Monitoring and early warning system and method for crossing blocking working rope

Abstract

The invention relates to a monitoring and early warning system and method for a working rope crossing a blocking net, and belongs to the technical field of rope state monitoring, and the system comprises a data collection and transmission module which is used for collecting multi-source state data of acoustic emission, tension and the like of the working rope and forming a structured data stream; the rope state intelligent analysis module is used for carrying out AE signal depth analysis and multi-modal feature fusion risk assessment based on the data stream, and constructing and updating a rope digital twinborn model; the damage feature and risk knowledge base is used for storing reference information to support intelligent analysis; the early warning decision and response module is used for performing graded early warning and residual service life (RUL) prediction according to the analysis result and knowledge base information and providing maintenance suggestions; the system optimization and self-learning module is used for coordinating system work and self-learning an optimization model and a knowledge base based on operation data; accurate monitoring, early damage identification, intelligent risk assessment and timely early warning of the state of the working rope can be realized, and the use safety and reliability are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of rope condition monitoring technology, specifically to a monitoring and early warning system for working ropes crossing a safety net. Background Technology

[0002] Working ropes are crucial for safety and reliability, especially in critical scenarios such as high-altitude operations, lifting and hoisting, cable structures, and netting applications (e.g., flexible boundary protection netting for large stadiums and airports). Traditionally, the inspection of working ropes has relied mainly on periodic visual inspections or simple mechanical parameter measurements, such as monitoring only the total tension. However, these methods have several shortcomings: Manual visual inspection is highly subjective and struggles to detect early, minor damage within ropes, such as single fiber breakage or microcrack propagation. Damage is often only detected when it has progressed to a significant degree, resulting in delayed warnings. Simple mechanical parameter measurements (such as tension) only reflect the macroscopic stress state of the rope and cannot reveal the initiation and development of localized damage. Traditional methods are ineffective in monitoring and responding promptly to damage accumulation and sudden incidents during dynamic operations (such as rope deployment and traction). Furthermore, the lack of scientific prediction of the rope's remaining service life (RUL) leads to maintenance strategies often based on experience or fixed cycles, potentially resulting in premature replacements that lead to waste or delayed replacements that pose safety hazards.

[0003] Acoustic emission (AE) technology, as an effective non-destructive testing method, can capture transient elastic waves released within materials due to damage (such as crack initiation and propagation, fiber breakage), making online monitoring of early rope damage possible. However, effectively extracting useful AE signals from complex background noise, combining them with other sensor information for intelligent analysis, and establishing a reliable early warning mechanism are current technical challenges.

[0004] Therefore, there is an urgent need for a system that can perform full life-cycle, multi-dimensional, and intelligent monitoring and early warning of working ropes crossing the sealing net, so as to improve the safety, reliability, and economy of working rope use. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a monitoring and early warning system for working ropes crossing safety nets to solve the aforementioned problems.

[0006] This invention provides the following technical solution: A monitoring and early warning system for crossing a work rope in a safety net includes: The data acquisition and transmission module is used to acquire multi-source status data of the working rope and process the multi-source status data to form a structured data stream. The intelligent rope condition analysis module is used to receive structured data streams and analyze and evaluate the condition of the working rope based on the structured data streams. Damage characteristics and risk knowledge base, used to store reference information for auxiliary working rope condition analysis and assessment; The early warning decision and response module is used to generate early warning information and execute response actions based on the status analysis and evaluation results of the working rope and reference information; The optimization and self-learning module, after communicating with the data acquisition and transmission module, the rope status intelligent analysis module, the damage characteristics and risk knowledge base, and the early warning decision and response module, is used to coordinate the work between the various modules and optimize system performance.

[0007] Furthermore, the data acquisition and transmission module includes: At least one distributed intelligent sensing node is used to acquire acoustic emission signal data, tension signal data, and attitude and vibration signal data along the length direction of the working rope; A dynamic operation monitoring unit is used to acquire acoustic emission signal data and tension signal data of the working rope during dynamic operation. The environmental parameter acquisition unit is used to acquire environmental parameter data at the work site. The data aggregation and synchronization management unit is used to perform time synchronization, format conversion, and aggregation of data obtained from distributed intelligent sensing nodes, dynamic operation monitoring units, and environmental parameter acquisition units to form a structured data stream.

[0008] Furthermore, both the distributed intelligent sensing node and the dynamic operation monitoring unit are equipped with acoustic emission AE sensors; The acoustic emission AE sensor installed at the distributed intelligent sensing node is used to capture the original crack AE electrical signal generated inside the rope due to fiber breakage, microcrack propagation or structural friction, and then amplifies, bandpass filters and converts the signal into a node-level acoustic emission AE digital signal. The acoustic emission (AE) sensor installed at the dynamic operation monitoring unit is used to monitor the original force-induced acoustic emission (AE) electrical signal generated by the internal acoustic activity of the rope under bending and dynamic stress. After preliminary amplification, bandpass filtering and analog-to-digital conversion, it is transformed into the dynamic monitoring unit's acoustic emission (AE) digital signal.

[0009] Furthermore, the intelligent rope status analysis module includes: The multi-source data preprocessing submodule preprocesses the input structured data stream; The acoustic emission signal depth analysis and damage identification submodule performs acoustic emission AE signal analysis and damage identification based on preprocessed acoustic emission AE signal data. The multimodal feature fusion and comprehensive risk assessment submodule combines various preprocessed sensor data and historical information to perform multimodal feature fusion and comprehensive risk assessment. The rope digital twin construction and dynamic update submodule is used to construct and update the digital twin model of the working rope.

[0010] Furthermore, the acoustic emission signal depth analysis and damage identification submodule is used to extract key feature parameters of acoustic emission AE events and to classify damage mechanisms using a deep learning model. The key feature parameters include the amplitude, duration, energy, rise time, count, and frequency characteristics of the acoustic emission AE events.

[0011] Furthermore, the damage characteristics and risk knowledge base is used to store a typical acoustic emission characteristic parameter set describing the damage evolution model of rope damage accumulation and development law, a risk assessment rule and threshold set for risk determination, and an environmental impact factor database and correction model describing the impact of environmental factors on ropes and monitoring signals. Among them, the typical acoustic emission characteristic parameter set includes the statistical distribution of key characteristic parameters of acoustic emission AE events representing specific damage mechanisms, waveform templates of typical acoustic emission AE events, and spectral patterns of typical acoustic emission AE events. Damage evolution models include mathematical models used to describe the accumulation and development of rope damage; The risk assessment rules and threshold set includes rules and values ​​used to map analysis results to predefined risk levels; The environmental impact factor database and correction model include quantitative relationship models or correction coefficients for the influence of environmental factors on rope performance and acoustic emission (AE) signal propagation characteristics.

[0012] Furthermore, the early warning decision and response module includes: The graded early warning logic judgment submodule performs graded early warning judgment based on the state analysis and evaluation results of the working rope and the early warning threshold in the reference information. The multi-channel alarm information generation and push submodule generates and pushes alarm information when an alert is triggered. The remaining service life prediction and maintenance recommendation submodule predicts the remaining service life of the working rope based on a digital twin model and provides maintenance recommendations. The visualization and human-computer interaction submodule is used for information display and interaction in the user interface.

[0013] Furthermore, the optimization and self-learning module includes: The system's main controller is used to coordinate the orderly operation of the above modules, manage data flow and control flow, handle communication and synchronization between modules, and respond to external commands and internal events. The closed-loop feedback adjustment submodule adjusts the acquisition parameters based on the monitored data, analysis results, and alarm information feedback. The knowledge base and model self-learning update submodule updates the analysis model and damage characteristics and risk knowledge base in the rope state intelligent analysis module based on data. Remote diagnostics and maintenance interface, used to provide remote access functionality.

[0014] The present invention also discloses a method for a monitoring and early warning system for crossing a working rope of a safety net, according to any one of the foregoing claims, comprising the following steps: S1. System Deployment and Initialization: S11. Install distributed intelligent sensing nodes on the working rope and install dynamic operation monitoring units at the winch or pulley; S12. The rope status intelligent analysis module creates a digital twin file for the current working rope and enters basic information; the damage characteristics and risk knowledge base loads the data corresponding to the preset rope model. S2. Real-time monitoring and data acquisition: S21. The sensor units of the data acquisition and transmission module begin to acquire raw data of the working rope in real time; S22. The data aggregation and synchronization management unit performs preliminary processing on the acquired raw data and sends the resulting data stream to the rope status intelligent analysis module. S3. Intelligent Analysis and Risk Assessment: S31. The multi-source data preprocessing submodule further preprocesses the received structured data stream; S32. The acoustic emission signal depth analysis and damage identification submodule performs advanced feature extraction and depth analysis on the preprocessed AE signal data output from S31; S33. The multimodal feature fusion and comprehensive risk assessment submodule combines the AE analysis results and refers to the damage characteristics and risk knowledge base to assess the comprehensive risk level of the current rope. S4. Early Warning Decision and Response: S41. The hierarchical early warning logic judgment submodule determines whether an early warning is triggered based on the analysis results of S3; S42. If an early warning is triggered, the multi-channel alarm information generation and push submodule sends alarm information to relevant personnel and the system; S43. The visualization and human-computer interaction submodule updates the rope status, risk level, and alarm information on the interface and records the event. S5. System Optimization and Self-Learning: S51. When the quality of monitoring data is poor, the closed-loop feedback adjustment submodule attempts to optimize the collection parameters; S52. The knowledge base and model self-learning update submodule regularly collects case data to optimize the AI ​​model and knowledge base.

[0015] Furthermore, the intelligent analysis and risk assessment in S3 also includes: S34, the rope digital twin construction and dynamic update submodule updates the analysis results and raw data to the corresponding rope digital twin file; S35, the remaining service life prediction and maintenance suggestion submodule updates the RUL prediction based on the latest digital twin data.

[0016] The present invention has the following beneficial technical effects: This invention integrates multiple sensors, including AE, tension, attitude and vibration sensors, and environmental sensors, and combines them with distributed sensing nodes and dynamic operation monitoring units to achieve comprehensive and real-time acquisition of state parameters of the working rope under static and dynamic working conditions. This invention utilizes acoustic emission technology and combines deep learning models to perform in-depth analysis of AE signals, which can effectively identify early micro-damage inside the rope (such as fiber breakage and micro-crack propagation) and assess the damage type, severity, and location. This invention utilizes multimodal feature fusion technology to integrate AE analysis results, mechanical parameters, vibration characteristics, environmental factors, and historical information provided by digital twins to calculate the instantaneous comprehensive risk index of the rope, resulting in a more comprehensive and accurate assessment. Based on the cumulative damage and real-time status recorded by the digital twin model, combined with damage evolution models or data-driven models, this invention can predict the remaining service life (RUL) of the working rope, providing a scientific basis for maintenance decisions. This invention triggers different levels of early warnings based on risk levels and RUL prediction results, and pushes alarm information through multiple channels to ensure that management and operational personnel can promptly grasp the risk situation and take corresponding measures. Through continuous data collection and analysis, the system can self-learn and update its analysis model and knowledge base, continuously improving the accuracy and intelligence of monitoring and early warning, and adjusting the collection parameters based on monitoring feedback. This invention prevents sudden accidents through early warnings, ensuring the safety of personnel and equipment, and optimizes maintenance strategies through RUL prediction and maintenance suggestions, extending rope lifespan and reducing operation and maintenance costs. Detailed Implementation

[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and 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 This embodiment provides a monitoring and early warning system for working ropes crossing a safety net, including a data acquisition and transmission module, a rope status intelligent analysis module, a damage characteristics and risk knowledge base, an early warning decision and response module, and an optimization and self-learning module, as detailed below: The data acquisition and transmission module includes distributed intelligent sensing nodes, dynamic operation monitoring units, environmental parameter acquisition units, and data aggregation and synchronization management units; Distributed intelligent sensing nodes are deployed at multiple points along the length of the working rope to collect key rope status parameters in real time. Internally, they integrate acoustic emission (AE) sensors, tension sensors, attitude and vibration sensors, as well as local processing and wireless communication units. The acoustic emission (AE) sensors employ high-sensitivity piezoelectric ceramic sensors (e.g., frequency response range 20kHz-1MHz), which are tightly bonded to the rope surface via acoustic coupling agent to capture weak acoustic emission signals (i.e., original crack AE electrical signals) generated within the rope due to fiber breakage, microcrack propagation, or structural friction. The captured original AE electrical signals undergo preliminary amplification and bandpass filtering (to remove significant environmental noise and electromagnetic interference) by the integrated processing circuitry within the node, followed by analog-to-digital conversion to form digitized node-level AE digital signals, which are then sent to the data aggregation point. The system receives AE signal data from one source, along with synchronization management unit data. A through-hole tension sensor is used to accurately measure the real-time tension of the rope at its location and output corresponding tension signal data for subsequent modules. Attitude and vibration sensors, employing MEMS triaxial accelerometers and gyroscopes, monitor the rope's vibration characteristics (frequency, amplitude) and static tilt angle, calculating sag and identifying abnormal swaying, and outputting attitude and vibration signal data. The local processing and wireless communication unit incorporates a microcontroller (MCU) to perform preliminary processing (such as preliminary filtering, feature extraction, and event packaging) on ​​the acquired raw data (raw crack AE electrical signals, tension signals, attitude and vibration signals, which will not be elaborated here), and wirelessly transmits the data via LoRaWAN or Zigbee near-field communication technology. It should be noted that, in order to facilitate the unified maintenance of the sensors, temperature sensors, humidity sensors, wind speed and direction sensors are also installed at the distributed intelligent sensing nodes, which can be selected as needed. The dynamic operation monitoring unit is used to monitor the status of the working rope during dynamic operations such as winding, releasing, and traction. It is integrated into the rope outlet of the winch or the key pulley structure (i.e., the structure at the traditional rope winding device). The dynamic operation monitoring unit includes an integrated dynamic tension sensor and an AE sensor. The integrated dynamic tension sensor is installed on the pulley shaft for measurement.

[0019] The real-time tension of the rope at this point is monitored; an AE sensor (using a small sensor) is embedded in the inner wall of the pulley groove, allowing it to make close contact with the rope surface as the rope passes through, in order to monitor the internal acoustic activity of the rope under bending and dynamic stress (i.e., the original stress AE electrical signal). The original stress AE electrical signal is also initially amplified and bandpass filtered by the internally integrated processing circuit, and then converted from analog to digital (the same processing method as the distributed intelligent sensing node). After processing, the acoustic signal constitutes the AE digital signal of the dynamic monitoring unit, which is suitable for capturing damage events in the dynamic process and constitutes another source of AE signal data sent to the data aggregation and synchronization management unit; a diameter monitoring sensor can also be added, whose output wear or diameter data can be used as auxiliary judgment information to assess the wear of the rope when it passes through; The environmental parameter acquisition unit consists of sensors (such as temperature sensors, humidity sensors, and wind speed and direction sensors) integrated into the distributed intelligent sensing nodes. It is used to collect environmental parameter data at the work site, providing environmental compensation basis for status analysis or assessing the impact of the environment on rope performance. The data aggregation and synchronization management unit is responsible for receiving node-level AE digital signals from distributed intelligent sensing nodes, dynamic monitoring unit AE digital signals from dynamic operation monitoring units, and other sensor data (such as tension signal data, attitude and vibration signal data, environmental parameter data, etc.). It ensures the synchronization of timestamps from different sources, performs format conversion, verification, and preliminary aggregation of the data, manages the transmission of data streams (e.g., aggregated through a gateway and then transmitted via 4G / 5G or wired network), and distributes the structured data stream with metadata (such as the ID, timestamp, and location information of each different sensor) to the subsequent rope status intelligent analysis module. The portion of the AE signal data in this structured data stream that has been synchronized and preliminarily aggregated by this unit is defined as the aggregated AE data.

[0020] In summary, the AE sensors in the distributed intelligent sensing nodes need to have a high signal-to-noise ratio and a wide dynamic range to capture signals ranging from weak single-filament breakage to strong fiber bundle breakage; the range and accuracy of the tension sensors need to meet the specific application scenarios (e.g., parameters of 0-100kN, accuracy of 0.5%FS); the attitude and vibration sensors can effectively reflect the macroscopic dynamic behavior of the rope; the dynamic operation monitoring unit should be able to withstand dynamic impacts and wear, and its sensor response speed should meet the requirements of dynamic monitoring; the data aggregation and synchronization management unit needs to have high reliability and a certain edge computing capability to process large amounts of real-time data.

[0021] The intelligent rope condition analysis module includes a multi-source data preprocessing submodule, an acoustic emission signal depth analysis and damage identification submodule, a multi-modal feature fusion and comprehensive risk assessment submodule, and a rope digital twin construction and dynamic update submodule. The multi-source data preprocessing submodule is used to perform noise reduction (such as bandpass filtering on aggregated AE data, Kalman filtering on tension signals, etc.), missing value interpolation, data alignment (based on timestamps), normalization and other operations on various sensor data input from the data acquisition and transmission module to improve data quality and facilitate subsequent analysis. It should be noted that the aggregated AE data processed by the multi-source data preprocessing submodule is the preprocessed AE signal data. The acoustic emission signal depth analysis and damage identification submodule uses a deep learning model trained on the rope damage mechanism (a CNN-based spectrogram analysis model or an RNN / LSTM-based AE event sequence analysis model, which can be selected as appropriate) to perform advanced feature extraction and damage identification on the preprocessed AE signal data. Advanced feature extraction specifically involves the following steps: First, by using a preset dynamic threshold, valid AE event segments (i.e., waveform segments with acoustic emission characteristics separated from background noise and irrelevant signals) are identified and segmented from the preprocessed AE signal data stream, enabling accurate identification of individual AE events. Then, for each identified valid AE event segment, key feature parameters are calculated, including amplitude (maximum amplitude of the AE event waveform), duration (length of time from the start to the end of the AE event), energy (calculated by integrating the envelope of the AE event waveform), rise time (time from the start of the event to reaching the peak amplitude), count (number of times the signal exceeds the threshold), and frequency characteristics (spectral features obtained by performing a fast Fourier transform on the AE event waveform, such as peak frequency, center frequency, and frequency component proportion). These extracted feature parameters will be used for subsequent damage identification and classification. Damage identification specifically involves classifying identified adverse event (AE) events into different damage mechanisms (such as single fiber breakage, multiple fiber breakage, fiber pull-out, matrix cracking, friction noise, etc.) and quantifying their severity. For example, a CNN model can be used to analyze the Mel-spectrum or short-time Fourier transform graph obtained from the AE signal data conversion to output the probability of each type of damage. Alternatively, the location of the AE source can be estimated by analyzing the time difference of arrival (TDOA) or signal strength differences between multiple AE sensors receiving the same event, thereby obtaining the location information of the damaged area. The multimodal feature fusion and comprehensive risk assessment submodule is based on the preprocessed AE signal data, tension signal data, vibration signal data, and environmental parameter data output by the multi-source data preprocessing submodule, as well as the historical information contained in the rope digital twin model (such as the cumulative fatigue damage degree calculated by analyzing historical tension data and preset fatigue models (such as Miner's rule), historical AE event statistics, etc.) obtained in real time from the rope digital twin construction and dynamic update submodule. It uses fusion algorithms (such as weighted scoring, fuzzy logic, Bayesian network or trained fusion neural network) to calculate the instantaneous comprehensive risk index of the current working rope. For example, when the AE sensor identifies a high-energy "fiber bundle breakage" event, even if the tension is still within the safe range, the comprehensive risk index will be significantly increased. The rope digital twin construction and dynamic update submodule creates a digital twin model for each monitored working rope, including its design parameters (material, specifications, strength, etc.), manufacturing information, historical usage records (number of operations, total duration, load spectrum), cumulative damage records (statistics based on AE events, fatigue damage estimation), maintenance records, and real-time status data (including currently collected data from various sensors), and updates the data in real time. The Damage Characteristics and Risk Knowledge Base stores typical acoustic emission characteristic parameter sets for ropes under different working conditions and damage states, damage evolution models describing the accumulation and development of rope damage, risk assessment rules and threshold sets for risk determination, and a database and correction model of environmental impact factors describing the impact of environmental factors on ropes and monitoring signals. Specifically, the typical acoustic emission characteristic parameter set consists of the statistical distribution of key characteristic parameters of AE events representing specific damage mechanisms (e.g., amplitude, duration, energy, rise time, typical range of counts or probability density function shape), typical AE event waveform templates (i.e., time-domain standard waveform shapes representing specific damage), and typical AE event spectrum patterns (i.e., representative spectrum features obtained after spectrum analysis of typical AE event waveforms, such as peak frequency, center frequency, and frequency component proportion). Damage evolution models are mathematical models or empirical formulas established based on fracture mechanics, material fatigue theory, or statistical analysis of large amounts of experimental data. They describe how damage (such as crack propagation rate and strength decay) accumulates and develops over time or with the number of uses under specific loads and environmental conditions. For example, the "preset fatigue accumulation model (using Miner's rule to correct the model)" used in the Remaining Service (RUL) prediction and maintenance recommendation submodule belongs to this type of model, and its parameters and the model itself can be stored here. Risk assessment rules and threshold sets are rules and specific values ​​set based on historical data, expert experience, and industry standards. They are used to analyze the damage identification results (such as damage type, severity, AE event frequency, and the proportion of specific types or high-energy AE events) output by the acoustic emission signal depth analysis and damage identification submodule. The comprehensive risk index calculated by the multimodal feature fusion and comprehensive risk assessment submodule, and the RUL predicted by the remaining useful life (RUL) prediction and maintenance recommendation submodule, are mapped to predefined risk levels (such as "attention", "warning", "danger") and trigger corresponding early warning levels. The "early warning threshold" referenced by the graded early warning logic judgment submodule comes from this. The environmental impact factor database and correction model refers to the storage of historical and real-time environmental parameter data (such as temperature, humidity, wind speed and direction), as well as quantitative relationship models or correction coefficients of the influence of these environmental factors on rope material properties (such as elastic modulus and strength) and the propagation characteristics of AE signals in ropes (such as sound speed change and signal attenuation coefficient). These models can be used to compensate for sensor readings or to consider factors such as accelerated aging in the environment when assessing damage. When performing AE signal analysis, risk assessment, or RUL prediction, the damage characteristics and risk knowledge base is queried to obtain relevant prior knowledge, model parameters, and reference baselines to guide the analysis process and improve accuracy. At the same time, the damage characteristics and risk knowledge base can be dynamically updated and optimized through experimental data, simulation results, industry standards, and cases accumulated during system operation (such as data collected by the knowledge base and model self-learning update submodule).

[0022] The early warning decision and response module includes a tiered early warning logic judgment submodule, a multi-channel alarm information generation and push submodule, a remaining useful life (RUL) prediction and maintenance suggestion submodule, and a visualization and human-computer interaction submodule. The tiered early warning logic judgment submodule, based on the damage identification results (such as damage type and severity), comprehensive risk index, and RUL prediction results output by the rope status intelligent analysis module, compares them with early warning thresholds set in the damage characteristics and risk knowledge base (such as AE event frequency threshold, specific type or high-energy AE event, comprehensive risk index level threshold, and RUL critical value) to trigger different levels of early warnings (such as "attention"). The alarm information generation and push module generates alarm information including the alarm level, possible causes (such as damage type identified by analysis of pre-processed AE signal data), location information of the damaged area, and timestamp when a warning is triggered. This information is then sent to relevant management personnel and on-site workers through various channels (such as on-site audible and visual alarms, monitoring center large screen pop-ups, SMS, email, mobile APP push, and voice calls). The Remaining Useful Life (RUL) prediction and maintenance recommendation submodule utilizes the rope digital twin to construct and dynamically update the cumulative damage records in the submodule. Combined with a preset fatigue accumulation model (using Miner's rule to correct the model) or a data-driven RUL prediction model (using an LSTM-based regression model), it predicts the remaining safe service life or remaining safe working hours of the working rope. When the RUL is lower than a preset threshold, it generates maintenance or replacement recommendations. The visualization and human-computer interaction submodule provides a user interface (such as a web platform, local console, or mobile APP) to display data such as the current comprehensive risk level, RUL prediction results, and alarm information, as well as human operations such as querying detailed data, configuring system parameters, and confirming alarm information.

[0023] It should be noted that although both the acoustic emission signal depth analysis and damage identification submodule and the multimodal feature fusion and comprehensive risk assessment submodule involve risk assessment, the former focuses more on interpreting microscopic and direct damage information (such as what kind of damage, where, and how severe) from the preprocessed AE signal data itself, while the latter, on the other hand, combines more macroscopic mechanical states (based on tension signal data), dynamic behavior (based on vibration signal data), historical background (fatigue), etc., to make a global comprehensive risk assessment (such as how long the rope can still be used). Therefore, the two are different and cannot be substituted for each other. The optimization and self-learning module includes a system main controller, a closed-loop feedback adjustment submodule, a knowledge base and model self-learning update submodule, and a remote diagnostic and maintenance interface, as follows: The system main controller coordinates the orderly operation of each module, manages data flow and control flow, handles communication and synchronization between modules, and responds to external commands and internal events; The closed-loop feedback adjustment submodule: When the monitoring effect is poor (e.g., low signal-to-noise ratio of preprocessed AE signal data leading to missed events, or excessive fluctuations in tension signal data), it can attempt to adjust certain parameters of the distributed intelligent sensing nodes (e.g., adjust the gain or filtering settings of the AE sensor) according to preset rules or operator instructions; The knowledge base and model self-learning update submodule continuously collects the data monitored in each operation (e.g., aggregated AE data used for model training, preprocessed AE signal data and their corresponding spectrograms or feature parameters, and other synchronized sensor data such as tension signal data), analyzes the results, and receives alarm information, thereby enabling periodic training of the CNN model to improve its accuracy and robustness in recognizing different damage patterns, and also improves the accuracy of the RUL prediction model; The remote diagnostic and maintenance interface provides the ability to remotely access the system.

[0024] The working process is as follows: S1. System Deployment and Initialization: S11. Install distributed intelligent sensing nodes on the working rope and dynamic operation monitoring units at the winch or pulley, and connect them to the data aggregation and synchronization management unit; S12. The rope status intelligent analysis module creates a digital twin file for the current working rope and enters basic information; the damage characteristics and risk knowledge base loads the data corresponding to the preset rope model; S2. Real-time Monitoring and Data Acquisition: S21. The sensor units of the data acquisition and transmission module (distributed intelligent sensing nodes, dynamic operation monitoring units, environmental parameter acquisition units) begin to collect various raw data such as the original crack AE electrical signal, tension, vibration, and environment of the working rope in real time; S22. After the aggregation and synchronization management unit performs preliminary processing on the various types of raw data collected, the resulting data stream is sent to the rope status intelligent analysis module. It should be noted that preliminary processing refers to first performing time alignment and synchronization based on the timestamps of each raw data source to ensure consistency of the multi-source data in the time series. This is followed by data format conversion (unifying data from different sensors or communication protocols into an internal standard format), data packet verification (checking data integrity and correctness), and preliminary aggregation (such as packaging and organizing data from multiple sensors from the same sensing node under the same timestamp), forming a structured data stream. S3, Intelligent Analysis and Risk Assessment: S31, Multi-Source Data Preprocessing Submodule processes the data from the aggregation and synchronization management unit... The structured data stream received by the synchronization management unit (which includes aggregated AE data, tension signal data, attitude and vibration signal data, environmental parameter data, etc.) undergoes further cleaning, noise reduction (such as more refined bandpass filtering of aggregated AE data, Kalman filtering of tension signals, etc.), missing value imputation, data alignment (ensuring accurate alignment at finer time scales), normalization, and other operations, outputting preprocessed AE signal data, preprocessed tension signal data, etc.; S32, the acoustic emission signal depth analysis and damage identification submodule, performs advanced feature extraction and depth analysis on the preprocessed AE signal data output by S31, identifying the type of internal damage event, estimating its severity, and... The damage source is located by combining information from multiple sensors (if applicable); S33, the multimodal feature fusion and comprehensive risk assessment submodule combines AE analysis results, tension signal data, vibration signal data, etc., and refers to the damage characteristics and risk knowledge base to assess the current comprehensive risk level of the rope; S34, the rope digital twin construction and dynamic update submodule updates the analysis results and raw data to the corresponding rope's digital twin file; S35, the remaining service life prediction and maintenance recommendation submodule updates the RUL prediction based on the latest digital twin data; S4, early warning decision and response: S41, the graded early warning logic judgment submodule determines whether to trigger an early warning based on the analysis results of S3 (damage status, risk level, RUL);S42. If an alert is triggered, the multi-channel alarm information generation and push submodule sends alarm information to relevant personnel and the system; S43. The visualization and human-computer interaction submodule updates the rope status, risk level, and alarm information on the interface and records the event; S5. System optimization and self-learning (continuously performed in the background): S51. When the monitoring data quality is poor, the closed-loop feedback adjustment submodule attempts to optimize the collection parameters; S52. The knowledge base and model self-learning update submodule regularly collects case data to optimize the AI ​​model and knowledge base.

[0025] The embodiments described above are merely illustrative of specific implementations of the present invention, and while the descriptions are detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A monitoring and warning system for crossing a closed network working line, characterized in that, The system comprises: a data acquisition and transmission module for acquiring multi-source state data of the working rope and processing the multi-source state data to form a structured data stream; a rope state intelligent analysis module for receiving the structured data stream and analyzing and evaluating the state of the working rope based on the structured data stream; a damage feature and risk knowledge base for storing reference information for assisting in the analysis and evaluation of the state of the working rope; a pre-warning decision and response module for generating pre-warning information and performing response actions according to the analysis and evaluation results of the state of the working rope and the reference information; an optimization and self-learning module, which, after being communicatively connected with the data acquisition and transmission module, the rope state intelligent analysis module, the damage feature and risk knowledge base, and the pre-warning decision and response module, is used for coordinating the work among the modules and optimizing the performance of the system.

2. The monitoring and warning system for crossing the closed wire rope according to claim 1, characterized in that, The data acquisition and transmission module comprises: at least one distributed intelligent sensing node for acquiring acoustic emission signal data, tension signal data, and attitude and vibration signal data along the length direction of the working rope; a dynamic operation monitoring unit for acquiring acoustic emission signal data and tension signal data of the working rope during dynamic operation; an environmental parameter acquisition unit for acquiring environmental parameter data of the operation site; a data aggregation and synchronization management unit for performing time synchronization, format conversion, and aggregation on the data acquired from the distributed intelligent sensing node, the dynamic operation monitoring unit, and the environmental parameter acquisition unit to form a structured data stream.

3. The monitoring and warning system of a crossing fence working wire according to claim 2, characterized in that, The distributed intelligent sensing node and the dynamic operation monitoring unit are both installed with acoustic emission AE sensors; The acoustic emission AE sensors installed at the distributed intelligent sensing node are used to capture original crack AE electrical signals generated inside the rope due to fiber breakage, micro-crack propagation, or structural friction, and form node-level acoustic emission AE digital signals through preliminary amplification, band-pass filtering, and analog-to-digital conversion; The acoustic emission AE sensors installed at the dynamic operation monitoring unit are used to monitor original stress acoustic emission AE electrical signals generated inside the rope due to internal acoustic activity under bending and dynamic stress, and form dynamic monitoring unit acoustic emission AE digital signals through preliminary amplification, band-pass filtering, and analog-to-digital conversion.

4. The monitoring and warning system of a crossing blind line working wire according to claim 1, characterized in that, The rope state intelligent analysis module comprises: a multi-source data preprocessing submodule for preprocessing the input structured data stream; an acoustic emission signal deep analysis and damage identification submodule for performing acoustic emission AE signal analysis and damage identification based on the preprocessed acoustic emission AE signal data; a multi-modal feature fusion and comprehensive risk evaluation submodule for combining various preprocessed sensor data and historical information to perform multi-modal feature fusion and comprehensive risk evaluation; a rope digital twin construction and dynamic update submodule for constructing and updating a digital twin model of the working rope.

5. The monitoring and warning system of a crossing blind line working wire according to claim 4, characterized in that, The acoustic emission signal deep analysis and damage identification submodule is used to extract key feature parameters of acoustic emission AE events and perform damage mechanism classification using a deep learning model, and the key feature parameters include amplitude, duration, energy, rise time, count, and frequency characteristics of the acoustic emission AE events.

6. The monitoring and warning system of a crossing blind wire working rope according to claim 4, characterized in that, The damage feature and risk knowledge base is used to store a typical acoustic emission feature parameter set describing the damage evolution model of the rope damage accumulation and development, a risk assessment rule and threshold set for risk judgment, and an environmental influence factor database and correction model describing the influence of environmental factors on the rope and monitoring signal; The typical acoustic emission feature parameter set includes statistical distribution of acoustic emission AE event key feature parameters representing a specific damage mechanism, a typical acoustic emission AE event waveform template, and a typical acoustic emission AE event spectrum mode; The damage evolution model includes a mathematical model for describing the accumulation and development of rope damage; The risk assessment rule and threshold set includes rules and numerical values for mapping analysis results to predefined risk levels; The environmental influence factor database and correction model includes a quantitative relationship model or correction coefficient of the influence of environmental factors on the performance of the rope and the propagation characteristics of the acoustic emission AE signal.

7. The monitoring and warning system of a crossing blind line working wire according to claim 4, characterized in that, The early warning decision and response module includes: A hierarchical early warning logic judgment submodule that performs hierarchical early warning judgment according to the state analysis and evaluation results of the working rope and the early warning threshold in the reference information; A multi-channel alarm information generation and pushing submodule that generates and pushes alarm information when early warning is triggered; A remaining useful life prediction and maintenance suggestion submodule that predicts the remaining useful life of the working rope based on the digital twin model and provides maintenance suggestions; A visualization display and human-computer interaction submodule for user interface information display and interaction.

8. The monitoring and warning system of a crossing blind wire working rope according to claim 6, characterized in that, The optimization and self-learning module includes: A system main controller for coordinating the orderly work of the above modules, managing data flow and control flow, processing inter-module communication and synchronization, and responding to external instructions and internal events; A closed-loop feedback adjustment submodule that adjusts the acquisition parameters based on the monitored data, analysis results, and alarm information monitoring effect feedback; A knowledge base and model self-learning update submodule that updates the analysis models in the rope state intelligent analysis module and the damage feature and risk knowledge base based on data; A remote diagnosis and maintenance interface for providing remote access functions.

9. The method of monitoring and warning system of crossing the closed wire rope according to any one of claims 1-8, characterized in that, The method includes the following steps: S1, system deployment and initialization: S11, install distributed intelligent sensing nodes on the working rope, and install dynamic operation monitoring units at the winch or pulley; S12, the rope state intelligent analysis module creates a digital twin file for the current working rope and enters the basic information; the damage feature and risk knowledge base loads the preset data corresponding to the rope model; S2, real-time monitoring and data collection: S21, the sensor units of the data collection and transmission module start real-time collection of the original data of the working rope; S22, the data aggregation and synchronization management unit performs preliminary processing on the collected original data to form a data stream sent to the rope state intelligent analysis module; S3, intelligent analysis and risk assessment: S31, a multi-source data preprocessing submodule further preprocesses the received structured data stream; S32, an acoustic emission signal depth analysis and damage identification submodule performs advanced feature extraction and depth analysis on the AE signal data output by S31 after preprocessing; S33, a multi-modal feature fusion and comprehensive risk assessment submodule combines the AE analysis results and refers to the damage feature and risk knowledge base to assess the comprehensive risk level of the current rope; S4, early warning decision and response: S41, a hierarchical early warning logic judgment submodule determines whether to trigger an early warning based on the analysis results of S3; S42, if the early warning is triggered, a multi-channel alarm information generation and pushing submodule sends alarm information to relevant personnel and systems; S43, a visual display and human-computer interaction submodule updates the rope state, risk level, alarm information on the interface, and records the event; S5, system optimization and self-learning: S51, when the monitoring data quality is poor, a closed-loop feedback adjustment submodule attempts to optimize the acquisition parameters; S52, a knowledge base and model self-learning update submodule periodically collects case data for optimizing the AI model and knowledge base.

10. The method of claim 9, wherein, The intelligent analysis and risk assessment in S3 further includes: S34, a rope digital twin construction and dynamic update submodule updates the analysis results and original data to the digital twin archive of the corresponding rope; S35, a remaining useful life prediction and maintenance recommendation submodule updates the RUL prediction based on the latest digital twin data.