current, and current open and close status (this is a dynamic property and needs to be updated in real time). For example, a circuit breaker switch has a rated voltage of 0.4kV and a rated current of 630A. The real-time open and close status can reflect the current line on and off status, affecting the topological connectivity analysis of the substation.

[0056] Line node: When a line is abstracted and used as a node in a graph model, its attributes can include the line length, conductor type, resistance, reactance and other electrical parameters, as well as the starting and ending points of the line (connection directions with other nodes). For example, a line with a length of 1km and LGJ-120 conductors, these attributes will play an important role in subsequent power flow calculations based on the graph model.

[0057] Energy storage node: When energy storage equipment is used as a node in the graph model, the corresponding node attributes include energy storage capacity, charge and discharge power, current state of charge (SOC), etc. For example, for an energy storage device with a total energy storage capacity of 100kW・h, its capacity attribute records this value, and the real-time change of SOC reflects the current power reserve of the energy storage device, which can be used to analyze the energy regulation capacity of the substation.

[0058] Photovoltaic node: When photovoltaic equipment is abstracted as a node in the graph model, its node attributes include installed capacity, current power generation (dynamic attribute), maximum power point tracking (MPPT) status, etc. For example, for a distributed photovoltaic with an installed capacity of 50kW, its installed capacity attribute records the value fixedly, while the real-time power generation will change with external conditions such as light, reflecting its real-time contribution to the power injection of the substation.

[0059] Charging pile node: When the charging pile is abstracted as a node in the graph model, its node attributes include rated charging power, charging interface type, whether it is currently charging (dynamic attribute), etc. For example, for a DC charging pile with a rated charging power of 30kW, its rated power attribute clearly defines its charging capacity, while the real-time charging status is related to the load distribution of the substation.

[0060] Flexible interconnection device node: When the flexible interconnection device is abstracted as a node in the graph model, its node identifier is assigned a unique node identifier for each flexible interconnection device, which is convenient for accurate distinction and reference in the graph model. Its basic characteristic attributes include rated capacity, which determines the scale of power interaction it can achieve. For example, a flexible interconnection device with a rated capacity of 500kW means that it can achieve a maximum power transmission or regulation of 500kW between different connected parts; there is also a voltage adjustment range, which clarifies the upper and lower limits of the voltage on the two or more sides of the connection that the device can adjust, such as being able to flexibly adjust the voltage between 0.38kV and 0.42kV to maintain the stability of the voltage in the substation. Its operating status attributes are: real-time power transmission direction and size are key dynamic attributes, because during the operation of the flexible interconnection device, power may flow in both directions and change in size in real time according to factors such as load conditions and power output in different areas of the substation. Accurately recording this attribute can clearly reflect its role in power mutual assistance and power flow optimization in the substation. In addition, the operating mode (such as voltage regulation mode, power balance mode, etc.) is also a dynamic attribute, which reflects the current functional operating state of the device, making it easier to understand its specific impact on the operation of the substation.

[0061] In this step, the connection relationship between each device is abstracted as an edge in the graph model, including:

[0062] Edge attribute definition: In addition to the basic meaning of representing the connection between devices, edges can also have other attributes, such as the impedance value of the connecting line (for line connection), power transmission direction (in the case of power flow, it can be determined by analyzing the power flow direction, which is a dynamic attribute), etc. For example, the impedance value of the edge connecting the transformer and a certain line reflects the power loss during the power transmission process, while the power transmission direction can reflect the state of power output from the transformer to the line or reverse flow.

[0063] Multiple connection relationships are reflected: the connection forms between different devices are diverse, such as the connection between transformers and lines through outgoing terminals, and the connection between switches and lines through switches or circuit breakers. These are uniformly abstracted as edges in the graph model, accurately depicting the electrical connection path of the physical network in the substation area, and providing a clear structural framework for subsequent analysis of topological connectivity, power flow, etc.

[0064] In this implementation, the distributed photovoltaic power station is abstracted as a "point" element in the graph model, and it is associated with the transformer nodes, switch nodes, line intersection nodes, etc. in the power grid based on the electrical connection relationship. For example, if a distributed photovoltaic is directly connected to a low-voltage bus in a certain area, a directed edge is created in the graph model from the photovoltaic node to the corresponding transformer low-voltage side node to clearly define the power injection direction and the grid level. Through a unique identification coding system, such as equipment asset number, global unique identifier (GUID), etc., the accuracy and uniqueness of the association between each photovoltaic node and the grid node are ensured to avoid duplication or incorrect connection, laying a solid foundation for subsequent complex power grid analysis.

[0065] In this implementation, the lines connecting the photovoltaic power station and the power grid and the various devices within the power grid are mapped to the "line" elements of the graph model, and a bidirectional association relationship is established between the line entity and the node. In addition to the electrical connection, the electrical parameter information such as the resistance, reactance, and current carrying capacity of the line, as well as the operation and maintenance data such as line inspection records and fault history are also associated. By using the topological connection sequence of the line, an orderly node-line sequence is formed in the graph structure to accurately depict the power transmission path. When the distributed photovoltaic output changes, it can quickly follow the line association to track the chain effect on the equipment along the line and other related photovoltaic power stations.

[0066] This implementation method also divides the grid "surface" elements into units such as substations and power supply areas, and includes distributed photovoltaics and power grid equipment in the same area into the corresponding surface management. An association model containing information such as subordination, capacity share, and load density is constructed between the surface and internal nodes and lines to control the cluster characteristics of distributed photovoltaics from a macro-regional perspective. For example, the proportion of the total installed capacity of distributed photovoltaics in a certain substation area to the load capacity of the substation area is counted to intuitively present the degree of support of photovoltaic resources in the region for local power supply, and assist in formulating targeted regional energy balance and scheduling strategies.

[0067] In this step, when analyzing the graph model, one or more algorithms such as a breadth-first search algorithm, a depth-first search algorithm, and a Dijkstra algorithm may be used.

[0068] Among them, the breadth-first search algorithm starts from a given starting node, first visits all the adjacent nodes of the node, and then visits the adjacent nodes of these adjacent nodes in turn, and so on, until all reachable nodes are traversed. In the topological search of low-voltage substations in the distribution network, the breadth-first search algorithm can be used to quickly determine other devices that can be reached from a certain substation device, thereby constructing the local topological structure of the substation. The depth-first search algorithm starts from the starting node and explores as deeply as possible along a path until it cannot continue or reaches the target node, then backtracks to the previous step and continues to explore other paths. The depth-first search algorithm has advantages in finding specific paths or detecting loops. For example, when analyzing whether there is a closed loop in the power supply path within the substation, the depth-first search algorithm can be used for deep exploration. The Dijkstra algorithm is mainly used to calculate the shortest path from a node to all other nodes in the graph. In the distribution network, it can be used to find the shortest electrical distance between different nodes in the substation, thereby determining the optimal path for power transmission, which is of great significance for evaluating the power mutual assistance capacity within the substation and optimizing power dispatching.

[0069] After completing the above analysis, it can be determined which substations have the power mutual assistance capability, so that the substations with mutual assistance capability can be grouped into substation groups. If a substation does not have the power mutual assistance capability with any other substation, then the substation is treated as a substation group. At this time, the substation group composed of multiple substations is called a multi-substation group, and the substation group composed of a single substation is called a single substation group.

[0070] Step 3: Input the net load prediction characteristic factor data of each substation into the net load prediction model to obtain the net load prediction value of each substation, and superimpose the net load prediction values ​​belonging to the same substation group to obtain the net load prediction values ​​of N substation groups.

[0071] In this step, a day-ahead net load forecast is made for all substations, and the day-ahead net load forecast values ​​of all substations in the same substation group are superimposed to obtain the net load forecast results of each substation group. The net load forecast results serve as the basis for the execution of the substation group energy management strategy.

[0072] When making predictions in this step, based on the historical net load measurement data of all substations, the corresponding meteorological conditions and forecast data, the net load forecast characteristic factors of the substations are analyzed, and a net load forecast model of the substation based on the long short-term memory neural network (LSTM) is established using the deep learning framework of artificial intelligence. The LSTM forecast model is optimized using the nonlinear weighted particle swarm optimization algorithm (IPSO) to obtain the IPSO-LSTM intelligent forecast model, and the IPSO-LSTM intelligent forecast model is used as the net load forecast model. Finally, the net load forecast model is used to forecast the net load of all substations. The specific process is as follows:

[0073] Firstly, the characteristic factors of net load prediction in the substation are analyzed based on multiple data sources.

[0074] The historical measurement data of the net load of the substation records the actual net load of the substation in various time periods in the past, including load change information in different seasons and time periods (such as working days and holidays, peak and trough periods, etc.), which is the basic data source for analyzing the load characteristics of the substation and exploring potential laws. Through statistical analysis of these historical data, such as calculating statistical quantities such as mean, variance, extreme value, and analyzing the changing trends of daily load curves, weekly load curves, and annual load curves, we can have a preliminary understanding of the periodicity and volatility of the substation load.

[0075] Meteorological factors have a significant impact on the net load of the substation. Actual meteorological data such as temperature, humidity, light intensity, and wind speed are closely related to the loads of cooling, heating, lighting, and photovoltaic power generation in the substation. For example, the cooling load will increase in hot weather, while the photovoltaic power generation output will decrease on cloudy days, thus affecting the net load of the substation. At the same time, meteorological forecast data can provide a basis for predicting future load changes in advance, so that the forecast can take into account the impact of upcoming changes in meteorological conditions on the load, and improve the foresight and accuracy of the forecast.

[0076] This step combines the historical measurement data of the net load of the substation and the corresponding meteorological conditions and forecast data, and uses data analysis methods such as correlation analysis and principal component analysis to screen out the characteristic factors that have a significant impact on the net load of the substation. For example, through correlation analysis, it is found that the correlation coefficient between temperature and the net load of a certain substation in summer is as high as 0.8, indicating that temperature is a key influencing factor of the net load of the substation in summer, and then it is identified as one of the important prediction characteristic factors. These characteristic factors will be used as input variables for the subsequent establishment of the prediction model, laying the foundation for accurate prediction of the net load of the substation.

[0077] Next, a substation net load prediction model based on long short-term memory neural network (LSTM) is established.

[0078] Long short-term memory neural network (LSTM) is a special recurrent neural network (RNN) that aims to solve the gradient vanishing or gradient exploding problems of traditional RNN when processing long sequence data, and can better capture the long-term dependencies in time series data. In the net load forecasting of the substation area, since the load data has obvious time series characteristics, such as daily, weekly, and annual periodicity and dynamic volatility over time, LSTM can remember the past load information through its unique memory unit structure (including input gate, forget gate, output gate, etc.), and predict the future net load value based on the current input and historical memory.

[0079] The construction process of the LSTM-based net load prediction model in this step is as follows:

[0080] First, determine the number of input layer nodes of the model. Its value usually corresponds to the number of characteristic factors selected for net load prediction of the substation. For example, if five characteristic factors such as temperature, date type (working day / holiday), and time are selected, the number of input layer nodes is set to 5. In the middle, set the appropriate number of hidden layers and the corresponding number of neurons, which often needs to be determined through multiple experiments and tuning to achieve better prediction performance. Finally, the output layer, the number of nodes in the output layer is generally 1, which is used to output the predicted net load value of the substation. Then train the model, take the historical characteristic factor data as input, and the corresponding historical net load data as output label, and use appropriate training algorithms (such as stochastic gradient descent method, etc.) to let the model learn the mapping relationship between input and output.

[0081] Then, the nonlinear weighted particle swarm optimization algorithm (IPSO) is used to optimize the LSTM-based substation net load forecasting model.

[0082] The nonlinear weighted particle swarm optimization algorithm (IPSO) is an optimization algorithm developed on the basis of the traditional particle swarm optimization (PSO) algorithm. The traditional PSO algorithm simulates the foraging behavior of bird flocks and uses the position and velocity information of particles in the flock to search for the optimal solution. The IPSO algorithm introduces a nonlinear weight adjustment mechanism, which enables particles to dynamically adjust their search strategies according to their own iteration times, fitness, etc. during the search process, which is more conducive to jumping out of the local optimal solution and converging towards the global optimal solution.

[0083] In this step, the LSTM model is optimized as follows:

[0084] In the net load forecasting scenario of the substation, the IPSO algorithm is used to optimize the LSTM forecasting model, mainly to optimize and adjust some key parameters in the LSTM model, such as the number of hidden layer neurons, learning rate, time step, etc. These parameters are regarded as the position vector elements of the particles in the IPSO algorithm, and the mean square error (MSE) of the forecasting model and other evaluation indicators are used as the fitness function. The position of the particles (i.e., parameter values) is continuously updated through the IPSO algorithm to find a set of parameter values ​​that optimizes the fitness function (i.e., minimizes the forecast error), thereby improving the accuracy and generalization ability of the LSTM forecasting model, and thus obtaining the IPSO-LSTM intelligent forecasting model, which is used as the net load forecasting model.

[0085] Finally, the net load prediction model is used to predict the net load of all substations. Specifically, the pre-processed prediction period characteristic factor data of each substation is input into the net load prediction model. The net load prediction model will output the net load prediction value of the corresponding substation according to the mapping relationship between input and output learned before. For all substations, such operations are performed one by one, and the net load prediction results of each substation in a certain time period in the future (such as the next 1 hour, the next 24 hours, etc.) can be obtained, providing strong data support for the subsequent formulation of substation energy management strategies.

[0086] According to the substation group topology, the net load forecast results of a single or multiple substations in the group are aggregated. If it is a single substation group (i.e., there is only one substation), the net load forecast result of the substation is the net load forecast result of the group; if it is a multi-substation group, this patent uses a common aggregation method to add, and the group net load forecast result is the sum calculated according to the net load forecast. Through such aggregation, the net load forecast at the substation group level can be obtained, which is convenient for performing power resource allocation, load balancing and other management operations on the substation group from a more macro perspective.

[0087] Step 4: Determine the energy management strategy of the nth substation group according to the net load forecast value of the nth substation group and the energy storage capacity of the substation group, and use the energy management strategy of the nth substation group to control the distributed resources in the nth substation group, where n is a positive integer not greater than N. Figure 2 As shown, for each station area group, this step specifically includes:

[0088] Determine the photovoltaic cumulative net reverse transmission capacity of the substation group according to the net load prediction value of the substation group, specifically: determine whether the net load prediction value of the substation group is less than zero; if the net load prediction value of the substation group is less than zero, determine the photovoltaic cumulative net reverse transmission capacity of the substation group as the absolute value of the net load prediction value of the substation group; if the net load prediction value of the substation group is greater than or equal to zero, determine the photovoltaic cumulative net reverse transmission capacity of the substation group as zero;

[0089] Determine whether the cumulative net photovoltaic reverse transmission capacity of the substation group exceeds the energy storage capacity of the substation group;

[0090] If the cumulative net reverse transmission capacity of the photovoltaic power generation group does not exceed the energy storage capacity of the photovoltaic power generation group, the distributed resources in the photovoltaic power generation group are controlled by using the maximum reverse load rate real-time energy regulation strategy;

[0091] If the cumulative net photovoltaic reverse transmission capacity of the substation group exceeds the energy storage capacity of the substation group, it is determined whether the maximum reverse transmission power of at least one substation in the substation group exceeds the maximum reverse load rate standard;

[0092] If the maximum reverse transmission power of at least one area in the area group exceeds the maximum reverse load rate standard, the distributed resources in the area group are controlled by adopting the reverse transmission period uniform reverse load rate energy regulation strategy;

[0093] If there is no situation in which the maximum reverse power of any area in the area group exceeds the maximum reverse load rate standard, an energy regulation strategy of uniformly reducing the reverse load rate during heavy load periods is adopted to control the distributed resources in the area group.

[0094] In this step, when the photovoltaic cumulative net reverse transmission capacity of the substation group does not exceed the energy storage capacity of the substation group, the maximum reverse load rate real-time energy regulation strategy is used to control the distributed resources in the substation group. In order to ensure that the actual available capacity of the energy storage is greater than the photovoltaic cumulative net reverse transmission capacity of the substation group, under the maximum reverse load rate real-time energy regulation strategy, the power generated by the reverse transmission can be selected from near to far according to the electrical distance, and charged according to the maximum charging power acceptable to the distributed resources, that is, ,in, is the charging power of the i-th distributed resource in the area group under the maximum reverse load reduction real-time energy regulation strategy, is the maximum charging power acceptable to the i-th distributed resource in the substation group.

[0095] In this step, when the photovoltaic cumulative net reverse transmission capacity of the substation group exceeds the energy storage capacity of the substation group, and the maximum reverse transmission power of at least one substation in the substation group exceeds the maximum reverse load rate standard, the uniform reverse load rate energy regulation strategy during the reverse transmission period is adopted to control the distributed resources in the substation group. In order to ensure that the energy storage is always charged during the period of severe reverse power, under the uniform reverse load rate energy regulation strategy during the reverse transmission period, the power generated by the reverse transmission can be selected from the distributed resources according to the electrical distance from near to far, and the distributed resources are charged at the smaller value of the constant power or the maximum charging power acceptable to the equipment during the entire reverse transmission period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the area group under the uniform reverse load rate energy regulation strategy during the reverse period, is the maximum acceptable charging power of the i-th distributed resource in the station group, S ES,i is the rated capacity of the ith distributed resource in the substation group, Tover This is the entire reverse transmission period.

[0096] In this step, when the cumulative net photovoltaic reverse transmission capacity of the substation group exceeds the energy storage capacity of the substation group, and there is no substation in the substation group whose maximum reverse transmission power exceeds the maximum reverse load rate standard, the energy regulation strategy of uniformly reducing the reverse load rate during heavy load periods is adopted to control the distributed resources in the substation group. In order to ensure that the energy storage is charged during the period of severe reverse power exceeding the maximum reverse load rate, under the energy regulation strategy of uniformly reducing the reverse load rate during heavy load periods, the power generated by the reverse transmission is selected from the nearest to the farthest according to the electrical distance, and the distributed resources are charged at the smaller value of the constant power or the maximum charging power acceptable to the equipment during the heavy load period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the cluster under the energy regulation strategy of uniformly decreasing the reverse load rate during the heavy load period, is the maximum acceptable charging power of the i-th distributed resource in the station group, S ES,i is the rated capacity of the ith distributed resource in the substation group, T LFDover This is the heavy load period.

[0097] In this implementation, each substation is equipped with high-precision smart meters, sensors and other equipment to monitor key data such as the power generation power of distributed photovoltaics, the power consumption of various loads, and the operating status of flexible interconnected devices in real time. These data are quickly transmitted to the real-time energy management system (EMS) of the substation via a high-speed communication network, such as a power fiber-optic private network or a 5G wireless private network. Based on the collected data, EMS accurately calculates the net load of the substation at each moment, that is, the load demand minus the photovoltaic output. When the net load is positive, it means that the load in the substation exceeds the photovoltaic power generation and there is a power gap. If the net load is negative, it means that there is surplus photovoltaic power. If there is surplus photovoltaic power, first try to find adjustable load resources in this substation. For example, for some interruptible loads, such as some industrial production equipment with timed operation functions or acceptable short-term power outages, non-critical lighting and air-conditioning systems in large commercial places, etc., according to the pre-set priorities and protocols, their operating status is appropriately adjusted, and the start-up is delayed or suspended for a period of time, so as to make the excess photovoltaic power be digested locally and achieve a preliminary source-load balance. When there is still surplus photovoltaic power in this area after internal adjustment, the flexible interconnection device will be activated. The flexible interconnection device has bidirectional power transmission capabilities and can accurately and quickly adjust the power flow and size. EMS identifies which surrounding areas have power demand, that is, the net load is positive and has the ability to accept additional power, based on the pre-set area group topology and the real-time load conditions of each area. Through the optimization algorithm, the electrical distance between each receiving area and the current area (reflecting the power transmission loss), the load margin of the receiving area (the remaining load that can be carried), and the transmission efficiency of the flexible interconnection device itself are comprehensively considered to determine the power value allocated to each receiving area, and transfer the surplus photovoltaic power in an efficient and low-loss manner, helping other areas meet their own load needs and further improve the energy utilization efficiency of the entire area group.

[0098] If the load in the substation has completely absorbed the photovoltaic power, and there is even a surplus, then consider storing the excess energy in the energy storage device. The setting of the energy storage power refers to the results calculated by the above-mentioned maximum reverse load rate reduction strategy, the reverse load rate strategy at a uniform speed during the reverse transmission period, or the reverse load rate strategy at a uniform speed during the heavy load period for charging. For example, if the net load predicts a large photovoltaic power generation, after calculation, the reverse load rate strategy at a uniform speed during the heavy load period should be used. The system will use the load reduction capacity of the energy storage as much as possible in the period when the reverse transmission exceeds the maximum reverse load rate standard, so as to uniformly reduce the impact on the substation under the condition of maximum photovoltaic absorption. If there is no restriction and the energy storage is used according to the conventional charging and discharging strategy, it is possible that the energy storage capacity will be exhausted before the maximum reverse load rate standard is reached, resulting in a waste of photovoltaic resources, and in the subsequent time period, the photovoltaic power will always be reversed at the maximum reverse load rate standard, which will cause a loss in the life of the transformer. Before performing energy storage operations, the EMS strictly compares the current state of charge (SOC) of the energy storage device with the rated capacity to ensure that the injected power will not cause the SOC to exceed the upper limit allowed by the device, preventing damage to the energy storage device due to overcharging. At the same time, combined with the rated charge and discharge power of the energy storage device, check whether the charging power calculated according to the established strategy is within the acceptable range of the device. If it exceeds the limit, the power value is automatically lowered to prioritize the safe and stable operation of the energy storage system.

[0099] After the above processing, if there is still unabsorbed power surplus in the substation, the excess power will be smoothly fed back to the upper-level power grid if the grid permits. At this time, closely monitor the acceptance capacity indicators on the grid side, such as real-time monitoring of the voltage, power flow distribution and other parameters of a certain node in the grid, and obtain the upper limit information of the reverse injection power acceptable to the grid in a timely manner with the help of communication with the grid dispatching center. Once it is found that the power fed back to the grid is about to or has exceeded the acceptable capacity of the grid, in order to maintain the safe and stable operation of the overall grid and avoid causing shocks to the grid, such as voltage exceeding the limit, line overload and other faults, the energy management system decisively decides to start the light abandonment mechanism. By accurately regulating the photovoltaic inverter, the power output of the photovoltaic system is quickly reduced until the grid acceptance requirements are met, ensuring that the energy interaction between the substation and the grid is within a safe and controllable range.

[0100] It is not difficult to find that this implementation method traverses and searches for the power mutual assistance capability between substations, groups substations with power mutual assistance capability, utilizes the power mutual assistance characteristics between substations, and adopts different substation group energy management strategies according to the results of substation net load prediction to adapt to the joint autonomy problem between substations and substations under the new distribution system. This energy management strategy has the advantages of predicting photovoltaic reverse overload substations to respond in advance, improving photovoltaic absorption, high overload control effect, fast speed, and applicable real-time energy management.

[0101] The second embodiment of the present invention relates to a distributed resource adjustment device in a substation area, comprising:

[0102] An acquisition module, used for acquiring power grid data resources and generating a power grid model based on the power grid data resources;

[0103] An analysis module, used to convert the power grid model into a graph model, analyze the graph model, obtain topological structure information between substations, and group substations with mutual assistance capabilities into substation groups to obtain N substation groups, where N is a positive integer;

[0104] A prediction module is used to input the net load prediction characteristic factor data of each substation into the net load prediction model to obtain the net load prediction value of each substation, and to superimpose the net load prediction values ​​belonging to the same substation group to obtain the net load prediction values ​​of N substation groups;

[0105] A control module is used to determine the substation group energy management strategy of the nth substation group based on the net load prediction value of the substation group and the energy storage capacity of the substation group, and use the substation group energy management strategy of the nth substation group to control the distributed resources within the nth substation group, wherein n is a positive integer not greater than N.

[0106] When the analysis module converts the power grid model into a graph model, each device in the power grid model is abstracted as a node in the graph model, and the connection relationship between each device is abstracted as an edge in the graph model.

[0107] When the analysis module analyzes the graph model, it adopts one or more of a breadth-first search algorithm, a depth-first search algorithm and a Dijkstra algorithm.

[0108] The net load prediction model is obtained in the following manner: based on the historical net load measurement data of all substations, the corresponding actual meteorological data and predicted data, the net load prediction characteristic factors of the substations are analyzed, a prediction model based on a long short-term memory neural network is established using a deep learning framework, and the prediction model is optimized using a nonlinear weighted particle swarm optimization algorithm to obtain a net load prediction model.

[0109] The control module comprises:

[0110] A determination unit, used to determine the photovoltaic cumulative net reverse transmission capacity of the nth substation group according to the net load prediction value of the nth substation group;

[0111] The first judgment unit is used to judge whether the photovoltaic cumulative net reverse transmission capacity of the nth substation group exceeds the energy storage capacity of the nth substation group;

[0112] The first control unit is used to control the distributed resources in the nth substation group by adopting the maximum reverse load rate real-time energy regulation strategy when the photovoltaic cumulative net reverse transmission capacity of the nth substation group does not exceed the energy storage capacity of the nth substation group;

[0113] The second judgment unit is used to judge whether the maximum reverse power of at least one area in the nth area group exceeds the maximum reverse load rate standard when the cumulative net reverse power of the photovoltaic power generation in the nth area group exceeds the energy storage capacity of the nth area group;

[0114] The second control unit is used to control the distributed resources in the nth substation group by adopting the reverse load rate uniform speed energy regulation strategy during the reverse transmission period when the maximum reverse transmission power of at least one substation in the nth substation group exceeds the maximum reverse load rate standard;

[0115] The third control unit is used to control the distributed resources in the nth substation group by adopting the energy regulation strategy of uniformly reducing the reverse load rate during the heavy load period when the maximum reverse power of no substation in the nth substation group exceeds the maximum reverse load rate standard.

[0116] The determining unit comprises:

[0117] A judgment subunit, used to judge whether the net load prediction value of the nth substation group is less than zero;

[0118] The first determination subunit is used to determine the photovoltaic cumulative net reverse transmission capacity of the nth substation group as the absolute value of the net load prediction value of the nth substation group when the net load prediction value of the nth substation group is less than zero;

[0119] The second determination subunit is used to determine the photovoltaic cumulative net reverse transmission capacity of the nth substation group as zero when the net load prediction value of the nth substation group is greater than or equal to zero.

[0120] When the first control unit adopts the maximum reverse load rate real-time energy regulation strategy to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged according to the maximum charging power acceptable to the distributed resources, that is, ,in, is the charging power of the i-th distributed resource in the n-th substation group under the maximum reverse load reduction real-time energy regulation strategy, is the maximum charging power acceptable to the i-th distributed resource in the n-th substation group.

[0121] When the second control unit adopts the uniform reverse load rate energy regulation strategy during the reverse transmission period to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged at a smaller value of a constant power or a maximum charging power acceptable to the device during the entire reverse transmission period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the n-th substation group under the uniform reverse load rate energy regulation strategy during the reverse period, is the maximum charging power acceptable to the i-th distributed resource in the n-th station group, S ES,i is the rated capacity of the ith distributed resource in the nth substation group, T over This is the entire reverse transmission period.

[0122] When the third control unit adopts the energy regulation strategy of uniformly decreasing the reverse load rate during the heavy load period to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged at a smaller value of a constant power or a maximum charging power acceptable to the device during the heavy load period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the n-th substation group under the energy regulation strategy of uniformly decreasing the reverse load rate during the heavy load period, is the maximum charging power acceptable to the i-th distributed resource in the n-th station group, S ES,i is the rated capacity of the ith distributed resource in the nth substation group, T LFDover This is the heavy load period.

[0123] The third embodiment of the present invention relates to an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the distributed resource regulation method of the first embodiment when executing the computer program.

[0124] The fourth embodiment of the present invention relates to a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the distributed resource adjustment method of the first embodiment are implemented.

[0125] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program codes.

[0126] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0127] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction method, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0129] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A distributed resource adjustment method for a substation, characterized in that: The following steps are involved: Acquire power grid data resources, and generate a power grid model based on the power grid data resources; Convert the power grid model into a graph model, analyze the graph model, obtain topological structure information between substations, and group substations with mutual assistance capabilities into substation groups to obtain N substation groups, where N is a positive integer; Input the net load prediction characteristic factor data of each substation into the net load prediction model to obtain the net load prediction value of each substation, and superimpose the net load prediction values ​​belonging to the same substation group to obtain the net load prediction values ​​of N substation groups; The substation group energy management strategy of the nth substation group is determined according to the net load forecast value of the nth substation group and the energy storage capacity of the substation group, and the substation group energy management strategy of the nth substation group is used to control the distributed resources within the nth substation group, wherein n is a positive integer not greater than N.

2. The distributed resource adjustment method according to claim 1, characterized in that: The converting the power grid model into a graph model specifically includes: abstracting each device in the power grid model as a node in the graph model, and abstracting the connection relationship between each device as an edge in the graph model.

3. The distributed resource adjustment method according to claim 1, characterized in that: When analyzing the graph model, one or more of a breadth-first search algorithm, a depth-first search algorithm and a Dijkstra algorithm are used.

4. The distributed resource adjustment method according to claim 1, characterized in that: The net load forecasting model is obtained in the following way: Based on the historical net load measurement data of all substations, the corresponding actual meteorological data and forecast data, the net load forecast characteristic factors of the substations are analyzed, and a forecast model based on long short-term memory neural network is established using a deep learning framework. The forecast model is optimized using a nonlinear weighted particle swarm optimization algorithm to obtain a net load forecast model.

5. The distributed resource adjustment method according to claim 1, characterized in that: The step of determining the energy management strategy of the nth substation group according to the net load prediction value of the nth substation group and the energy storage capacity of the substation group, and controlling the distributed resources in the nth substation group by using the energy management strategy of the nth substation group, specifically includes: Determine the accumulative net photovoltaic reverse transmission capacity of the nth substation group according to the net load forecast value of the nth substation group; Determine whether the accumulative net reverse transmission capacity of the nth grid area group exceeds the energy storage capacity of the nth grid area group; If the accumulative net reverse transmission capacity of the nth group of substations does not exceed the energy storage capacity of the nth group of substations, the maximum reverse load reduction real-time energy regulation strategy is used to control the distributed resources in the nth group of substations; If the accumulative net reverse transmission capacity of the nth grid area group exceeds the energy storage capacity of the nth grid area group, it is determined whether the maximum reverse transmission power of at least one grid area in the nth grid area group exceeds the maximum reverse load rate standard; If the maximum reverse transmission power of at least one area in the nth area group exceeds the maximum reverse load rate standard, the uniform reverse load rate energy regulation strategy during the reverse transmission period is adopted to control the distributed resources in the nth area group; If the maximum reverse power of no area in the nth area group exceeds the maximum reverse load rate standard, the energy regulation strategy of uniformly reducing the reverse load rate during the heavy load period is adopted to control the distributed resources in the nth area group.

6. The distributed resource adjustment method according to claim 5, characterized in that: The step of determining the photovoltaic cumulative net reverse transmission capacity of the nth group of substations according to the net load forecast value of the nth group of substations specifically includes: Determine whether the net load forecast value of the nth substation group is less than zero; If the net load forecast value of the nth group of grid stations is less than zero, the PV cumulative net reverse transmission capacity of the nth group of grid stations is determined as the absolute value of the net load forecast value of the nth group of grid stations; If the net load forecast value of the nth substation group is greater than or equal to zero, the photovoltaic cumulative net reverse transmission capacity of the nth substation group is determined to be zero.

7. The distributed resource adjustment method according to claim 5, characterized in that: When the maximum reverse load rate real-time energy regulation strategy is used to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged according to the maximum charging power acceptable to the distributed resources, that is, ,in, is the charging power of the i-th distributed resource in the n-th substation group under the maximum reverse load reduction real-time energy regulation strategy, is the maximum charging power acceptable to the i-th distributed resource in the n-th substation group.

8. The distributed resource adjustment method according to claim 5, characterized in that: When the energy regulation strategy of uniform reverse load rate during the reverse transmission period is adopted to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged at a smaller value of a constant power or a maximum charging power acceptable to the device during the entire reverse transmission period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the n-th substation group under the uniform reverse load rate energy regulation strategy during the reverse period, is the maximum charging power acceptable to the i-th distributed resource in the n-th station group, S ES,i is the rated capacity of the ith distributed resource in the nth substation group, T over This is the entire reverse transmission period.

9. The distributed resource adjustment method according to claim 5, characterized in that: When the energy regulation strategy of uniformly decreasing the reverse load rate during the heavy load period is adopted to control the distributed resources in the nth substation group, the power generated by the reverse transmission is selected from near to far according to the electrical distance, and the distributed resources are charged at the smaller value of the constant power or the maximum charging power acceptable to the equipment during the heavy load period, that is, ,in, is the actual charging power at the current time, is the average charging power of the i-th distributed resource in the n-th substation group under the energy regulation strategy of uniformly decreasing the reverse load rate during the heavy load period, is the maximum charging power acceptable to the i-th distributed resource in the n-th station group, S ES,i is the rated capacity of the ith distributed resource in the nth substation group, T LFDover This is the heavy load period.

10. A distributed resource adjustment device for a substation, characterized in that: include: An acquisition module, used for acquiring power grid data resources and generating a power grid model based on the power grid data resources; An analysis module, used to convert the power grid model into a graph model, analyze the graph model, obtain topological structure information between substations, and group substations with mutual assistance capabilities into substation groups to obtain N substation groups, where N is a positive integer; A prediction module is used to input the net load prediction characteristic factor data of each substation into the net load prediction model to obtain the net load prediction value of each substation, and to superimpose the net load prediction values ​​belonging to the same substation group to obtain the net load prediction values ​​of N substation groups; A control module is used to determine the substation group energy management strategy of the nth substation group based on the net load prediction value of the substation group and the energy storage capacity of the substation group, and use the substation group energy management strategy of the nth substation group to control the distributed resources within the nth substation group, wherein n is a positive integer not greater than N.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the distributed resource adjustment method for a substation are implemented as described in any one of claims 1-9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distributed resource adjustment method for a substation as described in any one of claims 1 to 9 are implemented.

Images