Collection, calculation and transmission integrated gateway for power distribution station house
By designing an integrated gateway for computing and transmission in distribution stations, using CPU+NPU hybrid processors and multiple communication interfaces, the existing gateways have solved the shortcomings of unstructured data acquisition and transmission, realizing dynamic data acquisition frequency adjustment and artificial intelligence reasoning acceleration, and improving identification accuracy and operation and maintenance efficiency.
Patent Information
- Application Number
- CN202421864089.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2025-06-24
- Estimated Expiration
- 2034-08-02
AI Technical Summary
The existing distribution station gateways lack support for unstructured data acquisition and transmission, cannot dynamically adjust the data acquisition frequency, and the accuracy of artificial intelligence inference recognition is not high, there are problems of missed and false alarms, and the remote operation and maintenance management functions are insufficient, resulting in large on-site operation and maintenance workload of equipment.
A integrated gateway for computing and transmission of distribution stations was designed, using CPU+NPU hybrid processors, supporting multiple communication interfaces such as micro-power wireless, LoRa transmissive transmission, RS485, RS232, Ethernet, etc., which can dynamically adjust the data acquisition frequency, and accelerate artificial intelligence inference through built-in NPU to achieve secondary accurate analysis and judgment. At the same time, it is connected to the IoT management platform through the 5G/4G full network access module, providing remote operation and maintenance management functions.
It realizes the acquisition and transmission of unstructured data such as videos and maps, dynamically adjusts the data acquisition frequency, improves the accuracy and practicality of artificial intelligence reasoning, reduces the on-site operation and maintenance workload of equipment, and improves the intelligent level of distribution station monitoring.
Smart Images

Figure CN223024431U_ABST
Abstract
Description
Technical Field
[0001] The utility model belongs to the technical field of digitalization of power grids, and specifically relates to an integrated data acquisition, calculation and transmission gateway for a power distribution station. Background Art
[0002] With the rapid development of information technology and new energy technology, distribution substations have begun to develop in the direction of digitalization, but there are a series of problems in the digitalization of distribution substations. Chinese patent 202222711479.2 proposes an edge intelligent gateway for distribution substations to achieve access to environmental sensing equipment and environmental control equipment in distribution substations with diversified links, but the patent has three shortcomings: first, it lacks support for the collection and transmission of unstructured data such as local discharge maps; second, it does not support management functions such as remote debugging and upgrading; third, it can only be fixedly identified from the video stream content, and it is impossible to dynamically adjust the collection frequency and the linkage platform for secondary precise judgment according to the recognition results, the accuracy is not high and the practicality is poor; for example, Chinese patent 202121306888.3 proposes a new type of artificial intelligence visualization gateway for distribution substations, but the patent lacks the NPU or GPU required for artificial intelligence calculations, and cannot run artificial intelligence algorithms efficiently and stably, and the application effect is limited; therefore, it is very necessary to provide a distribution substation collection, calculation and transmission integrated gateway that dynamically adjusts the data collection frequency, realizes secondary precise judgment through linkage, improves practicality, and reduces the workload of on-site equipment operation and maintenance. Summary of the invention
[0003] The purpose of the utility model is to overcome the shortcomings of the existing technology and to provide a distribution station room integrated data collection, calculation and transmission gateway which can dynamically adjust the frequency of data collection, realize secondary accurate analysis through linkage, improve practicality and reduce the workload of on-site operation and maintenance of equipment.
[0004] The purpose of the utility model is achieved as follows: a distribution station integrated collection, calculation and transmission gateway, which includes a core board, and the core board is respectively connected to a power module, at least one indicator light, a Console interface, at least one USB interface, at least one 100M Ethernet interface, at least one micro-power wireless, at least one LoRa transparent transmission, a 5G / 4G full network access module, at least one RS485 and at least one RS232.
[0005] The micro-power wireless adopts 2.4G / 470M dual-mode micro-power wireless, and 2.4G / 470M are each set up with 2 channels; the LoRa transparent transmission adopts 470M LoRa transparent transmission, and is set up with 2 channels.
[0006] The USB interface is provided with 2 channels, the 100M Ethernet interface is provided with 4 channels, the RS485 is provided with 4 channels, and the RS232 is provided with 4 channels.
[0007] The core board is composed of a CPU+NPU hybrid processor, memory, and storage.
[0008] When collecting data, the core board communicates with the terminal device through the network port, wireless, RS232, and RS485 communication interfaces, and collects the sensed data to the gateway according to the protocol format of the terminal device.
[0009] When calculating, the core board analyzes and processes the collected terminal device perception data, and analyzes and judges it through the artificial intelligence model. The NPU processor built into the core board can efficiently run the artificial intelligence model and accelerate the artificial intelligence reasoning.
[0010] When transmitting, the core board sends the collected original perception data, calculation analysis and artificial intelligence reasoning results to the IoT management platform through the 5G / 4G full network access module and the APN node.
[0011] The power module converts 220V AC power into DC power of multiple voltage levels such as 12V, 5V, and 3.3V to power all functional circuits of the gateway.
[0012] The indicator lights are used to indicate the working status of the gateway device, including RS232, RS485 serial port working indicator lights, Ethernet working indicator lights, and system working indicator lights; the system working indicator lights are divided into operation, error, timing, 4G, power 1, and power 2.
[0013] The Console interface and the 100M Ethernet interface both adopt RJ45 interface specifications, and the USB interface adopts USB Type A interface specifications.
[0014] The beneficial effects of the utility model are as follows: the utility model is an integrated data acquisition, calculation and transmission gateway for distribution station rooms. The integrated data acquisition, calculation and transmission gateway for distribution station rooms solves the information interaction between the data plane and the control plane and the data acquisition and transmission of unstructured data such as videos and graphs, realizes practical edge computing and edge AI framework, and accelerates the application of artificial intelligence in station scenes such as equipment defect identification and video violation identification. During use, the utility model solves the problem of data acquisition and transmission of unstructured data such as partial discharge spectra and video images. Since the existing gateway artificial intelligence reasoning and recognition adopts a fixed cycle, and the reasoning and recognition accuracy is not high, and the problem of missed reports and false alarms is serious, the utility model can dynamically adjust the data collection frequency according to the recognition results and trigger the artificial intelligence platform to link and accurately judge the secondary, so as to realize timely and accurate identification of problems and improve the practicality of scene artificial intelligence. The utility model solves the problem of remote operation and maintenance management of distribution station room gateways, reduces the workload of equipment on-site operation and maintenance, and improves the usability of the device. The utility model has the advantages of dynamically adjusting the data collection frequency, linking to realize secondary accurate judgment, improving practicality, and reducing the workload of equipment on-site operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 This is the front schematic diagram of an integration gateway for acquisition, calculation, and transmission in a substation building of the present utility model.
[0016] Figure 2 This is the back schematic diagram of an integration gateway for acquisition, calculation, and transmission in a substation building of the present utility model.
[0017] Figure 3 This is the composition architecture diagram of an integration gateway for acquisition, calculation, and transmission in a substation building of the present utility model.
[0018] Figure 4 This is the implementation and application schematic diagram of an integration gateway for acquisition, calculation, and transmission in a substation building of the present utility model. Specific implementation manners
[0019] The following further describes the present utility model with reference to the accompanying drawings. Embodiment
[0020] As Figures 1-4 shown, an integration gateway for acquisition, calculation, and transmission in a substation building includes a core board, and the core board is respectively connected to a power supply module, at least one indicator light, a Console interface, at least one path of USB interface, at least one path of gigabit Ethernet interface, at least one path of micro-power wireless, at least one path of LoRa transparent transmission, a 5G / 4G full-net communication module, at least one path of RS485, and at least one path of RS232.
[0021] The micro-power wireless uses a 2.4G / 470M dual-mode micro-power wireless, and 2 paths are set for each of 2.4G / 470M; the LoRa transparent transmission uses a 470M LoRa transparent transmission, and 2 paths are set.
[0022] In this embodiment, for the 4G / 470M dual-mode micro-power wireless interface, 2 paths are set for each mode, which meets the requirements of the communication protocol standard of the micro-power wireless network for power transmission and transformation equipment Internet of Things, and is suitable for connecting Internet of Things devices with short distance and low power consumption, such as sensors for contact temperature measurement, switch cabinet partial discharge, environmental temperature and humidity, etc.;
[0023] There are 2 paths of LoRa transparent transmission. Using LoRa (Long Range) low-power wide-area network technology, it realizes long-distance and low-power data transmission. The transparent transmission function ensures the direct transmission of data from one end to the other end, without additional data processing on the gateway, and is used for accessing data of non-standard wireless sensors (non-micro-power wireless);
[0024] There is 1 path of 5G / 4G full-net communication, which provides high-speed mobile network access capability, ensuring that in the case of no fiber optic internal network in the substation building, the gateway can still access the power grid information internal network platform (Internet of Things management platform, unified video platform, artificial intelligence platform, etc.) through 5G / 4G APN.
[0025] There are 2 USB interfaces, 4 100M Ethernet interfaces, 4 RS485 interfaces, and 4 RS232 interfaces.
[0026] In this embodiment, the 4 RS485 interfaces and the 4 RS232 interfaces support half-duplex multi-point communication and full-duplex point-to-point communication respectively, with strong anti-interference ability and long-distance transmission ability, and are mainly used to access controllers such as fans and water pumps.
[0027] The core board consists of a CPU+NPU hybrid processor, memory, and storage.
[0028] When the core board collects data, it communicates with the terminal device through network ports, wireless, RS232, and RS485 communication interfaces, and collects the sensed data into the gateway according to the protocol format of the terminal device.
[0029] When the core board calculates, it analyzes and processes the sensed data of the terminal device collected, and conducts analysis and judgment through an artificial intelligence model. The built-in NPU processor of the core board can efficiently run the artificial intelligence model to accelerate artificial intelligence reasoning.
[0030] When the core board transmits data, it sends the collected original sensed data, calculation analysis, and artificial intelligence reasoning results to the Internet of Things management platform through the 5G / 4G full-netcom module via the APN node.
[0031] In this embodiment, the core board module consists of a CPU+NPU hybrid processor, memory, and storage, can run the Linux operating system, supports containerized operation, and can deploy collection, calculation, and transmission applications according to different containers. The collection application is responsible for communicating with the terminal device through communication interfaces such as network ports, wireless, RS232, and RS485, and collecting the sensed data into the gateway according to the protocol format of the terminal device; the calculation application is responsible for analyzing and processing the sensed data of the terminal device collected, and conducting analysis and judgment through an artificial intelligence model. The built-in NPU of the core board can efficiently run the artificial intelligence model to accelerate artificial intelligence application reasoning; the transmission application is responsible for sending the collected original sensed data, calculation analysis, and artificial intelligence reasoning results to the Internet of Things management platform through the 5G / 4G full-netcom module via the APN node.
[0032] In this embodiment, the integrated acquisition and calculation gateway of the present utility model has the following functions: ① Acquisition function: The acquisition function supports access through wired and wireless interfaces such as micro-power wireless, LoRa transparent transmission, RS485, RS232, and Ethernet. For example, it can access 9 types of sensing terminals such as contact temperature sensors, switchgear partial discharge sensors, switchgear temperature and humidity, ambient temperature and humidity, water level, smoke, cameras, fans, and water pumps. In terms of protocols, it supports Modbus RTU, Modbus TCP, the communication protocol of the micro-power wireless network for power transmission and transformation equipment Internet of Things, Onvif protocol, RTSP protocol, etc., and can support custom protocols in the form of upgrading protocol library files. For the unstructured data of partial discharge spectrograms, the acquisition supports collecting data into the gateway through standard protocols and custom protocols. Then, the gateway converts the unstructured data into XML hypertext language for description, and performs GZIP compression processing on the results, converting the processed hypertext content into a string for subsequent transmission;
[0033] ② Calculation function: The calculation function is divided into two types. One is data calculation and analysis. Based on the data of the 9 types of sensing terminals collected, it analyzes the data change rate and threshold, alarms when the change rate and threshold exceed the limit, and executes linkage strategies. For example, when the ambient temperature and humidity exceed the limit, it alarms and turns on the fan to cool down and remove moisture; when the ambient water level exceeds the limit, it alarms and turns on the water pump to drain water; when the contact temperature rise exceeds the limit, it alarms, etc. The other is artificial intelligence reasoning. The gateway extracts image frames from the RTSP data stream of the camera at a minute frequency and inputs them into an artificial intelligence model (such as safety helmets, water accumulation, fires, people falling, small animal intrusion, etc.) for reasoning and analysis. If a suspected result is identified, it increases the image frame extraction frequency of the RTSP data stream to 5S / frame (can be set), continuously captures the suspected frames and converts them into the Base64 string format, and finally uploads the suspected result and the suspected frames to the Internet of Things management platform to trigger the large model of the artificial intelligence platform for secondary accurate judgment;
[0034] ③ Transmission function: The transmission function mainly transmits the original data of the 9 types of sensing terminals collected, the calculation analysis, and the results of artificial intelligence reasoning and identification to the power grid information intranet platform through 5G / 4G APN; 9 types of sensing data such as contact temperature sensors, switchgear partial discharge (spectrograms), switchgear temperature and humidity, ambient temperature and humidity, water level, smoke, fans, water pumps, image frames, etc., as well as the calculation analysis and the results of artificial intelligence reasoning and identification, are connected to the Internet of Things management platform through the MQTT protocol; The camera video data is connected to the unified video platform through the B interface protocol; In addition, it also supports receiving remote management commands issued by the Internet of Things management platform through the MQTT protocol, such as program debugging and upgrade management operations, to achieve remote operation and maintenance.
[0035] This utility model is an integrated data acquisition, calculation and transmission gateway for distribution station rooms. It solves the information interaction between the data plane and the control plane and the data acquisition and transmission of unstructured data such as videos and graphs, realizes practical edge computing and edge AI framework, and accelerates the application of artificial intelligence in station room scenarios such as equipment defect identification and video violation identification. During use, this utility model realizes the collection and transmission of unstructured data such as partial discharge spectra and video images on the basis of the existing gateway remote signaling and telemetry data collection and transmission, comprehensively covers the data acquisition and transmission needs of station room equipment and environment (structured and unstructured), and further improves the intelligent level of station room monitoring; this utility model realizes artificial intelligence application on the side, which is different from In the existing gateway artificial intelligence fixed-cycle reasoning and recognition mode, it can dynamically adjust the data collection frequency according to the recognition results and link with the artificial intelligence platform for secondary precise analysis, with higher timeliness and accuracy. Compared with the centralized cloud reasoning mode, it greatly reduces the gateway traffic consumption and improves the practicality of gateway artificial intelligence without sacrificing the timeliness and accuracy of recognition. The utility model uses the communication channel between the Internet of Things management platform and the gateway to realize remote operation and maintenance functions such as program debugging, upgrading, and restarting. The equipment on-site operation and maintenance workload is small, and the operation and maintenance is convenient and fast. The utility model has the advantages of dynamically adjusting the data collection frequency, linking to achieve secondary precise analysis, improving practicality, and reducing the workload of equipment on-site operation and maintenance. Example
[0036] like Figures 1-4 As shown, a distribution station integrated procurement, calculation and transmission gateway includes a core board, and the core board is respectively connected to a power module, at least one indicator light, a console interface, at least one USB interface, at least one 100M Ethernet interface, at least one micro-power wireless interface, at least one LoRa transparent transmission interface, a 5G / 4G full network access module, at least one RS485 interface and at least one RS232 interface.
[0037] The power module converts 220V AC power into DC power of multiple voltage levels such as 12V, 5V, and 3.3V to power all functional circuits of the gateway.
[0038] The indicator lights are used to indicate the working status of the gateway device, including RS232, RS485 serial port working indicator lights, Ethernet working indicator lights, and system working indicator lights; the system working indicator lights are divided into operation, error, timing, 4G, power 1, and power 2.
[0039] The Console interface and the 100M Ethernet interface both adopt RJ45 interface specifications, and the USB interface adopts USB Type A interface specifications.
[0040] In this embodiment, the Console interface adopts the RJ45 interface specification, which is a non-network debugging interface. Through this interface, the gateway can be locally debugged and upgraded; the USB interface adopts the USB Type A interface specification, supporting common USB devices such as U disks, keyboards, mice, etc.; the 100M Ethernet interface adopts the RJ45 interface specification, supports 10 / 100Mbps adaptive transmission rate, can be connected to a local area network (LAN), wide area network (WAN) or other network devices, provides a stable data transmission channel, and is suitable for a variety of network environments and application scenarios.
[0041] In this embodiment, the application structure diagram is as follows: Figure 4 As shown, in this implementation case, there are two application environments: one with a fiber-optic intranet and one without a fiber-optic intranet. The fiber-optic information intranet directly connects the power distribution station's procurement, calculation and transmission integrated gateway to the information intranet through a 100M network port. The fiber-optic information intranet connects the power distribution station's procurement, calculation and transmission integrated gateway to the information intranet through an APN through the 5G / 4G full-network access module inside the procurement, calculation and transmission integrated gateway.
[0042] ① Collection: contact temperature sensors, switch cabinet temperature and humidity sensors, ambient temperature and humidity sensors, water level sensors, and smoke sensors are connected to the integrated collection, calculation and transmission gateway through the LoRa transparent transmission interface; switch cabinet partial discharge sensors are connected to the integrated collection, calculation and transmission gateway through the 2.4G / 470M dual-mode micro-power wireless interface; cameras are connected to the integrated collection, calculation and transmission gateway through 100M Ethernet; fans and water pumps are connected to the integrated collection, calculation and transmission gateway through the RS485 interface, fully covering the collection needs of various wired and wireless sensors in the scene;
[0043] ② Computing: In this case, the integrated computing and transmission gateway supports data analysis such as abnormal temperature rise of switch cabinet contacts, temperature threshold exceeding the limit, and temperature and humidity exceeding the limit in the cabinet; supports linkage with fans when the ambient temperature and humidity exceed the limit; supports linkage with water pumps when the ambient water level exceeds the limit; supports linkage with environmental smoke alarms; in terms of artificial intelligence, it supports abnormal reasoning and recognition of safety helmets, water accumulation, fire, personnel falls, and small animal intrusions. Normally, the gateway will extract camera video stream images every minute for reasoning and analysis. If it is identified that there may be an abnormality, it will perform high-frequency acquisition of video stream images at 5S / frame, and upload the reasoning results and the collected images to the IoT management platform, triggering the artificial intelligence platform's secondary reasoning and judgment. The recognition results are more timely and accurate, reducing the problem of missed and false alarms in traditional modes, and the function is more practical;
[0044] ③ In terms of transmission: In this case, the integrated calculation and transmission gateway uploads the data of 9 types of sensing terminals and the calculation and analysis results to the Internet of Things management platform through the MQTT protocol, including unstructured data such as partial discharge spectrograms and video pictures. The camera video data accesses the unified video platform through the B interface protocol of the gateway, meeting the construction requirements of the intelligent Internet of Things scenario in the distribution substation room scenario; in addition, through the Internet of Things management platform, remote management operations such as program debugging, upgrading, and device restart can be performed on the integrated calculation and transmission gateway, realizing remote operation and maintenance, and greatly reducing the device operation and maintenance workload.
[0045] To sum up, the method for collecting and transmitting unstructured data such as partial discharge spectrograms and video pictures proposed by the present utility model realizes full coverage of the scenario data collection requirements; the present utility model proposes to dynamically adjust the data collection frequency according to the recognition result and trigger the artificial intelligence platform to link the secondary precise judgment mechanism, realizing timely and accurate problem recognition and improving the practicality of the scenario artificial intelligence; the present utility model proposes to realize remote operation and maintenance management of the integrated calculation and transmission gateway based on the channel of the Internet of Things management platform, reducing the device on-site operation and maintenance workload and improving the usability of the device.
[0046] The present utility model is an integrated calculation and transmission gateway for distribution substation rooms. Aiming at the integrated calculation and transmission gateway for distribution substation rooms, it solves the information interaction between the data plane and the control plane and the collection and transmission of unstructured data such as videos and spectrograms, realizes practical edge computing and edge AI frameworks, and accelerates the artificial intelligence applications in substation room scenarios such as equipment defect identification and video violation identification. In use, on the basis of the existing gateway remote signal and remote measurement data collection and transmission, the present utility model realizes the collection and transmission of unstructured data such as partial discharge spectrograms and video pictures, comprehensively covering the collection and transmission requirements of substation room equipment and environmental data (structured and unstructured), and further improving the intelligent level of substation room monitoring; the present utility model realizes artificial intelligence applications on the edge side, different from the existing gateway artificial intelligence fixed-cycle reasoning and recognition mode, can dynamically adjust the data collection frequency according to the recognition result and link with the artificial intelligence platform for secondary precise judgment, with higher timeliness and accuracy of judgment. Compared with the centralized cloud reasoning mode, without sacrificing the timeliness and accuracy of recognition, it greatly reduces the gateway traffic consumption and improves the practicality of the gateway artificial intelligence; the present utility model realizes remote operation and maintenance functions such as program debugging, upgrading, and restart through the communication channel between the Internet of Things management platform and the gateway, with less on-site operation and maintenance workload of the device and convenient and fast operation and maintenance; the present utility model has the advantages of dynamically adjusting the data collection frequency, realizing secondary precise judgment through linkage, improving practicality, and reducing the on-site operation and maintenance workload of the device.
Claims
1. A power distribution station integrated data acquisition, calculation and transmission gateway, which includes a core board and is characterized by: The core board is respectively connected to a power module, at least one indicator light, a Console interface, at least one USB interface, at least one 100M Ethernet interface, at least one micro-power wireless interface, at least one LoRa transparent transmission interface, a 5G / 4G full network access module, at least one RS485 interface and at least one RS232 interface.
2. According to claim 1, a power distribution station integrated data acquisition, calculation and transmission gateway is characterized in that: The micro-power wireless adopts 2.4G / 470M dual-mode micro-power wireless, and 2.4G / 470M are each set up with 2 channels; the LoRa transparent transmission adopts 470M LoRa transparent transmission, and is set up with 2 channels.
3. According to claim 2, a distribution station room data acquisition, calculation and transmission integrated gateway is characterized in that: The USB interface is provided with 2 channels, the 100M Ethernet interface is provided with 4 channels, the RS485 is provided with 4 channels, and the RS232 is provided with 4 channels.
4. The integrated data acquisition, calculation and transmission gateway for power distribution station according to claim 1 is characterized by: The core board is composed of a CPU+NPU hybrid processor, memory, and storage.
5. The integrated data acquisition, calculation and transmission gateway for power distribution station according to claim 4 is characterized by: When collecting data, the core board communicates with the terminal device through the network port, wireless, RS232, and RS485 communication interfaces, and collects the sensed data to the gateway according to the protocol format of the terminal device.
6. The integrated data acquisition, calculation and transmission gateway for power distribution station according to claim 5 is characterized by: When calculating, the core board analyzes and processes the collected terminal device perception data, and analyzes and judges it through the artificial intelligence model. The NPU processor built into the core board can efficiently run the artificial intelligence model and accelerate the artificial intelligence reasoning.
7. The integrated data acquisition, calculation and transmission gateway for power distribution station according to claim 6 is characterized by: When transmitting, the core board sends the collected original perception data, calculation analysis and artificial intelligence reasoning results to the IoT management platform through the 5G / 4G full network access module and the APN node.
8. The integrated data acquisition, calculation and transmission gateway for power distribution station according to claim 1 is characterized by: The power module converts 220V AC power into DC power of multiple voltage levels such as 12V, 5V, and 3.3V to power all functional circuits of the gateway.
9. The integrated data acquisition, calculation and transmission gateway for power distribution station according to claim 1, characterized in that: The indicator lights are used to indicate the working status of the gateway device, including RS232, RS485 serial port working indicator lights, Ethernet working indicator lights, and system working indicator lights; the system working indicator lights are divided into operation, error, timing, 4G, power 1, and power 2.
10. The integrated data acquisition, calculation and transmission gateway for power distribution station according to claim 1, characterized in that: The Console interface and the 100M Ethernet interface both adopt RJ45 interface specifications, and the USB interface adopts USB Type A interface specifications.
Citation Information
Patent Citations
Novel artificial intelligence visual gateway for power distribution station house
CN214670197U
Edge intelligent gateway for power distribution station house
CN218570248U