Crane operation state real-time monitoring and health assessment system based on 5G edge computing

Abstract

This invention discloses a real-time monitoring and health assessment system for tower crane operation status based on 5G edge computing, belonging to the field of engineering machinery safety monitoring technology. It includes: a sensing and acquisition unit configured to communicate with subsequent units via an industrial bus, used to collect tower crane operating condition data, mechanism operation data, attitude and environmental data, and structural damage data; and an edge data processing and control unit deployed at the tower crane site, including an embedded processor, local storage module, and I / O control interface, configured to perform local preprocessing, anomaly identification, and safety control logic execution on the data collected by the sensing and acquisition unit. This invention features low response latency, accurate structural damage identification, data security and compliance, and reliable multi-tower collaborative operation, effectively improving the safety and intelligent management level of tower crane operation, and is applicable to the field of intelligent monitoring of construction tower cranes.

Description

Technical Field

[0001] This invention belongs to the field of engineering machinery safety monitoring technology, specifically, it relates to a tower crane operation status real-time monitoring and health assessment system based on 5G edge computing. Background Technology

[0002] Tower cranes are the core lifting equipment on construction sites, and their operational safety and structural health directly affect construction safety and project progress.

[0003] Traditional tower crane monitoring systems generally suffer from problems such as limited data acquisition dimensions, inability to identify early structural damage, high latency due to reliance on centralized cloud processing, and vulnerability to loss of control during network outages. Furthermore, they lack edge-to-edge collaboration capabilities in multi-tower collaborative operation scenarios, increasing the risk of collisions. Data transmission security is insufficient, and there is no end-to-end compliance and closed-loop operation and maintenance management, failing to meet the safety supervision requirements for special equipment. In addition, traditional systems lack real-time mapping and full lifecycle health assessment capabilities based on digital twins, making it difficult to shift from passive alarms to proactive prediction. In view of these issues, this paper proposes a real-time monitoring and health assessment system for tower crane operation status based on 5G edge computing. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a 5G edge computing-based real-time monitoring and health assessment system for tower crane operation status that can overcome or at least partially solve the above problems.

[0005] To solve the above-mentioned technical problems, the basic concept of the technical solution adopted by the present invention is as follows: The tower crane real-time monitoring and health assessment system based on 5G edge computing includes: a sensing and acquisition unit configured to communicate with subsequent units via an industrial bus to collect tower crane operating data, mechanism operation data, attitude environment data, and structural damage data; an edge data processing and control unit deployed at the tower crane site, including an embedded processor, local storage module, and I / O control interface, configured to perform local preprocessing, anomaly identification, and safety control logic execution on the data collected by the sensing and acquisition unit, achieving a local closed loop for safety protection; a 5G communication transmission unit configured to communicate bidirectionally with the edge data processing and control unit and the cloud platform unit, with a built-in encryption chip to ensure data transmission security; a cloud platform unit including a data server and an AI computing module, configured to perform tower crane health assessment and life prediction based on received data; and a human-machine interaction unit configured to communicate with the edge data processing and control unit to achieve local visualization display and operation interaction.

[0006] Preferably, the sensing and acquisition unit includes: a working condition sensing submodule, including a tension sensor and a pressure sensor, for acquiring lifting weight and amplitude data; a mechanism operation acquisition submodule, including an encoder and a speed sensor, for acquiring operating parameters of the hoisting, luffing, and slewing mechanisms; an attitude and environment sensing submodule, including a gyroscope, an tilt sensor, and an environmental sensor, for acquiring tower crane attitude and on-site environmental data; a structural acoustic emission monitoring unit, configured to acquire stress wave signals of structural crack initiation and propagation; and a global clock synchronization submodule, configured to synchronize the data acquired by the above submodules in time.

[0007] Preferably, the structural acoustic emission monitoring unit consists of multi-channel acoustic emission sensors, which are deployed at the root of the tower crane boom, the standard section of the tower body, and the critical dangerous sections of the slewing support; the sampling frequency of the acoustic emission sensors is not less than 500kHz, and is used to identify early cracks in combination with time-frequency analysis algorithms.

[0008] Preferably, the edge data processing and control unit further includes: a fault self-diagnosis and self-healing control module, configured to complete fault root cause location and graded handling based on a preset fault detection algorithm; and a local storage module, configured to store sensor raw data and processing results, and configured to permanently store abnormal event data.

[0009] Preferably, the 5G communication transmission unit adopts a 5G standalone industrial-grade module, configured to support network slices with end-to-end unidirectional transmission latency of no more than 10ms and transmission reliability of no less than 99.9%, and has a built-in symmetric encryption algorithm and two-way authentication mechanism.

[0010] Preferably, the cloud platform unit further includes: a structural damage assessment model configured to combine digital twin stress mapping and fatigue accumulation calculation to output health level and remaining life prediction based on tower crane structural parameters and real-time monitoring data; and a lifting capacity and lifting frequency operation efficiency analysis model configured to analyze operation efficiency.

[0011] Preferably, the edge data processing and control unit and the cloud platform unit are constructed with a horizontal federated learning iterative framework, configured to complete a closed-loop iteration of local model training, encrypted parameter uploading, cloud aggregation calculation and edge model updating under the premise of prohibiting the original running data from leaving the domain.

[0012] Preferably, it also includes a full-element digital twin module, which includes: a lightweight digital twin module deployed on the edge data processing and control unit, configured to achieve local real-time mapping based on a simplified model; and a full-element digital twin module deployed on the cloud platform unit, configured to achieve global mapping and state backtracking based on a full-dimensional model.

[0013] Preferably, it also includes a tower crane collaborative control module, configured to establish a 5G local area network logical connection between the edge data processing and control units of multiple tower cranes in the same project, which can realize edge-to-edge collaboration and offline autonomy without relying on the cloud.

[0014] Preferably, the cloud platform unit further includes a full-process security and compliance management module, configured to generate an immutable security and compliance file based on a hash verification mechanism.

[0015] As a preferred embodiment of the present invention: By adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art: 1. This invention achieves synchronous acquisition of four types of data—operating conditions, mechanism operation, attitude environment, and structural acoustic emission—through a sensing and acquisition unit. Combined with global clock synchronization, it ensures accurate data timing, solving the problems of single monitoring dimensions and asynchronous data in traditional tower cranes. At the same time, relying on edge data processing and control unit, it achieves millisecond-level local real-time processing and safe closed-loop control, and can independently complete fault self-diagnosis, self-healing control, and safety protection. It can ensure operational safety without relying on the cloud, greatly improving system reliability and response speed.

[0016] 2. This invention employs a transmission scheme combining 5G industrial-grade communication, hardware encryption, and two-way authentication to ensure low-latency, high-reliability, and high-security data transmission, adapting to the complex and harsh environment of construction sites. Through structural acoustic emission monitoring units and time-frequency analysis algorithms, early fatigue microcracks can be accurately identified, enabling early warning of structural damage. The cloud platform integrates structural damage assessment, digital twin stress mapping, fatigue accumulation calculation, and remaining life prediction functions, upgrading tower crane management from passive early warning to proactive health assessment and enhancing its intelligence level.

[0017] 3. This invention achieves multi-machine model collaborative optimization through a horizontal federated learning iterative framework, ensuring data remains within its domain while balancing model accuracy and data privacy and security. It leverages a full-element digital twin module to form a dual-mapping system between the edge and cloud, offering strong visualization, traceability, and simulation capabilities. The tower crane collaborative control module supports edge-to-edge collaboration and offline autonomy, enabling multi-level anti-collision protection. The full-process safety and compliance management module achieves a closed-loop process for monitoring, auditing, and maintenance, meeting the regulatory requirements for special equipment. The overall system possesses outstanding advantages such as integration, low latency, high reliability, intelligence, and safety compliance, significantly improving tower crane operation safety, operational efficiency, and management level.

[0018] The specific embodiments of the present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0019] In the attached diagram: Figure 1This is a schematic diagram illustrating the overall architecture and interaction relationships of a tower crane operation status real-time monitoring and health assessment system based on 5G edge computing; Figure 2 This is a schematic diagram of the sensing and acquisition unit structure; Figure 3 This is a schematic diagram of the working process of the structural acoustic emission monitoring unit; Figure 4 This is a schematic diagram of the edge data processing and control unit structure; Figure 5 This is a schematic diagram of the cloud platform unit structure; Figure 6 A schematic diagram of a full-element digital twin module; Figure 7 This is a schematic diagram of the structure of the group tower collaborative control module. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below with reference to the accompanying drawings. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0021] Example: Refer to Figures 1-7 A real-time monitoring and health assessment system for tower crane operation status based on 5G edge computing includes: a sensing and acquisition unit configured to communicate with subsequent units via an industrial bus to collect tower crane operating condition data, mechanism operation data, attitude environment data, and structural damage data; an edge data processing and control unit deployed at the tower crane site, including an embedded processor, local storage module, and I / O control interface, configured to perform local preprocessing, anomaly identification, and safety control logic execution on the data collected by the sensing and acquisition unit, achieving a local closed loop for safety protection; a 5G communication transmission unit configured to communicate bidirectionally with the edge data processing and control unit and the cloud platform unit, with a built-in encryption chip to ensure data transmission security; a cloud platform unit including a data server and an AI computing module, configured to perform tower crane health assessment and life prediction based on received data; and a human-machine interaction unit configured to communicate with the edge data processing and control unit to achieve local visual display and operation interaction.

[0022] The industrial bus adopts industrial-grade bus protocols such as CAN, RS485, ModbusTCP or Profinet to ensure the stability and anti-interference of data transmission, and is suitable for the harsh environment of high and low temperature, strong vibration and strong electromagnetic interference at the construction site. The operating data includes core operating parameters such as lifting weight and amplitude; the mechanism operation data includes motion parameters such as displacement, speed, and acceleration of the three major mechanisms of hoisting, luffing, and slewing; the attitude and environment data includes parameters such as tower tilt angle, swing amplitude, wind speed, temperature, and humidity; and the structural damage data includes related signals such as structural cracks and fatigue damage. The embedded processor uses an industrial-grade high-performance chip and is equipped with a real-time operating system to ensure the real-time performance of data preprocessing, anomaly identification, and safety control logic, with a response latency of milliseconds. The I / O control interface uses an industrial-grade isolated interface, supports multi-channel sensor access and multi-channel control command output, and has short-circuit protection and surge protection functions. The local storage module uses industrial-grade solid-state drives, which are vibration-resistant and have wide operating temperature characteristics, and can meet the long-term stable storage needs in the field. The encryption chip uses a hardware encryption chip with an independent encryption operation unit, which can effectively prevent eavesdropping, tampering and forgery during data transmission; The data server uses a high-capacity, high-reliability industrial-grade server to support large-scale time-series data storage and fast retrieval. The AI ​​computing module uses a high-performance computing chip to efficiently run various intelligent evaluation and prediction models. The human-machine interaction unit adopts an industrial-grade high-brightness touch screen, which has anti-glare, waterproof and dustproof characteristics, adapts to the complex lighting and environmental conditions of the construction site, and supports functions such as parameter configuration, historical data query, early warning information viewing and emergency manual control.

[0023] In summary, by simultaneously collecting four core data categories—working conditions, mechanism operation, attitude and environment, and structural damage—by the sensing and acquisition unit, comprehensive monitoring of the tower crane's operating status can be achieved. This provides complete data support for subsequent safety control and health assessment, avoiding missed detection of safety hazards and assessment biases due to data gaps. It also solves the technical problems of traditional tower crane monitoring systems, such as single data acquisition dimensions and incomplete coverage. Local closed-loop processing is achieved through edge data processing and control unit, with response latency down to the millisecond level. All safety control logic is executed on the tower crane itself, without relying on the cloud or external network, avoiding safety risks caused by network latency, interruption or cloud failure, greatly improving the safety of tower crane operation, and solving the technical problems of traditional systems relying on centralized cloud processing and high safety response latency. The 5G communication transmission unit adopts industrial-grade modules and hardware encryption chips to achieve two-way secure communication between the edge and the cloud, prevent data eavesdropping, tampering and forgery, adapt to the harsh environment of the construction site, ensure the stability and security of data transmission, and solve the technical problems of poor reliability and insecure data transmission in traditional tower crane communication methods. The cloud platform unit uses AI computing modules to achieve tower crane health assessment and life prediction, breaking through the limitations of traditional monitoring which can only provide simple early warnings. This promotes the transformation of tower crane management from passive early warning to proactive assessment, and solves the technical problems of low intelligence and lack of professional health assessment capabilities in traditional systems. The industrial-grade human-machine interface unit features anti-glare, waterproof and dustproof characteristics, adapts to complex lighting and environment at construction sites, supports multiple operation functions, improves the ease of operation and intuitiveness for on-site operators, reduces the probability of operational errors, and solves the technical problems of poor adaptability and inconvenient operation of traditional human-machine interface equipment.

[0024] Reference Figure 2 The sensing and acquisition unit includes: a working condition sensing submodule, including a tension sensor and a pressure sensor, for acquiring lifting weight and amplitude data; a mechanism operation acquisition submodule, including an encoder and a speed sensor, for acquiring operating parameters of the hoisting, luffing, and slewing mechanisms; an attitude and environment sensing submodule, including a gyroscope, an tilt sensor, and an environmental sensor, for acquiring tower crane attitude and on-site environmental data; a structural acoustic emission monitoring unit, configured to acquire stress wave signals of structural crack initiation and propagation; and a global clock synchronization submodule, configured to synchronize the data acquired by the above submodules in time.

[0025] Among them, the tension sensor adopts a pin-type tension sensor, which is installed at the force position between the end of the lifting arm and the hook pulley block. After temperature compensation, linear correction and signal filtering, the measurement accuracy is not less than ±0.5%, and it can adapt to load changes during heavy load, light load and dynamic amplitude change process. Pressure sensors are used to assist in the acquisition of amplitude-related pressure signals, working in conjunction with tension sensors to improve the accuracy of operational data acquisition. The encoder adopts a multi-turn absolute encoder, with one unit configured in each of the three major mechanisms: hoisting, luffing, and slewing. It can collect the displacement, velocity, and acceleration parameters of the mechanism in real time, with a resolution of no less than 1000 lines, ensuring the accuracy of motion parameter acquisition. The speed sensor adopts a Hall effect or magnetoelectric speed sensor, which is installed at the motor output shaft end to monitor the actual speed of the motor. The measurement range is 0-3000 r / min, and the accuracy is not less than ±1 r / min. The gyroscope adopts a high-precision microelectromechanical gyroscope with a measurement range of ±2000° / s and a resolution of not less than 0.01° / s, which is used to collect the angular velocity signal of the tower body. The tilt sensor is a dual-axis tilt sensor with a measurement range of ±15° and an accuracy of no less than 0.01°, used to collect the tilt angle of the tower. The environmental sensors include wind speed sensors and temperature and humidity sensors. The wind speed sensor has a measurement range of 0-60m / s and an accuracy of no less than ±0.1m / s. The temperature and humidity sensor has a measurement range of -40℃~85℃ and 0-100%RH, and is used to monitor the environmental parameters of the construction site in real time. The structural acoustic emission monitoring unit uses a piezoelectric acoustic emission sensor, which has high sensitivity and can capture weak stress wave signals. The global clock synchronization submodule adopts a high-precision clock synchronization protocol and is paired with an Ethernet controller that supports hardware timestamps. The synchronization accuracy is better than 1 millisecond, ensuring that the data collected by each submodule is strictly aligned on the time axis, providing reliable timing guarantee for multi-source data fusion and accurate fault location.

[0026] Reference Figure 3 The structural acoustic emission monitoring unit consists of multi-channel acoustic emission sensors, which are deployed at the root of the tower crane boom, the standard section of the tower body, and the critical dangerous sections of the slewing support. The sampling frequency of the acoustic emission sensors is not less than 500kHz, which is used to identify early cracks in combination with time-frequency analysis algorithms.

[0027] The multi-channel acoustic emission sensor has no fewer than 8 channels, enabling it to simultaneously acquire stress wave signals from multiple critical sections. The sensor is fixed using magnetic attraction or structural adhesive, and an acoustic coupling agent is applied during installation to improve stress wave transmission efficiency and ensure the integrity of signal acquisition. Critical critical sections include fatigue-prone areas such as the weld at the root of the boom, the main weld of the standard tower section, and the connection surface between the inner and outer rings of the slewing support. These areas are key regions where the tower crane structure experiences concentrated stress and is prone to cracking. The sensor deployment density is adjusted according to the degree of danger, with one sensor deployed per meter at key critical sections. The sampling frequency is set to 500kHz-1MHz, which can effectively capture high-frequency, short-time, and weak stress wave signals generated by the initiation and propagation of microcracks inside the structure, avoiding signal loss. The time-frequency analysis algorithm adopts wavelet packet transform or short-time Fourier transform algorithm, which can extract multi-dimensional feature parameters such as impact number, rise time, duration, amplitude, energy, and centroid frequency from the original stress wave signal. Through preset judgment rules, early fatigue microcracks are accurately identified. The judgment rules include impact frequency stable in the 50-200Hz range, signal energy concentrated in the 100-300kHz frequency band, and signal amplitude mainly low amplitude of 0.1-1V with no obvious strong impact signal, ensuring the accuracy and reliability of early crack identification.

[0028] Reference Figure 4The edge data processing and control unit further includes: a fault self-diagnosis and self-healing control module, configured to complete fault root cause location and graded handling based on a preset fault detection algorithm; and a local storage module, configured to store sensor raw data and processing results, and configured to permanently store abnormal event data.

[0029] Among them, the fault detection algorithm adopts a combination of residual analysis and decision tree reasoning. Residual analysis is used to calculate the deviation between the actual value of the sensor and the theoretical value of the mechanism model. When the deviation exceeds the preset threshold (the threshold can be dynamically adjusted according to the tower crane model and working conditions) and the duration exceeds 3 seconds, it is judged as a fault. Decision tree reasoning is used to locate the root cause of a fault based on the correlation between multiple signals, and can accurately identify different types of faults such as sensor faults, mechanical faults, and communication faults. The tiered treatment is divided into three levels: Level 1 faults are minor faults such as short-term sensor abnormalities and signal interference. The system will automatically reset the sensor and reacquire the signal without manual intervention. Level 2 faults include general faults such as overload, over-amplitude, and over-tilt angle. The system will activate an audible and visual alarm and restrict the movement of the mechanism in the dangerous direction. Manual confirmation and reset are required. Level 3 faults are serious faults such as structural cracks, failure of key mechanisms, and power failure. The system immediately cuts off the power supply, executes a safety lock, remotely reports the fault information, and prohibits the tower crane from starting until the fault is completely eliminated. The local storage module has a storage capacity of no less than 128GB and can cyclically store the sensor raw data and processing results of the most recent 30 days. The cyclic storage adopts a first-in-first-out (FIFO) mechanism. Abnormal event data includes relevant data for various abnormal events such as overload, crack alarm, tilt angle exceeding limit, and mechanism failure. During storage, the original data, processing results, and alarm information for 5 minutes before and after the event are automatically written to a long-term non-erasable storage area to achieve permanent storage. It supports export via USB interface, which facilitates subsequent accident tracing, fault analysis, and compliance review.

[0030] Reference Figure 1 The 5G communication transmission unit adopts a 5G standalone industrial-grade module, configured to support network slices with end-to-end unidirectional transmission latency of no more than 10ms and transmission reliability of no less than 99.9%, and has a built-in symmetric encryption algorithm and two-way authentication mechanism.

[0031] Among them, the 5G standalone industrial-grade module supports mainstream 5G frequency bands (n41, n78, n79, etc.), is compatible with 4G fallback function, and ensures stable communication in areas with poor 5G signal coverage. The module has an industrial-grade protection level (IP67) and can adapt to the harsh environment of the construction site. Network slicing involves applying specifically to the operator to allocate dedicated network resources for system control commands and critical monitoring data, isolating other network services, ensuring that end-to-end one-way transmission latency is consistently ≤10ms, transmission reliability is ≥99.9%, and in complex obstructed environments, the average latency is controlled to within 5ms, with a packet loss rate of less than 0.1%. The symmetric encryption algorithm uses the AES-256 encryption algorithm, which has high-strength encryption characteristics and can encrypt all transmitted data. The two-way authentication mechanism is implemented through pre-installed device certificates. Edge devices and the cloud platform verify each other's identities. Only devices that have passed authentication can interact with data. The session key is automatically updated every hour, which further enhances communication security and effectively prevents security risks such as unauthorized access, data eavesdropping, tampering and forgery.

[0032] Reference Figure 5 The cloud platform unit also includes: a structural damage assessment model, configured to combine digital twin stress mapping and fatigue accumulation calculation, and output health level and remaining life prediction based on tower crane structural parameters and real-time monitoring data; and a lifting operation efficiency analysis model, configured to analyze operation efficiency.

[0033] Among them, the structural damage assessment model pre-establishes a three-dimensional finite element model of the tower crane, covering all key structures such as the tower crane boom, tower body, and slewing support. The complex finite element calculation is transformed into a lightweight neural network proxy model through the order reduction model method, which reduces the computational complexity and improves the assessment efficiency. Digital twin stress mapping can quickly calculate the stress distribution of each section of the structure by inputting real-time working condition data such as lifting weight, amplitude, rotation angle, and wind speed, and present it intuitively in the form of stress cloud map, which can accurately locate stress concentration areas. The fatigue cumulative calculation adopts the linear cumulative damage criterion (Miner criterion), combined with the rainflow counting method to statistically analyze the fatigue load of the critical section, and combined with the fatigue characteristic curve of the tower crane structural material to calculate the fatigue damage degree of the critical section in real time. The health level is divided into four levels based on indicators such as fatigue damage degree, crack detection results, overload records, and number of tilt angle exceedances: healthy (damage degree ≤ 20%), slightly damaged (20% < damage degree ≤ 40%), moderately damaged (40% < damage degree ≤ 60%), and severely damaged (damage degree > 60%). Different maintenance recommendations are corresponding to different health levels. The remaining service life prediction uses a time series model or degradation model, combined with historical monitoring data and real-time data, to predict the remaining service life of key tower crane structures. The prediction error is controlled within ±10%, which has high engineering practical value. The lifting weight and lifting frequency operation efficiency analysis model calculates efficiency indicators such as lifting efficiency, load rate, and equipment utilization rate by statistically analyzing data such as the lifting weight, lifting frequency, operation time, and idle time of tower cranes. It outputs an efficiency analysis report, providing data support for tower crane scheduling and operation optimization, and helping to improve construction efficiency and reduce operating costs.

[0034] Reference Figure 5 The edge data processing and control unit and the cloud platform unit are constructed with a horizontal federated learning iterative framework, which is configured to complete the closed-loop iteration of local model training, encrypted parameter uploading, cloud aggregation calculation and edge model updating under the premise of prohibiting the original running data from leaving the domain.

[0035] Among them, the horizontal federated learning iterative framework adopts a collaborative mode of edge training and cloud aggregation. Each tower crane's edge data processing and control unit acts as an edge node, using its own collected historical and real-time operating data to independently train a lightweight structural damage assessment model. The training process is executed only locally on the edge node, and the original operating data is always kept on the tower crane site and not uploaded to the cloud, effectively protecting data privacy and business security. Local model training uses the mini-batch gradient descent algorithm, and the training cycle is set to once a day based on the data update frequency to ensure that the model can adapt to changes in the tower crane's operating status in a timely manner. After the model parameters are encrypted using the AES-256 encryption algorithm, they are uploaded to the global aggregation node in the cloud via the 5G communication transmission unit. During the parameter upload process, a fragmented transmission and verification retransmission mechanism is adopted to ensure the integrity and security of parameter transmission. The cloud-based global aggregation node uses a federated averaging algorithm to aggregate encrypted model parameters uploaded by multiple tower crane edge nodes, generating a global model with stronger generalization capabilities. During the aggregation process, no original data is touched; only the encrypted parameters are processed. After the global model is encrypted, it is distributed to each edge node. The edge nodes replace the original local model with the global model to complete one round of iterative optimization. The iteration cycle is set to once a week. Through multiple rounds of iteration, the evaluation accuracy and generalization ability of the model are continuously improved to adapt to the operating needs of tower cranes of different models and under different working conditions.

[0036] Reference Figure 6 It also includes a full-element digital twin module, comprising: a lightweight digital twin module deployed at the edge data processing and control unit, configured to achieve local real-time mapping based on a simplified model; and a full-element digital twin module deployed at the cloud platform unit, configured to achieve global mapping and state backtracking based on a full-dimensional model.

[0037] Among them, the lightweight digital twin module is based on a simplified multi-rigid-body kinematics model, ignoring details such as structural elastic deformation, and focuses on presenting the three-dimensional attitude and key operating parameters of the tower crane. The model rendering frame rate is no less than 30fps, which meets the real-time monitoring needs of on-site operators. The tower crane's lifting height, luffing range, slewing angle, lifting weight and other parameters, as well as the tilt and sway status of the tower body, can be displayed intuitively through the human-computer interaction unit. This module interacts with the edge data processing and control unit in real time, with a data update frequency of 100ms / time, ensuring the real-time nature of the mapping; The full-element digital twin module constructs a high-fidelity virtual model that includes tower crane geometry, material properties, load conditions, wind load model, foundation stiffness, and structural characteristics. The model accuracy is no more than 1% different from the physical tower crane, and it can completely map all the operating and structural states of the physical tower crane. This module supports functions such as multi-view roaming browsing, historical status backtracking, real-time rendering of stress cloud maps, fault simulation and accident inversion. Historical status backtracking can query the tower crane's operating status and structural status for the past 30 days. Fault simulation can simulate the tower crane's response under different types of faults. Accident inversion can restore the entire process of an accident, providing strong support for fault investigation and accident analysis. The full-element digital twin module achieves data synchronization through a 5G communication transmission unit, ensuring data consistency between the edge lightweight model and the cloud full-element model, forming a collaborative mode of real-time edge display + high-precision cloud management, thereby improving the visualization and intelligence level of tower crane management.

[0038] Reference Figure 7 The system also includes a tower crane collaborative control module, which is configured to establish a 5G local area network logical connection between the edge data processing and control units of multiple tower cranes in the same project, enabling edge-to-edge collaboration and offline autonomy without relying on the cloud.

[0039] Among them, the 5G local area network logical connection is realized through the local area network function of the 5G module, without relying on external public networks. Multiple tower crane edge nodes in the same project form an edge computing cluster, and the communication latency between nodes does not exceed 5ms, ensuring the real-time performance of collaborative control. The tower crane collaborative control module incorporates a spatiotemporal conflict detection algorithm, which collects core parameters such as lifting height, luffing amplitude, slewing angle, lifting weight, and operating trajectory of each tower crane in real time. Through an edge computing cluster, it performs real-time comparison and conflict prediction of the operating status of multiple tower cranes. When it detects that the operating areas of two or more tower cranes overlap or that their operating trajectories may collide, a tiered early warning mechanism is immediately triggered. A Level 1 warning (safe distance ≥ 5m) will trigger an audible and visual alert to remind operators to adjust their working posture. Level 2 warning (3m ≤ safe distance < 5m) limits the tower crane's operating speed in the dangerous direction to reduce the risk of collision; Level 3 warning (safe distance <3m) immediately cuts off the power supply to the relevant tower cranes in the dangerous direction and implements temporary locking until the operators adjust their working positions and resolve the conflict, thus achieving active anti-collision protection for the group of tower cranes.

[0040] The offline autonomous function is configured so that when the 5G public network is interrupted or the cloud platform fails, the edge computing cluster can run independently. The tower crane collaborative control module relies on the locally stored tower crane parameters, work area planning data and preset collaborative rules to continuously realize spatiotemporal conflict detection, hierarchical early warning and anti-collision control to ensure that the tower crane operation is not interrupted. Meanwhile, the system has a built-in load balancing scheduling strategy. When it detects that a tower crane is in an overloaded state while other tower cranes are in an idle or lightly loaded state, it pushes scheduling suggestions through the human-machine interaction unit and cloud platform to guide operators to optimize job allocation, avoid equipment damage caused by overloaded operation of a single tower crane, and improve the overall operating efficiency of multiple tower cranes. In addition, the tower crane collaborative control module supports the function of dividing the work area. It can preset the working range and prohibited intersection area of ​​each tower crane according to the construction site layout. Through collaborative verification between edge nodes, it ensures that the tower cranes operate strictly within the designated area, further reducing the risk of collision and adapting to large-scale tower crane collaborative operation scenarios.

[0041] Reference Figure 5 It also includes a full-process safety and compliance management module, configured to communicate bidirectionally with the cloud platform unit and the edge data processing and control unit to realize tower crane operation compliance monitoring, closed-loop operation and maintenance process and remote intervention control.

[0042] The compliance management module pre-enters tower crane registration information, special equipment inspection reports, operator qualification information and work permit scope, and compares the actual operating parameters of the tower crane with compliance standards in real time. When violations such as operating beyond the qualification scope, operating beyond the permit scope, or failing to conduct inspections on time occur, a compliance warning is immediately triggered, and the warning information is simultaneously pushed to the cloud platform and the on-site human-machine interaction unit. At the same time, dangerous operation actions of the tower crane are restricted until the violation is rectified. Compliance data is automatically retained to form a complete compliance ledger, supporting retrieval and query by time period and violation type, adapting to special equipment regulatory requirements, and facilitating inspection and verification by regulatory authorities.

[0043] The remote operation and maintenance module supports remote access to the edge data processing and control unit via the cloud. It can view tower crane operation data, fault information, and health assessment results in real time, and remotely issue instructions such as parameter configuration, fault reset, and maintenance reminders. Basic operation and maintenance work can be completed without on-site personnel. The module has a built-in operation and maintenance work order management function, which can automatically generate operation and maintenance work orders based on the tower crane health level and fault type, assign them to the corresponding operation and maintenance personnel, track the work order execution progress, and form an operation and maintenance closed loop of early warning-work order-handling-closing. Meanwhile, the module supports remote debugging of sensors and calibration of parameters by maintenance personnel, reducing on-site maintenance workload, improving maintenance efficiency, and reducing maintenance costs. In addition, the remote maintenance module has a hierarchical permission management function, which divides different operation permissions into administrators, maintenance personnel, and operators to ensure operational security and data security and prevent unauthorized operations.

[0044] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-described technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A tower crane operation status real-time monitoring and health assessment system based on 5G edge computing, characterized in that, include: The sensing and acquisition unit is configured to communicate with subsequent units via an industrial bus to collect tower crane operating data, mechanism operation data, attitude and environmental data, and structural damage data. An edge data processing and control unit, deployed at the tower crane site, includes an embedded processor, a local storage module, and an I / O control interface. It is configured to perform local preprocessing, anomaly identification, and safety control logic execution on the data collected by the sensing and acquisition unit, thereby achieving a local closed loop for safety protection. The 5G communication transmission unit is configured to communicate bidirectionally with the edge data processing and control unit and the cloud platform unit, and has a built-in encryption chip to ensure data transmission security. The cloud platform unit, including a data server and an AI computing module, is configured to perform tower crane health assessment and life prediction based on the received data. The human-computer interaction unit is configured to communicate with the edge data processing and control unit to realize local visualization display and operation interaction.

2. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 1, characterized in that, The sensing and acquisition unit includes: The working condition sensing submodule includes a tension sensor and a pressure sensor, which are used to collect load and amplitude data; The mechanism operation acquisition submodule includes an encoder and a speed sensor, which are used to collect the operating parameters of the hoisting, luffing and slewing mechanisms; The attitude and environment sensing submodule includes a gyroscope, tilt sensor and environmental sensor, used to collect tower crane attitude and on-site environmental data; The structural acoustic emission monitoring unit is configured to acquire stress wave signals related to the initiation and propagation of structural cracks. The global clock synchronization submodule is configured to synchronize the time of the data collected by the above submodules.

3. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 2, characterized in that, The structural acoustic emission monitoring unit consists of multi-channel acoustic emission sensors, which are deployed at the root of the tower crane boom, the standard section of the tower body, and the critical dangerous sections of the slewing support. The sampling frequency of the acoustic emission sensors is not less than 500kHz, which is used to identify early cracks in combination with time-frequency analysis algorithms.

4. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 1, characterized in that, The edge data processing and control unit also includes: The fault self-diagnosis and self-healing control module is configured to complete fault root cause location and graded treatment based on a preset fault detection algorithm; The local storage module is configured to store raw sensor data and processing results, and is also configured to permanently store abnormal event data.

5. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 1, characterized in that, The 5G communication transmission unit adopts a 5G standalone industrial-grade module, configured to support network slices with end-to-end unidirectional transmission latency of no more than 10ms and transmission reliability of no less than 99.9%, and has built-in symmetric encryption algorithm and two-way authentication mechanism.

6. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 5, characterized in that, The cloud platform unit also includes: The structural damage assessment model is configured to combine digital twin stress mapping and fatigue accumulation calculation, and output health level and remaining life prediction based on tower crane structural parameters and real-time monitoring data. The lifting operation efficiency analysis model is configured to analyze operation efficiency.

7. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 6, characterized in that, The edge data processing and control unit and the cloud platform unit are constructed with a horizontal federated learning iterative framework, which is configured to complete the closed-loop iteration of local model training, encrypted parameter uploading, cloud aggregation calculation and edge model updating under the premise of prohibiting the original running data from leaving the domain.

8. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 1, characterized in that, It also includes a full-element digital twin module, which comprises: A lightweight digital twin module deployed at the edge data processing and control unit is configured to achieve local real-time mapping based on a simplified model; The full-element digital twin module deployed on the cloud platform is configured to achieve global mapping and state backtracking based on a full-dimensional model.

9. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 8, characterized in that, It also includes a tower crane collaborative control module, configured to establish a 5G local area network logical connection between the edge data processing and control units of multiple tower cranes in the same project, so as to achieve edge-to-edge collaboration and offline autonomy without relying on the cloud.

10. The tower crane operation status real-time monitoring and health assessment system based on 5G edge computing according to claim 1, characterized in that, The cloud platform unit also includes a full-process security and compliance management module, configured to generate tamper-proof security and compliance files based on a hash verification mechanism.

Images