Multi-target identification and safety early warning system of deep shaft hoisting system

Abstract

The invention discloses a multi-target identification and safety pre-warning system of a deep shaft hoisting system. The multi-target identification and safety pre-warning system comprises a lateral pressure type tension sensor arranged on a winch steel wire rope, a temperature and humidity sensor arranged on a shaft wall, a visual sensor arranged on a shaft portal frame, a data receiving and processing device arranged on a shaft mouth and a pre-warning executing mechanism. Each sensor is connected with the data receiving and processing device and is connected with the early warning execution mechanism, and a YOLOv11 lightweight model and a multi-mode neural network fusion model are built in the data receiving and processing device and are used for analyzing visual data and multi-source physical parameters. The limitation of function simplification of a traditional monitoring scheme is broken through through a multi-source data fusion framework, 30 frame / second real-time target identification is realized by relying on a lightweight model, a timestamp alignment mechanism and closed-loop early warning logic are combined, the security upgrade from post-event passive recording to pre-event active interception is realized in a low-illumination and high-dust deep shaft environment, and the safety of the deep shaft environment is improved. And life safety of operators and stable operation of equipment are effectively guaranteed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of construction safety monitoring, relates to the technical field of deep vertical shaft safety monitoring, and specifically relates to a multi-target identification and safety warning system of a deep vertical shaft lifting system. BACKGROUND

[0002] The deep vertical shaft construction environment has multiple physical challenges; for example, in the construction of a deep vertical shaft of a hydropower station tunnel or a metal mine, the dense metal structure in the construction environment causes serious attenuation of wireless signals, the excessively high dust concentration causes visual sensor vision to be blurred, and the low-illumination condition caused by insufficient underground light causes the target identification function to fail. At the same time, the system needs to synchronously monitor dynamic risk sources, including sudden changes in the mechanical state of the steel wire rope, the deviation of the hoist hook movement trajectory, the absence of personnel safety equipment, and environmental parameter abnormalities; however, the prior art cannot realize multi-target cooperative tracking, resulting in a high key risk missed detection rate.

[0003] Although some monitoring methods have been researched and designed at present, the existing monitoring scheme has many deficiencies; the function is single, only focuses on equipment position monitoring, and ignores personnel behavior and environmental state; the environmental adaptability is weak, wireless transmission is unstable in the vertical shaft scene; the warning mechanism is incomplete, and lacks multi-source data fusion and closed-loop control capability. These defects make the system unable to meet the requirements of risk prediction and immediate intervention of the construction safety regulations, causing response delay and protection vacuum. SUMMARY

[0004] In order to solve the deficiencies in the prior art described above, a multi-target identification and safety warning system of a deep vertical shaft lifting system is disclosed. The purpose of the present application is to realize real-time monitoring of the equipment state, personnel behavior and environmental parameters of the deep vertical shaft lifting system through multi-source data fusion and intelligent decision-making algorithm, and to realize accurate discrimination and immediate warning in multiple working conditions by using deep learning technology, so as to protect the life safety of the operating personnel and the stable operation of the construction equipment.

[0005] The application is implemented by the following technical solutions:

[0006] A multi-target identification and safety warning system of a deep vertical shaft lifting system; the deep vertical shaft lifting system comprises two groups of winding machines arranged outside the deep vertical shaft mouth, and the two groups of winding machines are respectively connected with the underground construction platform through the steel wire ropes suspended from the vertical shaft gantry truss at the top of the deep vertical shaft mouth; characterized in that:

[0007] The multi-target identification and safety warning system constructs a "perception-fusion-decision-execution" full-link collaborative architecture, including: a physical quantity perception layer composed of distributed side pressure type tension sensors deployed on the steel wire rope of the hoist, an environment perception layer composed of multiple sets of temperature and humidity sensors arranged at intervals on the shaft wall, an image perception layer composed of visual sensors arranged on the shaft gantry truss, a core decision layer composed of a multi-source data receiver and processor deployed at the shaft mouth, and a response execution layer composed of a warning execution mechanism integrated with a sound and light alarm and a cloud storage system; each perception layer device forms a real-time data interaction closed loop with the data receiver and processor through heterogeneous communication links to realize fusion, and the data receiver and processor drive the warning execution mechanism to complete the hierarchical response through intelligent decision algorithms.

[0008] Further, the side pressure type tension sensor adopts LCZ-401B-1ST-7-V5-P13 model with a range of 15T, which is fixed on the steel wire rope 5 meters away from the hoist drum through a bolt clamp, and forms rigid contact with the steel wire rope to realize dynamic capture of tension; the temperature and humidity sensor adopts COS-03USB type recorder, which is arranged on the shaft wall support structure at intervals of 50 meters to form a shaft cylinder longitudinal environment parameter monitoring belt, realizing collaborative perception of temperature and humidity gradient at different depths; the visual sensor is a 1080P high-definition camera, which is installed at the center point of the crossbeam of the gantry truss, and covers the whole area of the shaft upper platform and the shaft mouth cover through field angle optimization design, and forms a spatial monitoring range complementary to the physical sensor.

[0009] Further, each perception layer device adopts a differentiated communication collaboration scheme: the side pressure type tension sensor and the temperature and humidity sensor realize stable transmission of low-rate physical data through anti-interference RS485 shielded cable, the visual sensor realizes high-speed transmission of high-definition video stream through gigabit network cable, and the two types of links realize time alignment and protocol conversion through the multi-interface adaptation module of the data receiver and processor, ensuring the collaboration of heterogeneous data transmission.

[0010] Further, the data receiver and processor adopts Intel Core I5-4310M embedded development board as the core, integrates multi-source data fusion module, target detection algorithm module and trajectory prediction module to form a collaborative decision center; the development board is built-in YOLOv11 lightweight model and multi-modal neural network fusion model, which realizes parallel operation of the two types of models through dynamic allocation of hardware resources; the multi-modal neural network fusion model as a core collaborative unit performs spatio-temporal correlation analysis and joint identification on tension data, temperature and humidity data and visual features, eliminating errors of a single data source.

[0011] The YOLOv11 lightweight model extracts high-dimensional visual features and performs cross-modal collaborative fusion with low-dimensional physical features of steel wire rope tension (kN) and temperature and humidity (℃ / %RH) processed by a neural network fusion model. A multi-dimensional feature matrix containing "time-space-physical quantity-semantic" is generated through a feature weight dynamic allocation algorithm, supporting five types of target identification: a safety equipment identification module detects the personnel protection state in conjunction with a visual sensor, a steel wire rope positioning module optimizes the pixel coordinate capture accuracy combined with the tension data fluctuation characteristics, a steel wire rope state analysis module cooperatively judges the wire breakage anomaly through image texture and tension mutation, a steel feature extraction module identifies the material type based on the combination of visual contour and tension change curve, and an edge position calculation module quantifies the overlap of the hoisting object and the well mouth edge through the IoU algorithm and realizes risk prediction combined with trajectory prediction.

[0012] Further, the pre-warning execution mechanism adopts a hierarchical response collaborative design: the audible and visual alarm device adopts a red and yellow dual-color LED warning light and a voice broadcast module integrated architecture, and forms an instruction response closed loop through a GPIO interface and a data receiving and processing unit. The cloud storage system is built based on the Ali Cloud OSS service, and the local decision data and the cloud backup are cooperated through the 4G module to form a "local real-time response + cloud historical tracking" dual-track data management mode.

[0013] Further, the audible and visual alarm device and the multi-target identification module form a linkage response mechanism: when it is identified that the personnel do not wear safety helmets for more than 10 seconds, the voice prompt and yellow warning light are started to alarm cooperatively; when it is determined by the trajectory prediction module that the collision probability of the steel wire rope within 5 seconds exceeds 80%, the buzzer alarm, red warning light flashing and speed reduction instruction are triggered to output cooperatively; when the mechanical overload or environmental anomaly occurs, the differential cooperative response of the mechanical red rotating alarm and the environmental yellow flashing alarm is activated respectively to realize the accurate matching of risk type and alarm mode.

[0014] Further, the multi-target identification and safety pre-warning system adopts a mobile power independent power supply scheme, and realizes the power supply cooperation of the perception layer device and the core processor through a low-power management module to ensure the continuous operation of the system in the absence of power grid, and balances the data acquisition frequency and the endurance ability through an energy consumption dynamic adjustment algorithm.

[0015] Further, the data receiving and processing unit establishes a time stamp alignment mechanism to realize the spatio-temporal cooperation of multi-source data, and performs hierarchical judgment on the fusion data through a dynamic threshold algorithm: when it is detected that the steel wire rope tension exceeds the threshold value of 147kN, the historical data of the tension sensor and the visual image are verified in linkage, and the mechanical overload response is triggered after confirmation; when the environmental temperature and humidity exceeds the preset safety range, the environmental anomaly protocol is started combined with the multi-depth sensor data trend analysis to avoid false alarm of a single sensor.

[0016] The multi-target recognition and safety warning system of the deep vertical shaft hoisting system has the following beneficial effects:

[0017] Through the collaborative design of the multi-source heterogeneous sensor fusion architecture, the image perception of the visual sensor, the physical quantity monitoring of the side pressure type tension sensor, and the environmental perception of the temperature and humidity sensor are deeply integrated, and the time stamp alignment technology is used to realize the spatio-temporal correlation analysis of the three types of data. The fluctuation of the steel wire rope tension captured by the physical sensor and the trajectory features of the visual sensor form a state verification closed loop. The gradient distribution data of the temperature and humidity sensor provides multi-dimensional basis for environmental abnormality judgment. The three work together to break through the one-sidedness of traditional single sensor monitoring, completely eliminate the monitoring blind area in the complex environment of deep vertical shaft, and realize precise control of the running state of the hoisting system;

[0018] Relying on the adaptive collaboration of lightweight algorithms and hardware platforms, the YOLOv11n model and the multi-modal neural network fusion model are deployed on the Intel Core I5 embedded development board, and the parallel operation of visual high-dimensional feature extraction and physical low-dimensional feature analysis is realized through dynamic allocation of hardware resources. The cross-modal feature fusion mechanism between models makes the five special recognition functions (personnel equipment detection, steel wire rope positioning, broken wire recognition, material classification, and edge overlap calculation) form data complementation. The tension data and visual texture features work together to improve the accuracy of steel wire rope state judgment, and the temperature and humidity parameters provide environmental reference for personnel safety equipment wearing necessity, providing multi-dimensional and highly reliable precise data support for safety monitoring;

[0019] Based on the closed-loop collaborative mechanism of dynamic trajectory prediction and multi-level response, the data receiver and processor input the fused tension trend, visual trajectory and environmental parameters into the trajectory prediction module, realizing the advance prediction of the motion path of the steel wire rope within 5 seconds. When the collision probability is more than 80% or the IoU value is greater than 0.3, the system automatically links the warning executive mechanism, and through the collaborative output of the hierarchical warning (voice prompt, buzzer alarm, color light flicker) of the sound and light alarm device and the device speed reduction instruction, a complete closed loop of "perception-analysis-decision-execution" is built. Compared with traditional manual monitoring, the risk response time is shortened by 3-5 seconds, and major safety hazards such as collision and overload are avoided in time;

[0020] In terms of system collaborative optimization of low cost and high adaptability, the mobile power independent power supply scheme and the edge computing architecture are adopted. The core hardware realizes cost control through functional modular design, and the total cost is less than 1000 yuan. The standby power consumption is reduced by 40% by using low-power management module, which perfectly adapts to the deep vertical shaft non-power grid construction scene. At the same time, the cloud storage system and the local warning executive mechanism work together, only upload the key video clips 30 seconds before and after the alarm event, which reduces the bandwidth consumption by more than 90% compared with full video transmission, while ensuring data traceability and greatly saving deployment and operation cost;

[0021] Through the millisecond multidimensional collaborative response system, when mechanical overload (tension > 147kN), personnel violation (not wearing equipment for more than 10 seconds), collision risk or environmental abnormalities are detected, the data receiver and processor immediately link the sensor secondary verification (such as synchronous verification of the visual image of the steel wire rope deformation when the tension is abnormal), and after confirmation, the four-level response (power cut-off, speed reduction, voice prompt, sound and light alarm, short message push) is triggered within 200ms. The collaborative linkage of each functional component realizes the essential safety leap from "after-the-fact passive recording" to "before-the-fact active interception", significantly reducing the safety accident rate of the deep vertical shaft lifting system. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is the system framework diagram of the present application;

[0023] Figure 2 is the system arrangement schematic diagram of the embodiment of the present application.

[0024] In the figure: 1, 2 are hoists; 3, 4 are side pressure type tension sensors; 5, 6 are steel wire ropes; 7 is a pulley; 8 is a truss; 9 is a high-definition camera; 10 is a temperature and humidity sensor; 11 is a data receiver and processor; 12 is an uphole platform; 13 is a downhole construction platform; 14 is a vertical shaft wall. DETAILED DESCRIPTION

[0025] The present application will be further described below in conjunction with specific embodiments, which are further illustrations of the principles of the present application and do not limit the present application in any way. The same or similar technologies as the present application do not exceed the scope of protection of the present application.

[0026] Embodiment:

[0027] As shown in Figure 1 , Figure 2 , the present embodiment is applied to a multi-target recognition and safety warning system of a deep vertical shaft lifting system, which comprises: side pressure type tension sensors 3, 4 respectively arranged on the steel wire ropes 5, 6 pulled by the hoists 1, 2, a temperature and humidity sensor 10 arranged on the vertical shaft wall 14, a visual sensor 9 arranged on the vertical shaft portal truss 8, a data receiver and processor 11 arranged at the wellhead, and a warning execution mechanism;

[0028] The side pressure type tension sensors 3, 4 adopt LCZ-401B-1ST-7-V5-P13 type with a range greater than 15T, are fixed to the main steel wire ropes 5, 6 5 meters away from the hoist drums 1, 2 through bolt clamps, and the sensing end is closely attached to the surface of the steel wire ropes 5, 6, which can collect the tension data of the two steel wire ropes 5, 6 in real time and transmit them to the data receiver and processor 11 through RS485 shielded cable, ensuring stable transmission of the tension signal in the complex electromagnetic environment of the deep vertical shaft;

[0029] The temperature and humidity sensor 10 adopts a COS-03USB type recorder, and is arranged at intervals of 10 meters along the shaft wall 14 support structure, with a total of 5 sensors, the sensor probe is exposed to the support surface by 3-5 cm, avoiding being blocked by the shaft wall 14 attachments, collecting temperature and humidity parameters every minute and uploading through a shielded cable to form a longitudinal distribution monitoring of the shaft environment parameters;

[0030] The visual sensor 9 is a 1080P high-definition camera, which is installed on the center point of the beam of the portal truss 8 through a waterproof support, the lens axis is perpendicular to the shaft opening plane, the field of view covers the shaft opening and the platform 12 area above the shaft, and video streams are collected at a rate of 30 frames per second and transmitted to the data receiving and processor 11 through a gigabit industrial network cable, the lens cover is provided with an anti-fog coating and a dust screen to adapt to the dusty and humid environment in the shaft, and video streams are collected at a rate of 30 frames per second and transmitted to the data receiving and processor 11 through a gigabit industrial network cable, and special training is required for the steel wire to enhance the accuracy of the visual recognition algorithm in positioning the center point;

[0031] The data receiving and processor 11 takes an Intel Core I5-4310M embedded development board as the core, integrates a multi-source data fusion module, a target detection algorithm module and a trajectory prediction module, and the development board is built-in YOLOv11 lightweight model, the model is preloaded with five types of target recognition weight files: the safety equipment recognition module can detect the wearing state of the safety helmet, the reflective vest and the anti-dust mask of the operating personnel, the steel wire positioning module captures the real-time position of the center point of the steel wire 5, 6 through pixel coordinate conversion, the steel wire state analysis module identifies the broken wire and wear and other abnormalities based on image texture recognition, the steel feature extraction module judges the type of hoisted material (steel bar / concrete) through contour matching, and the edge position calculation module quantifies the overlap degree of the hoisted material and the edge of the platform 12 above the shaft by the IoU algorithm;

[0032] Each sensor and the data receiving and processor 11 adopt a hybrid transmission architecture: the by-pass tension sensors 3, 4 and the temperature and humidity sensor 10 are directly connected with the development board through an RS485 interface, and the communication baud rate is set to 9600 bps; the visual sensor 9 is connected with the development board gigabit Ethernet port through a USB-to-network module, and the video stream transmission adopts the RTSP protocol. The development board is built-in timestamp alignment module, and the multi-source data synchronization is realized through a dynamic timestamp alignment algorithm, taking the visual sensor 30 frames per second and about 33 ms interval as the reference, the high-frequency data of the tension is interpolated and smoothed, and the low-frequency data of the temperature and humidity is timestamp mapped and expanded, so that the time synchronization error of the multi-source data is less than 10 ms;

[0033] The multi-source data fusion module uses multimodal feature stitching technology to fuse low-dimensional physical data such as wire rope tension (kN) and temperature / humidity (°C / %RH) with high-dimensional information such as target coordinates, equipment status, and texture features extracted by visual sensors, generating a multi-dimensional state matrix containing "time-space-physical quantity-semantic features". The data receiver and processor 11 establishes multi-level decision rules, triggering a mechanical overload response when the wire rope tension exceeds the 147kN threshold, and initiating an environmental anomaly protocol when the ambient temperature and humidity exceed the preset safety range. The trajectory prediction module predicts the wire rope's movement path within 5 seconds based on the fused data, automatically activating the intervention process when the collision probability exceeds 80% or the IoU value is >0.3.

[0034] In the early warning execution mechanism, the audible and visual alarm device integrates red and yellow dual-color LED warning lights and a voice broadcast module. It is connected to the development board through the GPIO interface and installed in a conspicuous position in the wellhead control room. The cloud storage system is built on Alibaba Cloud OSS service. The development board communicates with the cloud through a 4G module (model EC20) and automatically uploads 30-second video clips before and after the alarm event and sensor abnormal data packets. The video compression format adopts H.265 to reduce data consumption.

[0035] In this embodiment, the system is powered by a 12V lithium battery with a capacity of 20Ah, which is integrated into the protective box of the data receiver and processor 11. The protective box adopts an IP65 waterproof design and is fixed in front of the winches 1 and 2. The contact between the side-pressure tension sensors 3 and 4 and the wire ropes 5 and 6 should be tightened as much as possible to improve the detection accuracy and sensitivity of the sensors. The temperature and humidity sensor 10 is fixed to the concrete surface of the well wall 14 support structure by expansion bolts, and the distance between it and the well wall is maintained at 5cm to prevent vibration interference.

[0036] The visual sensor 9 is installed 5 meters above the wellhead plane, and the lens focal length is set to 8mm to ensure that the field of view covers the entire trajectory of the steel wire ropes 5 and 6 at the wellhead. Targeted training is carried out on the steel wire ropes to enhance the accuracy of the visual recognition algorithm in locating their center point. The development board of the data receiver and processor 11 is pre-installed with the Windows operating system. The target detection algorithm occupies less than 512MB of memory when running, ensuring stability during multi-task parallel processing.

Claims

1. A multi-target identification and safety early warning system for a deep vertical shaft hoisting system; the deep vertical shaft hoisting system includes: Two sets of winches are installed outside the deep vertical shaft opening. Each set of winches is connected to the underground construction platform via steel wire ropes suspended from the vertical shaft gantry truss at the top of the shaft opening. The key feature is that: The multi-target recognition and safety early warning system constructs a full-link collaborative architecture of "perception-fusion-decision-execution", including: a physical quantity perception layer consisting of pressure-side tension sensors distributed on the winch wire rope; an environmental perception layer consisting of multiple temperature and humidity sensors spaced apart on the shaft wall; an image perception layer consisting of vision sensors installed on the shaft gantry truss; a core decision layer consisting of multi-source data receivers and processors deployed at the shaft opening; and a response execution layer consisting of an early warning execution mechanism integrating audible and visual alarms and cloud storage systems. Each perception layer device is integrated with the data receiver and processor through heterogeneous communication links to form a real-time data interaction closed loop. The data receiver and processor drives the early warning execution mechanism to complete hierarchical responses through intelligent decision-making algorithms.

2. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 1, characterized in that: Furthermore, the pressure-side tension sensor uses the LCZ-401B-1ST-7-V5-P13 model with a range of 15T. It is fixed to the wire rope 5 meters away from the winch drum using bolt clamps, forming a rigid contact with the wire rope to achieve dynamic tension capture. The temperature and humidity sensor uses a COS-03USB recorder, which is deployed at 50-meter intervals on the shaft wall support structure to form a longitudinal environmental parameter monitoring zone for the shaft, enabling coordinated perception of temperature and humidity gradients at different depths. The visual sensor is a 1080P high-definition camera, installed from above at the center point of the crossbeam of the gantry truss. Through optimized field-of-view design, it covers the entire area of ​​the shaft platform and wellhead cover, complementing the spatial monitoring range of the physical sensor.

3. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 1, characterized in that: Each sensing layer device adopts a differentiated communication collaboration scheme: the pressure-side tension sensor and the temperature and humidity sensor achieve stable low-speed physical data transmission through anti-interference RS485 shielded cable, and the vision sensor achieves high-speed transmission of high-definition video stream through gigabit network cable. The two types of links achieve timing alignment and protocol conversion through the multi-interface adaptation module of data receiver and processor, ensuring the collaboration of heterogeneous data transmission.

4. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 1, characterized in that: The data receiver and processor uses an Intel Core I5-4310M embedded development board as its core, integrating a multi-source data fusion module, a target detection algorithm module, and a trajectory prediction module to form a collaborative decision-making hub. The development board has a built-in YOLOv11 lightweight model and a multimodal neural network fusion model, which achieve parallel operation of the two types of models through dynamic allocation of hardware resources. The multimodal neural network fusion model serves as the core collaborative unit, performing spatiotemporal correlation analysis and joint identification on tension data, temperature and humidity data, and visual features to eliminate errors from a single data source.

5. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 4, characterized in that: After extracting high-dimensional visual features, the YOLOv11 lightweight model is cross-modal collaboratively fused with the low-dimensional physical features of wire rope tension (kN) and temperature and humidity (°C / %RH) processed by the neural network fusion model. A multi-dimensional feature matrix containing "time-space-physical quantity-semantics" is generated through a feature weight dynamic allocation algorithm, supporting the recognition of five types of targets: the safety equipment recognition module and the visual sensor work together to detect the protective status of personnel; the wire rope positioning module combines tension data fluctuation features to optimize the pixel coordinate capture accuracy; the wire rope status analysis module judges wire breakage abnormalities through image texture and tension change; the rebar feature extraction module identifies material type based on visual contour and tension change curve; and the edge position calculation module quantifies the overlap between the hoisted object and the wellhead edge through the IoU algorithm and combines it with trajectory prediction to achieve risk prediction.

6. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 1, characterized in that: The early warning execution mechanism adopts a hierarchical response and collaborative design: the audible and visual alarm device adopts an integrated architecture of red and yellow dual-color LED warning lights and voice broadcast module, and forms a command response closed loop with data receiver and processor through GPIO interface; the cloud storage system is built on Alibaba Cloud OSS service, and realizes the collaboration between local decision data and cloud backup through 4G module, forming a dual-track data management mode of "local real-time response + cloud historical traceability".

7. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 6, characterized in that: The sound and light alarm device and the multi-target recognition module form a linkage response mechanism: when a person is detected not wearing a safety helmet for 10 seconds, a voice prompt and a yellow warning light are activated in tandem; when the trajectory prediction module determines that the probability of collision with the wire rope within 5 seconds exceeds 80%, a buzzer alarm, a flashing red warning light, and a speed reduction command are triggered in tandem; in case of mechanical overload or environmental abnormality, a differentiated collaborative response is activated for a mechanical red rotation alarm and an environmental yellow flashing warning, respectively, to achieve precise matching of risk type and alarm mode.

8. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 1, characterized in that: The multi-target recognition and safety early warning system is powered independently by a mobile power supply. It achieves power supply coordination between the perception layer devices and the core processor through a low-power management module, ensuring continuous operation of the system in scenarios without a power grid. At the same time, it balances data acquisition frequency and battery life through a dynamic energy consumption adjustment algorithm.

9. The multi-target identification and safety early warning system for deep vertical shaft hoisting systems according to claim 1, characterized in that: The data receiver and processor establish a timestamp alignment mechanism to achieve spatiotemporal collaboration of multi-source data. The fused data is classified and judged through a dynamic threshold algorithm: when the wire rope tension is detected to exceed the 147kN threshold, the historical data of the tension sensor and the visual image are immediately linked for verification. After confirmation, the mechanical overload response is triggered. When the ambient temperature and humidity exceed the preset safety range, the environmental anomaly protocol is activated by combining the trend analysis of multi-depth sensor data to avoid false alarms from a single sensor.

Images