A method for fall prevention assessment and prediction of operating status signals for lifting machinery

Abstract

This application relates to the field of crane operation safety technology, and in particular to a method for fall prevention assessment and prediction of crane operation status signals. The method includes collecting multi-source signal features, constructing and verifying an analysis model and a verification model, and obtaining fall prevention risk assessment results. By integrating structural mechanics, dynamic operation, and environmental signals, and employing a fusion approach of statistical, temporal, geometric, and environmental features, dynamic modeling and cross-comparison are achieved based on neural networks, improving the comprehensiveness of monitoring and the accuracy of assessment. This application can optimize data adaptability, improve the false judgment rate, achieve trend-based early warning, and ensure the safe operation of cranes.

Description

Technical Field

[0001] This invention relates to the field of safety monitoring and fault prediction technology for lifting machinery, specifically to a method for assessing and predicting fall prevention based on the operating status signals of lifting machinery. Background Technology

[0002] With the increasing reliance on lifting machinery in construction, port logistics, and mining, its operational safety has become a crucial aspect of ensuring the safety of personnel and property. Fall prevention is a core requirement for lifting machinery safety monitoring. During operations such as lifting heavy objects, luffing, and slewing, structural failures, drive anomalies, braking malfunctions, or environmental interference can lead to serious accidents. Existing methods for monitoring lifting machinery safety have certain limitations: some schemes rely solely on single physical quantities (such as stress and displacement) for data collection and analysis, making it difficult to comprehensively reflect the complex operating conditions of the equipment and limiting their ability to capture the correlation between structural mechanical properties and dynamic operating conditions; while some schemes incorporate multi-source data, they lack efficient data fusion technology, easily leading to data redundancy or missing information, affecting the accuracy of risk assessment. Furthermore, existing models are mostly built based on static operating conditions, failing to fully incorporate lifting equipment-specific data (such as luffing angle, load weight, and braking response characteristics) during training, validation, and improvement, resulting in insufficient adaptability to dynamic parameter changes. Meanwhile, the multi-data cross-comparison method relies heavily on formula calculations and does not fully consider the correspondence and type differences between lifting equipment data, which may lead to a decrease in the accuracy of risk assessment and make it impossible to effectively achieve trend assessment and early warning of fall risk.

[0003] Traditional monitoring methods typically assess the safety status of lifting machinery through periodic inspections or manual patrols, which are time-consuming and have limited ability to detect potential minor faults. While some sensor-based monitoring systems can collect data and issue simple alarms, they fail to integrate data specific to the lifting equipment for in-depth analysis and risk prediction. Therefore, how to integrate multi-source operational data, construct a dynamic model highly correlated with lifting equipment data, and achieve cross-comparison through data correspondence to improve the operational safety of lifting machinery has become an urgent technical problem to be solved. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for assessing and predicting fall prevention based on the operating status signals of lifting machinery. By integrating multi-source operating data, constructing a dynamic model highly correlated with the data of the lifting equipment, and performing cross-comparison based on the data correspondence, the method improves the operating safety of lifting machinery.

[0005] To achieve the above objectives, the present invention provides a method for fall prevention assessment and prediction of operating status signals for lifting machinery, comprising: Collect multi-source signal characteristics during the operation of lifting machinery; A crane operation status analysis model is established based on a neural network using the features of the multi-source signals. A crane operation status verification model is established based on a neural network using the features of the multi-source signals. The fall risk assessment results are obtained by using the lifting machinery operation status analysis model and the lifting machinery operation status verification model.

[0006] The multi-source signal characteristics acquired during the operation of the lifting machinery include: Collect structural mechanical signals during the operation of lifting machinery, including stress tensor components, deformation, and load tension; Collect dynamic operating signals during the operation of lifting machinery, including hoisting speed, luffing speed, slewing speed, braking response time, and vibration speed; Collect position and environmental signals during the operation of lifting machinery, including real-time position coordinates, amplitude angle, slewing angle, wind speed, wind direction, and dust concentration; Based on the structural mechanics signal, dynamic operation signal, pose and environmental signal, the corresponding multi-source signal features are obtained, including statistical features, temporal features, geometric features and environmental features; The structural mechanical signals, dynamic operation signals, pose and environmental signals, and their corresponding multi-source signal characteristics are used as multi-source signal characteristics during the operation of lifting machinery.

[0007] The statistical characteristics of the structural mechanics signal include the maximum stress, the average deformation, and the rate of change of load tension; the time-domain characteristics of the dynamic operation signal include the variance of lifting speed, the average braking response time, and the peak vibration velocity; the geometric characteristics of the pose and environmental signal include the change in amplitude angle and the rate of change of rotation angle; and the environmental characteristics of the environmental signal include the average wind speed, wind direction stability, and the trend of dust concentration change.

[0008] Furthermore, the establishment of a crane operation status analysis model based on a neural network using the multi-source signal features includes: Using the structural mechanics signal, dynamic operation signal, and pose and environment signal from the multi-source signal features, a structural mechanics data analysis module, a dynamic operation data analysis module, and a pose and environment data analysis module are respectively established based on neural networks; The structural mechanics data analysis module, dynamic operation data analysis module, and pose and environmental data analysis module are used as the operation status analysis model for lifting machinery.

[0009] Furthermore, the structural mechanics data analysis module, which utilizes the structural mechanics signals, dynamic operation signals, and pose and environmental signals from the multi-source signal features, is established based on a neural network, including: Historical structural mechanical signal data are extracted based on the statistical characteristics of the structural mechanical signals, including historical maximum stress, historical average deformation, and historical load tensile force change rate. Using the historical maximum stress, historical average deformation, and historical load tensile change rate as inputs, and the current statistical characteristics of the structural mechanics signal as outputs, an initial structural mechanics data analysis module is established based on neural network training. The prediction results of historical structural mechanical signal data are obtained based on the initial structural mechanical data analysis module. Determine whether the predicted result is consistent with the historical structural mechanics signal data. If so, use the initial structural mechanics data analysis module as the structural mechanics data analysis module; otherwise, update the historical structural mechanics signal data and return to the previous steps.

[0010] Furthermore, establishing a crane operation status verification model based on a neural network using the multi-source signal features includes: A dynamic operation data verification module and a pose and environment data verification module are established based on a neural network using the dynamic operation signal, pose and environment signals in the multi-source signal features. The dynamic operation data verification module and the pose and environment data verification module are used as the verification model for the operation status of lifting machinery.

[0011] Furthermore, the dynamic operation signal, pose, and environmental signals from the multi-source signal features are used to establish a dynamic operation data verification module based on a neural network, which includes: Historical dynamic operation signal data is extracted based on the time-domain characteristics of the dynamic operation signal, including historical lifting speed variance, historical braking response time mean and historical vibration speed peak value. Using the historical lifting speed variance, historical braking response time mean, and historical vibration speed peak as inputs, and the current time-domain characteristics of the dynamic operation signal as outputs, an initial dynamic operation data verification module is established based on neural network training. The verification results of historical dynamic operation signal data are obtained based on the initial dynamic operation data verification module. Determine whether the verification result is consistent with the historical dynamic operation signal data. If so, use the initial dynamic operation data verification module as the dynamic operation data verification module; otherwise, update the historical dynamic operation signal data and return to the previous steps.

[0012] Furthermore, the fall risk assessment results obtained using the aforementioned crane operation status analysis model and crane operation status verification model include: Real-time multi-source signal characteristics during the operation of lifting machinery are acquired respectively; By using the real-time multi-source signal feature input to the crane machinery operation status analysis model, real-time structural mechanics data analysis results, real-time dynamic operation data analysis results, and real-time pose and environmental data analysis results are obtained sequentially. The real-time dynamic operation data verification result and the real-time pose and environmental data verification result are obtained by using the real-time multi-source signal feature input to the crane machinery operation status verification model. Determine whether the real-time structural mechanics data analysis result is consistent with the real-time dynamic operation data verification result. If so, proceed with the next steps. Otherwise, output the content that is inconsistent between the real-time structural mechanics data analysis result and the real-time dynamic operation data verification result as the fall risk assessment result. Determine whether the real-time pose and environmental data analysis results are consistent with the real-time pose and environmental data verification results. If they are consistent, output the real-time structural mechanics data analysis results, real-time dynamic operation data analysis results, real-time pose and environmental data analysis results, real-time dynamic operation data verification results, and real-time pose and environmental data verification results as the fall risk assessment results. Otherwise, output the content that is inconsistent between the real-time pose and environmental data analysis results and the real-time pose and environmental data verification results as the fall risk assessment results.

[0013] Compared with the closest existing technology, the present invention has the following advantages: Enhance the comprehensiveness of monitoring: By integrating structural mechanical signals, dynamic operation signals, and position and environmental signals, the complex operating state of the lifting machinery can be fully reflected, and the correlation between structural mechanical performance and dynamic working conditions can be captured. Enhance data fusion capabilities: Employ a multi-dimensional fusion approach that integrates statistical features, temporal features, geometric features, and environmental features to eliminate data redundancy and information gaps, thereby improving the accuracy of risk assessment; Optimize model adaptability: Build a dynamic model based on neural networks, and combine it with the specific data of the lifting equipment during the training and validation process to improve adaptability to changes in dynamic parameters; Improve cross-comparison methods: reduce false positive rates and improve the accuracy of risk assessment by comparing data correspondence, data type consistency and data value rationality from multiple dimensions; Achieve trend assessment and early warning: By combining secondary verification of environmental characteristics with trend analysis, potential risks can be identified in advance to ensure the safe operation of lifting machinery. Attached Figure Description

[0014] Figure 1 This is a flowchart of a method for assessing and predicting the fall prevention of operating status signals for lifting machinery, provided by the present invention. Figure 2 This is a modular schematic diagram of a fall prevention assessment and prediction method for operating status signals of lifting machinery provided by the present invention.

[0015] Figure label: 1. Multi-source signal feature acquisition module; 2. Structural mechanics data analysis module; 3. Dynamic operation data analysis module; 4. Pose and environmental data analysis module; 5. Dynamic operation data verification module; 6. Pose and environmental data verification module; 7. Fall risk assessment result output module. Detailed Implementation

[0016] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1: This invention provides a method for fall prevention assessment and prediction of operating status signals for lifting machinery, such as... Figure 1 As shown, the process includes: collecting multi-source signal features during the operation of lifting machinery; establishing a lifting machinery operation status analysis model based on a neural network using the multi-source signal features; establishing a lifting machinery operation status verification model based on a neural network using the multi-source signal features; and obtaining fall risk assessment results using the lifting machinery operation status analysis model and the lifting machinery operation status verification model.

[0019] like Figure 2 As shown, the system includes a multi-source signal feature acquisition module 1, a structural mechanics data analysis module 2, a dynamic operation data analysis module 3, a pose and environmental data analysis module 4, a dynamic operation data verification module 5, a pose and environmental data verification module 6, and a fall protection risk assessment result output module 7. The following will describe in detail the specific implementation process of each module and their interconnections and collaboration methods.

[0020] First, the multi-source signal feature acquisition module 1 is used to acquire the multi-source signal features during the operation of the lifting machinery. The core task of this module is to acquire various types of signals generated by the lifting machinery during operation, including structural mechanics signals, dynamic operation signals, and pose and environmental signals. For structural mechanics signals, the acquired data includes stress tensor components, deformation, and load tension; for dynamic operation signals, the acquired data includes hoisting speed, luffing speed, slewing speed, braking response time, and vibration velocity; for pose and environmental signals, the acquired data includes real-time pose coordinates, luffing angle, slewing angle, wind speed, wind direction, and dust concentration. These signals are directly acquired through sensor devices installed at key parts of the lifting machinery, such as the hook, boom, and slewing mechanism, to ensure the authenticity and accuracy of the signals. The acquired raw signals were preprocessed to extract statistical, temporal, geometric, and environmental features. Statistical features included maximum stress, mean deformation, and rate of change of tensile load. Temporal features included the variance of lifting speed, mean braking response time, and peak vibration velocity. Geometric features included the change in amplitude angle and the rate of change of slewing angle. Environmental features included mean wind speed, wind direction stability, and dust concentration trends. These features served as input data for subsequent analysis and validation models.

[0021] Next, a crane operation status analysis model is established based on a neural network using the structural mechanics data analysis module 2, dynamic operation data analysis module 3, and pose and environment data analysis module 4. These three modules correspond to the analysis needs of structural mechanics signals, dynamic operation signals, and pose and environment signals, respectively. Taking the structural mechanics data analysis module 2 as an example, its specific implementation process is as follows: First, historical structural mechanics signal data is extracted based on the statistical characteristics of the structural mechanics signals, including historical maximum stress, historical average deformation, and historical load tensile rate of change. This data comes from the historical operation records of the crane. Then, the historical maximum stress, historical average deformation, and historical load tensile rate of change are used as input data, and the current statistical characteristics are used as output data. The neural network is then trained to establish an initial structural mechanics data analysis module. During the training process, the backpropagation algorithm is used to adjust the weight parameters of the neural network so that the prediction results are as close as possible to the actual data. After training is completed, the initial structural mechanics data analysis module is used to predict historical structural mechanics signal data, and it is determined whether the prediction results are consistent with the historical data. If they are consistent, this module is used as the final structural mechanics data analysis module; if they are inconsistent, the historical data is updated and retrained until the consistency requirement is met. The implementation process of dynamic operation data analysis module 3 and pose and environment data analysis module 4 is similar to that of structural mechanics data analysis module 2, which respectively model dynamic operation signals and pose and environment signals.

[0022] After constructing the crane operation status analysis model, a crane operation status verification model needs to be established through the dynamic operation data verification module 5 and the pose and environment data verification module 6. The purpose of these two modules is to cross-validate the results of the analysis model to improve the reliability of the evaluation results. Taking the dynamic operation data verification module 5 as an example, its specific implementation process is as follows: First, historical dynamic operation signal data is extracted based on the time-domain characteristics of the dynamic operation signals, including historical lifting speed variance, historical braking response time mean, and historical vibration velocity peak value. Then, this historical data is used as input data, and the current time-domain characteristics are used as output data. Training is performed based on a neural network to establish the initial dynamic operation data verification module. After training, this module is used to verify the historical dynamic operation signal data and determine whether the verification result is consistent with the historical data. If consistent, this module is used as the final dynamic operation data verification module; if inconsistent, the historical data is updated and retrained until the consistency requirement is met. The implementation process of the pose and environment data verification module 6 is similar to that of the dynamic operation data verification module 5, mainly focusing on modeling pose and environment signals.

[0023] After completing the construction of the analysis model and the verification model, the system proceeds to the fall protection risk assessment result output module 7 to obtain the final fall protection risk assessment result. The workflow of this module is as follows: First, the multi-source signal feature acquisition module 1 acquires real-time multi-source signal features during the operation of the lifting machinery. These features include real-time statistical features, real-time temporal features, real-time geometric features, and real-time environmental features. Then, the real-time multi-source signal features are input into the lifting machinery operation state analysis model to sequentially obtain real-time structural mechanics data analysis results, real-time dynamic operation data analysis results, and real-time pose and environmental data analysis results. Simultaneously, the real-time multi-source signal features are input into the lifting machinery operation state verification model to obtain real-time dynamic operation data verification results and real-time pose and environmental data verification results. Next, a consistency judgment is performed on the analysis results and verification results. First, it is determined whether the real-time structural mechanics data analysis results are consistent with the real-time dynamic operation data verification results. If they are consistent, the consistency between the real-time pose and environmental data analysis results and the real-time pose and environmental data verification results is further determined. If both are consistent, the results of real-time structural mechanics data analysis, real-time dynamic operation data analysis, real-time pose and environmental data analysis, real-time dynamic operation data verification, and real-time pose and environmental data verification will be output as the fall safety risk assessment result. If any one of them is inconsistent, the inconsistent content will be output as the fall safety risk assessment result.

[0024] Throughout the process, the various modules are connected and collaborate via data flow. The multi-source signal feature acquisition module 1 provides the basic data, which is then passed to the structural mechanics data analysis module 2, the dynamic operation data analysis module 3, and the pose and environment data analysis module 4 for analysis and modeling. The output of the analysis model is further passed to the dynamic operation data verification module 5 and the pose and environment data verification module 6 for cross-validation. Finally, the fall risk assessment result output module 7 integrates all analysis and verification results to generate the final fall risk assessment report. This modular design not only improves the system's scalability and flexibility but also ensures the independence and accuracy of each step.

[0025] The above embodiments illustrate the specific implementation process of the present invention, which can effectively improve the comprehensiveness and accuracy of monitoring the operating status of lifting machinery. By integrating multi-source signal features, constructing a dynamic model, and performing cross-comparison, the present invention provides reliable technical assurance for the safe operation of lifting machinery.

[0026] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principle of this invention will be further explained below in conjunction with a specific application scenario.

[0027] In the actual operation of lifting machinery, taking a port logistics scenario as an example, the lifting equipment is performing the task of lifting containers. First, the initial data acquisition is completed through the multi-source signal feature acquisition module 1. Sensor devices are installed on key parts such as the hook, boom, and slewing mechanism to monitor the operating status of the lifting machinery in real time. For example, when the hook is raised, the sensors record changes in the lifting speed; when the boom luffs, the sensors capture the dynamic changes in the luffing angle; simultaneously, environmental sensors detect fluctuations in wind speed and direction. After preprocessing, these raw signals extract statistical features, temporal features, geometric features, and environmental features. For example, the maximum stress value reflects the maximum load borne by the hook, while the average wind speed is used to assess the impact of the environment on the stability of the lifting machinery.

[0028] Next, the structural mechanics data analysis module 2 analyzes the structural mechanics signals of the lifting machinery based on a neural network. Taking the stress tensor components of the hook as an example, the historical maximum stress, historical average deformation, and historical load tensile change rate are extracted as input data, and the current statistical characteristics are used as output data. The weight parameters of the neural network are adjusted through a backpropagation algorithm to make the prediction results as close as possible to the actual operating data. For example, at a certain moment, if the maximum stress of the hook exceeds the normal range of historical data, the system will determine whether there is a potential risk of structural failure based on the trained model. The dynamic operation data analysis module 3 and the pose and environment data analysis module 4 perform similar modeling for dynamic operation signals and pose and environment signals, respectively. For example, the dynamic operation data analysis module 3 determines whether there is a risk of drive abnormality or braking failure in the lifting machinery by analyzing the lifting speed variance and the braking response time mean.

[0029] After the analysis model is constructed, the dynamic operation data verification module 5 and the pose and environment data verification module 6 cross-validate the analysis results. Taking the dynamic operation data verification module 5 as an example, it verifies the current dynamic operation signal based on the historical lifting speed variance, the historical braking response time mean, and the historical vibration velocity peak value. If the verification result is inconsistent with the historical data, the historical data is updated and the model is retrained until the consistency requirements are met. For example, when the braking response time of the crane is significantly prolonged at a certain moment, and the verification module finds that the data deviates from the historical data, the system will further analyze whether there is a potential fault in the braking system.

[0030] Finally, the fall risk assessment output module 7 integrates all analysis and verification results to generate the final fall risk assessment report. For example, under a specific working condition, the system, through real-time structural mechanics data analysis, finds that the maximum stress of the hook is close to the critical value. Simultaneously, dynamic operation data analysis shows an extended braking response time, and posture and environmental data analysis indicates a significant increase in wind speed. By judging the consistency of these results, the system confirms a high fall risk and outputs corresponding warning information. If any analysis result is inconsistent with the verification result—for example, if the real-time posture and environmental data analysis result differs from the real-time posture and environmental data verification result—the system will output the specific inconsistency as a risk warning.

[0031] Throughout the process, collaboration between modules ensures efficient data transmission and processing. The multi-source signal feature acquisition module 1 provides foundational data for subsequent analysis; the structural mechanics data analysis module 2, dynamic operation data analysis module 3, and pose and environmental data analysis module 4 perform modeling and analysis for different types of signals, respectively; the dynamic operation data verification module 5 and the pose and environmental data verification module 6 cross-validate the analysis results; and the fall risk assessment result output module 7 finally integrates all results and generates an assessment report. This modular design not only improves the system's scalability and flexibility but also ensures the independence and accuracy of each step.

[0032] As can be seen from the implementation steps of the specific application scenarios described above, this invention can effectively improve the comprehensiveness and accuracy of monitoring the operating status of lifting machinery by integrating multi-source signal features, constructing a dynamic model, and performing cross-comparison. For example, in the event of a sudden increase in wind speed, the system can combine environmental characteristics with dynamic operating characteristics to detect potential risks in advance and issue warnings, thereby ensuring the safe operation of the lifting machinery.

[0033] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0034] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0035] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0036] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0037] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for fall prevention assessment and prediction of operating status signals for lifting machinery, characterized in that, include: S1. Collect multi-source signal characteristics during the operation of lifting machinery; S2. Utilize the multi-source signal features to establish a crane machinery operation status analysis model based on a neural network; S3. Utilize the multi-source signal features to establish a crane operation status verification model based on a neural network; S4. Obtain the fall risk assessment results using the lifting machinery operation status analysis model and the lifting machinery operation status verification model.

2. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 1, characterized in that, The multi-source signal characteristics acquired during the operation of the lifting machinery include: S1-1. Collect structural mechanical signals during the operation of the lifting machinery, including stress tensor components, deformation, and load tension; S1-2. Collect dynamic operating signals during the operation of the lifting machinery, including hoisting speed, luffing speed, slewing speed, braking response time, and vibration speed; S1-3. Collect the position and environmental signals of the lifting machinery during operation, including real-time position coordinates, amplitude angle, rotation angle, wind speed, wind direction and dust concentration. S1-4. Obtain the corresponding multi-source signal features, including statistical features, temporal features, geometric features, and environmental features, based on the structural mechanical signals, dynamic operation signals, and pose and environmental signals.

3. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 2, characterized in that, The statistical characteristics of the structural mechanics signal include the maximum stress, the mean deformation, and the rate of change of load tension. The time-domain characteristics of the dynamic operation signal include the variance of the lifting speed, the mean braking response time, and the peak vibration velocity. The geometric characteristics of the pose and environmental signal include the change in amplitude angle and the rate of change of rotation angle. The environmental characteristics of the environmental signal include the mean wind speed, wind direction stability, and the trend of dust concentration change.

4. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 1, characterized in that, The model for analyzing the operating status of lifting machinery based on neural networks, utilizing the features of the multi-source signals, includes: S2-1. Using the structural mechanics signal, dynamic operation signal, and pose and environment signal in the multi-source signal features, respectively, establish a structural mechanics data analysis module, a dynamic operation data analysis module, and a pose and environment data analysis module based on a neural network. S2-2, The structural mechanics data analysis module, dynamic operation data analysis module, and pose and environmental data analysis module are used as the operation status analysis model of the lifting machinery.

5. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 4, characterized in that, The structural mechanics data analysis module, which utilizes structural mechanics signals from the multi-source signal features and is based on a neural network, includes: S2-1-1. Extract historical structural mechanical signal data based on the statistical characteristics of the structural mechanical signals, including historical maximum stress, historical average deformation, and historical load tensile force change rate. S2-1-2. Using the historical maximum stress, historical average deformation, and historical load tensile change rate as inputs, and the current statistical characteristics of the structural mechanics signal as outputs, an initial structural mechanics data analysis module is established based on neural network training. S2-1-3. Obtain the prediction results of historical structural mechanical signal data based on the initial structural mechanical data analysis module; S2-1-4. Determine whether the prediction result is consistent with the historical structural mechanics signal data. If so, use the initial structural mechanics data analysis module as the structural mechanics data analysis module. Otherwise, update the historical structural mechanics signal data and return to S2-1-2.

6. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 1, characterized in that, The model for verifying the operating status of lifting machinery based on neural networks, utilizing the features of the multi-source signals, includes: S3-1. Using the dynamic operation signal, pose and environment signal in the multi-source signal features, a dynamic operation data verification module and a pose and environment data verification module are established based on a neural network. S3-2. The dynamic operation data verification module and the pose and environment data verification module are used as the operation status verification model of the lifting machinery.

7. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 6, characterized in that, The dynamic operation signal based on the multi-source signal features is used to establish a dynamic operation data verification module based on a neural network, which includes: S3-1-1. Extract historical dynamic operation signal data based on the time-domain characteristics of the dynamic operation signal, including historical lifting speed variance, historical braking response time mean and historical vibration speed peak value. S3-1-2. Using the historical lifting speed variance, historical braking response time mean and historical vibration speed peak as inputs, and the current time domain characteristics of the dynamic operation signal as outputs, an initial dynamic operation data verification module is established based on neural network training. S3-1-3. Obtain the verification results of historical dynamic operation signal data according to the initial dynamic operation data verification module; S3-1-4. Determine whether the verification result is consistent with the historical dynamic operation signal data. If yes, use the initial dynamic operation data verification module as the dynamic operation data verification module. Otherwise, update the historical dynamic operation signal data and return to S3-1-2.

8. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 1, characterized in that, The fall risk assessment results obtained using the aforementioned crane operation status analysis model and crane operation status verification model include: S4-1. Acquire the real-time multi-source signal characteristics during the operation of the lifting machinery; S4-2. Using the real-time multi-source signal characteristics input into the crane machinery operation status analysis model, the real-time structural mechanics data analysis results, the real-time dynamic operation data analysis results, and the real-time pose and environmental data analysis results are obtained sequentially. S4-3. Using the real-time multi-source signal characteristics input into the crane machinery operation status verification model, the real-time dynamic operation data verification results and the real-time pose and environment data verification results are obtained. S4-4. Determine whether the real-time structural mechanics data analysis result is consistent with the real-time dynamic operation data verification result. If yes, execute S4-5. Otherwise, output the content that is inconsistent between the real-time structural mechanics data analysis result and the real-time dynamic operation data verification result as the fall risk assessment result. S4-5. Determine whether the real-time pose and environmental data analysis results are consistent with the real-time pose and environmental data verification results. If so, output the real-time structural mechanics data analysis results, real-time dynamic operation data analysis results, real-time pose and environmental data analysis results, real-time dynamic operation data verification results, and real-time pose and environmental data verification results as the fall prevention risk assessment results. Otherwise, output the content that is inconsistent between the real-time pose and environmental data analysis results and the real-time pose and environmental data verification results as the fall prevention risk assessment results.

9. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 8, characterized in that, The real-time multi-source signal features are acquired through a multi-source signal feature acquisition module. The real-time multi-source signal features include real-time statistical features, real-time time-domain features, real-time geometric features, and real-time environmental features.

10. The method for fall prevention assessment and prediction of operating status signals for lifting machinery as described in claim 9, characterized in that, The real-time statistical features include the real-time maximum stress, the real-time average deformation, and the real-time load tensile change rate. The real-time time-domain features include the real-time lifting speed variance, the real-time braking response time average, and the real-time vibration velocity peak value. The real-time geometric features include the real-time amplitude angle change and the real-time slewing angle change rate. The real-time environmental features include the real-time average wind speed, the real-time wind direction stability, and the real-time dust concentration change trend.

Images