Fault early warning method oriented to loading arm and boarding ladder and based on WASM modular model

Abstract

The invention relates to a WASM modular model-based fault early warning method for a loading arm and a boarding ladder, and belongs to the technical field of intelligent monitoring and fault diagnosis of harbor machinery. The method comprises the following steps: constructing an edge dual-stack sensing node adaptive to working condition characteristics of two types of equipment, capturing mechanical dynamic data for a loading and unloading arm, filtering interference in combination with a medium type, recording operation state data for a boarding ladder, presetting a displacement threshold according to hydrological historical data, and carrying out parallel acquisition and preliminary cleaning on the data through a time sequence synchronization mechanism; configuring an exclusive diagnosis rule, outputting real-time equipment state data and adapting to different scene thresholds; establishing a unified data space-time reference, fusing working condition elements, and adding space-time labels to align data; a cloud edge cooperation mechanism is established, an edge node uploads abnormal data and the like to a cloud end, early warning information and an equipment state are associated, and a linkage signal is pushed to a central controller. Accurate recognition of faults of the loading and unloading arm and the boarding ladder is achieved, and the operation safety and the operation and maintenance efficiency of port equipment are improved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent monitoring and fault diagnosis technology for port machinery, specifically relating to a fault early warning method based on the WASM modular model for loading arms and boarding stairs. Background Technology

[0002] In port and dock operations, loading arms and boarding ladders are core equipment for ensuring the loading and unloading of ships and the boarding and disembarking of personnel; their operational status directly affects operational efficiency and safety. Currently, fault warnings for these two types of equipment largely rely on traditional centralized monitoring systems, which suffer from problems such as delayed response and poor adaptability to different scenarios. On the one hand, loading and unloading arms need to handle the transportation of different media (such as crude oil and liquefied natural gas), resulting in complex mechanical dynamic characteristics; boarding ladders are significantly affected by hydrology and waves, leading to large fluctuations in their operating status. Existing unified early warning models are unable to accurately capture the differentiated abnormal characteristics of the two types of equipment, which can easily lead to false alarms or missed alarms. On the other hand, traditional models are mostly deployed in the cloud, resulting in high data transmission latency and a lack of spatiotemporal alignment processing for collected data, leading to insufficient diagnostic accuracy. At the same time, model updates require overall reconstruction, making it difficult to quickly adapt to changes in terminal operating conditions and increasing maintenance costs. Summary of the Invention

[0003] To address the aforementioned problems in the existing technology, this invention provides a fault early warning method based on the WASM modular model for loading arms and boarding stairs. The objective of this invention can be achieved through the following technical solutions: S1: Construct edge dual-stack sensing nodes adapted to the operating characteristics of loading arms and boarding ladders, and preset acquisition schemes for the operational differences of the two types of equipment; for loading arms, capture mechanical dynamic data, and simultaneously preset interference filtering rules based on media type; for boarding ladders, record operation status data, and preset normal displacement thresholds based on historical hydrological data of the dock; based on the mechanical dynamic data and operation status data, perform parallel acquisition and preliminary cleaning of data from the two types of equipment through a time-series synchronization mechanism between nodes; S2: Configure diagnostic rules for abnormal issues of loading arms and boarding ladders; compile the core algorithm through WASM and deploy it directly to the edge node; the edge uses preprocessed real-time data to output real-time equipment status data and sets preset thresholds for different operating scenarios according to the equipment type parameter group.

[0004] S3: Establish a unified spatiotemporal benchmark for data that fits the operation scenarios of loading arms and boarding ladders, and define spatiotemporal labeling rules by integrating actual working conditions; add spatial and temporal labels to each piece of collected data, and align the time and spatial dimensions of the two types of equipment data through label association; S4: By establishing a cloud-edge collaborative management mechanism, edge nodes upload the aforementioned abnormal issues, real-time equipment status data, and operation logs to the cloud database via an encrypted transmission protocol; the generated early warning information is correlated and matched with the real-time operating status of the equipment, and a linkage signal containing the early warning level and suggested handling measures is pushed to the terminal control system.

[0005] Specifically, the construction of edge dual-stack sensing nodes adapted to the working conditions of loading arms and boarding ladders for data acquisition of the two types of equipment involves the following process: a node communication architecture is built using both wireless and wired Ethernet communication modes; capacity adaptation of node data storage is performed for different installation environments of loading arms and boarding ladders; and computing power allocation is performed for node data processing.

[0006] Specifically, the process of capturing mechanical dynamic data for the loading and unloading boom to obtain core operating status information of the loading and unloading boom is as follows: collecting the rotation angle of the boom joint, the real-time pressure value of the hydraulic pipeline, the vibration amplitude and frequency of different parts of the boom, and the medium transport temperature at the pipeline interface; collecting operating status information data in real time according to the operating cycle of the loading and unloading boom; and organizing the collected data in chronological order to obtain a continuous and complete mechanical dynamic data sequence.

[0007] Specifically, the combined medium type refers to the pre-set interference filtering rules of the loading and unloading arm to handle the interference of the environment and the medium on the data. The specific process is as follows: for crude oil transported by the loading and unloading arm, a moving average filtering algorithm is used to smooth the collected pressure data to eliminate pressure fluctuations caused by the flow of high-viscosity crude oil; for liquefied natural gas transported, a notch filter algorithm is used to filter specific frequency signals generated by electromagnetic interference in low-temperature environments; for chemical transported, a Kalman filter algorithm is used to correct the attenuation effect of corrosive media on data signals. Through the adaptation of different algorithms, the mechanical dynamic data truly reflects the operating status of the equipment.

[0008] Specifically, the operation status data recording for the boarding ladder is used to record the action information during the operation of the boarding ladder. The specific process is as follows: collect the lateral and longitudinal displacement of the boarding ladder body, the pitch and rotation angle of the ladder body, the lifting and rotation speed of the drive system, and the load-bearing weight of the ladder body steps. Record these data synchronously at the time point triggered by the boarding ladder operation command, and integrate the data of different operation stages to form an operation status data chain.

[0009] Specifically, the process of setting a normal displacement threshold for the boarding ladder based on the wharf's historical hydrological data to define the boarding ladder's operating status involves: collecting historical hydrological data from the wharf for a preset period, classifying and statistically analyzing the historical hydrological data, calculating the average displacement of the boarding ladder under different tidal periods and different wave levels, and using the average displacement as a benchmark, combined with the structural bearing capacity of the boarding ladder, determining the displacement threshold for each period.

[0010] Specifically, the timing synchronization mechanism is used to collect data from two types of devices in parallel. The specific process is as follows: a clock synchronization program is deployed in each edge node, a node close to the central control area of ​​the dock is designated as the clock reference node, and the remaining nodes receive the clock calibration signal sent by the reference node at a fixed period and correct their local clocks. The timestamp error of all nodes is controlled within a preset range, and data from two types of devices is collected simultaneously at a uniformly set collection frequency.

[0011] As a preferred technical solution of the present invention, S1 constructs an edge dual-stack sensing node adapted to the operating characteristics of the loading boom and boarding ladder. A node communication architecture is built using both wireless and wired Ethernet communication modes. The storage capacity of the node is adapted to the different operating environments of the loading boom and boarding ladder, and the computing power of the node's processing portion is allocated accordingly. Targeted acquisition schemes are designed for the two types of equipment: for the loading boom, its mechanical dynamic data is captured, specifically the rotation angle of the boom joints, the real-time pressure value of the hydraulic pipeline, the vibration amplitude and frequency of different parts of the boom, and the medium transport temperature at the pipeline interface. This data is collected in real-time according to the loading boom's operating cycle and organized chronologically. Simultaneously, interference filtering rules are preset based on the type of transported medium: for crude oil, a moving average filtering algorithm is used to smooth pressure fluctuations; for liquefied natural gas, a notch filter algorithm is used to filter low-temperature electromagnetic interference; and for chemical media, a Kalman filter is used. The algorithm corrects signal attenuation; for the boarding ladder, its operational status data is recorded, specifically the lateral and longitudinal displacement of the ladder body, pitch and rotation angles, lifting and rotation speeds of the drive system, and the load-bearing weight of the ladder treads. These data are recorded synchronously at the time of the operation command trigger, and integrated to form an operational status data chain; the average displacement under different tidal periods and wave levels is statistically analyzed, and the normal displacement threshold for each period is determined in combination with the structural bearing capacity of the boarding ladder; based on the mechanical dynamic data and operational status data, parallel acquisition and preliminary cleaning are achieved through a time-series synchronization mechanism between nodes: a clock synchronization program is deployed at each edge node, and the node closest to the central control area of ​​the dock is designated as the clock reference node. The remaining nodes receive the clock calibration signal from the reference node at a fixed period and correct their local clocks, keeping the acquisition timestamp error of all nodes within a very small range; then, data from both types of equipment are acquired simultaneously at a unified frequency, and the acquired data is preliminarily cleaned.

[0012] Specifically, the diagnostic rules configured for abnormal issues of the loading boom and boarding ladder are used to identify equipment abnormalities. The specific process is as follows: For the loading boom, the hydraulic system leakage is analyzed, real-time pressure data of the hydraulic pipeline is continuously collected, the pressure decay rate is calculated at fixed time intervals, and the calculation results are compared with the pressure decay safety range designed for the equipment. If the decay rate exceeds the pressure decay safety range, it is determined that the hydraulic system is leaking. For the boarding ladder, the drive motor overload is analyzed, real-time current data of the drive motor is collected, and the current data is compared with the rated current value of the motor. If the current continuously exceeds the rated current value and the time exceeds the set standard, it is determined that the drive motor is overloaded. The judgment process, data comparison standard, and time setting requirements for each abnormal situation are converted into logical algorithms that the model can directly execute.

[0013] Specifically, the process of compiling the core algorithm using WASM for direct deployment on edge nodes involves: compiling the core algorithm using a dedicated WASM tool, converting the algorithm code into bytecode format, compressing the compiled bytecode to fit the storage capacity of the edge nodes, transmitting the compressed bytecode to each edge node via wired communication, installing the WASM runtime component in the node's runtime environment, loading the bytecode and completing initialization settings, and directly calling the preprocessed real-time data in the edge nodes for calculation and analysis.

[0014] Specifically, the preset thresholds for adapting different operating scenarios based on equipment type parameter combinations are as follows: for loading and unloading booms, the preset thresholds are adjusted in conjunction with the medium being transported and the operating period; when transporting liquefied natural gas, the structural safety factor threshold is reduced to address the risk of low-temperature embrittlement; when operating at night, the mechanical dynamic deviation value threshold is reduced to address operational errors that may be caused by poor visibility; for boarding ladders, the preset thresholds are adjusted in conjunction with the tidal period and wind speed level at the dock; and specific thresholds for corresponding operating scenarios are generated through combinations of different parameters.

[0015] Specifically, the integration of actual operating condition elements is used to define spatiotemporal labeling rules. The specific process is as follows: integrate actual operating condition elements, including berth identification, equipment unique number, work shift, tide level, and wind speed level; define spatial labels that include berth identification and equipment installation location coordinates, and time labels that include collection time, tide level code, and wind speed level code; record the generation rules of the spatial and time labels, and automatically attach the corresponding spatial and time labels when each piece of collected data is generated.

[0016] Specifically, the encrypted transmission protocol is used to upload edge node data to the cloud database. The specific process is as follows: the abnormal problems, real-time device status data and operation logs in the edge node are encrypted to generate an encrypted data packet with verification information. A dedicated communication link is established between the edge node and the cloud database. During the data packet transmission process, the data integrity is verified by the verification information. After receiving the data packet, the cloud database decrypts it using the corresponding key and stores the decrypted data according to device type, spatial tag and time tag. At the same time, the data upload time and the corresponding edge node identifier are recorded.

[0017] The beneficial effects of this invention are as follows: (1) By setting up edge dual-stack sensor nodes adapted to loading arms and boarding ladders and using the WASM modular deployment mechanism, accurate acquisition and low-latency diagnosis of the operating data of the two types of equipment can be achieved. For the mechanical dynamic data (rotation angle, hydraulic pressure, etc.) of the loading arms, the edge dual-stack sensor nodes use differentiated filtering algorithms (moving average filtering for crude oil and notch filtering for liquefied natural gas) to eliminate interference based on the medium type; for the operating status data (displacement, angle, etc.) of the boarding ladders, the displacement threshold is preset based on hydrological historical data, and the data time consistency is ensured through time synchronization; the core diagnostic algorithm is deployed at the edge after being compiled by WASM, and can directly call the preprocessed data to quickly output the equipment status (such as hydraulic leakage and motor overload judgment), which not only solves the problem of delayed response of traditional centralized monitoring, but also avoids the insufficient adaptation of the unified model to the different operating conditions of the two types of equipment, and improves the accuracy and real-time performance of anomaly identification; (2) By setting up a unified data spatiotemporal benchmark and cloud-edge collaborative management mechanism, effective integration of equipment data and safe joint operation and maintenance can be achieved. The unified spatiotemporal benchmark integrates working condition elements such as berth identification and tide level, and adds spatial labels (berth and installation coordinates) and time labels (collection time and hydrological code) to each data, ensuring that the data of loading and unloading arms and boarding ladders are aligned in the spatiotemporal dimension, avoiding diagnostic deviations caused by data misalignment; the cloud-edge collaborative mechanism uploads abnormal data and operation logs from the edge to the cloud for classified storage through encrypted transmission, and at the same time associates the early warning information with the real-time operation status of the equipment, and pushes linkage signals containing early warning levels and handling suggestions to the central control system. This not only solves the problem of fragmented and difficult-to-trace traditional monitoring data, but also realizes collaborative safety management and control of terminal equipment, reduces the investigation cost of operation and maintenance personnel, and improves the overall safety and operation and maintenance efficiency. Attached Figure Description

[0018] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0019] Figure 1This is an architecture diagram of a fault early warning method based on the WASM modular model for loading arms and boarding ladders according to the present invention. Figure 2 This is a data flow diagram of a fault early warning method based on the WASM modular model for loading arms and boarding ladders according to the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0021] Please see Figure 1-2 A fault early warning method based on the WASM modular model for loading arms and boarding stairs. S1: Construct edge dual-stack sensing nodes adapted to the operating characteristics of loading arms and boarding ladders, and preset acquisition schemes for the operational differences of the two types of equipment; for loading arms, capture mechanical dynamic data, and simultaneously preset interference filtering rules based on media type; for boarding ladders, record operation status data, and preset normal displacement thresholds based on historical hydrological data of the dock; based on the mechanical dynamic data and operation status data, perform parallel acquisition and preliminary cleaning of data from the two types of equipment through a time-series synchronization mechanism between nodes; S2: Configure diagnostic rules for abnormal issues of loading arms and boarding ladders; compile the core algorithm through WASM and deploy it directly to the edge node; the edge uses preprocessed real-time data to output real-time equipment status data and sets preset thresholds for different operating scenarios according to the equipment type parameter group.

[0022] S3: Establish a unified spatiotemporal benchmark for data that fits the operation scenarios of loading arms and boarding ladders, and define spatiotemporal labeling rules by integrating actual working conditions; add spatial and temporal labels to each piece of collected data, and align the time and spatial dimensions of the two types of equipment data through label association; S4: By establishing a cloud-edge collaborative management mechanism, edge nodes upload the aforementioned abnormal issues, real-time equipment status data, and operation logs to the cloud database via an encrypted transmission protocol; the generated early warning information is correlated and matched with the real-time operating status of the equipment, and a linkage signal containing the early warning level and suggested handling measures is pushed to the terminal control system.

[0023] Specifically, the construction of edge dual-stack sensing nodes adapted to the working conditions of loading arms and boarding ladders for data acquisition of the two types of equipment involves the following process: a node communication architecture is built using both wireless and wired Ethernet communication modes; capacity adaptation of node data storage is performed for different installation environments of loading arms and boarding ladders; and computing power allocation is performed for node data processing.

[0024] Specifically, the process of capturing mechanical dynamic data for the loading and unloading boom to obtain core operating status information of the loading and unloading boom is as follows: collecting the rotation angle of the boom joint, the real-time pressure value of the hydraulic pipeline, the vibration amplitude and frequency of different parts of the boom, and the medium transport temperature at the pipeline interface; collecting operating status information data in real time according to the operating cycle of the loading and unloading boom; and organizing the collected data in chronological order to obtain a continuous and complete mechanical dynamic data sequence.

[0025] Specifically, the combined medium type refers to the pre-set interference filtering rules of the loading and unloading arm to handle the interference of the environment and the medium on the data. The specific process is as follows: for crude oil transported by the loading and unloading arm, a moving average filtering algorithm is used to smooth the collected pressure data to eliminate pressure fluctuations caused by the flow of high-viscosity crude oil; for liquefied natural gas transported, a notch filter algorithm is used to filter specific frequency signals generated by electromagnetic interference in low-temperature environments; for chemical transported, a Kalman filter algorithm is used to correct the attenuation effect of corrosive media on data signals. Through the adaptation of different algorithms, the mechanical dynamic data truly reflects the operating status of the equipment.

[0026] Specifically, the operation status data recording for the boarding ladder is used to record the action information during the operation of the boarding ladder. The specific process is as follows: collect the lateral and longitudinal displacement of the boarding ladder body, the pitch and rotation angle of the ladder body, the lifting and rotation speed of the drive system, and the load-bearing weight of the ladder body steps. Record these data synchronously at the time point triggered by the boarding ladder operation command, and integrate the data of different operation stages to form an operation status data chain.

[0027] Specifically, the process of setting a normal displacement threshold for the boarding ladder based on the wharf's historical hydrological data to define the boarding ladder's operating status involves: collecting historical hydrological data from the wharf for a preset period, classifying and statistically analyzing the historical hydrological data, calculating the average displacement of the boarding ladder under different tidal periods and different wave levels, and using the average displacement as a benchmark, combined with the structural bearing capacity of the boarding ladder, determining the displacement threshold for each period.

[0028] Specifically, the timing synchronization mechanism is used to collect data from two types of devices in parallel. The specific process is as follows: a clock synchronization program is deployed in each edge node, a node close to the central control area of ​​the dock is designated as the clock reference node, and the remaining nodes receive the clock calibration signal sent by the reference node at a fixed period and correct their local clocks. The timestamp error of all nodes is controlled within a preset range, and data from two types of devices is collected simultaneously at a uniformly set collection frequency.

[0029] As a preferred technical solution of the present invention, S1 constructs an edge dual-stack sensing node adapted to the operating characteristics of the loading boom and boarding ladder. A node communication architecture is built using both wireless and wired Ethernet communication modes. The storage capacity of the node is adapted to the different operating environments of the loading boom and boarding ladder, and the computing power of the node's processing portion is allocated accordingly. Targeted acquisition schemes are designed for the two types of equipment: for the loading boom, its mechanical dynamic data is captured, specifically the rotation angle of the boom joints, the real-time pressure value of the hydraulic pipeline, the vibration amplitude and frequency of different parts of the boom, and the medium transport temperature at the pipeline interface. This data is collected in real-time according to the loading boom's operating cycle and organized chronologically. Simultaneously, interference filtering rules are preset based on the type of transported medium: for crude oil, a moving average filtering algorithm is used to smooth pressure fluctuations; for liquefied natural gas, a notch filter algorithm is used to filter low-temperature electromagnetic interference; and for chemical media, a Kalman filter is used. The algorithm corrects signal attenuation; for the boarding ladder, its operational status data is recorded, specifically the lateral and longitudinal displacement of the ladder body, pitch and rotation angles, lifting and rotation speeds of the drive system, and the load-bearing weight of the ladder treads. These data are recorded synchronously at the time of the operation command trigger, and integrated to form an operational status data chain; the average displacement under different tidal periods and wave levels is statistically analyzed, and the normal displacement threshold for each period is determined in combination with the structural bearing capacity of the boarding ladder; based on the mechanical dynamic data and operational status data, parallel acquisition and preliminary cleaning are achieved through a time-series synchronization mechanism between nodes: a clock synchronization program is deployed at each edge node, and the node closest to the central control area of ​​the dock is designated as the clock reference node. The remaining nodes receive the clock calibration signal from the reference node at a fixed period and correct their local clocks, keeping the acquisition timestamp error of all nodes within a very small range; then, data from both types of equipment are acquired simultaneously at a unified frequency, and the acquired data is preliminarily cleaned.

[0030] Specifically, the diagnostic rules configured for abnormal issues of the loading boom and boarding ladder are used to identify equipment abnormalities. The specific process is as follows: For the loading boom, the hydraulic system leakage is analyzed, real-time pressure data of the hydraulic pipeline is continuously collected, the pressure decay rate is calculated at fixed time intervals, and the calculation results are compared with the pressure decay safety range designed for the equipment. If the decay rate exceeds the pressure decay safety range, it is determined that the hydraulic system is leaking. For the boarding ladder, the drive motor overload is analyzed, real-time current data of the drive motor is collected, and the current data is compared with the rated current value of the motor. If the current continuously exceeds the rated current value and the time exceeds the set standard, it is determined that the drive motor is overloaded. The judgment process, data comparison standard, and time setting requirements for each abnormal situation are converted into logical algorithms that the model can directly execute.

[0031] Specifically, the formula for calculating the pressure decay rate of the loading / unloading arm hydraulic system is as follows: , in, To determine the pressure decay rate of the hydraulic system, The initial pressure value (unit: MPa) is the hydraulic pipeline pressure collected at the beginning of a certain monitoring cycle. The pressure value at time t (unit: MPa) represents the hydraulic pipeline pressure collected at any time within the monitoring period. Initial time (unit: min); t represents time t (in minutes), and Δt represents the pressure monitoring time interval.

[0032] In this embodiment, for the No. 1 loading and unloading arm of a 200,000-ton crude oil terminal (transporting medium is crude oil, design working pressure is 16MPa, and hydraulic leakage safe decay rate is ≤0.2MPa / min), the hydraulic system leakage is determined according to the formula: monitoring cycle start time. =0min; Initial pressure =15.2MPa, which is the normal operating pressure of the crude oil loading / unloading arm; 5 minutes later =5min, acquisition pressure =14.1MPa; Substituting into the formula, it can be calculated that the pressure decay rate of the hydraulic system exceeds the safe decay rate.

[0033] Specifically, the formula for calculating the normal displacement threshold of the boarding ladder is as follows: , in, The normal displacement threshold of the boarding ladder during a certain tidal period. This represents the historical average displacement of the boarding stairs during tidal periods. This represents the standard deviation of displacement data during tidal periods, reflecting the degree of fluctuation in the displacement data.

[0034] In this embodiment, regarding the No. 3 boarding ladder at a container terminal (during high tide operation, hydrological data on high tide levels at the terminal over the past three years shows the average displacement). =0.5m, which is the typical displacement of the boarding ladder at high tide at this wharf (less affected by wind and waves); standard deviation σ=0.12m, calculated from 1200 sets of high tide displacement data in the past 3 years), and the normal displacement threshold is set by calculation.

[0035] In this embodiment, for the No. 5 loading and unloading boom (transporting medium is ethanol, using the cloud-edge collaborative management mechanism of this invention), the edge node and the cloud are linked to realize anomaly early warning; the edge node collects the hydraulic pressure of the loading and unloading boom in real time (calculating the attenuation rate) and boom vibration data (calculating the matching degree). When the pressure attenuation rate exceeds the safety threshold or the vibration matching degree exceeds the fatigue threshold, the abnormal data, real-time pressure, vibration peak frequency and other parameters are uploaded to the cloud through an encrypted transmission protocol (AES-256 encryption); after receiving the data, the cloud associates it with the real-time operation information of the loading and unloading boom "currently in unloading operation state", generates a "level two early warning (hydraulic leakage + structural fatigue)", and pushes a linkage signal of "immediately stop unloading and arrange maintenance" to the terminal's central control system; the transporting medium is ethanol (which is a flammable chemical and requires a more stringent anomaly response to avoid leakage causing safety accidents); the encryption protocol is AES-256 (which meets the data security requirements of chemical terminals); the early warning level is set to level two (between "reminder to pay attention" and "emergency shutdown", balancing operational efficiency and safety).

[0036] Specifically, the process of compiling the core algorithm using WASM for direct deployment on edge nodes involves: compiling the core algorithm using a dedicated WASM tool, converting the algorithm code into bytecode format, compressing the compiled bytecode to fit the storage capacity of the edge nodes, transmitting the compressed bytecode to each edge node via wired communication, installing the WASM runtime component in the node's runtime environment, loading the bytecode and completing initialization settings, and directly calling the preprocessed real-time data in the edge nodes for calculation and analysis.

[0037] Specifically, the preset thresholds for adapting different operating scenarios based on equipment type parameter combinations are as follows: for loading and unloading booms, the preset thresholds are adjusted in conjunction with the medium being transported and the operating period; when transporting liquefied natural gas, the structural safety factor threshold is reduced to address the risk of low-temperature embrittlement; when operating at night, the mechanical dynamic deviation value threshold is reduced to address operational errors that may be caused by poor visibility; for boarding ladders, the preset thresholds are adjusted in conjunction with the tidal period and wind speed level at the dock; and specific thresholds for corresponding operating scenarios are generated through combinations of different parameters.

[0038] Specifically, the integration of actual operating condition elements is used to define spatiotemporal labeling rules. The specific process is as follows: integrate actual operating condition elements, including berth identification, equipment unique number, work shift, tide level, and wind speed level; define spatial labels that include berth identification and equipment installation location coordinates, and time labels that include collection time, tide level code, and wind speed level code; record the generation rules of the spatial and time labels, and automatically attach the corresponding spatial and time labels when each piece of collected data is generated.

[0039] In this embodiment, for the No. 4 boarding ladder at a bulk cargo terminal (which is greatly affected by wind and waves and requires real-time displacement exceeding limits), the boarding ladder displacement diagnosis algorithm (threshold calculation logic) is compiled and deployed to the edge node using WASM. The displacement diagnosis algorithm is compiled into WASM bytecode (adapted to the edge node's storage capacity). After deployment, the edge node collects the boarding ladder displacement data in real time; calculates the threshold for the current tidal period; if the real-time displacement exceeds the threshold, a "Level 1 sway warning" is immediately triggered at the edge, and the warning information is uploaded to the cloud. The WASM bytecode size is 80KB (the edge node storage capacity is typically 128KB-512KB); the current tidal period... (Typical displacement during low tide at bulk cargo terminals, with minimal impact from wind and waves); Threshold calculation logic is embedded in the WASM module (reducing data interaction between edge nodes and the cloud, lowering latency, and meeting the real-time diagnostic requirements of boarding ladders).

[0040] Specifically, the encrypted transmission protocol is used to upload edge node data to the cloud database. The specific process is as follows: the abnormal problems, real-time device status data and operation logs in the edge node are encrypted to generate an encrypted data packet with verification information. A dedicated communication link is established between the edge node and the cloud database. During the data packet transmission process, the data integrity is verified by the verification information. After receiving the data packet, the cloud database decrypts it using the corresponding key and stores the decrypted data according to device type, spatial tag and time tag. At the same time, the data upload time and the corresponding edge node identifier are recorded.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A WASM modular model-based fault warning method for loading and unloading arms and boarding ladders, characterized by, The method comprises the following steps: S1: Constructing an edge double-stack sensing node adapting to the working condition characteristics of the loading and unloading arm and the boarding ladder, and presetting a collection scheme according to the operation differences of the two types of equipment; For the loading and unloading arm, the mechanical dynamic data is captured, and interference filtering rules are preset in combination with the medium type; for the boarding ladder, the operation state data is recorded, and the normal displacement threshold is preset according to the historical hydrological data of the wharf; Based on the mechanical dynamic data and the operation state data, through the time sequence synchronization mechanism between nodes, the data of the two types of equipment is collected and preliminarily cleaned in parallel; S2: Diagnosing rules are configured for abnormal problems of the loading and unloading arm and the boarding ladder; the core algorithm is compiled through WASM and directly deployed on the edge node for operation; the edge end outputs real-time equipment state data using the preprocessed real-time data, and adapts the preset threshold of different working scenes according to the equipment type parameter combination; S3: Establishing a unified data space-time reference adapting to the working scenes of the loading and unloading arm and the boarding ladder, and fusing the actual working condition elements to regulate the space-time label rules; adding space labels and time labels to each collected data, and aligning the time dimension and space dimension of the data of the two types of equipment through label association; S4: Through the cloud-edge collaborative management mechanism, the edge node uploads the abnormal problems, real-time equipment state data and operation logs to the cloud database through an encrypted transmission protocol; the generated early warning information is associated and matched with the real-time operation state of the equipment, and a linkage signal containing the early warning level and the recommended treatment measures is pushed to the wharf central control system.

2. The method according to claim 1, characterized in that, The edge double-stack sensing node adapting to the working condition characteristics of the loading and unloading arm and the boarding ladder is used for data collection of the two types of equipment, and the specific process is as follows: a node communication architecture is built by adopting wireless communication and wired Ethernet dual communication mode, the node data storage is adapted in view of the different installation environment characteristics of the loading and unloading arm and the boarding ladder, and the node data processing is allocated in terms of computing power.

3. The method according to claim 1, characterized in that, The mechanical dynamic data captured for the loading and unloading arm is used to obtain the core running state information of the loading and unloading arm, and the specific process is as follows: the rotation angle of the boom joint of the loading and unloading arm, the real-time pressure value of the hydraulic pipeline, the vibration amplitude and frequency of different parts of the boom, and the medium conveying temperature at the pipeline interface are collected, the running state information data is collected in real time according to the running cycle of the loading and unloading arm, the collected data is arranged in time sequence, and a continuous and complete mechanical dynamic data sequence is obtained.

4. The method according to claim 1, characterized by, The preset interference filtering rules for the loading and unloading arm in combination with the medium type are used to process the interference of the environment and the medium on the data, and the specific process is as follows: for the crude oil medium transported by the loading and unloading arm, a moving average filtering algorithm is used to smooth the collected pressure data, so as to eliminate the pressure fluctuation caused by the flow of high-viscosity crude oil; for the liquefied natural gas medium, a notch filter algorithm is enabled to filter the specific frequency signals generated by electromagnetic interference in the low-temperature environment; for the chemical medium, a Kalman filter algorithm is used to correct the attenuation effect of the data signal caused by the corrosive medium, and the mechanical dynamic data is truly reflected by the adaptation of different algorithms to reflect the equipment running state.

5. The method of claim 1, characterized in that, The operation state data of the boarding ladder is recorded to record the action information during the operation of the boarding ladder. The specific process is: collecting the transverse displacement and longitudinal displacement of the boarding ladder body, the pitch angle and rotation angle of the boarding ladder body, the lifting speed and rotation speed of the driving system, and the load weight of the boarding ladder pedal, synchronously recording these data at the time point triggered by the boarding ladder operation instruction, and integrating the data of different operation stages to form an operation state data chain.

6. The method of claim 1, characterized in that, The normal displacement threshold of the boarding ladder is preset according to the historical hydrological data of the wharf to define the running state of the boarding ladder. The specific process is: collecting the historical hydrological data of the wharf at a preset time, classifying and counting the historical hydrological data, calculating the displacement mean value of the boarding ladder under different tidal periods and different wind wave levels, and determining the threshold of the displacement of each period based on the displacement mean value and the structural load capacity of the boarding ladder.

7. The method of claim 1, characterized in that, The time sequence synchronization mechanism is used to collect two types of device data in parallel. The specific process is: deploying a clock synchronization program in each edge node, designating a node close to the wharf control area as a clock reference node, and the remaining nodes receiving the clock calibration signal sent by the reference node at a fixed period and correcting the local clock to control the timestamp error of all nodes within a preset range, and collecting two types of device data at a unified set collection frequency.

8. The method of claim 1, characterized in that, The diagnostic rules for abnormal problems of the loading and unloading arm and the boarding ladder are configured to identify device abnormalities. The specific process is: for the loading and unloading arm, analyzing the hydraulic system leakage, continuously collecting real-time pressure data of the hydraulic pipeline, calculating the pressure decay rate at a fixed time interval, comparing the calculation result with the pressure decay safety range designed by the device, and determining that the hydraulic system is leaking if the decay rate exceeds the pressure decay safety range; for the boarding ladder, analyzing the driving motor overload condition, collecting real-time current data of the driving motor, comparing the current data with the rated current value of the motor, and determining that the driving motor is overloaded if the current continuously exceeds the rated current value and the exceeding time length reaches the set standard; the judgment process, data comparison standard, and time length setting requirement of each abnormal condition are converted into a model executable logic algorithm.

9. The method of claim 1, characterized in that, The core algorithm is compiled through WASM for direct deployment and operation in edge nodes. The specific process is: using WASM special tools to compile the core algorithm, converting the algorithm code into bytecode format, volume compressing the compiled bytecode to adapt to the storage capacity of the edge node, transmitting the compressed bytecode to each edge node through wired communication, installing the WASM runtime component in the node's running environment, loading the bytecode and completing the initialization setting, and directly calling the preprocessed real-time data in the edge node for calculation and analysis.

10. The method of claim 1, characterized in that, The preset threshold values are adapted to different work scenarios according to the combination of the device type parameters, and the specific process is as follows: for the loading and unloading arm, the preset threshold values are adjusted in combination with the conveying medium and the work period; when conveying liquefied natural gas, the structural safety factor threshold value is reduced to cope with the risk of low-temperature embrittlement, and when working at night, the mechanical dynamic deviation value threshold is reduced to cope with the operation errors that may be caused by poor visibility; for the boarding ladder, the preset threshold values are adjusted in combination with the wharf tide period and the wind speed grade; and the exclusive threshold values corresponding to the work scenarios are generated through the combination of different parameters.

11. The method of claim 1, characterized in that, The actual working condition elements are fused to define the space-time tag rules, and the specific process is as follows: the actual working condition elements, including the wharf berth identification, the device unique number, the work shift, the tide grade, and the wind speed grade, are fused; the space label contains the berth identification and the device installation position coordinates, the time label contains the collection time, the tide grade code, and the wind speed grade code, and the generation rules of the space label and the time label are recorded, so that the corresponding space label and time label are automatically attached to each piece of collected data.

12. The method of claim 1, characterized in that, The encrypted transmission protocol is used to upload the edge node data to the cloud database, and the specific process is as follows: the abnormal problems, the real-time device state data, and the operation logs in the edge node are encrypted to generate encrypted data packets with check information, a special communication link between the edge node and the cloud database is established, the data integrity is verified through the check information during the data packet transmission process, the cloud database receives the data packet and uses the corresponding key to decrypt the data, the decrypted data is classified and stored according to the device type, the space label, and the time label, and the data upload time and the corresponding edge node identification are recorded.

Images