5G-based edge intelligent photovoltaic control method and device

Abstract

The embodiments of this specification provide a 5G-based edge intelligent photovoltaic control method and device, wherein the method includes: the photovoltaic area obtains photovoltaic data and sends the photovoltaic data to the intelligent fusion terminal of the area based on the 5G slice type; the intelligent fusion terminal of the area uses a photovoltaic prediction model to predict photovoltaic power based on the photovoltaic data and determines the photovoltaic power prediction result; based on the photovoltaic power prediction result and the operating status of the photovoltaic area, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, and the output curve is sent to each photovoltaic array; the cloud platform performs agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and sends the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the area. The terminal equipment not only effectively realizes the interconnection and data sharing of each photovoltaic array under the photovoltaic area, but also improves the efficiency and security of data processing, and provides support for the efficient implementation of power grid business management, safe production, quality service and other work.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of photovoltaic power generation technology, and in particular to a 5G-based edge intelligent photovoltaic control method. Background Art

[0002] With the construction of new power systems, the shortcomings of traditional monitoring systems in areas with a high proportion of photovoltaic power stations are becoming increasingly apparent. For device access, traditional wired communication methods require a large number of fiber optic cables and RS485 communication buses. However, the RS485 bus has disadvantages such as a limited number of connected devices, short transmission distances, high communication latency, and communication congestion. Regarding power plant operation and maintenance (O&M) and inspections, PV power plants lack platform-based, data-driven, and visualization-based O&M solutions. Furthermore, they lack capabilities for equipment fault assessment and data mining, hindering the implementation of truly intelligent O&M. For real-time control, under traditional wired communication solutions, the communication unit in the control center of the distribution network cloud master station converts TCP control messages into serial protocol messages before sending them to the target PV inverters. This process takes 8 to 15 seconds, resulting in a long and slow control path.

[0003] Therefore, a better solution is urgently needed. Summary of the Invention

[0004] In view of this, embodiments of this specification provide a 5G-based edge intelligent photovoltaic control method. One or more embodiments of this specification also involve a 5G-based edge intelligent photovoltaic control device, a computing device, a computer-readable storage medium, and a computer program to address technical deficiencies in the prior art.

[0005] According to a first aspect of the embodiments of this specification, a 5G-based edge intelligent photovoltaic control method is provided, which is applied to an intelligent photovoltaic control system. The system includes a cloud platform, a station intelligent fusion terminal, and a photovoltaic station. The method includes:

[0006] The photovoltaic substation acquires photovoltaic data and sends the photovoltaic data to the substation intelligent fusion terminal based on the 5G slice type; wherein the photovoltaic substation includes a 5G communication gateway and at least one photovoltaic array;

[0007] The intelligent fusion terminal in the substation uses a photovoltaic prediction model to predict photovoltaic power based on photovoltaic data and determine the photovoltaic power prediction results. Based on the photovoltaic power prediction results and the operating status of the photovoltaic substation, it determines the output curve of each photovoltaic array through deep reinforcement learning decision-making and sends the output curve to each photovoltaic array.

[0008] The cloud platform conducts agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and sends the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

[0009] In one possible implementation, agent training is performed based on historical PV power generation operation data to determine weight coefficients, and the weight coefficients are sent to the PV prediction model in the intelligent fusion terminal of the substation area, including:

[0010] The cloud platform builds a training set for the agent model based on historical photovoltaic power generation operation data;

[0011] Perform model training based on the training set and determine the weight coefficient of the proxy model;

[0012] The weight coefficient is sent to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

[0013] In one possible implementation, the 5G communication gateway includes a 5G communication module and a photovoltaic controller;

[0014] 5G communication modules and photovoltaic controllers are used to collect data from photovoltaic devices in the photovoltaic area and issue reverse control instructions;

[0015] Photovoltaic data includes the DC side voltage, current, power, AC side voltage and the status of the photovoltaic grid-connected switch of the photovoltaic equipment.

[0016] In one possible implementation, PV data is sent to the intelligent fusion terminal in the substation area based on the 5G slice type, including:

[0017] Based on time delay slicing, the grid-connected switch control data of the photovoltaic array is sent to the intelligent fusion terminal in the substation area;

[0018] Sending photovoltaic inverter power control instructions to the intelligent fusion terminal in the substation based on reliable slicing;

[0019] Low-frequency data is sent to the substation intelligent fusion terminal based on low-power wide-area slicing; the low-frequency data includes meter data, temperature and humidity, and light intensity.

[0020] In one possible implementation, photovoltaic power prediction is performed based on photovoltaic data using a photovoltaic prediction model to determine a photovoltaic power prediction result, including:

[0021] Determine power data, environmental data, and timestamp data within a target time period based on photovoltaic data;

[0022] Construct a multidimensional time series based on power data, environmental data and timestamp data;

[0023] The photovoltaic power prediction results are determined based on the multidimensional time series through the reset gate, update gate, candidate state and final state of the GRU network.

[0024] In one possible implementation, based on the photovoltaic power prediction results and the operating status of the photovoltaic area, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, including:

[0025] Based on the photovoltaic power prediction results, and combined with the load status of the photovoltaic area, current electricity price, energy storage status and time point as the input state of deep reinforcement learning, the photovoltaic power generation power at the next moment is determined.

[0026] In a possible implementation, the method further includes:

[0027] The cloud platform obtains the operating data of the photovoltaic area and generates a visual display interface based on the operating data;

[0028] The cloud platform is also used for photovoltaic power generation statistical reporting, operation and maintenance management, and dispatching instructions for distribution master stations.

[0029] According to a second aspect of the embodiments of this specification, a 5G-based edge intelligent photovoltaic control device is provided, which is applied to an intelligent photovoltaic control system. The system includes a cloud platform, a station intelligent fusion terminal, and a photovoltaic station; the device includes:

[0030] A data slicing module is configured to obtain photovoltaic data from a photovoltaic area and send the photovoltaic data to an intelligent fusion terminal in the area based on a 5G slicing type; wherein the photovoltaic area includes a 5G communication gateway and at least one photovoltaic array;

[0031] The data prediction module is configured as a smart fusion terminal in the substation area. It uses the photovoltaic prediction model to predict photovoltaic power based on photovoltaic data and determine the photovoltaic power prediction results. Based on the photovoltaic power prediction results and the operating status of the photovoltaic substation, it determines the output curve of each photovoltaic array through deep reinforcement learning decision-making and sends the output curve to each photovoltaic array.

[0032] The data control module is configured as a cloud platform to perform agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and send the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

[0033] According to a third aspect of an embodiment of this specification, a computing device is provided, including:

[0034] memory and processor;

[0035] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the above-mentioned 5G-based edge intelligent photovoltaic control method are implemented.

[0036] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, which, when executed by a processor, implement the steps of the above-mentioned 5G-based edge intelligent photovoltaic control method.

[0037] According to a fifth aspect of the embodiments of this specification, a computer program is provided, wherein, when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned 5G-based edge intelligent photovoltaic control method.

[0038] The embodiments of this specification provide a 5G-based edge intelligent photovoltaic control method and device, wherein the method includes: the photovoltaic area obtains photovoltaic data and sends the photovoltaic data to the intelligent fusion terminal of the area based on the 5G slice type; the intelligent fusion terminal of the area uses a photovoltaic prediction model to predict photovoltaic power based on the photovoltaic data and determines the photovoltaic power prediction result; based on the photovoltaic power prediction result and the operating status of the photovoltaic area, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, and the output curve is sent to each photovoltaic array; the cloud platform performs agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and sends the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the area. The terminal equipment not only effectively realizes the interconnection and data sharing of each photovoltaic array under the photovoltaic area, but also improves the efficiency and security of data processing, and provides support for the efficient implementation of power grid business management, safe production, quality service and other work. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of a 5G-based edge intelligent photovoltaic control method provided by an embodiment of this specification;

[0040] Figure 2 This is a system overall architecture diagram of a 5G-based edge intelligent photovoltaic control method provided by an embodiment of this specification;

[0041] Figure 3 This is a schematic diagram of the structure of a photovoltaic power generation system in a substation area, according to an embodiment of this specification, using a 5G-based edge intelligent photovoltaic control method;

[0042] Figure 4 This is a schematic diagram of a single power station communication architecture of a 5G-based edge intelligent photovoltaic control method provided by an embodiment of this specification;

[0043] Figure 5 This is a schematic diagram of 5G network slicing management from the device side to the edge side of a 5G-based edge intelligent photovoltaic control method provided by an embodiment of this specification;

[0044] Figure 6This is a schematic diagram of cloud-edge collaborative control of a 5G-based edge intelligent photovoltaic control method provided by an embodiment of this specification;

[0045] Figure 7 This is a structural diagram of a 5G-based edge intelligent photovoltaic control device provided by an embodiment of this specification;

[0046] Figure 8 This is a structural block diagram of a computing device provided by one embodiment of this specification. DETAILED DESCRIPTION

[0047] The following description sets forth many specific details to facilitate a thorough understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0048] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a," "an," and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0049] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0050] In this specification, a 5G-based edge intelligent photovoltaic control method is provided. This specification also involves a 5G-based edge intelligent photovoltaic control device, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0051] See also Figure 1 , Figure 1 A flowchart of a 5G-based edge intelligent photovoltaic control method provided according to an embodiment of this specification is shown, which is applied to an intelligent photovoltaic control system. The system includes a cloud platform, an intelligent fusion terminal in the substation and a photovoltaic substation; the method specifically includes the following steps.

[0052] Step 101: The photovoltaic area obtains photovoltaic data and sends the photovoltaic data to the area intelligent fusion terminal based on the 5G slice type; wherein the photovoltaic area includes a 5G communication gateway and at least one photovoltaic array.

[0053] In one possible implementation, the 5G communication gateway includes a 5G communication module and a photovoltaic controller; the 5G communication module and the photovoltaic controller are used to realize data collection and the issuance of reverse control instructions for each photovoltaic device under the photovoltaic area; the photovoltaic data includes the DC side voltage, current, power, AC side voltage and status of the photovoltaic equipment's grid-connected switch.

[0054] like Figure 2 As shown in the figure, the overall architecture of the intelligent photovoltaic control system consists of four components: a cloud-based master station, communication pipelines, edge computing, and device-side sensing. It features software-defined and cloud-edge collaboration, supports device-side devices, and uses 5G private networks and other communication methods to achieve real-time data collection and control command distribution between the edge and the device. On the edge, Docker is used to deploy the photovoltaic ultra-short-term prediction app, power quality diagnosis app, and control strategy adjustment app, enabling ultra-short-term prediction and precise control of photovoltaic power generation in the substation. Preprocessed data and fault information are uploaded to the cloud platform via 5G private networks, optical fiber, and satellite communications.

[0055] In practical applications, the structure of a typical high-proportion photovoltaic power generation system is as follows: Figure 3 As shown, within the photovoltaic array in the substation, smart meters, grid-connected switches, inverters and other equipment transmit power consumption data, equipment operating status and other information to the 5G communication gateway of the corresponding photovoltaic array through the RS485 bus, completing the preliminary integration of data in the area; the substation intelligent fusion terminal is installed on the outgoing line side of the transformer in the substation, and 5G slicing technology is used to communicate with the 5G communication gateway of each photovoltaic array in the substation.

[0056] Afterwards, the terminal transmits the integrated data to the platform through the operator's network via the 5G communication link, leveraging the high bandwidth and low latency characteristics of the 5G network to ensure efficient and stable data transmission;

[0057] After receiving the data, the platform pre-processes the data through the data management module and then distributes it to each back-end service unit through the core service module: the security management and data service module performs data security management and service cluster processing; the back-end service module implements agent prediction model training, statistical report generation and operation and maintenance management; finally, the data analysis and presentation platform conducts in-depth analysis of the data and presents the analysis results in the form of visual charts to provide support for power operation decisions.

[0058] In this embodiment, the communication architecture of a single photovoltaic array is as follows: Figure 4As shown, the PV array is connected to a combiner box via cables for current collection. The combiner box is then connected to an inverter via cables to convert DC to AC power. The inverter is then connected to a grid-connected meter cabinet via cables. Inside the grid-connected meter cabinet, smart meters and smart grid-connected switches connect to a 5G communication gateway via an RS485 bus. Among the environmental monitoring equipment, weather monitors connect to the 5G communication gateway via an RS485 link, while cameras exchange data with the 5G communication gateway via Ethernet links, forming a hardware connection system that integrates multiple communication links.

[0059] When the PV array is operating, the inverter, smart meter, and smart grid-connected switch collect power operation data such as current, voltage, and power, and transmit it to the 5G communication gateway via the RS485 bus. A meteorological monitor collects environmental data such as light intensity and temperature, and transmits it to the 5G communication gateway via the RS485 link. On-site image data captured by the camera is transmitted to the 5G communication gateway via Ethernet. Finally, the 5G communication gateway integrates power operation data and environmental monitoring data, and uploads it to the edge device via the 5G private network, achieving comprehensive perception and efficient transmission of the PV array's operating status.

[0060] In one possible implementation, photovoltaic data is sent to the substation intelligent fusion terminal based on the 5G slicing type, including: sending the grid-connected switch control data of the photovoltaic array to the substation intelligent fusion terminal based on the time delay slicing; sending the photovoltaic inverter power control instructions to the substation intelligent fusion terminal based on the reliable slicing; sending low-frequency data to the substation intelligent fusion terminal based on the low-power wide-area slicing; wherein the low-frequency data includes meter data, temperature and humidity, and light intensity.

[0061] In this embodiment, the 5G communication gateway divides the transmission data into the following sections using the 5G slicing technology: Figure 5 As shown in the figure, 5G slices are divided into three types of transmission based on data transmission requirements. Data such as grid-connection switching commands, power control commands, and fault alarms require extremely high real-time and reliable transmission, ensuring rapid command issuance and immediate response to fault alarms. To minimize end-to-end latency, the 5G network's DU, CU, and core network are all offloaded to the edge cloud. Monitoring data such as voltage, current, power generation, temperature and humidity, and light intensity are collected from numerous terminals and contain massive amounts of data, requiring the network to support large-scale connectivity. Furthermore, most sensors are stationary and do not require mobility management. Therefore, offloading the 5G network's DU, CU, and core network to the core cloud reduces 5G network power consumption. Video surveillance data transmission requires high bandwidth and smooth, uninterrupted high-definition video. Therefore, the 5G network's DU and some core network functions are virtualized, along with storage servers, and placed in the edge cloud. The remaining virtualized core network functions are placed in the core cloud to meet the high-quality visual transmission requirements of video surveillance services.

[0062] Step 102: The intelligent fusion terminal in the substation uses the photovoltaic prediction model to predict photovoltaic power based on the photovoltaic data and determines the photovoltaic power prediction result; based on the photovoltaic power prediction result and the operating status of the photovoltaic substation, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, and the output curve is sent to each photovoltaic array.

[0063] In one possible implementation, photovoltaic power prediction is performed based on photovoltaic data using a photovoltaic prediction model to determine a photovoltaic power prediction result, including: determining power data, environmental data, and timestamp data within a target time period based on the photovoltaic data; constructing a multidimensional time series based on the power data, environmental data, and timestamp data; and determining the photovoltaic power prediction result based on the multidimensional time series through a reset gate, an update gate, a candidate state, and a final state of a GRU network.

[0064] In actual applications, intelligent fusion terminals are deployed on the edge of the grid, and deep learning algorithms are used to achieve comprehensive state perception and precise coordinated control of each photovoltaic array in a high-proportion photovoltaic grid. The process includes:

[0065] The edge device subscribes to micro-meteorological data from the edge weather station every 5 minutes through the 5G network, and collects data such as the power generation power, temperature, humidity, and light intensity of the photovoltaic modules every 15 minutes; the collected photovoltaic power generation data is preprocessed, and the standard deviation method is used to eliminate sudden measurement values.

[0066]

[0067]

[0068] If |P(t)-μ(t)|>3σ(t), it is determined to be an outlier and replaced by:

[0069]

[0070] Where N is the size of the photovoltaic power data sample; The time interval for collecting PV data for edge devices; Leading the way for photovoltaics Power generation at the moment; is the mean value of the photovoltaic power generation data sample; is the standard deviation of the photovoltaic power generation data sample at time t.

[0071] Secondly, the photovoltaic power generation power, light intensity, temperature, humidity, wind direction and timestamp data within the past hour are used to form a multidimensional time series; time features and weather features (cloud cover, irradiance) are extracted to form a feature vector for each time step. The input sequence of its GRU network is;

[0072]

[0073] in is a multidimensional time series at time t. The GRU network’s reset gate, update gate, candidate state, and final state are used to fuse historical data with current information. Finally, the fully connected layer is used to map the predicted value of photovoltaics.

[0074]

[0075] in, is the weight parameter of the fully connected layer of the GRU network; is the number of neurons in the fully connected layer; is the bias parameter of the fully connected layer.

[0076] In one possible implementation, the output curve of each photovoltaic array is determined through deep reinforcement learning based on the photovoltaic power prediction results and the operating status of the photovoltaic station. This includes: based on the photovoltaic power prediction results, and combining the load status of the photovoltaic station, the current electricity price, the energy storage status and the time point as the input state of deep reinforcement learning, deciding the photovoltaic power generation power at the next moment.

[0077] In practical applications, based on the results of photovoltaic predictions, combined with the load status of the substation, current electricity prices, energy storage status and time points as the input state of deep reinforcement learning, the photovoltaic power generation power at the next moment is decided and instructions are sent to the 5G communication gateway.

[0078] Distributed photovoltaic intelligent sensing devices and 5G communication gateways are deployed on the edge to collect data from each photovoltaic array in the substation and receive control instructions. The collected data includes the DC side voltage, current, power, AC side voltage, and the status of the photovoltaic grid-connected switches.

[0079] Step 103: The cloud platform performs agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and sends the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

[0080] In one possible implementation method, agent training is performed based on historical photovoltaic power generation operation data to determine the weight coefficient, and the weight coefficient is sent down to the photovoltaic prediction model in the intelligent fusion terminal of the substation, including: the cloud platform constructs a training set of the agent model based on the historical photovoltaic power generation operation data; the model is trained based on the training set to determine the weight coefficient of the agent model; and the weight coefficient is sent down to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

[0081] In actual applications, the cloud platform constructs a training set for the proxy model based on historical PV power generation operation data and trains the PV prediction model. After training, the weight coefficients of the cloud-based PV prediction proxy model are sent to the PV prediction model in the edge device.

[0082] In one possible implementation, the cloud platform also includes obtaining the operating data of the photovoltaic area and generating a visual display interface based on the operating data; the cloud platform is also used to prepare photovoltaic power generation statistical reports, operation and maintenance management, and dispatch instructions for the distribution master station.

[0083] In practical applications, the cloud platform integrates real-time operating data from high-volume photovoltaic substations and provides a visual display, including photovoltaic power generation monitoring, environmental monitoring, video monitoring, and inverter status query. The cloud platform supports photovoltaic power generation statistical reporting, operation and maintenance management, and dispatch instructions from the distribution master station. Satellite communications can quickly locate the substation where a fault occurs, assisting operators in quickly completing on-site maintenance work.

[0084] In an overall embodiment, the cloud-edge collaborative control process is as follows Figure 6 As shown, in this embodiment, the cloud platform first receives the photovoltaic operation data uploaded by the edge and completes the visualization of the photovoltaic operation status. It then determines whether the distribution master station needs to adjust the power of the substation. If no adjustment is required, the cloud platform conducts model training based on historical data. After the photovoltaic prediction agent model training is completed, it will be sent to the smart device on the edge side of the substation; if adjustment is required, the cloud platform will send the substation power control instruction to the edge side. The smart terminal on the edge side combines the algorithm model to predict the power generation power of each photovoltaic array. The control strategy APP generates the control power of the corresponding photovoltaic array, and then determines the control properties of the target control value through the 5G communication gateway. If it is rigid control, the end side executes the grid-connected switch operation; if it is flexible control, the end side implements the flexible control strategy. Finally, the 5G communication gateway re-collects the photovoltaic array measurement data and uploads it to the edge side to form a complete collaborative control closed loop.

[0085] Furthermore, the edge-side operating modes are divided into substation autonomous mode and master station linkage mode. Substation autonomous mode performs substation active power autonomy based on the preset PV power control range, automatically generates control strategies, and issues control commands. If the master station command is not refreshed for a long period (settable to 15 minutes), it automatically switches to substation autonomous mode. Master station linkage mode automatically tracks and responds to control commands issued by the distribution master station. If the substation is currently in substation autonomous mode and the master station command is refreshed, it automatically switches to master station linkage mode. This enables comprehensive status awareness and cloud-edge coordinated control of each PV array in a high-proportion PV substation.

[0086] The embodiments of this specification provide a 5G-based edge intelligent photovoltaic control method and device, wherein the method includes: the photovoltaic area obtains photovoltaic data and sends the photovoltaic data to the intelligent fusion terminal of the area based on the 5G slice type; the intelligent fusion terminal of the area uses a photovoltaic prediction model to predict photovoltaic power based on the photovoltaic data and determines the photovoltaic power prediction result; based on the photovoltaic power prediction result and the operating status of the photovoltaic area, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, and the output curve is sent to each photovoltaic array; the cloud platform performs agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and sends the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the area. The terminal equipment not only effectively realizes the interconnection and data sharing of each photovoltaic array under the photovoltaic area, but also improves the efficiency and security of data processing, and provides support for the efficient implementation of power grid business management, safe production, quality service and other work.

[0087] Furthermore, the embodiments of this specification utilize the high capacity, high speed and high reliability of 5G technology to realize the interaction of massive heterogeneous data between cloud, edge and end. Through the collaboration of the cloud platform with intelligent fusion terminals and photovoltaic equipment, an end-to-end technical architecture of "cloud-edge-end three-body collaboration" is formed. Tasks such as network forwarding, storage, computing and intelligent data analysis are migrated to edge processing, thereby effectively reducing response latency, alleviating the burden on the cloud and reducing bandwidth costs. In addition, the cloud platform can provide cloud services such as photovoltaic area scheduling and computing power distribution to improve overall data processing efficiency and service quality.

[0088] The embodiments in this specification utilize edge computing to achieve precise power forecasting, group scheduling, and control in photovoltaic (PV) substations. This enables measurable, adjustable, and controllable distributed PV systems, improving the reliability of operations in high-proportion PV substations. This enables precise location and monitoring of fault points, improving operational efficiency, reducing time, and lowering costs.

[0089] The embodiments of this specification utilize a 5G communication gateway on the device side to collect data from photovoltaic power plants. 5G network slicing technology is used to prioritize channels for control instructions, monitoring data, and video transmission. This enables millisecond-level data collection and reverse control instruction distribution for a high-proportion photovoltaic power station, eliminating traditional communication cables, fiber optic switches, and other equipment, reducing costs while improving reliability, transmission capacity, and speed.

[0090] Corresponding to the above method embodiment, this specification also provides an embodiment of an edge intelligent photovoltaic control device based on 5G, Figure 7 The figure shows a schematic diagram of the structure of a 5G-based edge intelligent photovoltaic control device provided by an embodiment of this specification. Figure 7 As shown, the device is applied to the intelligent photovoltaic control system. The system includes a cloud platform, a smart fusion terminal for the substation, and a photovoltaic substation. The device includes:

[0091] The data slicing module 701 is configured to obtain photovoltaic data from a photovoltaic area and send the photovoltaic data to an intelligent fusion terminal in the area based on a 5G slicing type. The photovoltaic area includes a 5G communication gateway and at least one photovoltaic array.

[0092] The data prediction module 702 is configured as a substation intelligent fusion terminal to use a photovoltaic prediction model to perform photovoltaic power prediction based on photovoltaic data and determine the photovoltaic power prediction results; based on the photovoltaic power prediction results and the operating status of the photovoltaic substation, it determines the output curve of each photovoltaic array through deep reinforcement learning decision-making and sends the output curve to each photovoltaic array;

[0093] The data control module 703 is configured as a cloud platform to perform agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and send the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

[0094] In one possible implementation, agent training is performed based on historical PV power generation operation data to determine weight coefficients, and the weight coefficients are sent to the PV prediction model in the intelligent fusion terminal of the substation area, including:

[0095] The cloud platform builds a training set for the agent model based on historical photovoltaic power generation operation data;

[0096] Perform model training based on the training set and determine the weight coefficient of the proxy model;

[0097] The weight coefficient is sent to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

[0098] In one possible implementation, the 5G communication gateway includes a 5G communication module and a photovoltaic controller;

[0099] 5G communication modules and photovoltaic controllers are used to collect data from photovoltaic devices in the photovoltaic area and issue reverse control instructions;

[0100] Photovoltaic data includes the DC side voltage, current, power, AC side voltage and the status of the photovoltaic grid-connected switch of the photovoltaic equipment.

[0101] In one possible implementation, PV data is sent to the intelligent fusion terminal in the substation area based on the 5G slice type, including:

[0102] Based on time delay slicing, the grid-connected switch control data of the photovoltaic array is sent to the intelligent fusion terminal in the substation area;

[0103] Sending photovoltaic inverter power control instructions to the intelligent fusion terminal in the substation based on reliable slicing;

[0104] Low-frequency data is sent to the substation intelligent fusion terminal based on low-power wide-area slicing; the low-frequency data includes meter data, temperature and humidity, and light intensity.

[0105] In one possible implementation, photovoltaic power prediction is performed based on photovoltaic data using a photovoltaic prediction model to determine a photovoltaic power prediction result, including:

[0106] Determine power data, environmental data, and timestamp data within a target time period based on photovoltaic data;

[0107] Construct a multidimensional time series based on power data, environmental data and timestamp data;

[0108] The photovoltaic power prediction results are determined based on the multidimensional time series through the reset gate, update gate, candidate state and final state of the GRU network.

[0109] In one possible implementation, based on the photovoltaic power prediction results and the operating status of the photovoltaic area, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, including:

[0110] Based on the photovoltaic power prediction results, and combined with the load status of the photovoltaic area, current electricity price, energy storage status and time point as the input state of deep reinforcement learning, the photovoltaic power generation power at the next moment is determined.

[0111] In a possible implementation, the method further includes:

[0112] The cloud platform obtains the operating data of the photovoltaic area and generates a visual display interface based on the operating data;

[0113] The cloud platform is also used for photovoltaic power generation statistical reporting, operation and maintenance management, and dispatching instructions for distribution master stations.

[0114] The embodiments of this specification provide a 5G-based edge intelligent photovoltaic control method and device, wherein the device includes: the photovoltaic area obtains photovoltaic data and sends the photovoltaic data to the area intelligent fusion terminal based on the 5G slice type; the area intelligent fusion terminal uses a photovoltaic prediction model to predict photovoltaic power based on the photovoltaic data and determines the photovoltaic power prediction result; based on the photovoltaic power prediction result and the operating status of the photovoltaic area, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, and the output curve is sent to each photovoltaic array; the cloud platform performs agent training based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and sends the weight coefficient to the photovoltaic prediction model in the area intelligent fusion terminal. The terminal equipment not only effectively realizes the interconnection and data sharing of each photovoltaic array under the photovoltaic area, but also improves the efficiency and security of data processing, providing support for the efficient implementation of power grid business management, safe production, quality service and other work.

[0115] The above is a schematic scheme of a 5G-based edge intelligent photovoltaic control device of this embodiment. It should be noted that the technical scheme of the 5G-based edge intelligent photovoltaic control device and the technical scheme of the 5G-based edge intelligent photovoltaic control method described above are based on the same concept. For details not described in detail in the technical scheme of the 5G-based edge intelligent photovoltaic control device, please refer to the description of the technical scheme of the 5G-based edge intelligent photovoltaic control method described above.

[0116] Figure 8 8 shows a block diagram of a computing device 800 according to one embodiment of the present disclosure. Components of the computing device 800 include, but are not limited to, a memory 810 and a processor 820. The processor 820 is connected to the memory 810 via a bus 830, and a database 850 is used to store data.

[0117] Computing device 800 also includes an access device 840 that enables computing device 800 to communicate via one or more networks 860. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. Access device 840 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, or a near field communication (NFC) interface.

[0118] In one embodiment of the present specification, the above components of the computing device 800 and Figure 8 Other components not shown in the figure may also be connected to each other, for example, via a bus. Figure 8 The computing device structure block diagram shown is for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art may add or replace other components as needed.

[0119] Computing device 800 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, personal digital assistant, laptop computer, notebook computer, netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or personal computer (PC). Computing device 500 can also be a mobile or stationary server.

[0120] Among them, the processor 520 is used to execute the following computer-executable instructions, which, when executed by the processor, implement the steps of the above-mentioned 5G-based edge intelligent photovoltaic control method. The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical solution of the computing device and the technical solution of the above-mentioned 5G-based edge intelligent photovoltaic control method belong to the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the above-mentioned 5G-based edge intelligent photovoltaic control method.

[0121] An embodiment of this specification also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the above-mentioned 5G-based edge intelligent photovoltaic control method.

[0122] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of this storage medium and the technical scheme of the aforementioned 5G-based edge intelligent photovoltaic control method are based on the same concept. For details not described in detail in the technical scheme of the storage medium, please refer to the description of the technical scheme of the aforementioned 5G-based edge intelligent photovoltaic control method.

[0123] An embodiment of this specification also provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned 5G-based edge intelligent photovoltaic control method.

[0124] The above is a schematic scheme of a computer program of this embodiment. It should be noted that the technical scheme of this computer program and the technical scheme of the aforementioned 5G-based edge intelligent photovoltaic control method are based on the same concept. For details not described in detail in the technical scheme of the computer program, please refer to the description of the technical scheme of the aforementioned 5G-based edge intelligent photovoltaic control method.

[0125] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0126] The computer instructions include computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium may be appropriately increased or decreased based on the requirements of legislation and patent practice within a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0127] It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0128] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0129] The preferred embodiments disclosed above are intended only to help illustrate this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made based on the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that those skilled in the art can better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A 5G-based edge intelligent photovoltaic control method, characterized in that: Applied to an intelligent photovoltaic control system, the system includes a cloud platform, a substation intelligent fusion terminal, and a photovoltaic substation; the method includes: The photovoltaic area acquires photovoltaic data and sends the photovoltaic data to the area intelligent fusion terminal based on the 5G slice type; wherein the photovoltaic area includes a 5G communication gateway and at least one photovoltaic array; The intelligent fusion terminal in the substation area uses a photovoltaic prediction model to perform photovoltaic power prediction based on the photovoltaic data and determines a photovoltaic power prediction result; based on the photovoltaic power prediction result and the operating status of the photovoltaic substation, the output curve of each photovoltaic array is determined through deep reinforcement learning decision-making, and the output curve is sent to each photovoltaic array; The cloud platform performs agent training based on historical photovoltaic power generation operation data to determine weight coefficients, and sends the weight coefficients to the photovoltaic prediction model in the intelligent fusion terminal of the substation area; The sending the photovoltaic data to the substation intelligent fusion terminal based on the 5G slice type includes: Sending the grid-connected switch control data of the photovoltaic array to the substation intelligent fusion terminal based on the time delay slice; Sending photovoltaic inverter power control instructions to the intelligent fusion terminal in the substation based on reliable slices; Low-frequency data is sent to the substation intelligent fusion terminal based on low-power wide-area slicing; wherein the low-frequency data includes meter data, temperature and humidity, and light intensity.

2. The method according to claim 1, characterized in that The method of performing agent training based on historical photovoltaic power generation operation data to determine weight coefficients, and sending the weight coefficients to the photovoltaic prediction model in the substation intelligent fusion terminal, includes: The cloud platform constructs a training set of the agent model based on the historical operation data of photovoltaic power generation; Performing model training based on the training set to determine the weight coefficient of the proxy model; The weight coefficient is sent to the photovoltaic prediction model in the substation intelligent fusion terminal.

3. The method according to claim 1, characterized in that The 5G communication gateway includes a 5G communication module and a photovoltaic controller; The 5G communication module and the photovoltaic controller are used to realize data collection and the issuance of reverse control instructions for each photovoltaic device under the photovoltaic area; The photovoltaic data includes the DC side voltage, current, power, AC side voltage and status of the photovoltaic grid-connected switch of the photovoltaic equipment.

4. The method according to claim 1, wherein The photovoltaic power prediction is performed based on the photovoltaic data using a photovoltaic prediction model to determine a photovoltaic power prediction result, including: determining power data, environmental data, and timestamp data within a target time period based on the photovoltaic data; forming a multidimensional time series based on the power data, environmental data and timestamp data; A photovoltaic power prediction result is determined based on the multidimensional time series through a reset gate, an update gate, a candidate state, and a final state of a GRU network.

5. The method according to claim 1, wherein The method of determining the output curve of each photovoltaic array by deep reinforcement learning decision-making based on the photovoltaic power prediction result and the operating status of the photovoltaic area includes: Based on the photovoltaic power prediction results, and combined with the load status, current electricity price, energy storage status and time point of the photovoltaic area as the input state of deep reinforcement learning, the photovoltaic power generation power at the next moment is determined.

6. The method according to claim 1, wherein Also includes: The cloud platform obtains the operating data of the photovoltaic area and generates a visual display interface based on the operating data; The cloud platform is also used to generate photovoltaic power generation statistical reports, operation and maintenance management, and dispatch instructions for the distribution master station.

7. A 5G-based edge intelligent photovoltaic control device, characterized in that: The steps for implementing the 5G-based edge intelligent photovoltaic control method according to any one of claims 1 to 6 are applied to an intelligent photovoltaic control system, wherein the system includes a cloud platform, an intelligent fusion terminal in a substation, and a photovoltaic substation; and the device includes: A data slicing module is configured to obtain photovoltaic data for the photovoltaic area and send the photovoltaic data to the area intelligent fusion terminal based on the 5G slicing type; wherein the photovoltaic area includes a 5G communication gateway and at least one photovoltaic array; The data prediction module is configured to use a photovoltaic prediction model to perform photovoltaic power prediction on the intelligent fusion terminal of the substation based on the photovoltaic data to determine a photovoltaic power prediction result; based on the photovoltaic power prediction result and the operating status of the photovoltaic substation, determine the output curve of each photovoltaic array through deep reinforcement learning decision-making, and send the output curve to each photovoltaic array; The data control module is configured to perform agent training on the cloud platform based on the historical operation data of photovoltaic power generation to determine the weight coefficient, and send the weight coefficient to the photovoltaic prediction model in the intelligent fusion terminal of the substation.

8. A computing device, characterized in that include: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the 5G-based edge intelligent photovoltaic control method described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the 5G-based edge intelligent photovoltaic control method described in any one of claims 1 to 6.

Images