Bridge type hoisting machinery evaluation method, system and equipment based on multi-source information fusion

Abstract

The invention relates to the technical field of hoisting machinery safety monitoring and intelligent operation and maintenance, in particular to a bridge type hoisting machinery evaluation method, system and equipment based on multi-source information fusion, and the system comprises a distributed multi-source sensing network, a local edge computing node and a remote cloud analysis platform. The distributed multi-source sensing network is deployed on a main beam structural member, an end beam structural member, a hoisting mechanism, a cart running mechanism and a trolley running mechanism of the bridge crane; the local edge computing node is in communication connection with the distributed multi-source sensor network; and the remote cloud analysis platform is in communication connection with the local edge computing node and is used for receiving the structured data packet, executing deep analysis based on multi-source information fusion and outputting a health state evaluation report and decision suggestions including a diagnosis conclusion, a prediction result and a maintenance suggestion. Through the arrangement, multi-dimensional information synchronous perception and edge cloud collaborative intelligent processing can be realized, and real-time early warning and depth prediction capabilities are realized.

Description

Technical Field

[0001] This application relates to the field of safety monitoring and intelligent operation and maintenance technology for lifting machinery, and in particular to an evaluation method, system and equipment for bridge crane machinery based on multi-source information fusion. Background Technology

[0002] Bridge cranes are key material handling equipment in industrial production. They operate under heavy loads, frequent starts and stops, and complex working conditions for extended periods. As a result, structural components are prone to fatigue damage and deformation, and transmission mechanisms are susceptible to wear and failure. Their safety status directly affects production safety and efficiency.

[0003] The safety status assessment of existing bridge cranes mainly relies on two technical approaches: one is periodic manual inspection, which has a long inspection cycle, is highly subjective, and is difficult to capture early dynamic deterioration signals during equipment operation; the other is online monitoring based on a single or a few types of sensors, which has limited information dimensions and can usually only achieve simple threshold over-limit alarms, lacking the ability to intelligently fuse and analyze multi-source information and predict status. Specifically, current safety status assessments of bridge cranes suffer from the following shortcomings: Firstly, monitoring methods are incomplete; single-type sensors cannot simultaneously perceive multiple physical field states such as stress, vibration, deformation, and temperature, leading to distorted assessments of the overall health status of the equipment and a high rate of missed detections for complex faults. Secondly, traditional monitoring solutions suffer from large response delays, failing to provide immediate on-site alarms for sudden anomalies and lacking the ability for in-depth analysis based on multi-source data fusion, making it difficult to achieve accurate fault diagnosis and predictive assessment of remaining lifespan. Furthermore, the common practice of directly uploading all raw data to a central server for centralized processing results in high network bandwidth pressure and long data analysis delays, which not only fails to meet the needs of real-time safety monitoring but also restricts the effective deployment and application of complex intelligent analysis algorithms. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a bridge crane machinery evaluation method, system, and equipment based on multi-source information fusion, which can achieve multi-dimensional information synchronous perception, edge-cloud collaborative intelligent processing, and real-time early warning and deep prediction capabilities.

[0005] In a first aspect, embodiments of this application provide a bridge crane machinery evaluation system based on multi-source information fusion, including: A distributed multi-source sensor network is deployed on the main beam structural components, end beam structural components, hoisting mechanism, trolley traveling mechanism, and trolley traveling mechanism of the bridge crane. The distributed multi-source sensor network is used to synchronously collect stress, vibration, geometric deformation, and temperature field information of the bridge crane in operation, forming multi-modal operating state raw data. A local edge computing node is communicatively connected to the distributed multi-source sensor network. The local edge computing node is used to perform real-time processing and preliminary analysis on the raw data of the multimodal operating status, and generate a structured data packet containing feature data and preliminary analysis results. The real-time processing and preliminary analysis include feature extraction, threshold comparison and trend warning. A remote cloud-based analysis platform is communicatively connected to the local edge computing node. The remote cloud-based analysis platform is used to receive the structured data packets, perform deep analysis based on multi-source information fusion, and output a health status assessment report and decision recommendations that include diagnostic conclusions, prediction results, and maintenance suggestions. The deep analysis includes comprehensive fault diagnosis and remaining life prediction.

[0006] According to some embodiments of the first aspect of this application, the distributed multi-source sensor network includes a structure monitoring module, an mechanism monitoring module, and a unified data acquisition unit; The structural monitoring module is used to monitor the structural status of the main beam structural components and end beam structural components of the bridge crane. The structural monitoring module includes a fiber optic strain sensor group, a deformation monitoring unit, and a tilt monitoring unit. The mechanism monitoring module is used to monitor the status of the hoisting mechanism, trolley traveling mechanism and gantry traveling mechanism of the bridge crane. The mechanism monitoring module includes a vibration sensor group, a running impact monitoring unit, a position monitoring unit and a temperature rise monitoring unit. The unified data acquisition unit is connected to the structure monitoring module and the mechanism monitoring module respectively. The unified data acquisition unit is used to synchronously acquire, convert analog to digital and encapsulate the output signals of the sensors of each module in the structure monitoring module and the mechanism monitoring module.

[0007] According to some embodiments of the first aspect of this application, the sensors in the fiber optic strain sensor group are respectively fixed to the key stress points of the main beam structural member at the mid-span, quarter-span and end sections by bonding, and are connected to the unified data acquisition unit through series armored optical cables; The deformation monitoring unit includes a laser ranging device installed at a fixed reference point and a high-precision tilt sensor on the main beam structural member. The measurement data from the laser ranging device and the high-precision tilt sensor are fused to obtain the real-time deflection of the main beam structural member under load. The tilt monitoring unit includes two fixed inclinometers, which are installed on the top of the end beam structures on both sides of the bridge crane. The tilt monitoring unit is used to monitor the abnormal levelness of the trolley track or the tilt of the trolley body corresponding to the trolley running mechanism based on the measurement data of the two fixed inclinometers.

[0008] According to some embodiments of the first aspect of this application, the vibration sensor group includes a plurality of triaxial acceleration sensors, which are respectively fixed to the surface of the reducer housing of the hoisting mechanism, the motor bearing seat of the hoisting mechanism, and the housing near the gear transmission part of the hoisting mechanism by magnetic attraction or bolts. The operational impact monitoring unit includes an acceleration sensor installed on the trolley drive wheel set bearing housing of the trolley traveling mechanism, the trolley travel wheel set bearing housing of the trolley traveling mechanism, and the connection point of the lifting device frame. The position monitoring unit includes rotary encoders respectively installed on the drive shafts of the trolley running mechanism and the trolley running mechanism. The position monitoring unit also includes travel limit position sensors arranged at both ends of the trolley track corresponding to the trolley running mechanism and the trolley track corresponding to the trolley running mechanism. The temperature rise monitoring unit includes a temperature sensor attached near the brake friction pad of the lifting mechanism, a temperature measuring point on the motor winding of the lifting mechanism, a temperature sensor on the surface of the gearbox of the lifting mechanism, and at least one infrared thermal imager aligned with key heat-generating components in the lifting mechanism, the trolley traveling mechanism, and the trolley traveling mechanism.

[0009] According to some embodiments of the first aspect of this application, the local edge computing node is used to perform the following operations: Receive raw multimodal operating status data from the distributed multi-source sensor network; The original data of the multimodal operating state is preprocessed to obtain standardized preprocessed data. The preprocessing includes timestamp alignment, unit normalization and signal filtering. From the standardized preprocessed data, time-domain statistical features and frequency-domain spectral features are extracted in real time according to a preset sliding time window; The time-domain statistical features and the frequency-domain spectral features are compared with the corresponding preset safe operating thresholds. If either the time-domain statistical features or the frequency-domain spectral features exceed the corresponding safe operating threshold, a corresponding alarm event is generated and a local audible and visual alarm signal is triggered. Calculate the short-term rate of change of the preset feature values ​​used to characterize the state trend in the time-domain statistical features and frequency-domain spectral features, and generate early warning information when the short-term rate of change exceeds the preset trend threshold; The time-domain statistical features, frequency-domain spectral features, alarm events, and early warning information are encapsulated into the structured data packet and sent to the remote cloud analysis platform via a wireless or wired communication link.

[0010] According to some embodiments of the first aspect of this application, the remote cloud analytics platform is used to perform the following operations: Receive and store the structured data packets uploaded from the local edge computing node corresponding to the bridge crane; A multi-source information fusion algorithm is invoked to correlate and comprehensively analyze stress characteristics, vibration characteristics, deformation characteristics, and temperature characteristics within the same time period. The fault diagnosis model trained based on historical data performs pattern recognition on the fused stress characteristics, vibration characteristics, deformation characteristics and temperature characteristics, and outputs specific fault types, fault location information and corresponding confidence levels. Based on the accumulated load spectrum data and structural fatigue damage model, the remaining fatigue life of the main beam structural component of the bridge crane is predicted. Based on the historical trend of the mechanism's operating status characteristic data and the wear and degradation model, the remaining service life of the transmission components in the hoisting mechanism, the trolley traveling mechanism, and the trolley traveling mechanism is predicted. The health status assessment report and decision-making recommendations are output through a visual interface.

[0011] Secondly, embodiments of this application provide a bridge crane machinery evaluation method based on multi-source information fusion, applied to the bridge crane machinery evaluation system based on multi-source information fusion provided in the first aspect embodiment, the method comprising: By synchronously collecting stress, vibration, geometric deformation and temperature field information of bridge cranes under operating conditions through a distributed multi-source sensor network, raw data of multi-modal operating conditions are formed. The raw data of the multimodal operation status is processed and preliminarily analyzed in real time by local edge computing nodes, and a structured data package containing feature data and preliminary analysis results is generated. The real-time processing and preliminary analysis include feature extraction, threshold comparison and trend warning. The structured data packets are received by a remote cloud analysis platform, and a deep analysis based on multi-source information fusion is performed. The platform outputs a health status assessment report and decision recommendations that include diagnostic conclusions, prediction results and maintenance suggestions. The deep analysis includes comprehensive fault diagnosis and remaining life prediction.

[0012] According to some embodiments of the second aspect of this application, the real-time processing and preliminary analysis of the raw data of the multimodal operating status through a local edge computing node includes: Time-frequency analysis is performed on the vibration signals in the original data of the multimodal operation state to extract vibration state characteristic parameters, including effective values, peak factor, kurtosis index and specific frequency band energy. The strain signals in the original data of the multimodal operation state are subjected to cyclic counting and stress amplitude statistics to obtain stress spectrum data; The deformation signal in the original data of the multimodal operation state is calculated to obtain the offset and springback residual amount relative to the unloaded reference, and the offset and springback residual amount of the unloaded reference are used as deformation characteristic values. The vibration state characteristic parameters, the stress spectrum data, and the deformation characteristic values ​​are compared in real time with preset vibration intensity standards, allowable stress standards, and maximum allowable deflection standards. When any of the vibration state characteristic parameters exceeds the vibration intensity standard, or the maximum stress amplitude in the stress spectrum data exceeds the allowable stress standard, or the maximum offset in the deformation characteristic values ​​exceeds the maximum allowable deflection standard, an immediate over-limit alarm is executed. The slope of the preset key indicators in the vibration state characteristic parameters, stress spectrum data and deformation characteristic values ​​changes over time, and early warnings are issued for the trend of continuous deterioration.

[0013] According to some embodiments of the second aspect of this application, the step of receiving the structured data packet through a remote cloud analysis platform, performing deep analysis based on multi-source information fusion, and outputting a health status assessment report and decision recommendations including diagnostic conclusions, prediction results, and maintenance suggestions includes: Spatial correlation and feature-level fusion are performed on the time-aligned vibration spectrum features, stress cycle information, local temperature data, and overall deformation data from the structured data package to generate a fused feature vector; A deep learning classification model trained using labeled historical fault data is used to classify the fused feature vectors and output fault diagnosis results containing fault type identifiers, fault location information and corresponding confidence levels. Based on the actual working load history of the bridge crane, the modified fatigue cumulative damage theory is used to update and calculate the safe remaining cycle number of the bridge crane's metal structure, which is then used as the predicted value for structural life. By combining a reliability model, the long-term evolution trend of vibration energy of the target transmission component is analyzed, and the time when the target transmission component reaches the functional failure threshold is estimated as a predicted value of the mechanism's lifespan. Based on the predicted structural lifespan, the predicted mechanism lifespan, and the fault diagnosis results, the overall risk level of the machine is assessed, and decision recommendations including targeted maintenance plans are generated.

[0014] Thirdly, embodiments of this application provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement: the bridge crane machinery evaluation method based on multi-source information fusion as described in the first aspect above.

[0015] The beneficial effects of this application are reflected in: 1. A distributed multi-source sensor network deployed on the main beam, end beams and various operating mechanisms is used to synchronously collect four types of information: stress, vibration, geometric deformation and temperature field, forming raw data of multimodal operating status. This fundamentally expands the monitoring dimension from single to comprehensive, providing a complete data foundation for subsequent multi-source information fusion analysis and effectively overcoming the problem of inaccurate assessment caused by one-sided information. 2. An edge-cloud collaborative architecture is adopted, consisting of local edge computing nodes and a remote cloud analysis platform. The edge nodes are responsible for real-time processing and preliminary analysis (feature extraction, threshold comparison, and trend warning), enabling local instant alarms for sudden anomalies and solving the real-time problem. The remote cloud analysis platform is responsible for in-depth analysis based on multi-source information fusion, outputting comprehensive fault diagnosis and remaining life prediction, thus elevating the system's function from simple alarms to intelligent diagnosis and prediction, systematically solving the problem of lack of foresight. 3. Edge nodes process the raw data and only upload the refined structured data packets, which greatly reduces the network transmission pressure. At the same time, real-time processing (alarms) with high requirements is deployed at the edge, while analysis (diagnosis and prediction) that requires big data and complex models is deployed in the cloud. This achieves reasonable allocation and efficient collaboration of computing resources and solves the drawbacks of long processing latency and low efficiency in the traditional mode from the perspective of system architecture.

[0016] This application, through this configuration, enables multi-dimensional information synchronous perception, edge-cloud collaborative intelligent processing, and combines real-time early warning and deep prediction capabilities. Attached Figure Description

[0017] Figure 1 A schematic diagram of the connection relationships in a bridge crane machinery evaluation system based on multi-source information fusion, provided for the first aspect of this application; Figure 2 This is a partial structural schematic diagram of a bridge crane provided in the first aspect embodiment of this application; Figure 3 A schematic diagram of another part of the structure of the bridge crane provided in the first aspect embodiment of this application; Figure 4 This is a schematic diagram of the structure of the trolley running mechanism provided in the first aspect embodiment of this application; Figure 5 This is a flowchart illustrating the bridge crane machinery evaluation method based on multi-source information fusion provided in the second aspect embodiment of this application; Figure 6 This is a schematic diagram of the process for real-time processing and preliminary analysis via a local edge computing node, provided in the second aspect embodiment of this application. Figure 7This is a schematic diagram of the process of receiving structured data packets through a remote cloud analysis platform, provided in the second aspect embodiment of this application; Figure 8 This is a schematic diagram of the structure of the electronic device provided in the third aspect embodiment of this application.

[0018] Figure label: Main beam structural component 110, end beam structural component 120, trolley traveling mechanism 130, trolley traveling mechanism motor 131, trolley frame 140, trolley traveling mechanism 141, hoisting mechanism 142, trolley track 143, trolley frame motor 144, reducer 145, drum 146, brake 147, driver's cab 150, guardrail 160, hook head 170. Detailed Implementation

[0019] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0020] Bridge cranes are key material handling equipment in industrial production. They operate under heavy loads, frequent starts and stops, and complex working conditions for extended periods. As a result, structural components are prone to fatigue damage and deformation, and transmission mechanisms are susceptible to wear and failure. Their safety status directly affects production safety and efficiency.

[0021] The safety status assessment of existing bridge cranes mainly relies on two technical approaches: one is periodic manual inspection, which has a long inspection cycle, is highly subjective, and is difficult to capture early dynamic deterioration signals during equipment operation; the other is online monitoring based on a single or a few types of sensors, which has limited information dimensions and can usually only achieve simple threshold over-limit alarms, lacking the ability to intelligently fuse and analyze multi-source information and predict status. Specifically, current safety status assessments of bridge cranes suffer from the following shortcomings: Firstly, monitoring methods are incomplete; single-type sensors cannot simultaneously perceive multiple physical field states such as stress, vibration, deformation, and temperature, leading to distorted assessments of the overall health status of the equipment and a high rate of missed detections for complex faults. Secondly, traditional monitoring solutions suffer from large response delays, failing to provide immediate on-site alarms for sudden anomalies and lacking the ability for in-depth analysis based on multi-source data fusion, making it difficult to achieve accurate fault diagnosis and predictive assessment of remaining lifespan. Furthermore, the common practice of directly uploading all raw data to a central server for centralized processing results in high network bandwidth pressure and long data analysis delays, which not only fails to meet the needs of real-time safety monitoring but also restricts the effective deployment and application of complex intelligent analysis algorithms.

[0022] The following description, in conjunction with the accompanying drawings, details a method, system, and equipment for evaluating bridge crane machinery based on multi-source information fusion, provided by the embodiments of this application, through specific implementations and application scenarios.

[0023] To address the aforementioned problems, the first aspect of this application proposes a bridge crane machinery evaluation system based on multi-source information fusion. The embodiments of this application will be further described below with reference to the accompanying drawings.

[0024] Reference Figure 1 The first aspect of this application provides a bridge crane machinery evaluation system based on multi-source information fusion, including a distributed multi-source sensor network, local edge computing nodes, and a remote cloud analysis platform. The distributed multi-source sensor network is deployed on the main beam structure 110, end beam structure 120, hoisting mechanism 142, trolley traveling mechanism 130, and trolley traveling mechanism 141 of the bridge crane. The distributed multi-source sensor network is used to synchronously collect stress, vibration, geometric deformation, and temperature field information of the bridge crane during operation, forming multi-modal operating state raw data. The local edge computing nodes and the distributed multi-source sensor network... The system employs a multi-source sensor network communication connection. Local edge computing nodes are used to perform real-time processing and preliminary analysis of raw data on multimodal operating status, generating structured data packets containing feature data and preliminary analysis results. Real-time processing and preliminary analysis include feature extraction, threshold comparison, and trend early warning. A remote cloud analysis platform communicates with the local edge computing nodes to receive structured data packets, perform deep analysis based on multi-source information fusion, and output a health status assessment report and decision recommendations containing diagnostic conclusions, prediction results, and maintenance suggestions. Deep analysis includes comprehensive fault diagnosis and remaining life prediction.

[0025] In some embodiments, a distributed multi-source sensor network is used to synchronously collect stress, vibration, geometric deformation, and temperature field information of the bridge crane under operating conditions at a sampling frequency of not less than 100Hz, and to ensure the time synchronization of data acquisition through a unified timing signal, forming raw data of multimodal operating conditions. Local edge computing nodes are connected to the distributed multi-source sensor network via industrial Ethernet or a 5G industrial private network to perform real-time processing and preliminary analysis of the raw data of multimodal operating conditions, generating structured data packets containing feature data and preliminary analysis results. Real-time processing and preliminary analysis include at least time-domain and frequency-domain feature extraction, instant comparison based on preset safety thresholds, and trend analysis and early warning based on short-term historical data. A remote cloud analysis platform is connected to the local edge computing node via the Internet or an enterprise private network to receive structured data packets, perform deep analysis based on multi-source information fusion, and output a health status assessment report and decision recommendations in the form of a visualized report, including quantitative diagnostic conclusions, time-axis-based prediction results, and actionable maintenance suggestions. Deep analysis includes comprehensive fault diagnosis based on machine learning models and remaining life prediction based on physical models.

[0026] In some embodiments, refer to Figures 2 to 4 The main beam structure 110 and the end beam structure 120 are rigidly connected by high-strength bolts or welding to form the main bridge structure of the crane. The main beam structure 110 spans across both sides of the factory building. The trolley traveling mechanism 130 (including the trolley traveling mechanism motor 131, wheel set, etc.) is installed below or at the end of the end beam structure 120, driving the entire crane to run longitudinally along the trolley track laid on the factory building columns. The trolley track 143 is fixedly laid on the upper surface of the end beam structure 120, parallel to the main beam structure 110. The trolley traveling mechanism 141 (including the trolley frame, trolley frame motor 144, traveling wheels, etc.) is supported on the trolley track 143 by the traveling wheels and can move laterally back and forth along the track. The hoisting mechanism 142 (including a motor, reducer 145, drum 146, brake 147, etc.) is installed as a whole on the trolley frame of the trolley traveling mechanism 141 and moves with the trolley. The lifting device (including the hook 170) is connected to the drum 146 of the hoisting mechanism 142 through a wire rope to achieve lifting. The driver's cab 150 is fixedly suspended on one side or below the end beam structure 120, or installed on the trolley traveling mechanism 141 and moves with the trolley, providing the operator with a control and observation position. The guardrail 160 is usually set along both sides of the walkway or maintenance passage above the main beam structure 110 and the end beam structure 120. Its core function is to prevent operators or maintenance personnel from falling from heights, and it is a mandatory safety device.

[0027] The beneficial effects of this application are reflected in: 1. A distributed multi-source sensor network deployed on the main beam, end beams and various operating mechanisms is used to synchronously collect four types of information: stress, vibration, geometric deformation and temperature field, forming raw data of multimodal operating status. This fundamentally expands the monitoring dimension from single to comprehensive, providing a complete data foundation for subsequent multi-source information fusion analysis and effectively overcoming the problem of inaccurate assessment caused by one-sided information. 2. An edge-cloud collaborative architecture is adopted, consisting of local edge computing nodes and a remote cloud analysis platform. The edge nodes are responsible for real-time processing and preliminary analysis (feature extraction, threshold comparison, and trend warning), enabling local instant alarms for sudden anomalies and solving the real-time problem. The remote cloud analysis platform is responsible for in-depth analysis based on multi-source information fusion, outputting comprehensive fault diagnosis and remaining life prediction, thus elevating the system's function from simple alarms to intelligent diagnosis and prediction, systematically solving the problem of lack of foresight. 3. Edge nodes process the raw data and only upload the refined structured data packets, which greatly reduces the network transmission pressure. At the same time, real-time processing (alarms) with high requirements is deployed at the edge, while analysis (diagnosis and prediction) that requires big data and complex models is deployed in the cloud. This achieves reasonable allocation and efficient collaboration of computing resources and solves the drawbacks of long processing latency and low efficiency in the traditional mode from the perspective of system architecture.

[0028] This application, through this configuration, enables multi-dimensional information synchronous perception, edge-cloud collaborative intelligent processing, and combines real-time early warning and deep prediction capabilities.

[0029] Reference Figure 1 It is understandable that the distributed multi-source sensor network includes a structural monitoring module, a mechanism monitoring module, and a unified data acquisition unit. The structural monitoring module is used to monitor the structural status of the main beam structural component 110 and the end beam structural component 120 of the bridge crane. The structural monitoring module includes a fiber optic strain sensor group, a deformation monitoring unit, and a tilt monitoring unit. The mechanism monitoring module is used to monitor the status of the hoisting mechanism 142, the trolley traveling mechanism 130, and the trolley traveling mechanism 141 of the bridge crane. The mechanism monitoring module includes a vibration sensor group, a running impact monitoring unit, a position monitoring unit, and a temperature rise monitoring unit. The unified data acquisition unit is connected to the structural monitoring module and the mechanism monitoring module respectively. The unified data acquisition unit is used to synchronously acquire, convert analog to digital, and encapsulate the output signals of the sensors in each module of the structural monitoring module and the mechanism monitoring module.

[0030] It should be noted that the structural monitoring module includes a fiber optic strain sensor group for measuring micro-strain, a deformation monitoring unit for measuring macro-deformation, and a tilt monitoring unit for measuring spatial attitude; the mechanism monitoring module is used to monitor the dynamic and thermodynamic states of the hoisting mechanism 142, the trolley traveling mechanism 130, and the trolley traveling mechanism 141 of the bridge crane. The mechanism monitoring module includes a vibration sensor group for monitoring high-frequency vibration, a running impact monitoring unit for monitoring low-frequency impact, a position monitoring unit for monitoring kinematic parameters, and a temperature rise monitoring unit for monitoring thermal load; the unified data acquisition unit is connected to the structural monitoring module and the mechanism monitoring module via shielded cables or optical fibers. The unified data acquisition unit integrates a multi-channel synchronous acquisition card and an embedded processor, which is used to synchronously acquire the output signals of the sensors in each module of the structural monitoring module and the mechanism monitoring module with time stamp recording, perform high-precision analog-to-digital conversion, and encapsulate data in accordance with a predetermined communication protocol (such as Modbus TCP or OPC UA).

[0031] Understandably, the sensors in the fiber optic strain sensor group are fixed to the key stress points of the main beam structural member 110 at the mid-span, quarter-span, and end sections by bonding, and are connected to a unified data acquisition unit through a series of armored optical cables. The deformation monitoring unit includes a laser rangefinder and a high-precision tilt sensor installed on the main beam structural member 110, respectively. The measurement data from the laser rangefinder and the high-precision tilt sensor are fused to obtain the real-time deflection of the main beam structural member 110 under load. The tilt monitoring unit includes two fixed inclinometers installed on the top of the end beam structural members 120 on both sides of the bridge crane. The tilt monitoring unit is used to monitor the abnormal levelness of the trolley track or the tilt of the trolley body corresponding to the trolley running mechanism 130 based on the measurement data of the two fixed inclinometers.

[0032] Specifically, the sensors in the fiber optic strain sensor group are fixed to the mid-span lower flange surface, the quarter-span lower flange surface, and the stress concentration area at the connection between the end section and the end beam of the main beam structural member 110 using epoxy resin adhesive. They are connected to the fiber optic demodulator in the unified data acquisition unit via armored optical cables with metal sheaths in series. The deformation monitoring unit includes a laser ranging device based on the laser phase ranging principle, which is fixed at reference points on the ground or on the factory column, and a high-precision tilt sensor at the center position of the mid-span upper flange of the main beam structural member 110. The measurement data from the laser rangefinder and the high-precision tilt sensor are processed by coordinate transformation and data fitting to obtain the real-time deflection curve of the main beam structure 110 under load from no load to full load. The tilt monitoring unit includes two dual-axis fixed inclinometers, which are respectively installed at the top center of the end beam structure 120 on both sides of the bridge crane. The tilt monitoring unit is used to calculate the relative tilt angle based on the measurement data of the two fixed inclinometers and monitor the levelness abnormality or car body tilt caused by uneven settlement or wear of the trolley track corresponding to the trolley running mechanism 130.

[0033] For example, the structural condition monitoring setup is as follows: (1) Stress monitoring: Fiber optic strain gauges are used. Three sets of sensors are arranged at the mid-span section of the main beam structural member 110, two sets of sensors are arranged at the quarter-span (L / 4) section, and two sets of sensors are arranged at the connection with the end beam structural member 120 to comprehensively monitor the strain distribution and fatigue hotspots of the main beam under bending and shear loads; (2) Deflection and deformation monitoring: A laser radar and high-precision inclinometer fusion system is used. One set of measurement units is arranged at the mid-span of the main beam structural member 110, and one set of measurement units is arranged at each of the two ends. The static deflection and dynamic deformation curves of the main beam are calculated by data fusion; (3) Tilt monitoring: A box-type fixed inclinometer is used. Four measuring points are arranged at the top of the end beam structural members 120 on both sides (i.e., the support foundation of the trolley running mechanism 130) to monitor the tilt of the entire vehicle bridge frame caused by uneven settlement or wear of the trolley track.

[0034] Understandably, the vibration sensor group includes multiple triaxial acceleration sensors, which are respectively fixed to the surface of the reducer 145 housing of the hoisting mechanism 142, the motor bearing housing of the hoisting mechanism 142, and the housing near the gear transmission part of the hoisting mechanism 142 by magnetic attraction or bolts; the running impact monitoring unit includes acceleration sensors installed on the trolley drive wheel bearing housing of the trolley traveling mechanism 130, the trolley traveling wheel bearing housing of the trolley traveling mechanism 141, and the connection point of the lifting device upper frame; the position monitoring unit includes acceleration sensors respectively installed on the trolley traveling mechanism 130. The position monitoring unit includes a rotary encoder on the drive shaft of the trolley running mechanism 141, and stroke limit position sensors installed at both ends of the trolley track corresponding to the trolley running mechanism 130 and the trolley track 143 corresponding to the trolley running mechanism 141; the temperature rise monitoring unit includes temperature sensors attached near the friction pad of the brake 147 of the lifting mechanism 142, the temperature measuring point of the motor winding of the lifting mechanism 142, the surface of the gearbox of the lifting mechanism 142, and at least one infrared thermal imager aligned with the key heat-generating components in the lifting mechanism 142, the trolley running mechanism 130, and the trolley running mechanism 141.

[0035] Specifically, the vibration sensor group includes multiple IEPE type triaxial acceleration sensors, which are respectively fixed to the surface directly above the input / output shaft bearing housing of the reducer 145 housing of the hoisting mechanism 142, the motor drive end and non-drive end bearing housing of the hoisting mechanism 142, and the housing near the high-speed gear transmission part of the hoisting mechanism 142 by magnetic attraction or bolts; the running impact monitoring unit includes low-frequency response acceleration sensors or force sensors installed on the outside of the trolley drive wheel set bearing housing of the trolley traveling mechanism 130, the outside of the trolley traveling wheel set bearing housing of the trolley traveling mechanism 141, and the connection point between the upper frame of the hoisting device and the wire rope; the position monitoring unit includes the drive motors of the trolley traveling mechanism 130 and the trolley traveling mechanism 141 respectively, which are installed via flexible couplings. The absolute rotary encoder on the output shaft and the position monitoring unit also include magnetic or photoelectric travel limit position sensors installed at the extreme positions at both ends of the trolley track 143 corresponding to the trolley traveling mechanism 130 and the trolley track 143 corresponding to the trolley traveling mechanism 141; the temperature rise monitoring unit includes embedded or surface-mount temperature sensors, which are respectively attached near the brake shoe friction plate of the brake 147 of the hoisting mechanism 142, the PT100 temperature measuring point pre-embedded in the motor winding of the hoisting mechanism 142, and the surface of the gearbox of the hoisting mechanism 142 near the bearing; the temperature rise monitoring unit also includes at least one infrared thermal imager fixed to the maintenance platform or driver's cab 150 and aimed at the key heat-generating components in the brake 147 of the hoisting mechanism 142, the motor of the trolley traveling mechanism 130, and the motor of the trolley traveling mechanism 141.

[0036] In some embodiments, the sensors (fiber optic strain gauges, inclinometers, and tiltmeters) of the structural monitoring module are directly fixed to designated surfaces of the main beam structural member 110 and the end beam structural member 120; the sensors (accelerometers, encoders, and temperature sensors) of the mechanism monitoring module are fixed to specific components of the hoisting mechanism 142 (reducer 145 housing, motor base, and brake 147), the trolley traveling mechanism 130 (drive wheel set), and the trolley traveling mechanism 141 (drive wheel set and drive shaft) according to the monitoring object; all sensors are connected to a unified data acquisition unit located in the driver's cab 150 or on the walkway of the main beam structural member 110 via cables.

[0037] For example, the deployment of mechanism operation status monitoring is as follows: (1) Vibration monitoring: Using triaxial piezoelectric accelerometers, two sets of sensors are arranged on the surface of the housing near the input shaft and output shaft of the reducer of the hoisting mechanism; one set of sensors is arranged on the drive end bearing seat and non-drive end bearing seat of the drive motor of the hoisting mechanism; three sets of sensors are arranged on the outer wall of the housing corresponding to the gear meshing area of ​​the reducer to capture gear meshing impact and bearing vibration. (2) Operation impact monitoring: Using accelerometers, sensors are arranged on the bearing seats of the four drive wheel groups and the key points of the two balance beams of the trolley running mechanism 130; one sensor is arranged on the bearing seats of the two traveling wheels of the trolley running mechanism 141 and the upper frame connection point of the lifting device to monitor the impact load when starting and stopping the operation and crossing the rail joint. (3) Position and stroke monitoring: Rotary encoders and stroke limit position sensors are used. Rotary encoders are installed on the drive shafts of the trolley running mechanism 130 and the trolley running mechanism 141; at the same time, a positioning reference point sensor is set up every 10 meters on the trolley track, and two sets of limit sensors are set up at the front and rear travel limit positions of the trolley track. (4) Temperature rise monitoring: Temperature and humidity sensors and infrared thermal imagers are used. Two temperature measuring points are arranged near the friction pads of each brake 147; three temperature measuring points are preset inside the windings of the hoisting mechanism motor; one oil temperature sensor is arranged in the oil sump of each reducer; at the same time, an infrared thermal imager is used to monitor the surface temperature field of key heat-generating components such as the brake 147, motor housing and reducer housing.

[0038] Understandably, local edge computing nodes are used to perform the following operations: Step S21: Receive raw multimodal operating status data from the distributed multi-source sensor network.

[0039] Step S22: Preprocess the raw data of multimodal operation status to obtain standardized preprocessed data. The preprocessing includes timestamp alignment, unit normalization and signal filtering.

[0040] In this step, the raw data of multimodal operation is preprocessed to obtain standardized preprocessed data. Preprocessing includes timestamp alignment based on GPS or network protocols and conversion of voltage / frequency signals into engineering units (e.g., m / s). 2 Unit normalization of (MPa, ℃) and signal filtering using low-pass filters or wavelet denoising algorithms.

[0041] Step S23: Extract time-domain statistical features and frequency-domain spectral features from the standardized preprocessed data in real time according to a preset sliding time window.

[0042] In this step, time-domain statistical features (such as mean, root mean square value, peak value, and kurtosis) and frequency-domain spectral features (such as spectral centroid, characteristic frequency amplitude, and sideband energy) are extracted in real time from the standardized preprocessed data according to a preset sliding time window (such as a short window of 1 to 10 seconds and a long window of 1 to 10 minutes).

[0043] Step S24: Compare the time-domain statistical features and frequency-domain spectral features with the corresponding preset safe operating thresholds. If either the time-domain statistical features or the frequency-domain spectral features exceed the corresponding safe operating threshold, generate the corresponding alarm event and trigger the local audible and visual alarm signal.

[0044] In this step, the time-domain statistical features and frequency-domain spectral features are compared with the corresponding preset safe operating thresholds set according to the crane design specifications and historical operating data. If either the time-domain statistical feature or the frequency-domain spectral feature exceeds the corresponding safe operating threshold, a corresponding alarm event is generated, which includes the feature name, the over-limit value, and the occurrence time, and the locally connected audible and visual alarm is triggered to issue an audible and visual alarm signal.

[0045] Step S25: Calculate the short-term change rate of the preset feature values ​​used to characterize the state trend in the time domain statistical features and frequency domain spectral features. If the short-term change rate exceeds the preset trend threshold, generate early warning information.

[0046] In this step, the short-term rate of change of the preset characteristic values ​​(such as effective vibration value and temperature value) used to characterize the state trend in the time domain statistical characteristics and frequency domain spectral characteristics is calculated based on the data of the most recent 1 hour. If the short-term rate of change exceeds the preset trend threshold for multiple consecutive calculation cycles, early warning information of state deterioration is generated.

[0047] Step S26: Encapsulate the time-domain statistical features, frequency-domain spectral features, alarm events, and early warning information into a structured data packet, and send it to a remote cloud analysis platform via a wireless or wired communication link.

[0048] In this step, time-domain statistical features, frequency-domain spectral features, alarm events, and early warning information are encapsulated into structured data packets according to a predefined JSON or Protobuf format and sent to a remote cloud analysis platform via a 4G / 5G wireless network or industrial fiber optic communication link.

[0049] Understandably, the remote cloud analytics platform is used to perform the following operations: Step S31: Receive and store the structured data packets uploaded from the local edge computing node corresponding to the bridge crane.

[0050] In this step, structured data packets uploaded from the local edge computing nodes corresponding to one or more bridge cranes are received and stored in the time-series database.

[0051] Step S32: Invoke the multi-source information fusion algorithm to perform correlation and comprehensive analysis on the stress characteristics, vibration characteristics, deformation characteristics and temperature characteristics within the same time period.

[0052] In this step, a multi-source information fusion algorithm is invoked to perform spatiotemporal alignment and comprehensive analysis of stress characteristics, vibration characteristics, deformation characteristics and temperature characteristics within the same time period, thereby uncovering the coupling relationship between different physical quantities.

[0053] Step S33: Based on the fault diagnosis model trained with historical data, perform pattern recognition on the fused stress characteristics, vibration characteristics, deformation characteristics and temperature characteristics, and output the specific fault type, fault location information and corresponding confidence level.

[0054] In this step, a fault diagnosis model, such as a convolutional neural network or support vector machine trained on a large amount of historical data, performs pattern recognition on the fused stress characteristics, vibration characteristics, deformation characteristics, and temperature characteristics, and outputs specific fault types (such as gear pitting, bearing wear, and structural cracks), fault location information (such as the 145 high-speed shaft of the hoisting reducer and the 110 mid-span of the main beam structural component) and corresponding confidence levels.

[0055] Step S34: Based on the accumulated load spectrum data and structural fatigue damage model, predict the remaining fatigue life of the main beam structural component 110 of the bridge crane.

[0056] In this step, based on the load spectrum data obtained by the cumulative rainflow counting method and structural fatigue damage models such as Miner linear cumulative damage or Paris crack propagation, the remaining fatigue life (expressed as the number of working cycles or calendar time) of the main beam structural component 110 of the bridge crane under a specified reliability is predicted.

[0057] Step S35: Based on the historical trend of the mechanism's operating status characteristic data and the wear and degradation model, predict the remaining service life of the transmission components in the hoisting mechanism 142, the trolley traveling mechanism 130, and the trolley traveling mechanism 141.

[0058] In this step, based on the long-term (e.g., monthly, annual) historical trends of the mechanism's operational status characteristics (such as vibration energy and temperature) and wear degradation models such as Weibull distribution or degradation state space, the remaining service life (expressed in hours or mileage) of transmission components such as gears and bearings in the hoisting mechanism 142, the trolley traveling mechanism 130, and the trolley traveling mechanism 141 is predicted. Through a web or mobile visualization interface, a health status assessment report and decision recommendations are output in the form of dashboards, trend charts, and report documents.

[0059] Step S36: Output a health status assessment report and decision recommendations through a visual interface.

[0060] In this step, a health status assessment report and decision recommendations are output in the form of dashboards, trend charts, and report documents through a web or mobile visualization interface.

[0061] Secondly, referring to Figure 5 The second aspect of this application provides a method for evaluating bridge cranes based on multi-source information fusion. This method is applied to the bridge crane evaluation system based on multi-source information fusion provided in the first aspect embodiment. The method for evaluating bridge cranes based on multi-source information fusion includes, but is not limited to, the following steps: Step S100: The stress, vibration, geometric deformation and temperature field information of the bridge crane under the operation state are collected synchronously through a distributed multi-source sensor network to form raw data of multi-modal operation state.

[0062] In this step, stress, vibration, geometric deformation and temperature field information of the bridge crane under operating conditions are collected synchronously in a complete working cycle of the crane (such as hoisting-running-lowering) through a distributed multi-source sensor network, forming raw data of multi-modal operating conditions.

[0063] Step S200: The raw data of multimodal operation status is processed and preliminarily analyzed in real time through the local edge computing node, and a structured data package containing feature data and preliminary analysis results is generated.

[0064] In this step, real-time processing and preliminary analysis include feature extraction, threshold comparison, and trend alerts.

[0065] In this step, the raw data of multimodal operation status is processed and preliminarily analyzed in real time within a latency of milliseconds to seconds through local edge computing nodes, and a structured data package containing feature data and preliminary analysis results is generated. The real-time processing and preliminary analysis includes at least feature extraction, threshold comparison and trend warning.

[0066] In step S300, the structured data packet is received through the remote cloud analysis platform, a deep analysis based on multi-source information fusion is performed, and a health status assessment report and decision recommendations containing diagnostic conclusions, prediction results and maintenance suggestions are output.

[0067] In this step, in-depth analysis includes comprehensive fault diagnosis and remaining life prediction.

[0068] In this step, structured data packets are received asynchronously or periodically through a remote cloud analysis platform. Deep analysis based on multi-source information fusion is performed, and a health status assessment report and decision recommendations containing diagnostic conclusions, prediction results, and maintenance suggestions are output. The deep analysis includes at least comprehensive fault diagnosis and remaining life prediction.

[0069] Reference Figure 6 It is understandable that step S200 involves real-time processing and preliminary analysis of the raw data of the multimodal operating status through local edge computing nodes, including but not limited to the following steps: Step 210: Perform time-frequency analysis on the vibration signals in the original data of multimodal operation to extract vibration state characteristic parameters.

[0070] In this step, the vibration state characteristic parameters include RMS value, peak factor, kurtosis index, and specific frequency band energy.

[0071] In this step, the vibration signals in the original data of multimodal operation are subjected to time-frequency analysis such as fast Fourier transform or wavelet transform to extract vibration state characteristic parameters. The vibration state characteristic parameters include RMS value, peak factor, kurtosis index and specific frequency band energy such as gear meshing frequency and bearing fault characteristic frequency.

[0072] Step 220: Perform cycle counting and stress amplitude statistics on the strain signals in the original data of multimodal operation to obtain stress spectrum data.

[0073] In this step, the strain signal in the original data of multimodal operation is subjected to rainflow counting and stress amplitude statistics to obtain stress spectrum data for fatigue life assessment.

[0074] Step 230: Calculate the deformation signal in the original data of multimodal operation state to obtain the offset and springback residual amount relative to the unloaded reference, and use the offset and springback residual amount of the unloaded reference as deformation characteristic values.

[0075] In this step, the deformation signal in the original data of multimodal operation is calculated to obtain the maximum offset relative to the unloaded reference and the springback residual amount after unloading. The offset of the unloaded reference and the springback residual amount are used as deformation characteristic values.

[0076] Step 240: The vibration state characteristic parameters, stress spectrum data and deformation characteristic values ​​are compared with the preset vibration intensity standard, allowable stress standard and maximum allowable deflection standard in real time. When any of the vibration state characteristic parameters exceeds the vibration intensity standard, or the maximum stress amplitude in the stress spectrum data exceeds the allowable stress standard, or the maximum offset in the deformation characteristic value exceeds the maximum allowable deflection standard, an immediate over-limit alarm is executed.

[0077] In this step, the vibration state characteristic parameters, stress spectrum data, and deformation characteristic values ​​are compared in real time with preset vibration intensity standards based on standards such as ISO 10816, allowable stress standards based on the yield strength of materials, and maximum allowable deflection standards based on design specifications. When any of the vibration state characteristic parameters exceeds the vibration intensity standard, or the maximum stress amplitude in the stress spectrum data exceeds the allowable stress standard, or the maximum offset in the deformation characteristic values ​​exceeds the maximum allowable deflection standard, an immediate over-limit alarm is triggered.

[0078] Step 250: Monitor the slope of the changes of preset key indicators in vibration state characteristic parameters, stress spectrum data and deformation characteristic values ​​over time, and issue early warnings for the trend of continuous deterioration.

[0079] In this step, the slope of the preset key indicators in the vibration state characteristic parameters, stress spectrum data and deformation characteristic values ​​changes over time is monitored. When the slope is continuously positive and exceeds the threshold, an early warning is issued for the trend of continuous deterioration.

[0080] Reference Figure 7 It is understood that step S300 includes, but is not limited to, the following steps: Step S310 involves spatially correlating and feature-level fusing the time-aligned vibration spectrum features, stress cycle information, local temperature data, and overall deformation data from the structured data package to generate a fused feature vector.

[0081] In this step, the time-aligned vibration spectrum features, stress cycle information, local temperature data, and overall deformation data from the structured data package are spatially correlated and feature-level fused based on the physical location relationship of the measuring points to generate a high-dimensional fused feature vector for characterizing the overall state of the equipment.

[0082] Step S320: Using a deep learning classification model trained with labeled historical fault data, classify the fused feature vectors and output fault diagnosis results containing fault type identifiers, fault location information and corresponding confidence levels.

[0083] In this step, a deep neural network (such as CNN-LSTM) classification model trained with labeled historical fault data is used to classify the fused feature vectors and output fault diagnosis results containing fault type identifiers (such as fault codes), fault location information (such as component numbers), and corresponding confidence levels (0-100%).

[0084] Step S330: Based on the actual working load history of the bridge crane, the modified fatigue cumulative damage theory is used to update and calculate the safe remaining cycle number of the bridge crane metal structure as the predicted value of the structural life.

[0085] In this step, based on the actual working load history of the bridge crane, the modified fatigue cumulative damage theory, which considers the effects of load sequence and average stress, is used to update and calculate the safe remaining cycle number of the main beam and other metal structures of the bridge crane, as the predicted value of structural life.

[0086] Step S340: By combining a reliability model, analyze the long-term evolution trend of vibration energy of the target transmission component, estimate the time when the target transmission component reaches the functional failure threshold, and use it as a predicted value for the mechanism's lifespan.

[0087] In this step, by combining the Weibull distribution or Markov chain reliability model, the long-term evolution trend of vibration energy of target transmission components such as the 145 input shaft bearing of the hoisting reducer is analyzed, and the time when the target transmission components reach the preset vibration alarm threshold or functional failure standard is estimated as the predicted value of the mechanism life.

[0088] Step S350: Based on the predicted structural life and mechanical life, and the fault diagnosis results, assess the overall risk level of the machine and generate decision recommendations that include targeted maintenance plans.

[0089] In this step, based on the predicted structural life and mechanism life, as well as the fault diagnosis results, the overall risk level (e.g., low, medium, high) is assessed, and decision recommendations including targeted maintenance plans such as "It is recommended to replace the 145 high-speed shaft bearing of the lifting reducer within the next 30 days" are generated.

[0090] Optionally, such as Figure 8As shown, the third aspect of this application also provides an electronic device 10, including a processor 11 and a memory 12. The memory 12 stores a program or instructions that can run on the processor 11. When the program or instructions are executed by the processor 11, they implement the various processes of the first aspect of the bridge crane machinery evaluation method based on multi-source information fusion, and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0091] It should be noted that the devices in the embodiments of this application include: servers, terminals, or other devices besides terminals.

[0092] The above device structure does not constitute a limitation on the device. The device may include more or fewer components than illustrated, or combine certain components, or arrange different components. For example, the input unit may include a graphics processing unit (GPU) and a microphone, and the display unit may use a liquid crystal display, organic light-emitting diode, or other forms to configure the display panel. The user input unit includes at least one of a touch panel and other input devices. A touch panel is also called a touch screen. Other input devices may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0093] Memory can be used to store software programs and various data. Memory can primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area can store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, memory can include volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM).

[0094] The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly handles operations related to the operating system, user interface, and applications, while the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into the processor.

[0095] This application also provides a readable storage medium storing a program or instructions. When executed by a processor, the program or instructions implement the various processes of the first aspect of the bridge crane machinery evaluation method based on multi-source information fusion, and achieve the same technical effect. To avoid repetition, these will not be described again here. The processor is the processor in the device described above. The readable storage medium includes computer-readable storage media such as ROM, RAM, magnetic disk, or optical disk. It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatus in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in an order different from that described. In addition, features described with reference to certain examples may be combined in other examples.

[0096] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0097] In the description of the embodiments of this application, the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first," "second," "third," or "fourth" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0098] In the description of the embodiments of this application, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

Claims

1. A bridge crane machinery evaluation system based on multi-source information fusion, characterized in that, include: A distributed multi-source sensor network is deployed on the main beam structural components, end beam structural components, hoisting mechanism, trolley traveling mechanism, and trolley traveling mechanism of the bridge crane. The distributed multi-source sensor network is used to synchronously collect stress, vibration, geometric deformation, and temperature field information of the bridge crane in operation, forming multi-modal operating state raw data. A local edge computing node is communicatively connected to the distributed multi-source sensor network. The local edge computing node is used to perform real-time processing and preliminary analysis on the raw data of the multimodal operating status, and generate a structured data packet containing feature data and preliminary analysis results. The real-time processing and preliminary analysis include feature extraction, threshold comparison and trend warning. A remote cloud-based analysis platform is communicatively connected to the local edge computing node. The remote cloud-based analysis platform is used to receive the structured data packets, perform deep analysis based on multi-source information fusion, and output a health status assessment report and decision recommendations that include diagnostic conclusions, prediction results, and maintenance suggestions. The deep analysis includes comprehensive fault diagnosis and remaining life prediction.

2. The bridge crane machinery evaluation system based on multi-source information fusion according to claim 1, characterized in that, The distributed multi-source sensor network includes a structure monitoring module, an mechanism monitoring module, and a unified data acquisition unit. The structural monitoring module is used to monitor the structural status of the main beam structural components and end beam structural components of the bridge crane. The structural monitoring module includes a fiber optic strain sensor group, a deformation monitoring unit, and a tilt monitoring unit. The mechanism monitoring module is used to monitor the status of the hoisting mechanism, trolley traveling mechanism and gantry traveling mechanism of the bridge crane. The mechanism monitoring module includes a vibration sensor group, a running impact monitoring unit, a position monitoring unit and a temperature rise monitoring unit. The unified data acquisition unit is connected to the structure monitoring module and the mechanism monitoring module respectively. The unified data acquisition unit is used to synchronously acquire, convert analog to digital and encapsulate the output signals of the sensors of each module in the structure monitoring module and the mechanism monitoring module.

3. The bridge crane machinery evaluation system based on multi-source information fusion according to claim 2, characterized in that, The sensors in the fiber optic strain sensor group are fixed to the key stress points of the main beam structural member at the mid-span, quarter-span and end sections by bonding, and are connected to the unified data acquisition unit through a series of armored optical cables. The deformation monitoring unit includes a laser ranging device installed at a fixed reference point and a high-precision tilt sensor on the main beam structural member. The measurement data from the laser ranging device and the high-precision tilt sensor are fused to obtain the real-time deflection of the main beam structural member under load. The tilt monitoring unit includes two fixed inclinometers, which are installed on the top of the end beam structures on both sides of the bridge crane. The tilt monitoring unit is used to monitor the abnormal levelness of the trolley track or the tilt of the trolley body corresponding to the trolley running mechanism based on the measurement data of the two fixed inclinometers.

4. The bridge crane machinery evaluation system based on multi-source information fusion according to claim 2, characterized in that, The vibration sensor group includes multiple triaxial acceleration sensors, which are respectively fixed to the surface of the reducer housing of the lifting mechanism, the motor bearing seat of the lifting mechanism, and the housing near the gear transmission part of the lifting mechanism by magnetic attraction or bolts. The operational impact monitoring unit includes an acceleration sensor installed on the trolley drive wheel set bearing housing of the trolley traveling mechanism, the trolley travel wheel set bearing housing of the trolley traveling mechanism, and the connection point of the lifting device frame. The position monitoring unit includes rotary encoders respectively installed on the drive shafts of the trolley running mechanism and the trolley running mechanism. The position monitoring unit also includes travel limit position sensors arranged at both ends of the trolley track corresponding to the trolley running mechanism and the trolley track corresponding to the trolley running mechanism. The temperature rise monitoring unit includes a temperature sensor attached near the brake friction pad of the lifting mechanism, a temperature measuring point on the motor winding of the lifting mechanism, a temperature sensor on the surface of the gearbox of the lifting mechanism, and at least one infrared thermal imager aligned with key heat-generating components in the lifting mechanism, the trolley traveling mechanism, and the trolley traveling mechanism.

5. The bridge crane machinery evaluation system based on multi-source information fusion according to claim 1, characterized in that, The local edge computing node is used to perform the following operations: Receive raw multimodal operating status data from the distributed multi-source sensor network; The original data of the multimodal operating state is preprocessed to obtain standardized preprocessed data. The preprocessing includes timestamp alignment, unit normalization and signal filtering. From the standardized preprocessed data, time-domain statistical features and frequency-domain spectral features are extracted in real time according to a preset sliding time window; The time-domain statistical features and the frequency-domain spectral features are compared with the corresponding preset safe operating thresholds. If either the time-domain statistical features or the frequency-domain spectral features exceed the corresponding safe operating threshold, a corresponding alarm event is generated and a local audible and visual alarm signal is triggered. Calculate the short-term rate of change of the preset feature values ​​used to characterize the state trend in the time-domain statistical features and frequency-domain spectral features, and generate early warning information when the short-term rate of change exceeds the preset trend threshold; The time-domain statistical features, frequency-domain spectral features, alarm events, and early warning information are encapsulated into the structured data packet and sent to the remote cloud analysis platform via a wireless or wired communication link.

6. The bridge crane machinery evaluation system based on multi-source information fusion according to claim 1, characterized in that, The remote cloud analytics platform is used to perform the following operations: Receive and store the structured data packets uploaded from the local edge computing node corresponding to the bridge crane; A multi-source information fusion algorithm is invoked to correlate and comprehensively analyze stress characteristics, vibration characteristics, deformation characteristics, and temperature characteristics within the same time period. The fault diagnosis model trained based on historical data performs pattern recognition on the fused stress characteristics, vibration characteristics, deformation characteristics and temperature characteristics, and outputs specific fault types, fault location information and corresponding confidence levels. Based on the accumulated load spectrum data and structural fatigue damage model, the remaining fatigue life of the main beam structural component of the bridge crane is predicted. Based on the historical trend of the mechanism's operating status characteristic data and the wear and degradation model, the remaining service life of the transmission components in the hoisting mechanism, the trolley traveling mechanism, and the trolley traveling mechanism is predicted. The health status assessment report and decision-making recommendations are output through a visual interface.

7. A method for evaluating bridge crane machinery based on multi-source information fusion, characterized in that, The method, applied to the bridge crane machinery evaluation system based on multi-source information fusion as described in any one of claims 1 to 6, comprises: By synchronously collecting stress, vibration, geometric deformation and temperature field information of bridge cranes under operating conditions through a distributed multi-source sensor network, raw data of multi-modal operating conditions are formed. The raw data of the multimodal operation status is processed and preliminarily analyzed in real time by local edge computing nodes, and a structured data package containing feature data and preliminary analysis results is generated. The real-time processing and preliminary analysis include feature extraction, threshold comparison and trend warning. The structured data packets are received by a remote cloud analysis platform, and a deep analysis based on multi-source information fusion is performed. The platform outputs a health status assessment report and decision recommendations that include diagnostic conclusions, prediction results and maintenance suggestions. The deep analysis includes comprehensive fault diagnosis and remaining life prediction.

8. The bridge crane machinery evaluation method based on multi-source information fusion according to claim 7, characterized in that, The real-time processing and preliminary analysis of the raw data of the multimodal operating status through local edge computing nodes includes: Time-frequency analysis is performed on the vibration signals in the original data of the multimodal operation state to extract vibration state characteristic parameters, which include effective values, peak factor, kurtosis index and specific frequency band energy. The strain signals in the original data of the multimodal operation state are subjected to cyclic counting and stress amplitude statistics to obtain stress spectrum data; The deformation signal in the original data of the multimodal operation state is calculated to obtain the offset and springback residual amount relative to the unloaded reference, and the offset and springback residual amount of the unloaded reference are used as deformation characteristic values. The vibration state characteristic parameters, the stress spectrum data, and the deformation characteristic values ​​are compared in real time with preset vibration intensity standards, allowable stress standards, and maximum allowable deflection standards. When any of the vibration state characteristic parameters exceeds the vibration intensity standard, or the maximum stress amplitude in the stress spectrum data exceeds the allowable stress standard, or the maximum offset in the deformation characteristic values ​​exceeds the maximum allowable deflection standard, an immediate over-limit alarm is executed. The slope of the preset key indicators in the vibration state characteristic parameters, stress spectrum data and deformation characteristic values ​​changes over time, and early warnings are issued for the trend of continuous deterioration.

9. The bridge crane machinery evaluation method based on multi-source information fusion according to claim 7, characterized in that, The process involves receiving the structured data packets via a remote cloud analysis platform, performing deep analysis based on multi-source information fusion, and outputting a health status assessment report and decision recommendations that include diagnostic conclusions, prediction results, and maintenance suggestions. Spatial correlation and feature-level fusion are performed on the time-aligned vibration spectrum features, stress cycle information, local temperature data, and overall deformation data from the structured data package to generate a fused feature vector; A deep learning classification model trained using labeled historical fault data is used to classify the fused feature vectors and output fault diagnosis results containing fault type identifiers, fault location information and corresponding confidence levels. Based on the actual working load history of the bridge crane, the modified fatigue cumulative damage theory is used to update and calculate the safe remaining cycle number of the bridge crane's metal structure, which is then used as the predicted value for structural life. By combining a reliability model, the long-term evolution trend of vibration energy of the target transmission component is analyzed, and the time when the target transmission component reaches the functional failure threshold is estimated as a predicted value of the mechanism's lifespan. Based on the predicted structural lifespan, the predicted mechanism lifespan, and the fault diagnosis results, the overall risk level of the machine is assessed, and decision recommendations including targeted maintenance plans are generated.

10. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements: the bridge crane machinery evaluation method based on multi-source information fusion as described in any one of claims 7 to 9.

Images