Edge acquisition computing gateway and multi-stage water level edge intelligent monitoring and control method
The edge acquisition and computing gateway enables refined monitoring and control of five levels of water level status. By combining local linkage and remote control, it solves the problems of intelligence and localization of existing water level monitoring devices, improves the accuracy and stability of water level monitoring, and is suitable for unattended computer room environments.
Patent Information
- Application Number
- CN202611289019.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-21
- Publication Date
- 2026-11-10
AI Technical Summary
Existing water level monitoring devices cannot achieve refined monitoring, local intelligent decision-making, and remote centralized management. They cannot adapt to the needs of digital, networked, and intelligent environmental monitoring systems, and are prone to malfunctions, delayed responses, or control failures, making it difficult to meet the requirements of domestic production and independent control.
Design an edge acquisition and computing gateway that employs a microcontroller unit, a multi-level water level sensing interface, a local linkage execution unit, a communication module, and a power management module. It identifies five levels of water level status through optocoupler-isolated digital input signals and achieves refined water level monitoring and control by combining local linkage and remote control modes.
It enables refined monitoring and local closed-loop automatic control of water levels in the data center, supports intelligent switching between remote and local modes, has a backup capability for network interruption, improves the accuracy, stability and security level of water level monitoring, and is suitable for unattended data center scenarios.
Smart Images

Figure CN123067265A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of edge computing and data center environmental monitoring, and in particular to an edge acquisition and computing gateway and a multi-level water level edge intelligent monitoring and control method. Background Technology
[0002] In critical infrastructure such as launch rooms and data centers, environmental monitoring systems are an important support for ensuring the safe and stable operation of equipment, and water level monitoring and control are key to preventing equipment short circuits and downtime caused by water leakage and water accumulation.
[0003] Existing water level monitoring devices mostly employ pure hardware circuitry, with control logic comprised of hard-wired analog and digital circuits. They lack a main control chip and programmability, and can only achieve a binary judgment of water presence / absence through probe on / off states. These devices lack data communication interfaces, cannot upload status updates or receive remote commands, and their operational status is invisible, limiting them to local open-loop operation. Furthermore, their control logic is simplistic, lacking intelligent judgment and safety redundancy mechanisms, making them prone to malfunctions, delayed responses, or control failures. Moreover, many key components are imported, making it difficult to meet the requirements of domestic production and independent controllability.
[0004] Existing technologies are unable to achieve precise water level monitoring, local intelligent decision-making, and remote centralized management, and are no longer suitable for the needs of digital, networked, and intelligent environmental monitoring systems. Summary of the Invention
[0005] In view of this, this application proposes an edge acquisition computing gateway, including a microcontroller unit, a multi-level water level sensing interface, a local linkage execution unit, a communication module and a power management module; The multi-level water level sensing interface is electrically connected to the microcontroller unit to collect multiple optically isolated digital input signals and input them to the microcontroller unit. The microcontroller unit is used to receive multiple optocoupler-isolated digital input signals, identify the five water level states of empty, low, normal, high and full according to the signal combination state code, and issue linkage control commands based on the five water level states. The local linkage execution unit is electrically connected to the microcontroller unit and is used to receive linkage control commands issued by the microcontroller unit and execute local linkage control outputs. The communication module is electrically connected to the microcontroller unit and is used to send and receive remote control commands and report water level data and equipment operating status to the monitoring platform. The power management module provides power to the microcontroller unit, the multi-level water level sensing interface, the local linkage execution unit, and the communication module.
[0006] In one possible implementation, the microcontroller unit incorporates a five-level water level state machine and a local linkage control strategy. The five-level water level status machine identifies the five water level states (empty, low, normal, high, and full) based on the combined status codes of the five optocoupler-isolated digital input signals input from the multi-level water level sensor interface. The water level status is confirmed to be valid only when the water level status changes and the sampling signals are consistent for two consecutive times. The local linkage control strategy implements differentiated control based on different water level conditions: When the water level is empty, the local control unit starts the water replenishment pump to replenish water; when the water level is low, the water replenishment pump continues to run and the buzzer is activated with a low-frequency intermittent alarm; when the water level is normal, the water replenishment pump and drain valve are closed, and there is no alarm output; when the water level is high, the local control unit activates the buzzer with a fast alarm and starts the drain valve; when the water level is full, the local control unit activates the buzzer with a high-frequency continuous alarm and cuts off the power supply to the water pump; when the water level drops from high or full to normal, the water pump and drain valve are automatically shut off after a preset delay, and the alarm is cleared.
[0007] In one possible implementation, the microcontroller unit also incorporates logic for switching between local and remote control modes; The edge acquisition and computing gateway runs in local automatic mode by default. The microcontroller unit independently executes local linkage control strategies based on the five-level water level status, without relying on the network. When the communication module receives a valid remote control command, it switches to remote control mode and pauses the local linkage control strategy, only responding to the remote control command to control the local linkage execution unit. When the remote control command stream is interrupted for more than the preset timeout threshold, or when the communication module receives a command to switch back to local mode, the local automatic mode will be automatically restored.
[0008] In one possible implementation, the communication module uses a lightweight UDP communication protocol to interact with the remote monitoring platform; The lightweight UDP communication protocol uses a 7-byte fixed-length hexadecimal frame structure, which includes, in sequence, the data center identifier byte, the machine number byte, the remote / automatic mode code byte, the query / response code byte, the input data byte, the output data byte, and the checksum byte. The lightweight UDP communication protocol can realize four types of interactive functions: periodically and proactively reporting input and output status and working mode; issuing remote control commands for controlling solid-state relay outputs; issuing local mode switching commands; and responding to remote status queries and transmitting real-time status back.
[0009] In one possible implementation, the multi-level water level sensing interface uses an AC pulse combined with a zero-bias detection scheme to eliminate electrochemical corrosion of the electrodes in a pure water environment. The microcontroller unit has a built-in timer. The timer outputs a square wave signal, which is then sent to the water level probe electrode via a current-limiting resistor to complete the excitation. The signal receiving side is equipped with a non-polar signal conditioning circuit consisting of a full-bridge rectifier circuit, an RC filter circuit, and a high-speed optocoupler, which can eliminate the DC bias voltage at both ends of the probe electrodes. The water level probe electrode is made of 316L stainless steel, the probe lead is covered with a fluororubber insulation layer and is potted with epoxy resin, achieving an overall IP68 sealing protection level.
[0010] In one possible implementation, the communication module is a dual-mode communication module, integrating two electrically isolated RS485 interfaces and a UDP Ethernet interface, supporting parallel communication between local bus and remote IP network. The power management module is a dual-redundant power management module that supports DC24V wide voltage input, is equipped with overvoltage and reverse connection protection circuits, and outputs a 12V power supply branch to power peripherals.
[0011] This application also provides a multi-level intelligent edge monitoring and control method for water level in computer room environments. Water level probes are installed at multiple locations with different water levels. Each probe is connected to a single optically isolated digital channel within the multi-level water level sensing interface of the edge acquisition and computing gateway. The signal is transmitted to the gateway microcontroller unit via the multi-level water level sensing interface. The edge gateway performs water level monitoring and control using the following steps: It receives multiple optically isolated digital input signals and identifies five water level states—empty, low, normal, high, and full—based on the combined state codes of the multiple input signals. The corresponding local linkage control output is executed based on the identified five water level states; and each water level state corresponds to a different linkage control output action; the linkage control output is executed locally by the edge acquisition and computing gateway by default; When a valid remote control command is received, the system switches to remote control mode and pauses local linkage control. When the remote control command is interrupted for more than the preset timeout threshold, or when a command to switch back to local is received, the system will automatically revert to local automatic control mode.
[0012] In one possible implementation, receiving multiple optocoupler-isolated digital input signals and identifying five water level states (empty, low, normal, high, and full) based on the combined state codes of the multiple input signals includes the following steps: Acquire 5 channels of optocoupler-isolated digital input signals; The five signals are encoded according to a preset combination rule; Based on the coding results, the water level status is determined to be one of five levels: empty, low, normal, high, and full.
[0013] In one possible implementation, the five water level states—empty, low, normal, high, and full—are determined based on the encoding result, including: Perform consistency judgment on the encoding results of multiple consecutive samples; When two consecutive sampling results are consistent, the current level 5 water level status is confirmed to be valid.
[0014] In one possible implementation, executing the corresponding local linkage control output based on the identified five-level water level status includes the following steps: Match the preset linkage strategy according to the current water level status; Control the solid-state relay output according to the preset linkage strategy; The alarm devices are driven to execute multi-level alarms according to the linkage strategy.
[0015] The beneficial effects of this invention are: This edge data acquisition and computing gateway acquires multiple optically isolated digital water level signals through a multi-level water level sensor interface and transmits them to the microcontroller unit. The microcontroller identifies five water level levels (empty, low, normal, high, and full) based on the multi-signal combination encoding and outputs linkage commands. The local linkage execution unit completes the local control action. The communication module realizes command interaction and operating status reporting with the remote monitoring platform, and the power management module provides unified power supply for all modules. During operation, the front-end sensor interface first performs optical anti-interference isolation acquisition on the multi-channel water level probe signals. The microcontroller unit realizes fine-grained water level classification judgment through multi-signal encoding and directly generates data at the edge. The system drives peripheral devices to execute control commands and enables remote operation and maintenance interaction via communication links. The entire unit is stably powered by a dedicated power supply module. It breaks through the limitations of traditional water level monitoring, which can only make binary judgments. It achieves autonomous closed-loop management of water levels through edge local computing. It can still perform basic protection actions normally when the network is abnormal. It can also meet the needs of remote status viewing and control through communication modules. The optical isolation design reduces the impact of electromagnetic interference in the computer room on sampling accuracy. The modular power supply ensures the operational stability of each link. The system takes into account water level monitoring accuracy, local safety protection capabilities and remote operation and maintenance convenience, making it suitable for unattended water level monitoring scenarios in computer rooms.
[0016] Other features and aspects of this application will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description
[0017] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this application together with the specification and serve to explain the principles of this application.
[0018] Figure 1 This diagram illustrates a block diagram of an edge acquisition and computing gateway according to an embodiment of this application. Figure 2 A flowchart illustrating the multi-level intelligent monitoring and control method for water level edges according to an embodiment of this application is shown. Figure 3 This application illustrates a system architecture diagram of an edge acquisition and computing gateway according to an embodiment of the present application. Figure 4 This diagram illustrates the embedded system architecture of an edge acquisition and computing gateway according to an embodiment of this application. Detailed Implementation Various exemplary embodiments, features, and aspects of this application will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.
[0019] It should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application or to simplify the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.
[0020] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.
[0021] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.
[0022] Furthermore, to better illustrate this application, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that this application can be implemented without certain specific details. In some instances, methods, means, components, and circuits well-known to those skilled in the art have not been described in detail in order to highlight the main points of this application.
[0023] The edge acquisition computing gateway and multi-level water level edge intelligent monitoring and control method of this application is an edge intelligent control scheme with local intelligent decision-making and remote security management capabilities. It is applied in the fields of edge computing and data center environmental monitoring, and plays a role in realizing refined monitoring of data center water level, local closed-loop automatic control, intelligent switching between remote / local modes, and network interruption security backup.
[0024] The specific implementation of this application is referred to Figures 1-4 .like Figure 1As shown, this is a specific embodiment of an edge acquisition and computing gateway according to this application. The edge acquisition and computing gateway 100 includes a microcontroller unit 110, a multi-level water level sensing interface 120, a local linkage execution unit 130, a communication module 140, and a power management module 150.
[0025] Specifically, the edge acquisition and computing gateway 100 uses a microcontroller unit 110 as its computing core, a multi-level water level sensor interface 120 as a front-end signal acquisition channel, a local linkage execution unit 130 as a field action output terminal, a communication module 140 as a cloud data interaction channel, and a power management module 150 as the overall power supply hub. These five modules are electrically interconnected via onboard wiring and integrated within the same industrial housing, making it suitable for embedded installation in data center racks. Adopting a modular architecture for acquisition, computing, execution, communication, and power supply, the functional boundaries of each module are clearly defined, facilitating production assembly, fault identification and repair. The overall structure is adaptable to the limited installation space of data centers, meeting the stable deployment requirements of industrial scenarios.
[0026] The multi-level water level sensing interface 120 is electrically connected to the microcontroller unit 110 and is used to collect multiple optically isolated digital input signals and input them to the microcontroller unit 110.
[0027] Specifically, the multi-level water level sensor interface 120 leads out multiple independent detection channels. Each channel is equipped with a high-speed optocoupler. The level signal from the water level probe first undergoes strong and weak current isolation through the optocoupler to suppress electromagnetic interference from the inverter and power supply lines in the computer room. Then, it is converted into a standard digital level signal and transmitted to the microcontroller unit 110 through GPIO pins. Hardware-wise, electrical isolation between the acquisition circuit and the main control circuit is achieved. Optical coupling isolation blocks strong electrical interference and surge impacts in the field, preventing interference signals from causing misjudgments or even hardware damage to the main control chip. Multiple independent channels can be connected to water level probes of different heights, providing a signal foundation for subsequent multi-signal combination encoding to determine the water level level.
[0028] The microcontroller unit 110 is used to receive multiple optocoupler-isolated digital input signals, identify the five water level states of empty, low, normal, high and full according to the signal combination state code, and issue linkage control commands based on the five water level states.
[0029] Specifically, the microcontroller periodically polls and reads the levels of various digital signals, converting the combinations of high and low levels into binary codes. It then matches these codes to a pre-programmed encoding mapping table to determine five water level levels: empty, low, normal, high, and full. Simultaneously, debouncing logic is incorporated to confirm a valid water level only after two consecutive identical samples. Based on the currently matched water level, the built-in linkage strategy is invoked to generate corresponding control commands, which are then sent to the local linkage execution unit 130. This multi-channel signal encoding achieves five-level fine-grained water level differentiation, overcoming the limitations of traditional single-probe systems that can only determine the presence or absence of water using binary methods. The sampling debouncing mechanism filters out instantaneous interference, significantly reducing the probability of water level misjudgment and enabling precise output of graded control commands based on water level differences.
[0030] The local linkage execution unit 130 is electrically connected to the microcontroller unit 110 and is used to receive linkage control commands issued by the microcontroller unit 110 and execute local linkage control output.
[0031] Specifically, the local linkage execution unit 130 uses MOS-type solid-state relays as output devices. It receives switching commands from the main controller and controls the start and stop of peripheral devices such as the water supply pump, drainage valve, and buzzer alarm. The water supply, drainage, and graded alarm actions corresponding to different water levels are all directly driven by this unit, without the need for cloud-based computation. This enables local execution of control actions at the edge, and water level safety management can still be completed independently even when the network is disconnected. The contactless solid-state relay switches have a long lifespan and fast response, making them suitable for long-term unattended operation and frequent start-stop conditions in computer rooms.
[0032] The communication module 140 is electrically connected to the microcontroller unit 110 and is used to send and receive remote control commands and report water level data and equipment operating status to the monitoring platform.
[0033] Specifically, the communication module 140 uses a lightweight UDP protocol to establish data interaction with the remote monitoring platform, periodically reporting the current water level, relay output status, and local / remote operating mode. It also receives remote switch control and mode switching commands from the platform; these commands are verified before being transmitted to the microcontroller for processing. Maintenance personnel can remotely monitor the water level and equipment status in the data center without on-site inspections. It supports remote takeover control, accommodating both routine automated management and remote maintenance needs, thus improving the convenience of data center operations and maintenance.
[0034] The power management module 150 provides power to the microcontroller unit 110, the multi-level water level sensing interface 120, the local linkage execution unit 130, and the communication module 140.
[0035] Specifically, the power management module 150 connects to the DC24V industrial power supply in the computer room. Internally, it performs voltage conversion, filtering, and regulation, providing matching and stable operating voltages for the main control, signal acquisition circuit, relay output circuit, and communication circuit. It is equipped with overvoltage and reverse connection protection circuits, and a high-end version can use dual redundant input power supplies. The power management module 150 provides a stable and clean power supply for all modules of the entire unit, avoiding the risk of hardware burnout caused by fluctuations in the computer room power supply or reverse wiring. The redundant power supply design allows for seamless switching in the event of a single power supply failure, ensuring uninterrupted water level monitoring and control and improving equipment operational reliability.
[0036] Furthermore, such as Figure 1 As shown, the edge acquisition and computing gateway 100 uses an embedded microcontroller unit 110 as its core computing hub. The multi-level water level sensing interface 120 continuously acquires multiple digital water level signals processed by optical coupling and transmits the acquired signals to the microcontroller unit 110. The microcontroller unit 110 performs combined state encoding operations on the multiple input signals, identifying five water level states: empty, low, normal, high, and full. Based on the matched water level level, it generates corresponding linkage control commands and sends them to the local linkage execution unit 130, which then performs local control outputs for peripherals such as water pumps and alarm devices. Simultaneously, the communication module 140 establishes data interaction with the remote monitoring platform. On the one hand, it periodically reports real-time water levels and equipment operating conditions; on the other hand, it receives remote control commands from the platform and forwards them to the microcontroller unit 110 for processing, realizing remote management and mode switching. All functional modules of the machine are powered by the 150 power management module, which provides stable voltage to ensure the stable operation of the acquisition, processing, execution and communication processes. The entire process of acquisition and identification, local control and remote interaction is continuously cycled to complete the intelligent monitoring and control of the water level in the computer room.
[0037] In one possible implementation, the microcontroller unit 110 integrates a five-level water level state machine and a local linkage control strategy. The five-level water level state machine identifies five water level states—empty, low, normal, high, and full—based on the combined state codes of the five optocoupler-isolated digital input signals input from the multi-level water level sensor interface 120. The water level state is confirmed as valid only when two consecutive sampled signals match. The local linkage control strategy performs differentiated control according to different water level states: when the water level is empty, the local linkage execution unit 130 is controlled to start the water replenishment pump. When the water level is low, the water replenishment pump is maintained and a buzzer is activated for low-frequency intermittent alarm. When the water level is normal, the water replenishment pump and drain valve are closed, and no alarm output is detected. When the water level is high, the local linkage execution unit 130 is controlled to activate the buzzer for a rapid alarm and start the drain valve. When the water level is full, the local linkage execution unit 130 is controlled to activate the buzzer for a high-frequency continuous alarm and cut off the power supply to the water pump. When the water level drops from high or full to normal, the water pump and drain valve will automatically shut off and the alarm will be cleared after a preset delay.
[0038] Furthermore, by integrating a five-level water level state machine and a hierarchical local linkage control strategy into the microcontroller unit 110, and relying on the combination encoding of 5-channel signals to achieve refined water level classification, coupled with a verification mechanism that ensures consistent sampling twice, the system filters out water level misjudgments caused by electromagnetic interference and instantaneous signal fluctuations in the computer room. Differentiated handling logics are set for the five water levels of empty, low, normal, high, and full, including water replenishment, hierarchical alarms, drainage, and pump shutdown protection. After the water level falls back to normal after being too high, a delayed reset mechanism is added to avoid frequent start-stop jitter of water pumps and alarm devices when the water level is critically fluctuating. It can actively replenish water and provide low-level warnings during water shortage stages, and promptly drain water and cut off the water inlet to achieve overflow protection during water level over-limit stages. It can also reduce device wear caused by repeated actions of peripheral devices. The entire process relies on the gateway edge side to autonomously complete state judgment and control output, forming a closed-loop water level safety management system without relying on the cloud network, thus improving the accuracy, stability, and safety protection level of water level monitoring and control in the computer room water tank.
[0039] In an alternative implementation, the water level identification method can use an analog water level sensor (such as a float-type variable resistance type) in conjunction with ADC sampling, and achieve multi-level judgment through threshold comparison, but its anti-interference capability is weaker than that of digital isolated input.
[0040] In one possible implementation, the microcontroller unit 110 also incorporates local and remote control mode switching logic. The edge acquisition and computing gateway 100 operates in local automatic mode by default, and the microcontroller unit 110 independently executes the local linkage control strategy based on the five-level water level status, without relying on the network. When the communication module 140 receives a valid remote control command, it switches to remote control mode and suspends the local linkage control strategy, only responding to remote control commands to control the local linkage execution unit 130. When the remote control command stream is interrupted for more than a preset timeout threshold, or when the communication module 140 receives a command to switch back to local mode, it automatically resumes local automatic mode.
[0041] Furthermore, by setting up automatic switching logic for local and remote dual control modes, the gateway defaults to relying on the local automatic mode to autonomously execute linkage strategies based on the water level status. Even if the network is interrupted abnormally, it can independently complete water level safety management and avoid monitoring failure after network outage. Upon receiving a valid remote command that has passed verification, it can promptly take over the device and suspend the local automatic logic to prevent malfunctions caused by command conflicts between local policies and remote operations. At the same time, it adds two automatic fallback recovery mechanisms: remote command timeout and disconnection, and receiving a switchback command. This balances the operational convenience of remote debugging and management by maintenance personnel with the security of local automatic protection in unattended scenarios, solving the problem that traditional water level devices can only be controlled locally or remotely, and lack a safety fallback when switching modes.
[0042] In an alternative implementation, the preset timeout threshold for remote control commands can be flexibly set to different durations such as 10 seconds, 30 seconds, and 60 seconds, depending on the network environment and security requirements of the computer room, to adapt to the security backup requirements of different scenarios.
[0043] In one possible implementation, the communication module 140 interacts with the remote monitoring platform using a lightweight UDP communication protocol. The lightweight UDP communication protocol employs a 7-byte fixed-length hexadecimal frame structure, which sequentially includes bytes for the equipment room identifier, machine number, remote / automatic mode code, query / response code, input data, output data, and checksum. The lightweight UDP communication protocol can implement four types of interactive functions: periodically and proactively reporting input / output status and operating mode; issuing remote control commands for controlling solid-state relay outputs; issuing local mode switching commands; and responding to remote status queries and transmitting real-time status back.
[0044] Furthermore, a lightweight UDP protocol with a fixed 7-byte frame length is used for data interaction. The simplified frame structure reduces data transmission overhead and packet loss probability in weak network environments in the data center. The data center identifier, device number, working mode, data information, and check bit are set within the frame. This can distinguish the device identity of multiple data centers and gateways, avoid missent of commands, and reduce data transmission errors caused by electromagnetic interference through the check bit. At the same time, it fully covers four core interaction requirements: periodic status reporting, remote relay control, mode switching, and status query. While meeting all communication functions of data center water level operation and maintenance, it achieves lightweight communication process and reliable data transmission, adapting to the low-power and high-efficiency data transmission requirements of edge gateways.
[0045] In an alternative implementation, a fixed-length lightweight UDP frame can be used, with fixed frame structures of different lengths such as 10 bytes and 16 bytes. The frame length can be flexibly adjusted according to the amount of communication data and transmission requirements to adapt to different network environments and communication needs.
[0046] In one possible implementation, the multi-level water level sensing interface 120 employs an AC pulse combined with a zero-bias detection scheme to eliminate electrochemical corrosion of the electrodes in a pure water environment. The microcontroller unit 110 has a built-in timer that outputs a square wave signal, which is then transmitted to the water level probe electrode via a current-limiting resistor to complete the excitation. The signal receiving side features a non-polarity signal conditioning circuit consisting of a full-bridge rectifier circuit, an RC filter circuit, and a high-speed optocoupler, which eliminates the DC bias voltage across the probe electrode. The water level probe electrode is made of 316L stainless steel, and the probe leads are covered with a fluororubber insulation layer and encapsulated with epoxy resin, achieving an overall IP68 sealing protection rating.
[0047] Furthermore, an AC pulse excitation combined with a zero-bias detection scheme is adopted to fundamentally eliminate the DC electrochemical corrosion problem of the water level probe electrodes under pure water conditions, extending the probe's service life. The receiver is equipped with a conditioning circuit consisting of a full-bridge rectifier, RC filter, and high-speed optocoupler, which can achieve both non-polarity signal adaptation and isolation from strong electromagnetic interference in the equipment room. The probe uses 316L stainless steel electrodes, combined with fluororubber insulation and epoxy resin potting to achieve IP68 sealing protection, enabling stable operation in humid and waterlogged environments for extended periods. This solves the pain points of traditional DC water level probes, such as easy corrosion, poor sealing and waterproofing, and unstable signal sampling under equipment room interference, significantly improving the stability of front-end water level signal acquisition and device durability.
[0048] In one possible implementation, the communication module 140 is a dual-mode communication module, integrating two electrically isolated RS485 interfaces and a UDP Ethernet interface, supporting parallel communication via local bus and remote IP network. The power management module 150 is a dual-redundant power management module, supporting a wide DC24V input voltage, equipped with overvoltage and reverse connection protection circuits, and outputting a 12V power supply branch to power peripherals.
[0049] Furthermore, the communication side integrates electrically isolated RS485 and Ethernet dual communication links, which can simultaneously connect to the field local bus and remote IP monitoring network. Even if one communication line fails, the other link can still ensure data exchange. The electrical isolation design also blocks surge interference on the bus side. The power supply side uses a dual redundant power supply with a wide DC 24V input, coupled with overvoltage and reverse connection protection, and an independent 12V power supply branch for external devices. This not only withstands hardware damage caused by fluctuations in the computer room power supply voltage and wiring errors, but also allows for seamless switching to ensure uninterrupted equipment operation in the event of a single power supply failure. From both communication and power supply perspectives, this enhances the reliability of the gateway for long-term unattended operation and reduces the safety risk of water level monitoring interruptions caused by line or power supply failures.
[0050] In an alternative implementation, the communication protocol can be replaced with Bluetooth or Wi-Fi for data transmission.
[0051] This application also provides a multi-level water level edge intelligent monitoring and control method suitable for computer room environment monitoring. Corresponding water level probes are set at multiple locations with different water level heights. Each water level probe is connected to a single optically isolated digital channel within the multi-level water level sensing interface 120 of the edge acquisition and computing gateway. The signal is transmitted to the gateway microcontroller unit 110 via the multi-level water level sensing interface 120. The edge gateway performs water level monitoring and control using the following steps: 100. Receive multiple optically isolated digital input signals and identify five water level states (empty, low, normal, high, and full) based on the combined state codes of the multiple input signals.
[0052] Specifically, water level probes are deployed at different liquid levels in the water tank. Each probe signal is connected to an independent optocoupler-isolated channel within a multi-level water level sensing interface (120mm) to eliminate electromagnetic interference in the equipment room. The microcontroller continuously polls and collects five digital signals, each signal representing whether the corresponding liquid level probe has been triggered. The five high and low level signals are concatenated into a binary code, and each code is pre-calibrated to uniquely correspond to one of the following water level levels: empty, low, normal, high, or full. The water level is not directly determined after a single sampling and encoding; only when two consecutive sampling codes match is a valid water level locked, filtering out false triggers caused by probe jitter and momentary interference. This multi-probe combination encoding achieves five levels of fine-grained liquid level differentiation, unlike traditional single-probe methods that only provide a binary judgment of water presence / absence. Optical isolation and secondary sampling verification provide dual noise reduction, significantly reducing the probability of water level misjudgment in the complex electromagnetic environment of the equipment room, and significantly improving monitoring stability.
[0053] 200. Execute the corresponding local linkage control output based on the identified five water level levels. Each water level level corresponds to a different linkage control output action. The linkage control output is executed locally by the edge acquisition and computing gateway 100 by default.
[0054] Specifically, the microcontroller unit 110 has a built-in hierarchical linkage strategy table, with dedicated execution logic matched to different water levels: Empty water level activates the water replenishment pump; low water level maintains water replenishment with a low-frequency intermittent buzzer alarm; normal water level shuts off the pump without alarm; high water level opens the drain valve with a rapid buzzer alarm; full water level cuts off the inlet pump with a high-frequency continuous alarm; after the water level drops to normal, the drain valve is shut off and the alarm is triggered again after a delay. The entire control logic is embedded within the microcontroller unit 110 of the edge acquisition and computing gateway 100, requiring no platform commands and operating independently even when the network is disconnected. This enables differentiated handling based on water level, adapting to all scenarios of water tank replenishment, drainage, and hierarchical alarms. Edge-based local autonomous computing and control, independent of the cloud network, ensures basic safety management of the water tank even in network outage scenarios, preventing overflow and water shortage accidents caused by network failures.
[0055] 300. When a valid remote control command is received, switch to remote control mode and pause local linkage control.
[0056] Specifically, the edge acquisition and computing gateway 100 receives commands from the monitoring platform via the UDP communication module 140, verifying the server room number, device address, and checksum byte to complete the legality verification. Upon recognizing a valid remote control command, it automatically disables the built-in water level linkage strategy, and the relay output of the local linkage execution unit 130 is entirely controlled directly by the remote command. During this period, it no longer automatically starts or stops the water pump or issues alarms based on the water level status. It supports remote manual takeover of the equipment from the platform, facilitating remote debugging and maintenance by operations and maintenance personnel. Local control is temporarily locked to avoid conflicts between local automatic logic and remote manual commands, eliminating equipment malfunctions caused by conflicting control logic.
[0057] 400. When the remote control command is interrupted for more than the preset timeout threshold, or when a command to switch back to local is received, the system will automatically revert to local automatic control mode.
[0058] Specifically, the edge acquisition and computing gateway 100 has a built-in timing module that continuously monitors remote data packets after entering remote mode. If no valid command is received from the platform for 30 consecutive seconds, the remote connection is considered interrupted. Alternatively, upon receiving a local mode switching command from the platform, it immediately exits remote mode, re-enables the five-level water level linkage strategy, and restores the autonomous control logic based on the water level. It has an automatic fallback mechanism for remote control timeouts to prevent the equipment from being under manual control for extended periods and losing automatic protection after the platform goes offline. It achieves seamless switching between local and remote modes, balancing the flexibility of remote operation and maintenance with the security of unattended local operation, and improving the overall fault tolerance of the data center's water level monitoring.
[0059] In an alternative implementation, the control mode switching can adopt a "remote priority, local backup" strategy, where remote commands always override local logic, but the security is lower than the "automatic timeout switchback" mechanism of this solution.
[0060] Furthermore, based on the same inventive concept, this application also provides a multi-level intelligent monitoring and control method for water level edges. For example... Figure 2 As shown, the edge acquisition and computing gateway 100 first executes step 100, continuously receiving multiple optically isolated digital input signals transmitted from the multi-level water level sensing interface 120. It performs encoding calculations based on the combined states of each signal to identify five water level states: empty, low, normal, high, and full. Next, in step 200, the edge acquisition and computing gateway 100 defaults to local automatic operation mode. For different water level levels, it calls a preset differentiated strategy to output corresponding local linkage control actions, autonomously driving peripherals such as water pumps and alarm devices to complete closed-loop water level management. If the verification passes and a valid remote control command is received during operation, step 300 is executed, and the gateway switches to remote control mode, temporarily suspending the built-in local water level linkage control logic and handing over direct control of the execution unit to the remote platform. The next step is to continuously monitor the status of the remote link in step 400. Once the remote control command interruption time exceeds the preset timeout threshold, or the platform issues a command to switch back to local mode, the gateway will exit the remote mode and automatically resume the local automatic control process. It will then re-execute autonomous linkage control based on the five-level water level status. The entire process is repeated to achieve continuous monitoring and reliable management of the data center water level.
[0061] In one possible implementation, receiving multiple optocoupler-isolated digital input signals and identifying five water level states (empty, low, normal, high, and full) based on the combined state encoding of the multiple input signals includes the following steps: acquiring five optocoupler-isolated digital input signals; encoding the five signals according to a preset combination rule; and determining the five water level states (empty, low, normal, high, and full) based on the encoding results.
[0062] In an alternative implementation, the multi-channel optocoupler isolated digital input signal can support different numbers of inputs, such as 3 or 8 channels, to achieve different levels of water level status identification, such as three or eight levels. The number of channels and levels can be flexibly adjusted according to the actual size of the water tank in the computer room and the monitoring requirements.
[0063] In one possible implementation, determining the five-level water level status (empty, low, normal, high, full) based on the encoding result includes: performing a consistency check on the encoding results of multiple consecutive samples. When two consecutive sampling results are consistent, the current five-level water level status is confirmed to be valid.
[0064] In one possible implementation, executing the corresponding local linkage control output based on the identified five water level states includes the following steps: matching a preset linkage strategy according to the current water level state; controlling the solid-state relay output according to the preset linkage strategy; and driving the alarm device to execute multi-level alarms according to the linkage strategy.
[0065] Specifically, this application proposes an edge acquisition and computing gateway 100, the core of which lies in achieving highly reliable, highly secure, and fully domestically produced edge intelligent closed-loop control through hardware and software co-design, multi-level intelligent water level monitoring and collaborative control. The specific technical solution is as follows: (I) System Architecture: The edge acquisition and computing gateway 100 includes: Microcontroller Unit 110 (MCU): Uses a domestically produced STM32 series main control chip.
[0066] Multi-level water level sensing interface 120: Supports 5-channel optocoupler-isolated digital input (DI), each corresponding to one water level probe. To avoid electrochemical corrosion caused by DC detection in pure water environments, this design adopts an AC pulse excitation + zero-bias detection scheme: The MCU generates a 1kHz square wave signal through a timer, which drives the probe electrode through a current-limiting resistor. The receiving end uses a full-bridge rectification + RC filter + high-speed optocoupler to form a non-polarity signal conditioning circuit, ensuring that there is no net DC bias voltage across the electrodes. Simultaneously, the probe electrodes are made of 316L stainless steel, and the leads are insulated with fluororubber and potted with epoxy resin to achieve IP68 sealing.
[0067] Local linkage execution unit 130: 8-channel MOS solid-state relays (DO) are used to control peripherals such as water pumps, solenoid valves, and audible and visual alarms.
[0068] Specifically, the 8-channel MOS solid-state relays are divided according to function: the first 6 channels respectively control the water supply pump, drain valve, low water level low-frequency alarm, high water level fast-response alarm, full water level continuous alarm, and water supply pump power failure protection circuit, directly implementing water supply, drainage, graded alarm, and overflow power failure protection corresponding to the five water levels; the remaining 2 channels are expansion channels, one of which can be linked to the dehumidification and ventilation equipment, and the other is used to output fault alarm dry contacts to the computer room environmental protection system, which not only fully covers all water level linkage actions of this solution, but also reserves space for future functional expansion.
[0069] Dual-mode communication module: Includes dual electrically isolated RS485 interfaces and UDP Ethernet interfaces, supporting parallel communication between local bus and remote IP network.
[0070] Dual-redundant power management module: supports DC24V wide voltage input, has overvoltage and reverse connection protection, and has 12V power supply.
[0071] The edge acquisition and computing gateway 100 can be used as a general edge node and is suitable for various dynamic and environmental sensing scenarios such as water level, temperature and humidity, and smoke detection. This embodiment takes water level monitoring as an example for explanation.
[0072] Furthermore, such as Figure 3 As shown, the power supply link uses a 24V input as the main power input, passing through a reverse polarity protection circuit for reverse polarity protection, a DC-DC conversion unit for graded voltage conversion, and finally dividing into three outputs: a 24V to 12V branch for powering actuators such as solenoid valves, a 12V to 5V branch for powering peripherals such as serial port sensors, and a 5V to 3.3V branch for powering the gateway's main control chip. The main control chip, as the core control unit of the gateway, is connected to two RS485 interfaces, an optocoupler isolation input circuit, a MOS solid-state relay output circuit, an Ethernet control module, and a buzzer, thereby achieving a full-function closed loop for multi-channel signal acquisition, local linkage control, remote data interaction, and alarm output.
[0073] (II) Embedded software control logic: 1. Five-level water level status machine identification: The gateway reads the status of 5 DI inputs in real time and analyzes the current water level according to the preset encoding rules (such as 00001=empty, 00011=low, 00111=normal, 01111=high, 11111=full) to form a five-level discrete status.
[0074] It should be noted that this system is used in a closed water tank environment, with the probe fully submerged and the water level stable without significant fluctuations or interference. Therefore, signal jitter mainly originates from contact noise or weak electromagnetic interference at the probe-circuit interface, rather than from liquid surface movement.
[0075] To ensure the reliability of state transitions, the software layer employs a lightweight state maintenance debouncing mechanism: any state change must satisfy "two consecutive samples are consistent" to be considered valid (i.e., if the current state is 00111, the state machine is only updated if the subsequent two samples are also 00111. Otherwise, the original state is maintained).
[0076] 2. Local closed-loop linkage control strategy: Pre-defined local action strategies corresponding to multiple water levels, for example: High water level: The buzzer will sound an alarm quickly.
[0077] When the water level is "full": the buzzer sounds an alarm quickly and the relay cuts off the power.
[0078] When the water level returns to "normal": the water pump will automatically shut off after a delay, and the alarm will be deactivated.
[0079] The strategy is executed independently in the gateway's local MCU, without relying on network connectivity, demonstrating its edge computing capabilities.
[0080] 3. Local / remote control mutual exclusion mechanism: The gateway operates in local automatic mode by default. When a valid remote control command is received, it automatically switches to remote mode, at which point the local water level linkage logic is suspended, and it fully responds to remote DO control commands.
[0081] If the remote command stream is interrupted for more than the preset timeout threshold (e.g., 30 seconds), or if a "switch back to local" command is received, the gateway will automatically resume local closed-loop control to ensure a safety backup.
[0082] (III) Lightweight UDP communication protocol: A UDP communication protocol with a 7-byte hexadecimal frame structure is defined for interaction between the edge data acquisition and computing gateway 100 and the remote monitoring platform. The frame format is shown in Table 1. Table 1: Frame structure definition of the 7-byte lightweight UDP communication protocol The following command types are supported: 0xEE: Active Status Reporting - Periodically (every 10ms) reports the current input / output status and operating mode.
[0083] 0xBB: Remote DO control - Switch to remote mode and remotely specify the status (ON / OFF) of a DO output.
[0084] 0xAA: Mode switching command - Switch to local mode.
[0085] 0xCC: Remote Status Query - Remote query requests instantly return the current status.
[0086] This protocol features short frame length, fast parsing, and reliable verification, which greatly reduces MCU resource consumption and adapts to resource-constrained edge computing environments.
[0087] Furthermore, such as Figure 4 As shown, the entire process of serial port data processing and control execution of the edge acquisition computing gateway 100 in this application is as follows: Input data from multiple RS485 sensors and the software host computer are first stored in the corresponding USART1 data buffer pool to complete data caching, and then uniformly sent to the USART data preprocessing unit to complete data parsing and verification. The preprocessed data is sent to the control information generation unit to generate corresponding alarm control signals to drive the buzzer to work, and to the working status monitoring unit to finally realize three core output functions: DI / DO status upload, query reply, and automatic and remote control, forming a complete closed loop from data acquisition, parsing and processing to control execution.
[0088] (iv) Full localization achieved: With a domestic content rate exceeding 95%, the main control unit, DC power supply, isolation devices, and communication components all utilize products from domestic manufacturers. The PCB design was completed using JLCPCB EDA and has passed engineering verification, demonstrating its capability for mass deployment. This Edge Acquisition and Computing Gateway 100 can be widely used in environmental monitoring systems within a domestically developed IT environment.
[0089] Furthermore, after the implementation of the technical solution of this application, the following significant technical effects can be achieved: 1. Improve water level monitoring accuracy: Achieve refined identification of 5 levels of water level, making alarms and responses more accurate and reducing false alarms / missed alarms.
[0090] 2. Enhanced system reliability: The local closed-loop control of the edge acquisition and computing gateway 100 ensures that it can still handle the situation autonomously and safely when the network is interrupted, and the MTBF (Mean Time Between Failures) is significantly improved.
[0091] 3. Avoid control conflicts: Completely eliminate the risk of command conflicts through local / remote mutual exclusion mechanism and timeout switchback strategy.
[0092] 4. Reduced communication overhead: The custom 7-byte UDP protocol reduces single-frame transmission latency to <1ms and CPU utilization to <5%, making it suitable for high-concurrency edge deployments.
[0093] 5. Meets the requirements of domestic IT innovation: The entire machine is domestically produced, breaking the dependence on foreign chips, and is suitable for key fields such as government affairs, finance, and energy.
[0094] 6. Easy to integrate and maintain: The protocol is open and the interface is standardized, which can be seamlessly connected to the main flow loop monitoring platform, reducing operation and maintenance costs.
[0095] 7. Possesses general edge intelligence capabilities: This gateway architecture can be extended to various sensing scenarios such as temperature and humidity, access control, and smoke detection, realizing "one gateway for multiple uses".
[0096] This application's edge acquisition computing gateway and multi-level intelligent edge monitoring and control method for water level can achieve refined water level monitoring, autonomous local control, secure remote management and control, lightweight communication transmission, and long-term probe detection through a combination of core steps including multi-level water level status coding and recognition, local linkage control execution, remote / local mode mutual exclusion switching, and automatic recovery after remote command timeout. This is further enhanced by auxiliary optimizations such as signal sampling debouncing, lightweight UDP communication, and AC pulse detection without DC bias. It fundamentally solves the technical problems of existing technologies, such as coarse water level judgment, rigid control logic, easy command conflicts, unreliable network outages, invisible status, and high maintenance difficulty. This significantly improves the monitoring accuracy, control reliability, and operational safety of the computer room environmental monitoring system.
[0097] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.
Claims
1. An edge acquisition and computing gateway, characterized in that, It includes a microcontroller unit, a multi-level water level sensing interface, a local linkage execution unit, a communication module, and a power management module; The multi-level water level sensing interface is electrically connected to the microcontroller unit and is used to collect multiple optically isolated digital input signals and input them to the microcontroller unit. The microcontroller unit is used to receive multiple optically isolated digital input signals, identify five water level states (empty, low, normal, high, and full) according to the signal combination state code, and issue linkage control commands based on the five water level states. The local linkage execution unit is electrically connected to the microcontroller unit and is used to receive linkage control commands issued by the microcontroller unit and execute local linkage control output. The communication module is electrically connected to the microcontroller unit and is used to send and receive remote control commands and report water level data and equipment operating status to the monitoring platform. The power management module provides power to the microcontroller unit, the multi-level water level sensing interface, the local linkage execution unit, and the communication module.
2. The edge acquisition and computing gateway according to claim 1, characterized in that, The microcontroller unit has a built-in five-level water level state machine and a local linkage control strategy; The five-level water level status machine identifies the five water level states (empty, low, normal, high, and full) based on the combined status codes of the five optically isolated digital input signals input from the multi-level water level sensing interface. The water level status is confirmed to be valid only when the water level status changes and the sampling signals are consistent for two consecutive times. The local linkage control strategy implements differentiated control based on different water level states: When the water level is empty, the local control unit starts the water replenishment pump to replenish water; when the water level is low, the water replenishment pump continues to run and the buzzer is activated with a low-frequency intermittent alarm; when the water level is normal, the water replenishment pump and drain valve are closed, and there is no alarm output; when the water level is high, the local control unit activates the buzzer with a fast alarm and starts the drain valve; when the water level is full, the local control unit activates the buzzer with a high-frequency continuous alarm and cuts off the power supply to the water pump; when the water level drops from high or full to normal, the water pump and drain valve are automatically shut off after a preset delay, and the alarm is cleared.
3. The edge acquisition and computing gateway according to claim 2, characterized in that, The microcontroller unit also has built-in logic for switching between local and remote control modes; The edge acquisition and computing gateway runs in local automatic mode by default, and the microcontroller unit independently executes the local linkage control strategy according to the five-level water level status without relying on the network. When the communication module receives a valid remote control command, it switches to remote control mode and pauses the local linkage control strategy, only responding to the remote control command to control the local linkage execution unit. When the remote control command stream is interrupted for more than the preset timeout threshold, or when the communication module receives a command to switch back to local mode, the local automatic mode will be automatically restored.
4. The edge acquisition and computing gateway according to claim 1, characterized in that, The communication module uses a lightweight UDP communication protocol to interact with the remote monitoring platform; The lightweight UDP communication protocol adopts a 7-byte fixed-length hexadecimal frame structure, which includes, in sequence, the data center identifier byte, the machine number byte, the remote / automatic mode code byte, the query / response code byte, the input data byte, the output data byte, and the check bit byte; The lightweight UDP communication protocol can realize four types of interactive functions: periodically and actively reporting input and output status and working mode, issuing remote control commands for controlling the output of solid-state relays, issuing local mode switching commands, responding to remote status queries and transmitting real-time status back.
5. The edge acquisition and computing gateway according to claim 1, characterized in that, The multi-level water level sensing interface adopts an AC pulse combined with a zero-bias detection scheme to eliminate electrochemical corrosion of electrodes in a pure water environment. The microcontroller unit has a built-in timer, which outputs a square wave signal and sends it to the water level probe electrode through a current-limiting resistor to complete the excitation. The signal receiving side is equipped with a non-polar signal conditioning circuit consisting of a full-bridge rectifier circuit, an RC filter circuit, and a high-speed optocoupler, which can eliminate the DC bias voltage at both ends of the probe electrodes. The water level probe electrode is made of 316L stainless steel, the probe lead is covered with a fluororubber insulation layer and is potted with epoxy resin, achieving an overall IP68 sealing protection level.
6. The edge acquisition and computing gateway according to claim 1, characterized in that, The communication module is a dual-mode communication module, integrating two electrically isolated RS485 interfaces and a UDP Ethernet interface, supporting parallel communication between local bus and remote IP network; The power management module is a dual-redundant power management module that supports DC24V wide voltage input, is equipped with overvoltage and reverse connection protection circuits, and outputs a 12V power supply branch to power peripherals.
7. A multi-level intelligent monitoring and control method for water level edges suitable for computer room environmental monitoring, characterized in that, Water level probes are installed at multiple locations with different water levels. Each probe is connected to a single optically isolated digital channel within the multi-level water level sensing interface of the edge acquisition and computing gateway. The signal is transmitted to the gateway microcontroller unit via the multi-level water level sensing interface. The edge gateway performs water level monitoring and control using the following steps: It receives multiple optically isolated digital input signals and identifies five water level states—empty, low, normal, high, and full—based on the combined state codes of the multiple input signals. The corresponding local linkage control output is executed based on the identified five water level states; and each water level state corresponds to a different linkage control output action; the linkage control output is executed locally by the edge acquisition and computing gateway by default; When a valid remote control command is received, the system switches to remote control mode and pauses local linkage control. When the remote control command is interrupted for more than the preset timeout threshold, or when a command to switch back to local is received, the system will automatically revert to local automatic control mode.
8. The multi-level intelligent monitoring and control method for water level edges according to claim 7, characterized in that, The process of receiving multiple optically isolated digital input signals and identifying the five water level states (empty, low, normal, high, and full) based on the combined state codes of the multiple input signals includes the following steps: Acquire 5 channels of optocoupler-isolated digital input signals; The five signals are encoded according to a preset combination rule; Based on the coding results, the water level status is determined to be one of five levels: empty, low, normal, high, and full.
9. The multi-level intelligent monitoring and control method for water level edges according to claim 8, characterized in that, The determination of the five water level states—empty, low, normal, high, and full—based on the coding results includes: Perform consistency judgment on the encoding results of multiple consecutive samples; When two consecutive sampling results are consistent, the current level 5 water level status is confirmed to be valid.
10. The multi-level intelligent monitoring and control method for water level edges according to claim 7, characterized in that, The step of executing the corresponding local linkage control output based on the identified five-level water level status includes the following steps: Match the preset linkage strategy according to the current water level status; Control the solid-state relay output according to the preset linkage strategy; The alarm devices are driven to execute multi-level alarms according to the linkage strategy.
