A cloud-edge-device collaborative intelligent high-speed active control device

Abstract

This utility model relates to the field of highway traffic management technology, and in particular to a cloud-edge-device collaborative intelligent highway active control device, comprising: a cloud control host unit, including a central processing unit and an AI accelerator card; an edge computing node cluster, installed in cabinets along the highway, including a central processing unit and an AI accelerator card; and an end-side device cluster, including data acquisition devices and execution devices. The data acquisition devices include a traffic flow detection sensor, an image acquisition device, and an optical distance measurement device; the execution devices include an information dissemination device, a traffic signal control device, an audio prompt device, and a physical isolation device. The cloud control host unit is connected to the edge computing node cluster, and the end-side device cluster is also connected to the edge computing node cluster. This utility model significantly improves the response speed, reliability, and maintainability of highway active control through the built-in protocol gateway of the edge computing nodes and the end-to-end high-speed redundant communication link.

Description

Technical Field

[0001] This utility model relates to the field of intelligent highway traffic management technology, and in particular to a cloud-edge-device collaborative intelligent highway active control device. Background Technology

[0002] Existing highway active control projects typically involve multiple vendors supplying millimeter-wave radar, checkpoint cameras, LiDAR, variable message signs, traffic lights, and automatic barrier gates. Most of these devices use proprietary communication protocols or incompatible industry protocol versions, necessitating multi-level protocol conversion gateways for communication between the sensing and control links. This "information silo" model not only introduces additional parsing and encapsulation latency, weakening the real-time performance of congestion identification and handling, but also results in fragmented and slow-responding collaborative control of multiple devices under the same traffic event. Furthermore, equipment upgrades or expansions require the repeated development of adapters, leading to high integration and maintenance costs and severely restricting system scalability and cross-vendor collaboration.

[0003] To overcome the aforementioned technical deficiencies, the cloud-edge-device collaborative intelligent highway active management and control device provided by this utility model is committed to establishing a unified interface system and standardized protocol conversion mechanism at the hardware level. Through the protocol gateway built into the edge computing node and the end-to-end high-speed redundant communication link, it breaks down the protocol barriers between heterogeneous devices, realizes low-latency fusion of multi-source data and rapid linkage of execution devices, thereby significantly improving the response speed, reliability and maintainability of highway active management and control. Utility Model Content

[0004] In view of at least one of the above technical problems, the present invention provides a cloud-edge-device collaborative intelligent high-speed active management and control device, which adopts an improved support structure to achieve active yielding.

[0005] According to a first aspect of this utility model, a cloud-edge-device collaborative intelligent high-speed active management and control device is provided, comprising:

[0006] The cloud-controlled host unit includes a central processing unit and an AI accelerator card;

[0007] Edge computing node clusters are installed in cabinets along highways, including central processing units and AI accelerator cards;

[0008] The end-side equipment cluster includes data acquisition devices and execution devices. The data acquisition devices include a traffic flow detection sensing mechanism, an image acquisition mechanism, and an optical distance measurement mechanism. The execution devices include an information dissemination mechanism, a traffic signal control mechanism, an audio prompt mechanism, and a physical isolation mechanism.

[0009] The cloud control host group is connected to the edge computing node cluster, and the terminal device cluster is connected to the edge computing node cluster.

[0010] In some embodiments of this utility model, the cloud control host group includes a central processing unit with at least 64 cores and at least 8 AI accelerator cards; the edge computing node cluster includes a central processing unit with at least 16 cores and at least 4 AI accelerator cards.

[0011] In some embodiments of this utility model, the edge computing node cluster adopts a dual-machine hot standby structure and has a built-in protocol conversion gateway, which supports at least 12 industrial communication protocols.

[0012] In some embodiments of this utility model, the traffic flow detection sensing mechanism is a millimeter-wave radar with a detection range of at least 200 meters and a speed measurement accuracy of ±0.1 km / h, and two adjacent millimeter-wave radars are installed with a spacing of 200 meters; the image acquisition mechanism is set to 4K resolution, includes event detection hardware components, and two adjacent image acquisition mechanisms are installed with a spacing of 500 meters; the optical distance measurement mechanism is a lidar with a detection accuracy of ±2 cm and a field of view of not less than 120°.

[0013] In some embodiments of this utility model, the data acquisition device is connected to the edge computing node cluster via an industrial Ethernet network and is powered by PoE.

[0014] In some embodiments of this utility model, the information publishing mechanism is a variable message sign with a display area of ​​not less than 2m² and a resolution of not less than 1920×960; the luminous intensity of the traffic signal control mechanism is not less than 8000cd / m²; the output power of the audio prompt mechanism is not less than 50W and the sound pressure level is not less than 90dB; and the lifting and lowering time of the physical isolation mechanism is not greater than 1.5 seconds.

[0015] In some embodiments of this utility model, the terminal device cluster and the edge computing node cluster are connected by a Cat5e network cable with a distance of no more than 100 meters, or by a single-mode optical fiber with a distance of more than 100 meters between them; the edge computing nodes in the edge computing node cluster are interconnected by a single-mode optical fiber ring network with a node spacing of no more than 2 kilometers; the edge computing node cluster and the cloud control host group form a dual-channel redundant communication link through a 5G dedicated network channel and a GPON optical fiber channel.

[0016] In some embodiments of this utility model, the chassis of each edge computing node in the edge computing node cluster meets the IP65 protection level and the operating temperature range is -40℃ to 70℃.

[0017] In some embodiments of this utility model, the key devices in the edge computing node cluster and the end-side device cluster all adopt dual power input and are connected to an uninterruptible power supply (UPS). The UPS can provide continuous power supply for no less than 4 hours after the mains power is cut off.

[0018] In some embodiments of this utility model, an IPSec VPN encryption module is provided in the communication link of the device, each device in the end-side device cluster is configured with a two-way certificate authentication module, and an intrusion detection component is provided at the edge computing node cluster.

[0019] The beneficial effects of this utility model are as follows: This utility model proposes a three-layer hardware integrated communication closed loop of "cloud control host group - edge computing node cluster - end-side device cluster": the cloud control host group undertakes global strategy calculation and situational simulation with high-density central processing unit and AI acceleration card; the edge computing node cluster is deployed in cabinets along the line, maintaining real-time synchronization with the cloud through high-speed redundant links, and aggregating and integrating data from the end side nearby; various sensing devices and execution devices are directly connected to the edge nodes through a unified interface to realize a closed loop of data collection, decision distribution and action execution in the same domain. Compared with the traditional distributed solution that relies on field control boxes, this structure, on the one hand, relies on the elastic computing power of the cloud to achieve second-level congestion prediction and strategy recalculation, and on the other hand, uses the local broadcasting capability of the edge nodes to compress the end-to-end latency to hundreds of milliseconds. It can also realize the synchronous linkage of multiple devices in scenarios such as accident warning, traffic light switching, and barrier gate operation, thereby significantly improving response speed and traffic safety. Therefore, this utility model not only ensures low latency and high reliability of active control, but also supports plug-and-play devices and on-demand capacity expansion, truly overcoming the bottlenecks of isolated computing power, fragmented protocols, and difficulties in system expansion in the prior art. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this utility model or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this utility model. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the structure of the cloud-edge-device collaborative intelligent high-speed active control device. Detailed Implementation

[0022] The technical solutions of the present utility model will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present utility model. Obviously, the described embodiments are only some embodiments of the present utility model, and not all embodiments.

[0023] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0025] like Figure 1 The cloud-edge-device collaborative intelligent high-speed active management and control device shown includes:

[0026] The cloud-controlled host unit includes a central processing unit and an AI accelerator card;

[0027] Edge computing node clusters are installed in cabinets along highways, including central processing units and AI accelerator cards;

[0028] The end-side equipment cluster includes data acquisition devices and execution devices. The data acquisition devices include traffic flow detection sensing mechanisms, image acquisition mechanisms, and optical distance measurement mechanisms; the execution devices include information dissemination mechanisms, traffic signal control mechanisms, audio prompt mechanisms, and physical isolation mechanisms.

[0029] The cloud-based control host group is connected to the edge computing node cluster, and the terminal device cluster is connected to the edge computing node cluster.

[0030] This invention proposes a three-layer hardware-integrated communication closed loop: a cloud-based control host unit, an edge computing node cluster, and an end-side device cluster. The cloud-based control host unit, equipped with a high-density central processing unit and AI accelerator cards, handles global policy calculations and situational analysis. The edge computing node cluster is deployed in racks along the route, maintaining real-time synchronization with the cloud through high-speed redundant links and aggregating and integrating data from the edge. Various sensing and execution devices are directly connected to the edge nodes via a unified interface, achieving a closed loop of data acquisition, decision dissemination, and action execution within the same domain. Compared to traditional distributed solutions relying on field control boxes, this structure leverages the elastic computing power of the cloud to achieve second-level congestion prediction and policy recalculation. Furthermore, it utilizes the local broadcasting capabilities of the edge nodes to compress end-to-end latency to hundreds of milliseconds. It can also achieve synchronized linkage of multiple devices in scenarios such as accident warnings, traffic light switching, and barrier gate actions, thereby significantly improving response speed and traffic safety. Therefore, this utility model not only ensures low latency and high reliability of active control, but also supports plug-and-play devices and on-demand capacity expansion, truly overcoming the bottlenecks of isolated computing power, fragmented protocols, and difficulties in system expansion in the prior art.

[0031] In some embodiments of this invention, the cloud control host unit includes at least a 64-core central processing unit and at least 8 AI accelerator cards; the edge computing node cluster includes at least a 16-core central processing unit and at least 4 AI accelerator cards. This invention proposes a structured improvement to address computing power bottlenecks, aiming to enable the cloud to perform batch processing of massive amounts of data, global strategy deduction, and elastic scaling, while allowing the edge to perform multi-source data fusion, millisecond-level algorithm inference, and instantaneous command issuance locally. When massive amounts of high-definition video streams, laser point clouds, and millimeter-wave vector data simultaneously flood in, the cloud can still generate optimal traffic control strategies in real time, and the edge can broadcast a single command to the execution device after completing AI inference in milliseconds, with the overall closed-loop latency controlled within hundreds of milliseconds. Simultaneously, computing power redundancy ensures that equipment upgrades or algorithm iterations only require software-level modifications, eliminating the need for frequent hardware replacements. Compared with traditional stand-alone industrial control solutions, this dual-layer high-configuration architecture, which combines large-scale centralized computing power in the cloud with AI inference at the edge, not only solves the problems of isolated computing power and difficulty in expansion, but also significantly improves congestion prediction accuracy, accident response speed, and system lifecycle maintainability.

[0032] In some embodiments of this invention, the edge computing node cluster adopts a dual-machine hot standby structure and incorporates a built-in protocol conversion gateway that supports at least 12 industrial communication protocols. This invention employs a dual-machine hot standby structure within the edge computing node cluster: two servers mirror the operating status in real time, and if one fails, the other can seamlessly take over within seconds, ensuring that the computing power along the line never goes offline; simultaneously, the protocol conversion function is integrated into the chassis, with the built-in gateway simultaneously adapting to at least twelve industrial communication protocols such as MQTT, Modbus-TCP, and OPC UA, condensing the originally scattered external conversion boxes into board-level hardware. The direct effect of this design is that regardless of node hardware failure or differences in vendor protocols, data flow will not be interrupted or command issuance will be slowed down, and the edge-cloud closed loop will maintain millisecond-level response; furthermore, adding or replacing any sensing or execution device only requires selecting the corresponding protocol in the gateway for "plug and play," significantly reducing operational complexity and expansion costs. Compared with the traditional single-machine + multiple gateway approach, this dual-machine hot standby parallel redundancy, plus multi-protocol integrated edge node, not only improves system availability from immediate shutdown upon failure to uninterrupted operation during failure, but also lowers the barrier to entry for heterogeneous devices from redeveloping adapters to out-of-the-box use, significantly enhancing the real-time performance, reliability, and sustainable evolution capabilities of the highway active management and control system.

[0033] In some embodiments of this utility model, the traffic flow detection sensing mechanism is a millimeter-wave radar with a detection range of at least 200 meters and a speed measurement accuracy of ±0.1 km / h. The two adjacent millimeter-wave radars are installed with a spacing of 200 meters. The image acquisition mechanism is set to 4K resolution and includes event detection hardware components. The two adjacent image acquisition mechanisms are installed with a spacing of 500 meters. The optical distance measurement mechanism is a lidar with a detection accuracy of ±2 cm and a field of view of not less than 120°. This invention employs three types of collaborative sensing hardware on the end side: a traffic flow detection sensor mechanism using millimeter-wave radar with a detection range of at least 200 meters and a speed measurement accuracy of ±0.1 km / h, installed in a continuous chain at 200-meter intervals, enabling any sudden speed change in any lane to be captured within two seconds; an image acquisition mechanism using cameras with 4K resolution and embedded event detection hardware components, one every 500 meters, achieving pixel-level identification of lane obstruction, littering, or pedestrian intrusion; and an optical distance measurement mechanism equipped with a lidar with a ranging accuracy of ±2 cm and a field of view ≥120°, providing a high-density point cloud reference for joint calibration of the radar and camera. Through this high-precision, short-interval, and multimodal complementary deployment, the system shortens the sampling period from tens of seconds in traditional schemes to sub-seconds in the longitudinal direction, and eliminates blind spots between lanes in the lateral direction, thus enabling the cloud model to obtain a continuous, complete, and high-confidence data stream. The ultimate result is that the time margin for congestion prediction can be increased to 1-2 km in advance, and accident detection is triggered more than 30% in advance on average. The instruction is sent to the execution device with only millisecond-level link delay, realizing true preventive active control and lane-level fine guidance.

[0034] In some embodiments of this invention, the data acquisition device and the edge computing node cluster are connected via an industrial Ethernet network and powered by PoE. This invention unifies all data acquisition devices to an industrial Ethernet network supporting gigabit speeds and uses PoE to transmit data and DC power simultaneously on the same twisted-pair cable. This avoids the construction and maintenance burden caused by traditional dual-bundle laying of data and power cables. Furthermore, leveraging the port isolation, overcurrent protection, and remote restart functions of industrial switches, the power-on, power-off, and link status of each sensor can be monitored and remotely controlled in real time by the edge computing nodes. Therefore, when edge devices need hot-swapping, replacement, or firmware upgrades, the power supply to a single port can be directly cut off and reactivated in the background without affecting the normal operation of other devices on the same network segment. Compared to traditional solutions relying on distributed power modules and multi-standard interfaces, this integrated industrial Ethernet + PoE design significantly simplifies cabling and power distribution, reduces the number of failure points, and reduces the average repair time to one-third of the original. It also reserves standard plug-and-play interfaces for future additions of sensors, improving the overall scalability, reliability, and maintenance efficiency of the system.

[0035] In some embodiments of this utility model, the information dissemination mechanism is a variable message sign with a display area of ​​not less than 2m² and a resolution of not less than 1920×960; the luminous intensity of the traffic signal control mechanism is not less than 8000cd / m²; the output power of the audio prompt mechanism is not less than 50W and the sound pressure level is not less than 90dB; and the lifting and lowering time of the physical isolation mechanism is not greater than 1.5 seconds. In traditional highway control systems, the variable message sign area is too small and the pixel matrix is ​​sparse, causing drivers to have to drive close to it to read the content in nighttime, rainy, or foggy conditions; insufficient signal light intensity, limited loudspeaker power, and slow barrier gate operation often make accident warnings and diversion instructions "visible but not in time to avoid," thus delaying the response time. To address the lag issues in the response of these three execution chains—visual, audible, and actionable—this invention, through the coordinated configuration of high brightness, high power, and high-speed operation, enables the system to synchronously transmit warnings or diversion instructions generated by the cloud-edge system to the driver within hundreds of milliseconds in a way that is "visible, audible, and effective." Compared to traditional low-end execution devices, this can advance the accident scene handling window by at least 3–5 seconds, significantly improving road safety and traffic efficiency in adverse weather and emergencies.

[0036] In some embodiments of this utility model, the end-side device cluster and the edge computing node cluster are connected by a Cat5e network cable with a distance not exceeding 100 meters, or by a single-mode optical fiber with a distance greater than 100 meters between them; the edge computing nodes in the edge computing node cluster are interconnected by a single-mode optical fiber ring network with a node spacing not exceeding 2 kilometers; the edge computing node cluster and the cloud control host group form a dual-channel redundant communication link through a 5G dedicated network channel and a GPON optical fiber channel. In existing highway monitoring systems, end-side sensing and execution devices are often randomly connected to 100 Mbps copper cables or single-link optical fibers, and attenuation and packet loss occur over slightly longer distances. The edge cabinets often use a tree-star topology, and once the backbone is damaged, the entire link is interrupted. At the same time, there is usually only one optical fiber or public network APN from the roadside to the cloud. In the event of construction line cuts, base station failures, or sudden extreme weather, cloud-edge policy synchronization and video backhaul are often forced to be interrupted. Based on these pain points, this utility model improves link availability from a single point of failure (i.e., network-wide degradation) to maintaining full service online even if any route fails, significantly enhancing the real-time performance, reliability, and disaster resistance of congestion warning and incident handling.

[0037] In some embodiments of this invention, the chassis of each edge computing node in the edge computing node cluster meets the IP65 protection rating and has an operating temperature range of -40℃ to 70℃. Cabinets along highways are typically exposed to extreme outdoor environments such as wind, rain, dust, snowmelt salt spray, and intense sunlight. Traditional industrial control computers are often installed in semi-enclosed control boxes using ordinary IT chassis or weakly protected aluminum shells with IP30-IP40 ratings. When encountering heavy rain, sand intrusion, or extreme heat or cold, they are prone to motherboard short circuits, fan jamming, and component thermal failure, causing edge computing power to stop abruptly and even requiring frequent on-road disassembly and maintenance. This invention, through the above-mentioned design, ensures that even in the event of typhoons, heavy rain, desert sandstorms, or extreme cold in the north, the interior of the chassis remains dry and clean, and the processor and AI accelerator card do not throttle or disconnect. The equipment lifecycle is extended from the traditional 3-5 years to over 8-10 years, significantly reducing the on-site failure rate and simultaneously lowering inspection and spare parts costs, providing higher reliability and availability for the highway active management system than ever before.

[0038] In some embodiments of this invention, key devices in the edge computing node cluster and the end-side device cluster all employ dual power inputs and are connected to an uninterruptible power supply (UPS). The UPS can provide continuous power for at least 4 hours after a mains power outage. Relying on a single power source makes the system susceptible to power outages or interruptions, especially during severe weather or construction. A momentary mains power failure will trigger a complete power outage, restart, and database self-check, causing an instantaneous interruption of the front-end monitoring and back-end control links, rendering accident warnings and congestion induction ineffective. This invention introduces a dual power input + 4-hour UPS power supply architecture design for the edge computing node cluster and key end-side devices: the two mains power sources act as hot backups for each other in real time via an ATS, and when one power fails, the other can carry the entire load within milliseconds, while the high-capacity UPS provides the entire node with at least four hours of continuous DC output, sufficient to cover the time window for emergency repairs or the arrival of emergency generators after a large-scale power outage. Therefore, even if the external power distribution cabinet loses power completely, the AI ​​inference, protocol gateway, and various radars, cameras, and information boards on the edge server can still maintain full-load operation, and the cloud policy distribution and data backhaul are unaffected; all control logic can be seamlessly continued without restarting, avoiding the link disconnection, data loss, and equipment lifespan reduction caused by sudden power outages in traditional single-power-supply architectures, bringing higher continuous availability and operational safety margins to the intelligent high-speed proactive management and control system than ever before.

[0039] In some embodiments of this invention, the communication link of the device is equipped with an IPSec VPN encryption module, each device in the end-side device cluster is configured with a two-way certificate authentication module, and an intrusion detection component is set at the edge computing node cluster. Deploying the intrusion detection component at the edge computing node cluster performs deep packet inspection and abnormal behavior analysis on north-south and east-west traffic. Once brute-force attacks, flooding scans, or protocol malformations are detected, the port can be blocked locally and the event can be synchronously reported to the cloud. Compared with traditional approaches that rely solely on VLAN isolation or application-layer passwords, this "three-in-one" hardware security system moves link protection, identity authentication, and threat awareness to the network edge, achieving 24 / 7, millisecond-level proactive defense. Even when encountering disconnection, spoofing, or zero-day vulnerability attacks, the system can maintain the continuity of policy distribution and data feedback, significantly improving the security reliability and operational confidence of intelligent high-speed proactive management.

[0040] Those skilled in the art should understand that this utility model is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this utility model. Various changes and modifications can be made to this utility model without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed utility model. The scope of protection of this utility model is defined by the appended claims and their equivalents.

Claims

1. A cloud-edge-device collaborative intelligent high-speed active control device, characterized in that, include: The cloud-controlled host unit includes a central processing unit and an AI accelerator card; Edge computing node clusters are installed in cabinets along highways, including central processing units and AI accelerator cards; The end-side equipment cluster includes data acquisition devices and execution devices. The data acquisition devices include a traffic flow detection sensing mechanism, an image acquisition mechanism, and an optical distance measurement mechanism. The execution devices include an information dissemination mechanism, a traffic signal control mechanism, an audio prompt mechanism, and a physical isolation mechanism. The cloud control host group is connected to the edge computing node cluster, and the terminal device cluster is connected to the edge computing node cluster.

2. The cloud-edge-device collaborative intelligent high-speed active control device according to claim 1, characterized in that, The cloud control host unit includes a central processing unit with at least 64 cores and at least 8 AI accelerator cards; the edge computing node cluster includes a central processing unit with at least 16 cores and at least 4 AI accelerator cards.

3. The cloud-edge-device collaborative intelligent high-speed active control device according to claim 1, characterized in that, The edge computing node cluster adopts a dual-machine hot standby structure and has a built-in protocol conversion gateway, which supports at least 12 industrial communication protocols.

4. The cloud-edge-device collaborative intelligent high-speed active control device according to claim 1, characterized in that, The traffic flow detection sensing mechanism is a millimeter-wave radar with a detection range of at least 200 meters and a speed measurement accuracy of ±0.1 km / h. The millimeter-wave radars are installed with a spacing of 200 meters between adjacent units. The image acquisition mechanism is set to 4K resolution and includes event detection hardware components. The image acquisition mechanisms are installed with a spacing of 500 meters between adjacent units. The optical distance measurement mechanism is a lidar with a detection accuracy of ±2 cm and a field of view of not less than 120°.

5. The cloud-edge-device collaborative intelligent high-speed active control device according to claim 1, characterized in that, The data acquisition device is connected to the edge computing node cluster via an industrial Ethernet network and is powered by PoE.

6. The cloud-edge-device collaborative intelligent high-speed active management and control device according to claim 1, characterized in that, The information dissemination mechanism is a variable message sign with a display area of ​​not less than 2m² and a resolution of not less than 1920×960; the luminous intensity of the traffic signal control mechanism is not less than 8000cd / m²; the output power of the audio prompt mechanism is not less than 50W and the sound pressure level is not less than 90dB; the lifting and lowering time of the physical isolation mechanism is not greater than 1.5 seconds.

7. The cloud-edge-device collaborative intelligent high-speed active management and control device according to claim 1, characterized in that, The terminal device cluster and the edge computing node cluster are connected by a Cat5e network cable with a distance not exceeding 100 meters, or by a single-mode optical fiber with a distance greater than 100 meters between them; the edge computing nodes in the edge computing node cluster are interconnected by a single-mode optical fiber ring network with a node spacing not exceeding 2 kilometers; the edge computing node cluster and the cloud control host group form a dual-channel redundant communication link through a 5G dedicated network channel and a GPON optical fiber channel.

8. The cloud-edge-device collaborative intelligent high-speed active control device according to claim 1, characterized in that, The chassis of each edge computing node in the edge computing node cluster meets the IP65 protection level and has an operating temperature range of -40℃ to 70℃.

9. The cloud-edge-device collaborative intelligent high-speed active control device according to claim 1, characterized in that, The key devices in the edge computing node cluster and the end-side device cluster all adopt dual power input and are connected to an uninterruptible power supply (UPS). The UPS can provide continuous power supply for no less than 4 hours after the mains power is cut off.

10. The cloud-edge-device collaborative intelligent high-speed active management and control device according to claim 1, characterized in that, The device is equipped with an IPSec VPN encryption module in its communication link, each device in the end-side device cluster is equipped with a two-way certificate authentication module, and an intrusion detection component is set at the edge computing node cluster.

Images