An artificial intelligence-based portal crane safety monitoring management system

Abstract

The application discloses a gantry crane safety monitoring management system based on artificial intelligence, and belongs to the technical field of crane safety monitoring. The system comprises a data acquisition module, a video acquisition module, a wireless transmission module, an artificial intelligence analysis module and a remote monitoring terminal. The data acquisition module acquires crane operation data and standardizes and encapsulates the crane operation data, and the video acquisition module completes global video acquisition of work. Various types of data are encrypted and transmitted to the artificial intelligence analysis module through an industrial-grade wireless network bridge. The module relies on an edge intelligent gateway, calls a cloud multi-modal large model through a standard interface, carries out data flow analysis and video flow analysis, and outputs safety early warning and linkage control instructions in stages. The remote monitoring terminal realizes remote centralized management and control, data query and report generation. The application provides a more perfect and intelligent gantry crane safety monitoring management system based on artificial intelligence.

Description

Technical Field

[0001] This invention relates to the field of crane safety monitoring technology, specifically to an artificial intelligence-based safety monitoring and management system for gantry cranes. Background Technology

[0002] Gantry cranes are core equipment for lifting heavy objects in ports, docks, freight yards, and industrial and mining enterprises. Their operational safety directly affects the personal safety of operators, equipment availability, and production efficiency. Current traditional gantry crane safety monitoring systems have many technical shortcomings, and existing publicly available gantry crane safety monitoring and management technologies also have significant limitations, as detailed below: Firstly, existing technologies, such as the patent with publication number CN208814530U, disclose a safety monitoring and management system for gantry cranes. This system collects basic operating parameters by setting up sensors such as wind speed, weight, and pressure, and combines them with cameras and upper-level monitoring components to achieve basic data monitoring and display. However, this type of technology can only achieve simple collection of basic parameters such as lifting weight, wind speed, and pressure, and provide over-limit alarms, resulting in limited monitoring data. Secondly, existing monitoring systems generally rely on passive alarm mechanisms with preset thresholds, lacking the ability to deeply learn and analyze equipment operation data and videos. Taking the CN208814530U patent as an example, it only achieves basic early warning through the linkage of sensor data and host computer, and cannot achieve early fault prediction; at the same time, the existing technology does not introduce artificial intelligence algorithms, cannot adapt to complex and ever-changing working environments, and is difficult to accurately identify dynamic risks (such as personnel intrusion and violations of operating procedures).

[0003] The present invention aims to overcome the shortcomings of the existing technology and provide a more complete and intelligent artificial intelligence-based safety monitoring and management system for gantry cranes. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention discloses an artificial intelligence-based safety monitoring and management system for gantry cranes, comprising a data acquisition module, a video acquisition module, a wireless transmission module, an artificial intelligence analysis module, and a remote monitoring terminal.

[0005] Preferably, the signal output terminals of the data acquisition module and the video acquisition module are electrically connected to the wireless transmission module, the wireless transmission module is electrically connected to the artificial intelligence analysis module, and the artificial intelligence analysis module is electrically connected to the remote monitoring terminal.

[0006] Preferably, the data acquisition module includes a PLC acquisition unit, which is used to acquire system status data, trolley operation data, hoist operation data, and main hook operation data of the gantry crane, and encapsulate them into a structured safety monitoring data package.

[0007] Preferably, the video acquisition module is equipped with multiple industrial-grade explosion-proof cameras for acquiring real-time video streams from the gantry crane's cab, electrical room, and work area.

[0008] Preferably, the wireless transmission module adopts a point-to-point networking mode and uses an industrial-grade wireless bridge to complete the encrypted wireless transmission of sensor data packets and video streams, adapting to outdoor open-air and strong electromagnetic interference lifting operation environments.

[0009] Preferably, the artificial intelligence analysis module is configured with an edge intelligent gateway, which calls a cloud-based multimodal large model through a standardized interface. The cloud-based multimodal large model simultaneously performs two types of intelligent analysis: 1. Data Stream Analysis: Continuously learns the timing characteristics of load changes, current fluctuations, voltage deviations, and inverter operation data to accurately identify potential problems such as main hook overload, motor overload, and power supply phase loss; 2. Video stream analysis: Based on multimodal visual understanding, it can detect potential hazards in real time, such as track workers not wearing safety helmets and drivers using mobile phones while on duty; It can predict equipment failures, identify operational risks and determine risk levels, and output local early warnings, interlocking controls and remote push instructions.

[0010] Preferably, the remote monitoring terminal includes a computer client with functions such as real-time video preview, historical data retrieval, alarm information push, and automatic generation of equipment health reports.

[0011] Compared with existing conventional monitoring devices and similar publicly disclosed patents, this invention has the following outstanding and substantial beneficial technical effects: 1. By collecting data from the original control system of the crane, including system status data, trolley operation data, hoist operation data, and main hook operation data, the problem of single monitoring data in traditional methods is completely solved; 2. Continuously learn load changes, current fluctuations, voltage deviations, and inverter parameter timing characteristics to accurately identify potential hazards such as main hook overload, motor overload, and power supply phase loss; 3. Based on multimodal visual understanding, it can detect potential hazards in the work area in real time, such as personnel not wearing safety helmets and drivers using mobile phones while on duty; 4. Relying on the powerful multimodal computing power of cloud-based large models, there is no need to deploy large-scale AI equipment locally. It achieves proactive prediction and intelligent identification through lightweight API calls, upgrading from traditional passive alarms to pre-emptive prevention and control, and significantly reducing the probability of safety accidents in lifting operations. Attached Figure Description

[0012] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of the structure of an artificial intelligence-based safety monitoring and management system for gantry cranes in one embodiment of the present invention; Figure 2 This is a schematic diagram of the workflow of the artificial intelligence analysis module in one embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] Reference Figure 1 As shown, an artificial intelligence-based safety monitoring and management system for gantry cranes includes a data acquisition module, a video acquisition module, a wireless transmission module, an artificial intelligence analysis module, and a remote monitoring terminal. The signal output terminals of the data acquisition module and the video acquisition module are electrically connected to the wireless transmission module, the wireless transmission module is electrically connected to the artificial intelligence analysis module, and the artificial intelligence analysis module is electrically connected to the remote monitoring terminal. In this embodiment, the data acquisition module uses a Siemens 1200 PLC, which is connected to the original electrical control system PLC of the gantry crane to read system status data, trolley operation data, gantry operation data, and main hook operation data from the original PLC.

[0015] Further system status data includes main power circuit breaker, main power contactor, main power phase sequence protection, main power voltage imbalance, main power undervoltage, main power overvoltage, main power overcurrent, foot switch, emergency stop pressed, electrical anti-shake start indicator, wind speed over-warning, alarm bell, wind speed over-speed alarm, windproof brake normal, wind speed sensor fault, windproof controller power supply fault, windproof controller coil fault, windproof brake release status, anchor brake (released) status, rail clamp (released) status, wheel-side brake (released) status, door (not closed) magnetic status, laser rangefinder collision avoidance device data, current wind speed, main power three-phase voltage A, main power three-phase voltage B, main power three-phase voltage C, main power three-phase current A, main power three-phase current B, main power three-phase current C, main power power, and main power total power.

[0016] Further data on the trolley's operation includes: trolley left travel limit, trolley right travel limit, trolley left pre-deceleration limit, trolley right pre-deceleration limit, trolley inverter fault, trolley encoder fault (front), trolley encoder fault (rear), trolley working brake status, trolley thermal relay fault, trolley operating master control status, trolley working brake fault, trolley position, trolley inverter fault ID, trolley inverter output frequency, trolley inverter output voltage, trolley inverter output current, and trolley motor running time.

[0017] Further data on the trolley's operation includes: trolley forward limit, trolley rear limit, trolley forward pre-deceleration limit, trolley rear pre-deceleration limit, trolley inverter fault, trolley encoder fault (left), trolley encoder fault (right), trolley working brake status, trolley thermal relay fault, trolley operating master control status, trolley working brake fault, trolley position, trolley inverter fault ID, trolley inverter output frequency, trolley inverter output voltage, trolley inverter output current, and trolley motor running time.

[0018] Further main hook operation data includes main hook rising limit, main hook falling limit, main hook rising pre-deceleration limit, main hook falling pre-deceleration limit, main hook inverter fault, main hook light load, main hook overload, main hook encoder fault (1), main hook encoder fault (2), main hook weighing sensor fault, main hook working brake status, main hook thermal relay fault, main hook master control status, main hook brake fault, main hook inverter fault ID, main hook position, main hook gross weight, main hook tare weight, main hook inverter output frequency, main hook inverter output voltage, main hook inverter output current, main hook cumulative lifting weight, and main hook motor running time.

[0019] Furthermore, the data acquisition module collects system status data, trolley operation data, trolley operation data, and main hook operation data, and encapsulates them into a structured safety monitoring data package.

[0020] The structured safety monitoring data packet uses a data structure consisting of a message header, system status data, trolley operation data, gantry operation data, and main hook operation data.

[0021] The message header further includes the vehicle-mounted CMS version, crane ID, message number, message type, data length, current working time, and cumulative working time.

[0022] In this embodiment, the video acquisition module uses Hikvision cameras, with a total of eight cameras installed in the crane cab, power distribution room, trolley room, trolley room, both sides of the end beam, and both sides of the trolley travel track. The data is then transmitted to the artificial intelligence analysis module via a wireless transmission module.

[0023] The wireless transmission module is installed on the end beam of the crane and the roof of the office. It adopts an industrial-grade GU-5800 wireless bridge to receive the structured sensor data packets from the data acquisition module and the compressed video stream from the video acquisition module in real time. After unified encryption and encapsulation, the data is uploaded to the artificial intelligence analysis module in real time through point-to-point or point-to-multipoint wireless bridge links.

[0024] Reference Figure 2 As shown, in one specific embodiment, the artificial intelligence analysis module uses an industrial edge smart gateway as its local core. It calls a preset cloud-based multimodal large model through a standardized encrypted interface to perform parallel analysis on data streams and video streams, thereby achieving intelligent identification and hierarchical control of potential hazards such as main hook overload, power phase loss, workers not wearing safety helmets, and drivers playing with mobile phones on duty.

[0025] First, the edge intelligent gateway locally performs data parsing, video decoding, image enhancement, and keyframe extraction, providing structured input data for subsequent cloud-based analysis. The gateway, carrying a unique API-Key and Token identity credentials, synchronously submits structured time-series data and on-site monitoring keyframe video to the cloud-based multimodal large-scale model via an HTTPS encrypted interface. The cloud-based multimodal large-scale model then performs parallel dual-path analysis: In the data flow analysis stage, the cloud-based multimodal large model continuously learns the time-series characteristics of main hook load changes, three-phase current / voltage fluctuations, power supply phase shifts, and inverter operation data to establish a benchmark model for normal equipment operation. When the actual load of the main hook exceeds 90% of the rated value, it is identified as an overload warning; when the load exceeds 105% of the rated value, it is judged as an overload risk. At the same time, through real-time monitoring and phase comparison of three-phase voltage and current, electrical hazards such as power supply phase loss and voltage imbalance can be accurately identified. Combined with cross-validation of motor operating current and temperature trends, it provides early warning of abnormal equipment conditions.

[0026] In the video stream analysis phase, the cloud-based multimodal big data model, based on multimodal visual understanding, performs real-time target detection and behavior recognition on the work site footage: for track workers, by extracting and comparing features of the helmet wearing status, it detects violations such as workers not wearing helmets in real time; for the driver's cab footage, by continuously tracking the driver's posture, hand movements, and facial state, it identifies violations such as the driver using a mobile phone while on duty, leaving the post, and being inattentive; the system timestamps and labels the identified violations and generates alarm evidence fragments.

[0027] After the large model completes the analysis, the risk level, hazard type, confidence level, and handling suggestions are transmitted back to the edge gateway through an encrypted interface. The gateway performs hierarchical control according to the risk level: for general warnings, local audible and visual prompts are triggered and alarm information is pushed to remote terminals; for more serious or emergency warnings, deceleration, locking, or emergency shutdown commands are output in conjunction with the system, and all data and alarm evidence are stored in the safety data black box to achieve closed-loop control of hazards.

[0028] In one specific embodiment, the remote PC terminal is equipped with a centralized monitoring platform, which supports group management of multiple gantry cranes, real-time video split-screen preview, operation curve plotting, and alarm pop-up highlighting; the system automatically generates equipment health reports, violation statistics reports, and hidden danger rectification ledgers on a daily / weekly / monthly basis, and exports and archives them with one click.

[0029] The present invention also provides a computer-readable storage medium storing a computer program; When the computer program is executed by the processor, it implements the gantry crane safety monitoring and management system as described in any embodiment of this application.

[0030] Storage media include USB flash drives, solid-state drives, industrial control computer built-in storage chips, server disk arrays, etc., which can independently store programs and historical monitoring data, facilitating equipment migration, program upgrades, and data backup and archiving.

[0031] Compared with existing conventional monitoring devices and similar publicly disclosed patents, this invention has the following outstanding and substantial beneficial technical effects: 1. By collecting data from the original control system of the crane, including system status data, trolley operation data, hoist operation data, and main hook operation data, the problem of single monitoring data in traditional methods is completely solved; 2. Continuously learn load changes, current fluctuations, voltage deviations, and inverter parameter timing characteristics to accurately identify potential hazards such as main hook overload, motor overload, and power supply phase loss; 3. Based on multimodal visual understanding, it can detect potential hazards in the work area in real time, such as personnel not wearing safety helmets and drivers using mobile phones while on duty; 4. Relying on the powerful multimodal computing power of cloud-based large models, there is no need to deploy large-scale AI equipment locally. It achieves proactive prediction and intelligent identification through lightweight API calls, upgrading from traditional passive alarms to pre-emptive prevention and control, and significantly reducing the probability of safety accidents in lifting operations.

[0032] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A safety monitoring and management system for gantry cranes based on artificial intelligence, characterized in that, It includes a data acquisition module, a video acquisition module, a wireless transmission module, an artificial intelligence analysis module, and a remote monitoring terminal. The signal output terminals of the data acquisition module and the video acquisition module are electrically connected to the wireless transmission module. The wireless transmission module is electrically connected to the artificial intelligence analysis module, and the artificial intelligence analysis module is electrically connected to the remote monitoring terminal.

2. The system according to claim 1, characterized in that, The data acquisition module includes a PLC acquisition unit, which is used to collect system status data, trolley operation data, hoist operation data, and main hook operation data of the gantry crane, and encapsulate them into a structured safety monitoring data package.

3. The system according to claim 1, characterized in that, The video acquisition module is equipped with multiple industrial-grade explosion-proof cameras to acquire real-time video streams from the gantry crane's cab, electrical room, and work area.

4. The system according to claim 1, characterized in that, The wireless transmission module adopts a point-to-point networking mode and uses an industrial-grade wireless bridge to complete the encrypted wireless transmission of sensor data packets and video streams.

5. The system according to claim 1, characterized in that, The artificial intelligence analysis module is configured with an edge intelligent gateway, which calls a preset cloud multimodal large model through a standardized interface. The cloud multimodal large model simultaneously performs two types of intelligent analysis, including equipment fault prediction, operation risk identification and risk level classification, and outputs local early warning, interlocking control and remote push instructions. The two types of intelligent analysis include data stream analysis and video stream analysis. Data stream analysis is used to continuously learn the timing characteristics of load changes, current fluctuations, voltage deviations, and inverter operation data to accurately identify potential hazards such as main hook overload, motor overload, and power supply phase loss. Video stream analysis is used to detect potential hazards in real time, such as track workers not wearing safety helmets and drivers using mobile phones while on duty, based on multimodal visual understanding.

6. The system according to claim 1, characterized in that, The remote monitoring terminal includes a computer client for real-time video preview, historical data retrieval, alarm information push, and automatic generation of equipment health reports.

Images