A routing optimization-based distribution area optical storage and charging control method and system

Abstract

This invention discloses a method and system for controlling photovoltaic (PV), energy storage, and charging (ESC) devices in distribution network areas based on route optimization, relating to the field of distribution network technology. The method includes: dividing an initial autonomous group (AAG) according to the distribution network topology; introducing a dynamic adjustment mechanism in the initial AAG to trigger temporary group formation based on service overload events, thus obtaining a dynamic AAG; real-time monitoring of the PV, ESC, and charging operation status of each node in the dynamic AAG; when the PV, ESC, and charging operation status meets preset conditions, semantically encoding the corresponding PV, ESC, and charging operation status to generate a compressed command stream; responding to the compressed command stream, driving a preset federated learning model to search for the optimal path for transmitting PV, ESC, and charging adjustment commands in each area of ​​the dynamic AAG; binding the compressed command stream and the optimal path to generate a joint control vector; and using the joint control vector to adjust the power of PV, ESC, and charging devices in the distribution network area to ensure stable grid operation.

Description

Technical Field

[0001] This invention relates to the field of distribution network technology, and in particular to a method and system for controlling the optical, energy storage and charging of distribution network substations based on routing optimization. Background Technology

[0002] Against the backdrop of energy structure transformation and high-proportion distributed energy access, user-side inverters, as the core node for coordinated control of source (photovoltaic / energy storage) and load (flexible load), have become crucial for supporting the flexible adjustment of new power systems due to their real-time communication, reliability, and low power consumption characteristics.

[0003] However, existing collaborative control schemes that rely on traditional communication paradigms mainly depend on centralized architectures (such as cloud platforms or centralized gateways) in terms of communication architecture. They often suffer from excessive end-to-end latency due to data relay. In terms of routing mechanisms, traditional wireless protocols (such as ZigBee) rely on static routing strategies. These limitations are no longer sufficient to meet the current demands for millisecond-level response and dynamic adjustment of source-load collaboration.

[0004] Therefore, how to effectively regulate the equipment in the distribution network area to ensure the stable operation of the power grid has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides a method and system for controlling photovoltaic, energy storage and charging equipment in distribution network areas based on routing optimization. It solves the problem of how to reliably regulate the power of photovoltaic, energy storage and charging equipment, so that its routing strategy can adapt to the needs of safe and stable operation of the power grid and improve the energy efficiency of the power grid.

[0006] To address the aforementioned technical problems, embodiments of the present invention provide a method for controlling the optical storage and charging of distribution network areas based on routing optimization, comprising: An initial autonomous group is defined based on the distribution network topology. A dynamic adjustment mechanism for the formation of a temporary group is introduced into the initial autonomous group to trigger the business over-limit event, thereby obtaining a dynamic autonomous group. The optical storage and charging operation status of each node in the dynamic autonomous group is monitored in real time. When the optical storage and charging operation status meets the preset conditions, the corresponding optical storage and charging operation status is semantically encoded to generate a compressed instruction stream. In response to the compressed command stream, a preset federated learning model is driven to search for the optimal path for transmitting optical storage and charging regulation commands in each region of the dynamic autonomous group. The compressed instruction stream and the optimal path are bound together to generate a joint control vector; The power of photovoltaic, energy storage and charging equipment in the distribution network area is regulated using the joint control vector.

[0007] Furthermore, the dynamic adjustment mechanism includes: When a voltage over-limit event is detected in the target area of ​​the initial autonomous group, the energy storage node in the target area is used as the master control node, and the photovoltaic node and load node associated with the voltage over-limit event are extracted to construct a temporary service group; The power of each node in the temporary service group is adjusted until the voltage over-limit event ends and the group is automatically disbanded.

[0008] Furthermore, the step of semantically encoding the corresponding optical storage and charging operating state to generate a compressed instruction stream includes: Based on the different types of changes in the optical storage and charging operation status, they are mapped to corresponding operation codes; The various changes in the optical storage and charging operation status are converted into corresponding parameter codes. The operation codes and parameter codes are combined to obtain the compression command stream.

[0009] Furthermore, after semantically encoding the corresponding optical storage and charging operating state to generate a compressed instruction stream, the method further includes: Extract the real-time neighbor status table from the dynamic autonomous group and select the target transmission path; The compressed instruction stream is input into the federated learning model using the target transmission path.

[0010] Furthermore, the pre-defined federated learning model searches for the optimal path for each region in the dynamic autonomous group to perform photoelectric storage-charging regulation, including: A state space is constructed using the communication status of each region in the dynamic autonomous group and the operation status of the optical storage and charging system as indicators. A reward function is designed with the goal of optimizing the business security level of each region in the dynamic autonomous group. The state space and the reward function are input into the federated learning model to calculate the reward function value of the optical storage and charging equipment adjustment command transmitted through different paths in each region; Based on the calculation results, the optimal path for each region is determined.

[0011] Furthermore, the step of inputting the compressed instruction stream into the federated learning model using the target transmission path further includes: The transmission priority is designed based on the urgency of each instruction in the compressed instruction stream; The instructions in the compressed instruction stream are input into the federated learning model according to the transmission priority using the target transmission path.

[0012] Furthermore, after using the joint control vector to regulate the power of photovoltaic, energy storage, and charging equipment in the distribution network area, the method further includes: The communication performance of each node in each region of the dynamic autonomous group is obtained in real time; the communication performance includes the proportion of communication energy consumption, communication latency, and event success rate. The communication performance is compared and analyzed with the pre-recorded initial communication performance to quantify the adjustment effect and send it to the operation and maintenance terminal of the distribution network.

[0013] Another embodiment of the present invention provides a distribution network area optical storage and charging control system based on routing optimization, comprising: The autonomous group construction module is used to divide the initial autonomous groups according to the distribution network topology, and introduce a dynamic adjustment mechanism for the formation of temporary groups triggered by business over-limit events into the initial autonomous groups to obtain dynamic autonomous groups; The instruction generation module is used to monitor the optical storage and charging operation status of each node in the dynamic autonomous group in real time. When the optical storage and charging operation status meets the preset conditions, the corresponding optical storage and charging operation status is semantically encoded to generate a compressed instruction stream. The path selection module is used to respond to the compressed command stream and drive a preset federated learning model to search for the optimal path for transmitting optical storage and charging adjustment commands in each region of the dynamic autonomous group. A control vector generation module is used to bind the compressed instruction stream and the optimal path to generate a joint control vector; The power regulation module is used to regulate the power of photovoltaic, energy storage and charging equipment in the distribution network area using the joint control vector.

[0014] Another embodiment of the present invention provides a computer device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the distribution network area optical storage and charging control method based on routing optimization as described above.

[0015] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the distribution network area optical storage and charging control method based on routing optimization as described above.

[0016] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following: This invention achieves deep coupling between communication topology and power grid status by using a dynamic autonomous group with the function of triggering temporary service groups for service overrun events, thereby improving the system's response speed to local power grid problems. By using semantic encoding to compress the command stream of the group node's operating status, the transmission load can be significantly reduced. By driving a federated learning model to search for the comprehensive optimal path, a collaborative routing decision is formed to improve power grid stability. The compressed command stream is bound to the optimal path to generate a joint control vector to effectively regulate the photovoltaic, energy storage, and charging equipment, thereby reducing communication energy consumption, improving command transmission efficiency, and maintaining the stability of distribution network area control. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of a distribution network area optical storage and charging control method based on routing optimization in one embodiment of the present invention; Figure 2 This is a schematic diagram of the optical-storage-charging control system structure of a distribution network area based on routing optimization in one embodiment of the present invention; Figure 3 This is a structural block diagram of a preferred embodiment of a computer device provided by the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. The terms "vertical," "horizontal," "left," "right," "upper," "lower," and similar expressions used herein are for illustrative purposes only and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0022] Photovoltaic, energy storage, and charging loads (PV, energy storage, and charging loads) are often used as flexible adjustment units with dual source and load attributes in distribution network areas. Through coordinated control, they achieve power balance and voltage support, effectively enhancing the area's ability to absorb high proportions of distributed energy and its operational resilience. Based on this, one embodiment of the present invention provides a distribution network PV-storage-charging control method based on route optimization. For details, please refer to... Figure 1 , Figure 1 The diagram shown is a schematic flowchart of a distribution network area optical storage and charging control method based on routing optimization in one embodiment of the present invention, including the following steps: S1. Divide the initial autonomous group according to the distribution network topology, and introduce a dynamic adjustment mechanism in the initial autonomous group to trigger the formation of temporary groups by business over-limit events, so as to obtain the dynamic autonomous group.

[0023] Preferably, taking a residential community distribution transformer system implementation project as an example, this distribution transformer includes 15 rooftop photovoltaic inverters (single unit capacity 5-10kW), 5 sets of household energy storage converters (capacity 10-20kWh), and 20 adjustable load nodes (smart air conditioning clusters). This step aims to construct a dynamic autonomous group based on electrical coupling relationships and business event states, using the distribution transformer topology as the implementation carrier, and deploying a dynamic autonomous group architecture with a dynamic adjustment mechanism. Specifically, according to the power grid topology, devices connected to the low-voltage side bus of the same transformer are divided into initial autonomous groups, forming 3 basic groups (East Group, West Group, and Central Group).

[0024] When a service over-limit event is detected in the target area of ​​the initial autonomous group, such as a voltage over-limit event, the energy storage node in the target area is used as the master node, and the photovoltaic node and load node associated with the voltage over-limit event are extracted to construct a temporary service group.

[0025] It should be understood that the voltage problem raised in this step has obvious regional characteristics. The voltage rise in the eastern region is mainly caused by the excess output of photovoltaic power in this region, and its electrical impact is mainly limited to this power supply branch. Therefore, the voltage over-limit area is first located by using smart meter measurement data. Then, clustering algorithms (such as K-means algorithm) can be used to automatically identify several photovoltaic nodes that have the greatest impact on the voltage in the eastern region, using electrical distance (line impedance parameter) and real-time power data as feature vectors. In the identification process, this embodiment sets dual screening conditions: first, the geographical location must be within the voltage over-limit area; second, the power sensitivity (dU / dP) of the node must exceed a set threshold.

[0026] For example, if a voltage exceedance (>235V) occurs in the eastern cluster, a temporary service cluster is dynamically formed, centered on the eastern energy storage node ESS01, comprising 5 photovoltaic nodes and 3 load nodes within the area. The selection process for the 5 photovoltaic nodes is as follows: First, a first-level screening is initiated based on the topological physical location, directly excluding equipment in the western and central areas with greater electrical distances, focusing only on the 8 photovoltaic inverters physically connected to the eastern busbar. Then, a second-level screening is performed using an improved K-means algorithm, calculating the power sensitivity (dU / dP) of these 8 nodes to the exceedance point based on the line impedance matrix, setting an effective threshold of 0.03 V / kW. At this point, node PV_01, located near the transformer outlet, is automatically eliminated due to its low line impedance and sensitivity of only 0.01 V / kW (i.e., adjusting its power has a negligible effect on voltage improvement); while node PV_05, located at the end of the line, has a measured sensitivity of 0.05 V / kW and is identified as a strongly correlated node. Finally, the 5 nodes, including PV_05, are clustered to generate a temporary service cluster.

[0027] Then, the power of each node in the temporary service group is adjusted until the voltage over-limit event ends and the group is automatically disbanded. Once the voltage returns to normal, the group is automatically disbanded and each node returns to the initial group to standby.

[0028] For example, in this embodiment, this adjustment is automatically triggered when the voltage detection value exceeds 235V for three consecutive sampling cycles. The temporary service group uses the East Zone energy storage node ESS01 as the control center, calculates the total required power adjustment based on the real-time voltage deviation, and then allocates it according to the capacity ratio of each photovoltaic node, generating specific power adjustment instructions. The instructions include key parameters such as adjustment direction (increase / decrease), adjustment magnitude (percentage), and execution sequence. If the East Zone voltage is detected to rise to 245V (10V over the limit), the main control node ESS01 needs to reduce the total active power by ΔP = 20kW. Assume that the temporary service group contains two inverters, PV_01 (rated 10kW) and PV_02 (rated 30kW), with a total capacity of 40kW. Based on this, the allocation ratio is established as "single unit rated capacity / total group capacity," with PV_01 having a weight of 25% and PV_02 having 75%. Accordingly, PV_01 needs to be reduced by 5kW, and PV_02 by 15kW. Ultimately, the control center maps the above values ​​to a "power down" opcode and the corresponding percentage parameter (e.g., down by 50%).

[0029] It should be understood that this embodiment aims to introduce a dynamic and adjustable mechanism into the initially divided autonomous group, making it a dynamic autonomous group that can adaptively adjust according to real-time service events within each area of ​​the distribution network, thereby maintaining the safety and stability of the power grid. Specifically, some signaling structures within this group are shown in Table 1 below: Table 1. Signaling Structure of Dynamic Autonomous Groups S2. Monitor the optical storage and charging operation status of each node in the dynamic autonomous group in real time. When the optical storage and charging operation status meets the preset conditions, the corresponding optical storage and charging operation status is semantically encoded to generate a compressed instruction stream.

[0030] The system monitors in real time the various changes in the photovoltaic, energy storage, and charging operation status of each node within the dynamic autonomous group, such as photovoltaic power fluctuations, energy storage SOC changes, voltage deviations, and load change responses. When the changes in these indicators exceed preset thresholds (photovoltaic power fluctuation threshold is 5%, energy storage SOC change threshold is 10%), corresponding power adjustment commands or energy storage charging and discharging commands will be generated and mapped to the corresponding operation codes. If the generated command is a power increase, it will be mapped to 0x3, and if the generated command is a power decrease, it will be mapped to 0x4.

[0031] Simultaneously, the various changes in the photovoltaic storage and charging operation status are converted into corresponding parameter codes. For example, the changes are encoded into a compact numerical representation using a non-linear formula. For instance, the power command uses S = 2.5 × (k + 1), where k is the parameter code value, and a 10% power adjustment is encoded as 0xA. Table 2 below provides some examples of the encoding for these commands: Table 2 Core Instruction Set Encoding Table Finally, the corresponding opcodes and parameter codes are combined to obtain the compression command stream. This compression command stream is specifically in the form of 8-bit compression command packets (e.g., 0x3A indicates a 10% power increase).

[0032] It is worth noting that this command stream will trigger path calculation in subsequent routing decisions. Therefore, the compressed command stream needs to be transmitted via a specific transmission path. Specifically, the real-time neighbor state table is extracted from the dynamic autonomous group, and a target transmission path is selected. It should be understood that when a temporary service group is established within the dynamic autonomous group due to "Eastern Area voltage exceeding limits," the members recorded in its neighbor state table are the photovoltaic, energy storage, and controllable load nodes that meet the requirements of the Eastern Area, selected through a clustering algorithm. The generated power regulation command 0x3A will then be sent to each node in this table as the target. That is, this target transmission path actually represents the optimal multi-hop or direct path for the planned compressed command stream to be transmitted from the current node to the target member node.

[0033] Meanwhile, transmission priorities are designed based on the urgency of each instruction in the compressed instruction stream to ensure preemptive transmission of urgent instructions (delay ≤ 20ms) and suppress redundant data. For example, overvoltage shutdown instructions can be set as the highest priority.

[0034] The instructions in the compressed instruction stream are input into the federated learning model according to their transmission priority using the target transmission path, and these instructions serve as perception conditions to influence the path decision-making process.

[0035] S3~S4, in response to the compressed command stream, drive the preset federated learning model to search for the optimal path for transmitting optical storage and charging regulation commands in each region of the dynamic autonomous group, bind the compressed command stream and the optimal path, and generate a joint control vector.

[0036] This step is the routing decision optimization process, which is specifically implemented by selecting the optimal path for the adjustment commands of each optical storage and charging device in the transmission area.

[0037] Due to a sudden change in the state of the optical storage and charging equipment in some nodes of the group, a command stream (compressed command stream) for adjustment is generated. The generation of this command stream is a key event that directly triggers the construction of a state space based on the communication status and optical storage and charging operation status of each region in the dynamic autonomous group. The reward function is designed with the goal of optimizing the service security level of each region in the dynamic autonomous group.

[0038] The state space and reward function are input into the federated learning model to calculate the reward function value of the optical storage and charging equipment adjustment command transmitted through different paths in each region.

[0039] For example, the state space S is reconstructed into a fused state S =<C, P> Where C represents the communication status (load within the group, backbone link quality), and P represents the optical storage and charging operation status (local voltage deviation Δ). V , frequency change rate d f / d t (etc.). Among them, the load within the group can be calculated as the ratio of the current node's data cache queue length to the total cache capacity, reflecting the degree of congestion; the backbone link quality is normalized based on the link's real-time signal-to-noise ratio (SNR) or packet delivery rate (PDR); the local voltage deviation ΔV is the absolute value of the difference between the real-time collected voltage and the nominal voltage; the frequency change rate df / dt is obtained by differentiating the frequency measurement value within a set time window.

[0040] The reward function is designed as R = 0.4* R com + 0.5* R phy + 0.1* R penalty ,in, R phy As a reward for physical security, when Δ V When the time limit is exceeded, the path that successfully transmits the voltage regulation command will receive an extremely high reward; R penalty As a physical constraint penalty, a huge negative reward is imposed on paths that cause line power to exceed limits; R com The communication performance reward is a positive incentive function for low-latency, high-throughput paths. For example, when the voltage in the eastern region rises to 238V, the energy storage node ESS01 selects the path with the shortest electrical distance to prioritize transmitting commands. After receiving the over-limit information, the decision-making networks of other energy storage nodes ESS03 and ESS05 in the group coordinate to avoid paths that may increase the load on the eastern region's lines.

[0041] In this embodiment, empirical weights determined through extensive simulation experiments using the Analytic Hierarchy Process (AHP) can be selected. The core logic is that physical security takes precedence (weight 0.5 is the highest), followed by communication quality (0.4), and finally, penalty for exceeding limits (0.1).

[0042] The action space is defined as the set of operations that an agent (a node in a dynamic autonomous swarm) performs to select the next-hop node from the neighbor state table. In this embodiment, Q-learning is used as the local decision engine to explore the optimal path, while the federated learning model is responsible for aggregating the Q-tables or network parameters of each node to achieve global collaborative optimization.

[0043] Based on the calculation results, the optimal path for each region is determined. After determining the optimal path, it is bound to the previously generated compressed command stream carrying adjustment instructions, generating a routing strategy that can precisely adjust the equipment in the distribution network area, such as binding "path PV01→ESS03" to "transmit power 60%". This joint control vector is the core execution command connecting communication optimization and physical control, and it directly participates in the real-time power adjustment control of the optical storage and charging equipment.

[0044] For example, through federated learning, the optimal path for the energy storage node ESS01 in the eastern region where the business event exceeded the limit was determined to be ESS01->PV02->ESS03. The system then sends the energy storage charge / discharge adjustment command to ESS03. Simultaneously, based on the real-time link quality (e.g., signal-to-noise ratio) of this path, a transmit power adjustment parameter (e.g., 60% power) is generated. At this point, the system deeply binds the 'optimal path,' 'transmit power parameter,' and 'energy storage charge / discharge adjustment command' to generate a joint control vector. This vector drives the node to adaptively adjust the RF module's transmit power while sending the command to ESS03 along the optimal path, thereby achieving cross-layer collaborative reduction of communication energy consumption while ensuring command delivery.

[0045] S5. Utilize joint control vectors to regulate the power of photovoltaic, energy storage, and charging equipment in the distribution network area.

[0046] This step is the final photovoltaic-storage-charging regulation. For example, the photovoltaic equipment receives commands to increase photovoltaic power by 10% or decrease it by 8%, and adjusts the inverter output power in real time to participate in voltage regulation and power balancing in the distribution area. The energy storage equipment receives commands to increase charging power by 50% or discharging power by 80%, and precisely controls the charging and discharging power to achieve peak-valley arbitrage or emergency power support.

[0047] In some embodiments of the present invention, an evaluation mechanism for adjusting performance is also designed. Specifically, the communication performance of each node in each region of the dynamic autonomous group is acquired in real time. In this embodiment, the communication performance includes the proportion of communication energy consumption, communication latency, and event success rate, wherein the proportion of communication energy consumption... η com Defined as the sum of the transmit power of all nodes Σ P tx,i Total power consumption of the device Σ Ptotal,i The percentage is expressed as follows: Communication delay T end2end End-to-end delay is represented as follows: It consists of three parts: transmission delay: L packet (Data packet length) / B k (Link bandwidth), queuing delay: Q k (Queue depth) / μ k (Service rate), semantic parsing latency t semantic .

[0048] Based on network reliability theory, self-healing success rate R selfheal =Number of nodes successfully recovered N recover Total number of faulty nodes N fail , means as follows: The communication performance is compared and analyzed with the pre-recorded initial communication performance to quantify the adjustment effect and send it to the distribution network's operation and maintenance terminal. Specifically, before deploying the dynamic autonomous groups, path selection, and photovoltaic-storage-charging control strategies of the distribution network substations in this embodiment, the initial communication performance (initial communication energy consumption ratio, communication latency, and event success rate) of each area of ​​the initial autonomous group is collected and recorded in real time. The adjustment effect between these initial communication performance and the actual communication performance is analyzed. For example, it is quantified by (initial value - real-time value) / initial value × 100%. Based on the calculation results, the improvement is determined and the results are sent to the distribution network's integrated operation and maintenance management platform. For example, the communication energy consumption ratio is 15.2% → 4.7% (an improvement of 69.1%), indicating a significant optimization effect of the communication strategy, achieving the expected goal.

[0049] In summary, this embodiment first divides the initial autonomous group based on the power grid topology and introduces a dynamic adjustment mechanism triggered by service over-limit events to generate a dynamic service group, achieving dynamic alignment between the communication topology and the power grid status. Second, by monitoring the operating status of each node within the group in real time, when the state change exceeds a preset threshold, semantic encoding is used to map it into a compressed command stream. On this basis, a state space and reward function are constructed to drive the distributed group under the federated learning framework to make collaborative routing decisions and determine the optimal transmission path. Subsequently, the optimal path is bound to the compressed command stream to generate a joint control vector for precise power adjustment of the photovoltaic, energy storage, and charging equipment, significantly improving the reliability and energy efficiency of the collaborative control of the distribution network area.

[0050] One embodiment of the present invention provides a distribution network area optical storage and charging control system based on route optimization. For details, please refer to [link to relevant documentation]. Figure 2 , Figure 2 The diagram shown is a schematic representation of a distribution network area optical storage and charging control system based on routing optimization in one embodiment of the present invention, comprising: The autonomous group construction module M1 is used to divide the initial autonomous group according to the distribution network topology, and introduce a dynamic adjustment mechanism for the formation of temporary groups triggered by business over-limit events into the initial autonomous group to obtain a dynamic autonomous group; The instruction generation module M2 is used to monitor the optical storage and charging operation status of each node in the dynamic autonomous group in real time. When the optical storage and charging operation status meets the preset conditions, the corresponding optical storage and charging operation status is semantically encoded to generate a compressed instruction stream. The path selection module M3 is used to respond to the compressed command stream and drive a preset federated learning model to search for the optimal path for transmitting optical storage and charging adjustment commands in each region of the dynamic autonomous group. The control vector generation module M4 is used to bind the compressed instruction stream and the optimal path to generate a joint control vector; The power regulation module M5 is used to regulate the power of photovoltaic, energy storage and charging equipment in the distribution network area using the joint control vector.

[0051] like Figure 3 As shown, this embodiment of the invention also provides a computer device. Figure 3 This is a structural block diagram of a preferred embodiment of a computer device provided by the present invention. The computer device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method described above.

[0052] Preferably, the computer program can be divided into one or more modules / units (such as computer program 1, computer program 2, ...), and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device.

[0053] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can be any conventional processor. The processor is the control center of the terminal device, connecting various parts of the terminal device through various interfaces and lines.

[0054] The memory mainly includes a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc., while the data storage area can store related data, etc. Furthermore, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard drive, a SmartMedia Card (SMC), a Secure Digital (SD) card, and a Flash Card, or other volatile solid-state storage devices.

[0055] It should be noted that the aforementioned terminal devices may include, but are not limited to, processors and memory, as will be understood by those skilled in the art. Figure 3The structural block diagram is merely an example of a terminal device and does not constitute a limitation on the terminal device. It may include more or fewer components than shown, or combine certain components, or use different components. Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0056] Accordingly, embodiments of the present invention provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the steps in the method of the above embodiments, for example... Figure 1 Steps S1 to S5 as described above.

[0057] The technical features and effects of the distribution network area optical-storage-charging control system based on route optimization proposed in this embodiment of the invention are the same as those of the distribution network area optical-storage-charging control method based on route optimization proposed in this embodiment of the invention, and will not be repeated here.

[0058] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A control method for optical storage and charging in distribution network areas based on route optimization, characterized in that, include: An initial autonomous group is defined based on the distribution network topology. A dynamic adjustment mechanism for the formation of a temporary group is introduced into the initial autonomous group to trigger the business over-limit event, thereby obtaining a dynamic autonomous group. The optical storage and charging operation status of each node in the dynamic autonomous group is monitored in real time. When the optical storage and charging operation status meets the preset conditions, the corresponding optical storage and charging operation status is semantically encoded to generate a compressed instruction stream. In response to the compressed command stream, a preset federated learning model is driven to search for the optimal path for transmitting optical storage and charging regulation commands in each region of the dynamic autonomous group. The compressed instruction stream and the optimal path are bound together to generate a joint control vector; The power of photovoltaic, energy storage and charging equipment in the distribution network area is regulated using the joint control vector.

2. The distribution network area optical storage and charging control method based on route optimization as described in claim 1, characterized in that, The dynamic adjustment mechanism includes: When a voltage over-limit event is detected in the target area of ​​the initial autonomous group, the energy storage node in the target area is used as the master control node, and the photovoltaic node and load node associated with the voltage over-limit event are extracted to construct a temporary service group; The power of each node in the temporary service group is adjusted until the voltage over-limit event ends and the group is automatically disbanded.

3. The distribution network area optical storage and charging control method based on route optimization as described in claim 1, characterized in that, The step of semantically encoding the corresponding optical storage and charging operating status to generate a compressed instruction stream includes: Based on the different types of changes in the optical storage and charging operation status, they are mapped to corresponding operation codes; The various changes in the optical storage and charging operation status are converted into corresponding parameter codes. The operation codes and parameter codes are combined to obtain the compression command stream.

4. The distribution network area optical storage and charging control method based on route optimization as described in claim 1, characterized in that, After semantically encoding the corresponding optical storage and charging operating state to generate a compressed instruction stream, the method further includes: Extract the real-time neighbor status table from the dynamic autonomous group and select the target transmission path; The compressed instruction stream is input into the federated learning model using the target transmission path.

5. The distribution network area optical storage and charging control method based on route optimization as described in claim 1, characterized in that, The pre-defined federated learning model searches for the optimal path for each region in the dynamic autonomous group to perform photoelectric storage-charging regulation, including: A state space is constructed using the communication status of each region in the dynamic autonomous group and the operation status of the optical storage and charging system as indicators. A reward function is designed with the goal of optimizing the business security level of each region in the dynamic autonomous group. The state space and the reward function are input into the federated learning model to calculate the reward function value of the optical storage and charging equipment adjustment command transmitted through different paths in each region; Based on the calculation results, the optimal path for each region is determined.

6. The distribution network area optical storage and charging control method based on route optimization as described in claim 4, characterized in that, The step of inputting the compressed instruction stream into the federated learning model using the target transmission path further includes: The transmission priority is designed based on the urgency of each instruction in the compressed instruction stream; The instructions in the compressed instruction stream are input into the federated learning model according to the transmission priority using the target transmission path.

7. The distribution network area optical storage and charging control method based on route optimization as described in claim 1, characterized in that, After using the joint control vector to regulate the power of photovoltaic, energy storage, and charging equipment in the distribution network area, the method further includes: The communication performance of each node in each region of the dynamic autonomous group is obtained in real time; the communication performance includes the proportion of communication energy consumption, communication latency, and event success rate. The communication performance is compared and analyzed with the pre-recorded initial communication performance to quantify the adjustment effect and send it to the operation and maintenance terminal of the distribution network.

8. A distribution network area optical-storage-charging control system based on route optimization, characterized in that, include: The autonomous group construction module is used to divide the initial autonomous groups according to the distribution network topology, and introduce a dynamic adjustment mechanism for the formation of temporary groups triggered by business over-limit events into the initial autonomous groups to obtain dynamic autonomous groups; The instruction generation module is used to monitor the optical storage and charging operation status of each node in the dynamic autonomous group in real time. When the optical storage and charging operation status meets the preset conditions, the corresponding optical storage and charging operation status is semantically encoded to generate a compressed instruction stream. The path selection module is used to respond to the compressed command stream and drive a preset federated learning model to search for the optimal path for transmitting optical storage and charging adjustment commands in each region of the dynamic autonomous group. A control vector generation module is used to bind the compressed instruction stream and the optimal path to generate a joint control vector; The power regulation module is used to regulate the power of photovoltaic, energy storage and charging equipment in the distribution network area using the joint control vector.

9. A computer device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the distribution network area optical storage and charging control method based on routing optimization as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the distribution network area optical storage and charging control method based on routing optimization as described in any one of claims 1 to 7.

Images