Remote control system and method for water taking bolt of intelligent fire hydrant based on 5G communication

Abstract

The invention relates to the crossing field of communication technology and fire-fighting Internet of Things, discloses a 5G communication intelligent fire hydrant water taking hydrant remote control system and method, and aims to solve the problems that an existing system is high in communication delay, limited in bandwidth, ineffective in emergency control and single in state perception. The method comprises the steps that a front-end terminal collects multi-dimensional state data and carries out edge intelligent processing; respectively transmitting a control instruction and a video stream through the 5G dual-network slice; the cloud platform performs simulation verification and predictive maintenance based on the digital twin model, and issues an accurate positioning control instruction; the terminal executing mechanism achieves millimeter-level valve adjustment through closed-loop servo control. The system comprises a multi-dimensional sensing module, a high-precision electric hydraulic execution mechanism, an edge calculation unit, a 5G double-slice communication module and a cloud digital twin platform. According to the application, ultra-low-delay, high-reliability and high-precision remote control and comprehensive state sensing are realized, and the safety and the intelligent level of a fire-fighting water supply network are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the cross field of communication technology and intelligent control of fire-fighting facilities, and particularly relates to a 5G communication intelligent fire hydrant water taking hydrant remote control system and method. BACKGROUND

[0002] With the continuous improvement of urban public safety system, fire-fighting water supply facilities as the key infrastructure for emergency rescue, the intelligent management level thereof directly affects the fire fighting efficiency and rational use of water resources. The traditional fire hydrant and water taking hydrant generally adopt mechanical structure, rely on manual inspection and on-site operation, and lack real-time perception and remote intervention ability of key parameters such as valve state, water pressure and flow, and illegal water taking. In the event of sudden fire or emergency water taking scene, the operation and maintenance personnel are difficult to quickly locate the available hydrant, judge the water supply state or remotely open and close the valve, which is easy to delay the rescue opportunity. At the same time, the existing part of the electronic transformation scheme is mostly based on 2G / 3G or narrowband Internet of Things communication, which has low transmission bandwidth, high time delay and weak concurrent ability, etc., and cannot support high-order intelligent functions such as high-definition video monitoring, multi-sensor fusion data backhaul and millisecond-level remote control instruction issuing.

[0003] Among them, the 5G communication technology provides a new technical path for the global perception and real-time linkage control of urban fire-fighting facilities with ultra-high bandwidth, ultra-low latency and massive connection characteristics. The remote control system based on 5G can theoretically realize the second-level reporting of fire hydrant state, high-definition video assisted decision and remote precise opening and closing operation, and significantly improve the emergency response ability and operation and maintenance efficiency.

[0004] In the prior art, although attempts have been made to integrate a wireless communication module into a fire hydrant control system, there are still multiple technical bottlenecks: the system architecture mostly adopts one-way data reporting mode, lacks a bidirectional closed-loop control mechanism, and cannot realize real-time feedback and verification of instruction execution state; the control logic and communication protocol have high coupling degree, and it is difficult to adapt to different manufacturers' valve actuator and sensor interface; in a high concurrency scene (such as multiple fire alarms triggered at the same time), the existing communication scheduling strategy is easy to cause instruction loss or response delay; in addition, the system generally does not integrate security authentication and abnormal behavior identification mechanism, and it is difficult to effectively prevent security risks such as illegal opening and destructive water taking. The above defects make the existing scheme difficult to meet the comprehensive management and control needs of high reliability, high security and high real-time in actual urban deployment, and there is an urgent need for a fire hydrant remote control system and method deeply integrating 5G communication capability, having closed-loop control logic and intelligent security protection. SUMMARY

[0005] In order to solve the technical problems of high communication delay, limited bandwidth, network congestion leading to control failure in emergency, single state sensing dimension and rough control precision existing in the existing fire hydrant remote monitoring system, the application provides a 5G communication intelligent fire hydrant water taking hydrant remote control system and method.

[0006] The application integrates multi-dimensional state sensing, edge intelligent computing, 5G dual-network slice communication, cloud digital twin simulation and closed-loop precise positioning control into a new technical architecture, aiming to realize ultra-low latency, high reliability and high bandwidth remote monitoring and millimeter-level accurate control of urban fire hydrant water taking hydrants, ensure absolute priority execution of fire water scheduling instructions and comprehensive real-time control of the state under any network load conditions, especially in extreme scenarios such as major disasters, and realize predictive maintenance through deep analysis of equipment operating state, thereby fundamentally improving the safety and intelligent management level of urban fire water supply network.

[0007] In one aspect, the application provides a 5G communication intelligent fire hydrant water taking hydrant remote control method, comprising the following steps: S1, the front-end intelligent sensing execution terminal collects and processes the multi-dimensional state data of the water taking hydrant in real time, and the multi-dimensional state data includes: Water taking hydrant internal pipe network water pressure time series data obtained by a high-frequency pressure sensor; Instantaneous flow and cumulative flow data obtained by an ultrasonic flowmeter; Pipe fluid acoustic vibration frequency spectrum data obtained by a piezoelectric ceramic acoustic sensor array built into the inner wall of the valve body structure; Water supply turbidity value obtained by a water quality turbidity sensor; Hydrant body vibration and tilt attitude data obtained by a three-axis acceleration sensor; Geographic position coordinates obtained by a global satellite navigation system module; And real-time video stream data of the water taking hydrant surrounding environment obtained by a high-definition wide-angle image acquisition module.

[0008] S2, the edge computing and control unit in the front-end intelligent sensing execution terminal performs localized fusion processing and feature extraction on the collected multi-dimensional state data.

[0009] Specifically, the edge computing and control unit executes a preset convolutional neural network model to perform real-time analysis on the acoustic vibration spectrum data to identify the characteristics of the micro leakage of the pipeline; at the same time, the edge computing and control unit performs correlation analysis on the water pressure time series data and the instantaneous flow data, establishes a hydraulics characteristic baseline, and is used for instant detection of abnormal water use behavior or pipe burst event; the edge computing and control unit classifies and labels the processed structured state data and real-time video stream data.

[0010] S3, the front-end intelligent perception execution terminal establishes and maintains two parallel and physically isolated dedicated network slice links to the cloud digital twin and scheduling platform through a 5G communication module.

[0011] The first network slice is an ultra-reliable low-latency communication network slice, and the second network slice is an enhanced mobile broadband network slice; the 5G communication module encapsulates the structured state data, response feedback data of control instructions and emergency alarm signals as first type data packets, and transmits them through the ultra-reliable low-latency communication network slice with an end-to-end delay of not more than 5ms; the 5G communication module encapsulates the real-time video stream data as second type data packets, and transmits them through the enhanced mobile broadband network slice.

[0012] S4, the cloud digital twin and scheduling platform receives the first type data packets and the second type data packets, and performs the following operations: real-time update the state parameters of the digital twin model corresponding to the front-end intelligent perception execution terminal using the structured state data; The digital twin model is a multi-physics field coupling model integrating fluid mechanics equations, material fatigue cumulative damage models and equipment kinematics models; The cloud digital twin and scheduling platform continuously monitors and health degree evaluates the current state of the digital twin model in a virtualized urban pipe network topology; The predictive maintenance module of the cloud digital twin and scheduling platform predicts the remaining useful life of valve seals, actuators and batteries based on the historical state data sequence of the digital twin model using a long short-term memory network algorithm, and generates a maintenance work order.

[0013] S5, when receiving a remote control instruction from a human-computer interaction interface, the intelligent scheduling and control decision engine of the cloud digital twin and scheduling platform first performs pre-execution simulation on the remote control instruction on the digital twin model.

[0014] The simulation calculation includes the transient effect on the downstream pipe network pressure distribution after the instruction execution, the strength of the water hammer effect, and the stress change during the valve opening or closing process. Only when the simulation result shows that all key parameters are within the preset safety threshold range, the intelligent scheduling and control decision engine converts the remote control instruction into a specific execution parameter sequence.

[0015] S6, the cloud digital twin and scheduling platform transmits the execution parameter sequence to the target front-end intelligent sensing execution terminal through the ultra-reliable low-latency communication network slice.

[0016] The execution parameter sequence defines the target position, maximum angular velocity, and segmented motion acceleration curve of the valve opening or closing action.

[0017] S7, the edge computing and control unit of the front-end intelligent sensing execution terminal receives and analyzes the execution parameter sequence, and drives the high-precision electro-hydraulic actuator to execute the valve control action.

[0018] The high-precision electro-hydraulic actuator is built-in an absolute value photoelectric encoder for real-time detection of the actual rotation position of the valve core.

[0019] The edge computing and control unit constitutes a closed-loop servo control system based on a proportional-integral-derivative control algorithm, continuously compares the real-time position feedback of the photoelectric encoder with the target position in the execution parameter sequence, dynamically adjusts the voltage and current applied to the drive motor until the error between the actual position of the valve core and the target position is less than the preset precision threshold; the precision threshold is 0.1mm.

[0020] During execution, the front-end intelligent sensing execution terminal real-time returns the execution state data such as the real-time position of the valve core, the motor drive current, and the hydraulic system pressure to the cloud digital twin and scheduling platform through the ultra-reliable low-latency communication network slice to complete the overall closed loop of the control process.

[0021] On the other hand, the present application provides an intelligent fire hydrant water taking hydrant remote control system for 5G communication, comprising: A front-end intelligent sensing execution terminal is installed on the body structure of the fire hydrant water taking hydrant or integrated in it, and the front-end intelligent sensing execution terminal comprises: A multi-dimensional state sensing module, which is composed of a piezoresistive pressure sensor integrated in the pipeline with a measurement range of 0-2.5 MPa and an accuracy of 0.25, an ultrasonic flowmeter clamped on the surface of the pipeline using time difference method, an acoustic sensor array composed of four piezoelectric ceramic acoustic sensors evenly distributed on the inner wall of the valve body structure at an interval of 90°, an optical water quality turbidity sensor installed on the inside of the water outlet, a three-axis acceleration sensor fixed to the internal reinforcing rib of the plug with a range of plus or minus 16 times the gravitational acceleration, a dual-mode satellite navigation receiving chip supporting Beidou and global positioning system, and a fixed-focus high-definition wide-angle image acquisition module with an infrared night vision function and a resolution of not less than 192 million pixels; the multi-dimensional state sensing module is used to comprehensively collect the working state and environmental information of the water taking plug.

[0022] A high-precision electric hydraulic actuator for driving the opening and closing of the water taking plug valve; the high-precision electric hydraulic actuator includes a brushless DC motor, a miniature bidirectional gear pump, a hydraulic valve block integrated with an electromagnetic reversing valve, a linear actuator cylinder, and an absolute value photoelectric encoder with a resolution of 4096 lines / revolution mechanically connected with the oil cylinder push rod; the brushless DC motor drives the hydraulic oil through the miniature bidirectional gear pump, controls the extension and retraction of the linear actuator cylinder through the hydraulic valve block, thereby accurately controlling the opening stroke of the valve.

[0023] An edge computing and control unit, the hardware core of which is a heterogeneous system-on-chip integrated with a four-core central processor, a neural network processing unit and a digital signal processor; the edge computing and control unit is electrically connected and in data communication with each sensor of the multi-dimensional state sensing module through a serial peripheral interface bus, an internal integrated circuit bus and an analog-to-digital conversion interface, and is responsible for running local data fusion algorithms, device state diagnosis models and closed-loop servo control algorithms; the edge computing and control unit interacts with the drive controller of the high-precision electric hydraulic actuator through a universal asynchronous receiver-transmitter interface.

[0024] A 5G communication module, which internally integrates a baseband chip supporting the new radio standard of the fifth generation mobile communication and is configured with two independent radio frequency front-end channels and double physical user identification card slots; the 5G communication module is configured to simultaneously register and activate two independent network slices reserved by the operator core network side for resource reservation and quality of service guarantee, respectively used for high reliability data transmission and high bandwidth video transmission; the 5G communication module establishes a wireless connection with the 5G base station through an external high-gain omnidirectional antenna.

[0025] An independent power supply management module, which is composed of a single-crystal silicon solar panel with a peak power of 50W and a lithium iron phosphate battery pack with a capacity of forty ampere-hours and a built-in maximum power point tracking charging controller, provides stable and long-lasting power supply for all electronic components of the front-end intelligent sensing execution terminal.

[0026] A cloud digital twin and scheduling platform deployed on a data center server cluster, the cloud digital twin and scheduling platform comprising: A data access and decoding gateway serving as a data aggregation entrance of all front-end intelligent sensing execution terminals, responsible for handling massive concurrent connections, decrypting and parsing data packets transmitted through different network slices.

[0027] A digital twin modeling engine that creates a virtual entity model corresponding to each connected front-end intelligent sensing execution terminal in the platform, including geometric models, physical properties, kinematic relationships, and behavior logic, and continuously drives and synchronizes the virtual entity model based on real-time state data.

[0028] A state monitoring and predictive maintenance module with various fault diagnosis and life prediction algorithm models, which analyzes historical and current operating parameters of the digital twin model to provide early warning of potential failures and dynamic optimization of key component maintenance cycles.

[0029] An intelligent scheduling and control decision engine that provides a geographic information system visualization interface, receives scheduling instructions from authorized operators, and performs safety simulation and impact assessment of control instructions before issuing instructions using the digital twin modeling engine to generate the optimal device execution strategy.

[0030] A 5G network slice management interface that interfaces with the core network management platform of the mobile communication network operator through standardized application programming interfaces, used for on-demand application, configuration, monitoring, and adjustment of resource parameters of the dedicated network slice allocated to the system, ensuring end-to-end communication service quality.

[0031] In summary, the present application includes at least one of the following beneficial technical effects: 1. By using 5G dual-network slice communication technology, the key control signaling and high-bandwidth video data stream are physically isolated, and the ultra-reliable low-latency communication network slice is used to ensure deterministic low-latency transmission of control instructions under any network congestion, solving the core pain point of unreliable communication in emergency situations in the prior art; the enhanced mobile broadband network slice realizes real-time high-definition video backhaul of the scene, providing unprecedented remote situational awareness capability.

[0032] 2. A new architecture is introduced that cooperates edge computing with cloud digital twin. The edge computing unit at the front end realizes real-time local processing of multi-dimensional sensor data and immediate abnormal response, greatly reducing the dependence on cloud computing resources and the bandwidth requirement of data transmission. The digital twin model in the cloud provides a high-fidelity simulation verification environment for complex control decisions, avoiding the impact and damage of misoperation on physical devices and pipe network systems, and achieving the predictability and safety of control.

[0033] 3. A high-precision electric hydraulic actuator with an absolute value encoder feedback is used, combined with a closed-loop servo control algorithm realized by an edge computing unit, to upgrade valve control from traditional rough "on-off" two-state control to millimeter-level stroke positioning control, making fine adjustment of flow and pressure a reality, and providing key execution capability for intelligent and fine scheduling of urban water supply pipe networks.

[0034] 4. A comprehensive and multi-dimensional state monitoring system is built from bottom-level sensors and actuators to the upper-layer cloud platform, and a predictive maintenance algorithm based on artificial intelligence is used to realize continuous tracking and active management of the health status of fire hydrant equipment, fundamentally changing the passive and lagging inspection and maintenance mode, significantly improving the integrity and reliability of the equipment, and reducing the life cycle operation and maintenance cost. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 is the overall technical scheme architecture diagram of the present application. DETAILED DESCRIPTION

[0036] To further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the specific embodiments according to the present application are described in detail below with reference to the drawings and preferred embodiments.

[0037] Referring to Figure 1 , the present application provides a 5G communication intelligent fire hydrant water taking hydrant remote control system and method, aiming to solve the technical defects of existing fire hydrant remote monitoring systems in communication delay, bandwidth limitation, network congestion, control failure, single state perception dimension, and rough control precision. By building a new technical architecture that integrates multi-dimensional state perception, edge intelligent computing, 5G dual-network slice communication, cloud digital twin simulation, and closed-loop precise positioning control, the present application realizes ultra-low latency, high reliability, and high bandwidth remote monitoring and millimeter-level accurate control of urban fire hydrant water taking hydrants, ensures absolute priority execution of fire water scheduling instructions and comprehensive real-time control of state in extreme scenarios such as major disasters, and realizes predictive maintenance through deep analysis of equipment operating status, thereby improving the safety and intelligent management level of urban fire water supply networks.

[0038] The application discloses a remote control method for an intelligent fire hydrant water taking hydrant of 5G communication. S1, a front-end intelligent sensing execution terminal collects and processes multi-dimensional state data of a water taking hydrant in real time, wherein the multi-dimensional state data comprises: water pressure time series data of an internal pipe network of the water taking hydrant acquired by a high-frequency pressure sensor; instantaneous flow and cumulative flow data acquired by an ultrasonic flow meter; pipe fluid acoustic vibration frequency spectrum data acquired by a piezoelectric ceramic acoustic sensor array embedded in an inner wall of a valve body structure; water supply turbidity value acquired by a water quality turbidity sensor; hydrant body vibration and inclination posture data acquired by a three-axis acceleration sensor; geographical position coordinates acquired by a global satellite navigation system module; and real-time video stream data of a surrounding environment of the water taking hydrant acquired by a high-definition wide-angle image acquisition module.

[0039] S2, an edge computing and control unit in the front-end intelligent sensing execution terminal performs localized fusion processing and feature extraction on the collected multi-dimensional state data.

[0040] Specifically, the edge computing and control unit performs a preset convolutional neural network model, analyzes the acoustic vibration frequency spectrum data in real time, identifies the micro leakage characteristics of the pipe, meanwhile, the edge computing and control unit performs correlation analysis on the water pressure time series data and the instantaneous flow data, establishes a hydraulics characteristic baseline, and is used for instant detection of abnormal water use behavior or pipe explosion event; the edge computing and control unit classifies and labels the processed structured state data and real-time video stream data.

[0041] S3, a 5G communication module in the front-end intelligent sensing execution terminal establishes and maintains two parallel, physically isolated special network slice links connected to a cloud digital twin and dispatch platform.

[0042] The first network slice is an ultra-reliable and low-latency communication network slice, and the second network slice is an enhanced mobile broadband network slice; the 5G communication module encapsulates the structured state data, response feedback data of a control instruction and an emergency alarm signal into a first type of data packet, and transmits the first type of data packet through the ultra-reliable and low-latency communication network slice with an end-to-end delay of not higher than 5 ms; the 5G communication module encapsulates the real-time video stream data into a second type of data packet, and transmits the second type of data packet through the enhanced mobile broadband network slice.

[0043] S4, the cloud digital twin and dispatch platform receives the first type of data packet and the second type of data packet, and performs the following operations: The structured state data is used to update state parameters of a digital twin model corresponding to the front-end intelligent sensing and execution terminal in real time; The digital twin model is a multi-physical field coupling model integrating fluid mechanics equations, material fatigue cumulative damage models, and equipment kinematics models; The cloud digital twin and scheduling platform continuously monitors and assesses the health of the current state of the digital twin model in a virtualized urban pipe network topology; The predictive maintenance module of the cloud digital twin and scheduling platform predicts the remaining useful life of valve seals, actuators, and batteries based on historical state data sequences of the digital twin model using a long short-term memory network algorithm and generates maintenance work orders.

[0044] S5, when receiving remote control instructions from the human-computer interaction interface, the intelligent scheduling and control decision engine of the cloud digital twin and scheduling platform first performs pre-execution simulation on the digital twin model for the remote control instructions.

[0045] The simulation calculation includes the transient effect on downstream pipe network pressure distribution, the intensity of water hammer effect, and the stress change during valve opening or closing. Only when the simulation results show that all key parameters are within the preset safety threshold range, the intelligent scheduling and control decision engine converts the remote control instructions into specific execution parameter sequences.

[0046] S6, the cloud digital twin and scheduling platform transmits the execution parameter sequences to the target front-end intelligent sensing and execution terminal through the ultra-reliable low-latency communication network slice.

[0047] The execution parameter sequence defines the target position, maximum angular velocity, and acceleration curve of segmented motion of the valve opening or closing action.

[0048] S7, the edge computing and control unit of the front-end intelligent sensing and execution terminal receives and analyzes the execution parameter sequence, and drives the high-precision electro-hydraulic actuator to execute valve control actions.

[0049] The high-precision electro-hydraulic actuator is built-in an absolute value photoelectric encoder for real-time detection of the actual rotation position of the valve core.

[0050] The edge computing and control unit constitutes a closed-loop servo control system based on proportional-integral-derivative control algorithm, which continuously compares the real-time position feedback of the photoelectric encoder with the target position in the execution parameter sequence, dynamically adjusts the voltage and current applied to the driving motor until the error between the actual position of the valve core and the target position is less than the preset precision threshold; the precision threshold is 0.1 mm.

[0051] During the execution process, the front-end intelligent sensing execution terminal will return the real-time position of the valve core, the motor driving current, and the execution state data such as the hydraulic system pressure to the cloud digital twin and scheduling platform through the ultra-reliable low-latency communication network slice in real time, to complete the overall closed loop of the control process.

[0052] The following will be expanded in detail.

[0053] In step S1, the front-end intelligent sensing execution terminal is deployed on the fire hydrant body structure of various fire-fighting water intake points such as urban roads, squares, and industrial areas, or is directly integrated into the internal cavity. The front-end intelligent sensing execution terminal realizes all-around sensing of the running state of the water intake hydrant and the surrounding environment through a multi-dimensional state sensing module.

[0054] The high-frequency pressure sensor adopts a piezoresistive principle, with a measurement range of 0-2.5MPa, an accuracy level of 0.25, and a sampling frequency of not less than 100Hz, which is used to continuously obtain the dynamic change curve of the internal water pressure of the pipe network.

[0055] The ultrasonic flowmeter adopts an external clamping installation method and is based on the time difference method measurement principle, which can synchronously output instantaneous flow and cumulative flow data, with a measurement accuracy of better than ±0.05%, suitable for accurate measurement under different pipe diameters and flow rates.

[0056] The piezoelectric ceramic acoustic sensor array is composed of four sensor units, which are evenly distributed on the inner wall of the valve body structure at an interval of 90°. Each sensor unit has an effective frequency response range of 20kHz-100kHz, which is used to collect acoustic vibration signals generated by the fluid flowing in the pipeline and form a complete ring acoustic spectrum.

[0057] The water quality turbidity sensor is installed inside the water outlet and adopts an optical scattering principle, with a measurement range of 0-1000NTU and a resolution of 0.1NTU, which is used to monitor the cleanliness of the water quality in real time.

[0058] The three-axis acceleration sensor is fixed to the internal reinforcing rib of the hydrant body, with a range of ±16 times the gravitational acceleration, which is used to detect the vibration and inclination attitude change of the hydrant body under external impact, earthquake or human damage.

[0059] The global satellite navigation system module supports Beidou and GPS dual-mode positioning, with a positioning accuracy of better than 2m, which is used to obtain the accurate geographic position coordinates of the water intake hydrant.

[0060] The high-definition wide-angle image acquisition module has infrared night vision function, with a resolution of 1920 pixels and a field of view angle of not less than 120°, which is used to collect real-time video stream data of the surrounding environment of the water intake hydrant in all-weather, with a video frame rate of not less than 25fps.

[0061] In step S2, the edge computing and control unit serves as the core processing unit of the front-end intelligent sensing terminal, and its hardware platform is a heterogeneous system-on-chip integrating a four-core central processing unit, a dedicated neural network processing unit, and a digital signal processor.

[0062] The edge computing and control unit establishes high-speed data channels with the pressure sensor, ultrasonic flow meter, acoustic sensor array, turbidity sensor, acceleration sensor, and image acquisition module through a serial peripheral interface bus, internal integrated circuit bus, and analog-to-digital conversion interface.

[0063] The edge computing and control unit first performs timestamp alignment and format standardization processing on the raw sensor data to ensure that all data is fused and analyzed under a unified time reference.

[0064] Subsequently, the edge computing and control unit loads a pre-trained lightweight convolutional neural network model, which takes a two-dimensional time-frequency graph composed of acoustic vibration spectrum data as input and outputs a binary classification result of whether there is a small leak and a rough estimate of the leak location. The lightweight convolutional neural network model has been trained and verified on a large amount of real leak scene data before shipment, and the inference delay is controlled within 10 ms.

[0065] At the same time, the edge computing and control unit performs cross-correlation analysis on the water pressure time series data and instantaneous flow data, calculates the cross-correlation coefficient at different time lags, and establishes a hydraulic characteristic baseline based on historical normal operating data. When the real-time data deviates from the hydraulic characteristic baseline by more than a preset threshold, the system determines that it is an abnormal water use behavior or a pipe burst event, and immediately triggers a local alarm.

[0066] The processed structured state data is packaged into a standardized data frame containing device identification, timestamp, sensor values, abnormal flag, and confidence score, and the real-time video stream data is additionally attached with geographic coordinates and device status labels, ready for the subsequent network transmission stage.

[0067] In step S3, the 5G communication module serves as the only communication bridge between the front-end intelligent sensing terminal and the cloud platform, and its internal integration is a baseband chip conforming to the new radio standard of the fifth generation mobile communication, and it is configured with two independent radio frequency front-end channels and dual physical user identification card slots.

[0068] 5G communication module initiates two registration requests to the operator's core network during the device power-on initialization phase, the first slice is the ultra-reliable low-latency communication network slice, its quality of service parameters are configured as end-to-end delay not higher than 5ms, reliability not less than 99.999%, bandwidth guarantee not less than 100kbps; the second slice is the enhanced mobile broadband network slice, its quality of service parameters are configured as downlink peak rate not less than 500Mbps, uplink peak rate not less than 100Mbps, the two slices achieve resource isolation at the physical layer, link layer and network layer, and do not interfere with each other.

[0069] The structured state data, control instruction response feedback and emergency alarm signal are packaged as the first type of data packet, protected by a lightweight encryption protocol, and transmitted through the ultra-reliable low-latency communication network slice.

[0070] The real-time video stream data is encoded into high-efficiency video coding format, packaged as the second type of data packet, and transmitted through the enhanced mobile broadband network slice with high throughput.

[0071] The 5G communication module has a built-in link quality monitoring module that can evaluate the signal strength, bit error rate and delay jitter of the two slices in real time, and automatically trigger the reconnection or switching mechanism when link degradation is detected, ensuring the continuous availability of communication.

[0072] In step S4, the cloud digital twin and the scheduling platform are deployed on a data center server cluster with high availability and elastic expansion capability.

[0073] The platform first receives data packets from all front-end terminals through the data access and decoding gateway, the decoding gateway decrypts, parses and checks the first type of data packet, extracts the structured state data and writes it into the real-time database; the decoding gateway decodes and stores the second type of data packet and establishes an association index with the corresponding device.

[0074] The digital twin modeling engine creates a unique virtual entity model for each connected front-end intelligent sensing execution terminal, which not only contains the three-dimensional geometric structure of the water tap, but also integrates fluid mechanics equations for simulating the dynamic changes of water pressure and flow, material fatigue cumulative damage models for evaluating the structural integrity of the valve body and sealing, and device kinematics models for describing the mechanical motion relationship during the opening and closing process of the valve.

[0075] The platform uses the structured state data uploaded in real time to drive the state synchronization of the digital twin model at a frequency of not less than 10Hz.

[0076] The state monitoring and predictive maintenance module continuously monitors the health indicators of each digital twin model, including pressure fluctuation amplitude, flow stability, acoustic anomaly energy, turbidity trend, vibration spectrum entropy, etc. The state monitoring and predictive maintenance module embeds a long short-term memory network model, which takes the historical state data sequence of the past 30 days as input and outputs the remaining effective life prediction value of the valve seal, high-precision electro-hydraulic actuator, and lithium iron phosphate battery pack.

[0077] When the predicted life is lower than the preset threshold, the system automatically generates a maintenance work order containing the device location, fault type, recommended replacement component, and priority, and pushes it to the operation and maintenance management system.

[0078] In step S5, the intelligent scheduling and control decision engine provides a visual human-computer interaction interface based on geographic information system technology, and the authorized operator can select the target water intake plug on the map and issue remote control instructions to open, close, or adjust to a specific opening degree.

[0079] Before the instruction is officially issued, the intelligent scheduling and control decision engine calls the digital twin modeling engine to perform pre-execution simulation on the digital twin model of the target device. The simulation process solves the transient fluid mechanics equation, calculates the pressure change curve of each node in the downstream pipe network after the instruction is executed, and evaluates whether negative pressure or overpressure phenomenon will occur; at the same time, it calculates the pressure wave peak value caused by water hammer effect to determine whether it exceeds the allowable stress of the pipe material; in addition, finite element analysis is also performed on the mechanical stress of the valve actuator during movement to ensure that structural damage will not occur due to overload.

[0080] All simulation results are compared with the preset safety threshold. Only when all key parameters are within the safe range, the engine will convert the original control instruction into a specific execution parameter sequence, which includes the angle value corresponding to the target opening degree of the valve, the maximum angular velocity limit during execution, and the acceleration curve defined in three sections, which is used to achieve smooth and impact-free valve movement.

[0081] In step S6, the execution parameter sequence is issued to the target front-end intelligent sensing execution terminal through the ultra-reliable low-latency communication network slice.

[0082] The low-latency and high-reliability characteristics of the ultra-reliable low-latency communication network slice ensure that the instruction is delivered within 5ms and still guarantees successful transmission under extreme network congestion conditions. The execution parameter sequence uses a structured binary format, including a check code and a version number, to prevent data tampering or parsing errors.

[0083] In step S7, after the edge computing and control unit of the front-end intelligent perception execution terminal receives the sequence of execution parameters, it first performs integrity check and version matching check. After the check is passed, the edge computing and control unit loads the target position, speed limit and acceleration curve parameters into the memory buffer of the closed-loop servo control algorithm.

[0084] The high-precision electric hydraulic actuator is composed of a brushless DC motor, a miniature bidirectional gear pump, a hydraulic valve block integrated with an electromagnetic reversing valve, a linear actuator cylinder and an absolute value photoelectric encoder.

[0085] The brushless DC motor drives the miniature bidirectional gear pump to generate a high-pressure hydraulic oil flow. The electromagnetic reversing valve switches the oil path direction according to the control signal to control the extension and retraction of the linear actuator cylinder. The linear actuator cylinder push rod is connected to the valve core through a mechanical linkage to achieve the opening or closing of the valve. The absolute value photoelectric encoder has a resolution of 4096 lines / revolution and can directly output the absolute angular position of the valve core without the need for power zero reset.

[0086] The edge computing and control unit runs a discretized proportional-integral-derivative control algorithm, and the control law is expressed as: ; wherein, is the motor drive voltage output in the kth control period, is the error between the actual position of the valve core at the current time and the target position, are the proportional, integral and derivative gain coefficients, respectively, whose values are adjusted online according to the valve load characteristics and the hydraulic system response characteristics.

[0087] The control algorithm runs at a frequency of no less than 1000Hz to adjust the motor drive current in real time and ensure that the valve core position error always converges.

[0088] When the absolute value of the error is less than the angular displacement corresponding to 0.1mm, it is determined that the control action is completed. During the entire execution process, the real-time position of the valve core, the motor current, the hydraulic system pressure and other data are continuously collected and returned to the cloud platform through a super-reliable low-latency communication network slice for updating the digital twin model and recording operation logs, thereby forming a complete control loop.

[0089] An intelligent fire hydrant water taking hydrant remote control system for 5G communication includes a front-end intelligent perception execution terminal and a cloud digital twin and dispatch platform.

[0090] The front-end intelligent perception execution terminal is installed on the body structure of the fire hydrant water taking hydrant or integrated inside, including a multi-dimensional state perception module, a high-precision electric hydraulic actuator, an edge computing and control unit, a 5G communication module and an independent power supply management module.

[0091] The multi-dimensional state sensing module is composed of a piezoresistive pressure sensor, an ultrasonic flow meter, a piezoelectric ceramic acoustic sensor array, an optical water quality turbidity sensor, a three-axis acceleration sensor, a dual-mode satellite navigation receiving chip and a high-definition wide-angle image acquisition module.

[0092] The high-precision electric hydraulic actuator includes a brushless DC motor, a miniature bidirectional gear pump, a hydraulic valve block integrated with an electromagnetic reversing valve, a linear actuating oil cylinder and an absolute value photoelectric encoder.

[0093] The edge computing and control unit has a heterogeneous system-level chip as the core, is connected with sensors and actuators through various bus interfaces, and a 5G communication module supports dual-network slice concurrent connection; an independent power supply management module is composed of a 50-watt monocrystalline silicon solar panel and a 40-ampere-hour lithium iron phosphate battery pack, and a maximum power point tracking charging controller is built-in.

[0094] The cloud digital twin and scheduling platform is deployed in a data center and includes a data access and decoding gateway, a digital twin modeling engine, a state monitoring and predictive maintenance module, an intelligent scheduling and control decision engine and a 5G network slice management interface; the cloud digital twin and scheduling platform is connected with an operator core network through a standardized application programming interface, and realizes dynamic management of network slice resources.

[0095] The embodiment realizes intelligent, refined and highly reliable remote control of city fire hydrant water taking hydrants through the above method and system, effectively solves the communication and control bottleneck of the prior art in an emergency scenario, and provides solid technical support for the digital transformation of city public safety infrastructure.

[0096] It is apparent for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application, and therefore, the embodiments should be regarded as exemplary and non-limiting in any respect.

[0097] In addition, it should be understood that although the present specification is described in terms of embodiments, each embodiment does not include only one independent technical solution, and the description manner of the specification is only for the sake of clarity, and those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can be combined appropriately to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for remote control of an intelligent hydrant water taking hydrant for 5G communication, characterized in that, Comprise: S1, through the front-end intelligent perception to execute terminal real-time collection of water tap multi-dimensional state data; S2, through the edge computing and control unit to carry out the fusion processing and feature extraction of the multi-dimensional state data locally; S3, through the 5G communication module, two parallel physical isolation special network slice links are established and maintained to connect to the cloud digital twin and scheduling platform, and the processed data is transmitted through the corresponding network slice; S4, through the cloud digital twin and scheduling platform, the state parameters of the digital twin model are updated in real time using structured state data, and the digital twin model is continuously monitored and health degree is evaluated, and predictive maintenance is carried out based on historical state data sequences; S5, when receiving remote control instructions, the intelligent scheduling and control decision engine is used to pre-execute simulation on the remote control instructions on the digital twin model; S6, through the ultra-reliable low-latency communication network slice, the execution parameter sequence is downloaded to the front-end intelligent perception execution terminal; S7, through the edge computing and control unit, the execution parameter sequence is received and analyzed, the high-precision electric hydraulic actuator is driven to execute the valve control action, and the execution state data is returned to the cloud digital twin and scheduling platform through the closed-loop servo control system to complete the control loop. 2.The intelligent hydrant water taking hydrant remote control method of 5G communication according to claim 1, wherein, In step S1, the multi-dimensional state data includes water tap internal pipe network water pressure time series data obtained by a high-frequency pressure sensor, instantaneous flow and cumulative flow data obtained by an ultrasonic flowmeter, pipe fluid acoustic vibration frequency spectrum data obtained by a piezoelectric ceramic acoustic sensor array built into the inner wall of the valve body structure, water supply turbidity value obtained by a water quality turbidity sensor, tap body vibration and inclination attitude data obtained by a three-axis acceleration sensor, geographic position coordinates obtained by a global satellite navigation system module, and real-time video stream data of the water tap surrounding environment obtained by a high-definition wide-angle image acquisition module. 3.The method of claim 2, wherein, The high-frequency pressure sensor is a piezoresistive pressure sensor, the measurement range is 0-2.5MPa, the accuracy is 0.25, and the sampling frequency is not less than 100Hz; The ultrasonic flowmeter is an external clamp type time difference flowmeter, and the measurement accuracy is better than ±0.05%; The piezoelectric ceramic acoustic sensor array is four sensor units evenly distributed at 90° intervals on the inner wall of the valve body structure, and the effective frequency response range is 20kHz-100kHz; The water quality turbidity sensor is an optical scattering type sensor, the measurement range is 0-1000NTU, and the resolution is 0.1NTU; The three-axis acceleration sensor has a range of ±16 times the gravitational acceleration; The global satellite navigation system module supports Beidou and GPS dual-mode positioning, and the positioning accuracy is better than 2m; The high-definition wide-angle image acquisition module has a resolution of 192 million pixels, a field of view angle of not less than 120°, an infrared night vision function, and a video frame rate of not less than 25fps. 4.The method of claim 1, wherein, In step S2, the preset convolutional neural network model is executed to analyze the acoustic vibration spectrum data in real time to identify the characteristics of the pipeline micro-leakage, and the correlation analysis is performed on the water pressure time series data and the instantaneous flow data to establish the baseline of the hydraulic characteristics for detecting abnormal water use behavior or pipe burst events, and the processed structured state data is classified and labeled with the real-time video stream data. 5.The method of claim 4, wherein, In step S2, it also includes: Data channels are established with each sensor through a serial peripheral interface bus, an internal integrated circuit bus, and an analog-to-digital conversion interface; The original sensor data is timestamped and standardized; A pre-trained lightweight convolutional neural network model is loaded, taking a two-dimensional time-frequency diagram composed of acoustic vibration spectrum data as input, and outputting binary classification results and rough estimates of micro-leakage locations, with a reasoning delay controlled within 10 ms; The cross-correlation coefficients of water pressure time series data and instantaneous flow data at different time lags are calculated, and the baseline of the hydraulic characteristics is established based on historical normal operating data; The processed structured state data is packaged into a standardized data frame containing device identification, timestamp, sensor values, abnormal flag, and confidence score. 6.The method of claim 1, wherein, In step S3, the first network slice is an ultra-reliable low-latency communication network slice, and the second network slice is an enhanced mobile broadband network slice. The structured state data, control instruction response feedback data, and emergency alarm signals are packaged into a first type of data packet and transmitted through the ultra-reliable low-latency communication network slice with an end-to-end delay of not more than 5 ms. The real-time video stream data is packaged into a second type of data packet and transmitted through the enhanced mobile broadband network slice with high throughput. 7.The method of claim 6, wherein, The 5G communication module integrates a baseband chip conforming to the new radio standard of the fifth generation mobile communication, and is equipped with two independent radio frequency front-end channels and two physical user identification card slots; The service quality parameters of the ultra-reliable low-latency communication network slice are configured as an end-to-end delay of not more than 5 ms, a reliability of not less than 99.999%, and a bandwidth guarantee of not less than 100 kbps; The service quality parameters of the enhanced mobile broadband network slice are configured as a downlink peak rate of not less than 500 Mbps and an uplink peak rate of not less than 100 Mbps; The first type of data packet is protected by a lightweight encryption protocol, and the second type of data packet is in a high-efficiency video coding format; The 5G communication module has a built-in link quality monitoring module that evaluates signal strength, bit error rate, and delay jitter in real time and automatically triggers a reconnection or switching mechanism when the link deteriorates. 8.The method of claim 6, wherein, In step S4, the first type of data packet is decrypted, parsed, and verified by the data access and decoding gateway and written into a real-time database, and the second type of data packet is video stream decoded and stored and an equipment association index is established; The digital twin model includes a three-dimensional geometric structure of the water intake plug, and integrates fluid mechanics equations, material fatigue cumulative damage models, and equipment kinematics models; The digital twin model state is synchronized using real-time structured state data at a frequency of not less than 10 Hz; The state monitoring and predictive maintenance module monitors the pressure fluctuation amplitude, flow stability, acoustic anomaly energy, turbidity change trend, and vibration spectrum entropy health index. 9.The intelligent hydrant water-fetching hydrant remote control method of 5G communication according to claim 1, wherein, In step S5, the transient fluid mechanics equation is solved on the digital twin model of the target device to calculate the pressure change curve of each node in the downstream pipe network after the execution of the instruction; whether the pressure peak value generated by the water hammer effect exceeds the allowable stress of the pipe material; performing finite element analysis on the mechanical stress of the valve actuator during movement; comparing the simulation results with the preset safety threshold, and only when all key parameters are within the safety range, generating an execution parameter sequence containing the target angle value, maximum angular velocity limit and three-section acceleration curve.

10. A smart hydrant water taking hydrant remote control system for 5G communication, characterized in that, Comprise: Front-end intelligent sensing and execution terminal, installed on the hydrant water taking plug body structure or integrated inside, comprising: Multi-dimensional state sensing module, used to collect multi-dimensional state data of the water taking plug; High-precision electro-hydraulic actuator, used to drive the opening and closing of the water taking plug valve; Edge computing and control unit, used for local fusion processing and feature extraction of multi-dimensional state data, and constitutes a closed-loop servo control system; 5G communication module, used to establish and maintain two parallel physically isolated dedicated network slice links connected to the cloud digital twin and scheduling platform; Independent power supply management module, composed of monocrystalline silicon solar cell panel and lithium iron phosphate battery pack; Cloud digital twin and scheduling platform, deployed in a data center server cluster, used to receive data packets transmitted through different network slices, and update the state parameters of the digital twin model in real time.

Images