Loading equipment operation state monitoring method and system based on cloud side-end cooperation

Abstract

The invention provides a loading equipment operation state monitoring method and system based on cloud side-end collaboration, and relates to the technical field of cloud side-end collaboration, side-end nodes self-organize sensing tasks according to adjacent node sensing states and loading equipment operation association information, and generate self-organizing results including task allocation, cooperation modes and time sequences; the cloud receives the result and generates a dynamic coordination rule in combination with equipment operation characteristics; the side end node optimizes a self-organizing process according to a rule, collects equipment operation state feedback data and uploads the data to the cloud; the cloud side fuses the data updating rule to generate logic and outputs an updating instruction; and the side end node receives an instruction iteration self-organizing process and uploads a result to form a bidirectional driving cycle. According to the invention, the execution efficiency and pertinence of the monitoring task are improved, the collaboration and stability of the system are enhanced, and the equipment state can be accurately and comprehensively monitored in real time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud edge-end collaboration, in particular to a loading equipment running state monitoring method and system based on cloud edge-end collaboration. BACKGROUND

[0002] In modern industrial production, loading equipment is a key component of logistics transportation and warehousing links, and the stability and reliability of its running state directly affect the efficiency and safety of the entire production process. The traditional loading equipment running state monitoring method mainly relies on manual regular inspection and local monitoring of single equipment. The manual inspection method not only consumes a lot of manpower and time, but also is difficult to achieve real-time and comprehensive monitoring, and is easy to miss some potential fault hidden dangers. Although the local monitoring of single equipment can obtain part of the running data of the equipment itself, due to the lack of collaborative analysis with other related equipment and environment, it is often difficult to accurately judge the overall running state and fault root cause of the equipment.

[0003] With the development of Internet of Things technology, some distributed monitoring-based systems have begun to be applied in the field of loading equipment monitoring. However, most of these systems adopt a centralized control architecture, all data are uploaded to a central server for processing, and the autonomy and flexibility of edge nodes are poor, and the requirements for network bandwidth and server performance are high. At the same time, the existing monitoring methods lack dynamic collaboration and adaptive adjustment capabilities when dealing with complex and variable loading equipment running scenarios, and cannot optimize monitoring tasks and collaborative rules in a timely manner according to the actual running state of the equipment and environmental changes, resulting in unsatisfactory monitoring effect and difficulty in meeting the needs of modern industrial production for efficient and reliable monitoring of loading equipment. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a loading equipment running state monitoring method and system based on cloud edge-end collaboration.

[0005] According to the first aspect of the present application, a loading equipment running state monitoring method based on cloud edge-end collaboration is provided, which comprises: The edge node self-organizes the perception task based on the adjacent node perception state and the loading equipment running associated information, and generates a perception task self-organization result, wherein the perception task self-organization result contains the perception task allocation of each edge node, the adjacent node cooperation mode and the task execution time sequence; The cloud receives the perception task self-organization result uploaded by each edge node, generates a dynamic collaboration rule in combination with the running characteristics of the loading equipment, and the dynamic collaboration rule defines the edge node perception task adjustment boundary, the data interaction mode and the abnormal response process; The edge node optimizes the self-organizing process of the perception task according to the dynamic coordination rule, collects and uploads the loading device operation state feedback data to the cloud, and the loading device operation state feedback data records the perception task execution effect and the device operation state change; The cloud fuses the loading device operation state feedback data of all edge nodes, updates the dynamic coordination rule generation logic, and outputs the rule update instruction. The edge node receives the rule update instruction, iterates the self-organizing process of the perception task, and generates a self-organizing process according to the second aspect of the present application, a loading device operation state monitoring system based on cloud edge cooperation is provided, the loading device operation state monitoring system based on cloud edge cooperation includes a machine readable storage medium and a processor, the machine readable storage medium stores machine executable instructions, and the processor executes the machine executable instructions, and the loading device operation state monitoring system based on cloud edge cooperation realizes the foregoing loading device operation state monitoring method based on cloud edge cooperation.

[0006] According to the third aspect of the present application, a computer readable storage medium is provided, and the computer readable storage medium stores computer executable instructions, and when the computer executable instructions are executed, the foregoing loading device operation state monitoring method based on cloud edge cooperation is realized.

[0007] According to any one of the above aspects, the technical effect of the present application is that: Firstly, the edge node self-organizes the perception task based on the adjacent node perception state and the loading device operation related information, generates a self-organizing result containing task allocation, cooperation mode and execution time sequence, fully gives play to the autonomy and flexibility of the edge node, can dynamically adjust the perception task according to the actual situation, and improves the execution efficiency and pertinence of the monitoring task. Secondly, the cloud generates dynamic coordination rules combined with the operation characteristics of the loading device, defines the edge node perception task adjustment boundary, data interaction mode and abnormal response process, and thirdly, the edge node optimizes the self-organizing process of the perception task according to the dynamic coordination rule, collects and uploads the loading device operation state feedback data, and the cloud fuses these data to update the dynamic coordination rule generation logic and output the rule update instruction, which can continuously optimize and adjust itself according to the actual operation state of the device and the environmental change, realizes real-time, accurate and comprehensive monitoring of the operation state of the loading device, effectively improves the operation reliability and production efficiency of the loading device, and reduces the risk of failure and maintenance cost. BRIEF DESCRIPTION OF DRAWINGS

[0008] Figure 1 The flowchart of the loading device operation state monitoring method based on cloud edge cooperation provided by the embodiment of the present application is shown; Figure 2This illustration shows a schematic diagram of the component structure of a cloud-edge-device collaborative vehicle loading equipment operation status monitoring system provided in this application embodiment for implementing the above-described cloud-edge-device collaborative vehicle loading equipment operation status monitoring method. Detailed Implementation

[0009] Figure 1 This paper illustrates a flowchart of a cloud-edge-device collaborative method and system for monitoring the operational status of loading equipment, as provided in an embodiment of this application. The detailed steps include: Step S110: The edge node self-organizes the sensing task and generates the sensing task self-organization result based on the sensing status of the adjacent nodes and the information related to the operation of the loading equipment. The sensing task self-organization result includes the sensing task allocation of each edge node, the cooperation mode of adjacent nodes, and the task execution sequence.

[0010] This embodiment takes port bulk cargo loading equipment as the application scenario. The port bulk cargo loading equipment includes mechanical transmission parts (such as belt conveyors and telescopic chutes), hydraulic control parts (such as hydraulic pumps, cylinders, and hydraulic valve groups), and electrical drive parts (such as motors, frequency converters, and PLC controllers). Edge nodes are deployed in various key parts of the equipment to realize distributed monitoring of the equipment's operating status.

[0011] Step S111: Each edge node collects its own sensing module's operating status and current sensing task information, including the sensing module's working mode, task execution progress, and remaining carrying capacity, and generates its own sensing capability data.

[0012] Edge nodes incorporate various sensing modules, such as vibration sensors, temperature sensors, current sensors, and pressure sensors. Taking an edge node deployed next to a belt conveyor motor as an example, its vibration sensor module currently operates in continuous sampling mode, while its temperature sensor module operates in intermittent sampling mode. The current sensing tasks include motor vibration signal acquisition (execution progress is calculated as the ratio of the acquired time to the total task time, e.g., the proportion of the acquired time to the total task time is A) and motor temperature acquisition (execution progress ratio is B). Remaining capacity is determined by comprehensively evaluating the percentage of CPU resources used (currently C), the percentage of memory resources used (currently D), and the remaining battery power (currently E). The generated self-sensing capability data includes parameters such as the operating mode of the aforementioned modules, the task execution progress ratio, and the remaining capacity ratio.

[0013] Step S112: The edge node establishes a perception state interaction link with the adjacent edge node, configures the link transmission protocol and data interaction format, transmits its own perception capability data and receives the own perception capability data of the adjacent edge node, forming a perception capability interaction set between nodes.

[0014] Adjacent edge nodes refer to nodes whose physical location is within a preset range, such as the edge node next to the belt conveyor motor and the adjacent edge nodes of the belt tension sensor and the reducer temperature sensor. These nodes establish a sensing status interaction link through a wireless communication module. The link transmission protocol is configured as a low-power wide-area network protocol, and the data interaction format adopts a structured text format, which specifies the names and data types of fields such as sensing module type, working mode, task ID, execution progress ratio, and remaining carrying capacity ratio. When establishing a link, node discovery is performed first by sending a broadcast packet containing the node ID and device type. After receiving a response, a handshake authentication is performed. After successful authentication, a stable connection is established. Subsequently, each node periodically transmits its own sensing capability data according to the configured protocol and format, and receives data sent by adjacent nodes. All received data is summarized with its own data to form a sensing capability interaction set between nodes.

[0015] Step S113: Retrieve the operation association information of the loading equipment, covering the operation dependencies and mutual influence paths of the mechanical transmission part, hydraulic control part, and electrical drive part, and generate the equipment operation association dataset.

[0016] The operational information of the loading equipment is stored in the local equipment information database of the edge nodes, which was imported during the initial equipment commissioning phase. The operational dependencies of the mechanical transmission components include: the operation of the belt conveyor depends on the output torque of the drive motor, which is affected by the meshing state of the reducer gears; the telescopic chute's extension and retraction depends on the extension and retraction of the hydraulic cylinder, which in turn depends on the oil supply pressure of the hydraulic pump. Interacting paths include: when the reducer gears wear, the load on the drive motor increases, leading to a rise in motor current and potentially increased vibration of the belt conveyor; excessively high hydraulic oil temperature reduces hydraulic viscosity, causing a decrease in the control accuracy of the hydraulic valve group and affecting the positioning accuracy of the telescopic chute. These dependencies and influence paths are described in the form of a directed graph, with each equipment component as a node and the dependencies and influence paths as directed edges. Edge attributes include influence coefficients and delay times, generating an equipment operation correlation dataset.

[0017] Step S114: Connect the perception capability interaction set between associated nodes with the device operation associated dataset, analyze the perception requirements of each operating part of the device and the adaptation relationship of the node perception capabilities, identify the operating parts of the device that require multi-node collaborative perception, and obtain a collaborative perception requirement list.

[0018] Step S114: Connect the perception capability interaction set between associated nodes with the device operation associated dataset, analyze the perception requirements of each operating part of the device and the adaptation relationship of the node perception capabilities, identify the operating parts of the device that require multi-node collaborative perception, and obtain a collaborative perception requirement list.

[0019] Step S1141: Analyze the interaction set of perception capabilities between nodes, extract the perception module type, perception range, data processing capability and remaining carrying capacity of each edge node, classify and organize them according to node identifier, mark the specific performance of each capability indicator, and obtain the node perception capability details.

[0020] A deep analysis of the interaction sets of sensing capabilities between nodes is performed, and the data is categorized according to node identifiers. For each node, the types of sensing modules it is equipped with are extracted in detail, such as vibration sensor modules, temperature sensor modules, pressure sensor modules, etc.; the sensing range of each module is clarified, such as the frequency range that the vibration sensor module can sense, and the measurement range of the temperature sensor module; the data processing capabilities of the nodes are sorted out, including functions such as filtering of raw data and feature extraction (such as peak value and mean value calculation); the remaining carrying capacity is evaluated by comprehensively judging the current CPU utilization, memory utilization, and communication bandwidth usage. The above information is arranged in order according to node identifiers to form a detailed node sensing capability list, in which each node's capability indicators are specifically described, such as "Node M01: Vibration sensor module (frequency range A-Hz), temperature sensor module (range BC℃), capable of time domain feature extraction, CPU utilization of remaining carrying capacity D%".

[0021] Step S1142: Analyze the equipment operation association dataset, extract the operation dependencies of the mechanical transmission part, hydraulic control part, and electrical drive part, determine the impact path and correlation strength of any part's operation status change on other parts, mark the specific links and correlation manifestations of the impact path, and obtain the equipment part association details.

[0022] From the equipment operation correlation dataset, the internal connections between mechanical transmission, hydraulic control, and electrical drive are broken down layer by layer. For the mechanical transmission part, the power transmission path between each component is clarified, such as the motor output shaft connecting to the reducer input shaft through a coupling, the reducer output shaft driving the drum, etc., and how a component failure (such as bearing wear) affects the operating state of subsequent components (such as drum speed fluctuations) through this path. For the hydraulic control part, the oil circuit connections and control relationships between hydraulic pumps, hydraulic cylinders, and hydraulic valves are analyzed. For example, how changes in hydraulic pump pressure affect the extension and retraction speed of hydraulic cylinders through the oil circuit, thereby changing the action of the actuator. For the electrical drive part, the transmission process of control signals from PLC output to frequency converter and then to motor is traced, as well as the impact of changes in parameters such as current and voltage on motor output power. For each change in operating state, the specific components and links involved in its impact path are identified, and the correlation strength is evaluated, such as "an abnormal increase in motor current (+E%) will lead to an increase in reducer temperature (+F℃), correlation strength level G", ultimately forming a detailed correlation list for the equipment part.

[0023] Step S1143: Associate the node perception capability details with the device part association details according to the device operation part, analyze the coverage of the perception capability of a single edge node to the perception requirements of the corresponding device operation part, mark the specific perception content with insufficient coverage, and obtain the single node coverage evaluation result.

[0024] On a per-equipment-operational basis, the node sensing capability details are matched with the equipment-specific association details. For example, for the "hydraulic pump" operating component, its sensing requirements include parameters such as outlet pressure, flow rate, pump body temperature, and vibration. Looking at the node sensing capability details, node N05, deployed near the hydraulic pump, has a pressure sensor module (covering pressure parameters) and a temperature sensor module (covering temperature parameters), but lacks a flow sensor module and a dedicated vibration sensor module. Therefore, it is determined that node N05's coverage of the hydraulic pump's sensing requirements is H%, with insufficient coverage for "flow rate" and "vibration" parameters. This type of analysis is performed for each equipment operating component, recording the coverage status and missing sensing content for each node in detail to form a single-node coverage assessment result.

[0025] Step S1144: For the equipment operation part with insufficient sensing capability in the single node coverage assessment result, analyze the sensing capability of other related equipment operation parts and corresponding edge nodes, screen adjacent nodes with supplementary sensing capability, label the screened nodes and supplementary sensing content, and obtain the collaborative node screening result.

[0026] For equipment operating parts with insufficient sensing capabilities in the single-node coverage assessment results, such as the missing flow and vibration parameters for the "hydraulic pump," other equipment operating parts associated with the hydraulic pump are identified based on the equipment part association details, such as node N06 deployed on the "hydraulic pipeline." Checking the sensing capability details of node N06 reveals that it is equipped with a flow sensor module, which can supplement the flow parameter collection needs of the hydraulic pump. Simultaneously, the adjacent node N07 within a 5-meter radius of the hydraulic pump is identified, as it is equipped with a vibration sensor module, which can supplement vibration parameter collection. Nodes N06 (supplementing flow) and N07 (supplementing vibration) are selected as candidate nodes for collaborative sensing of the hydraulic pump, and their respective supplementary sensing content is defined, forming the collaborative node selection results.

[0027] Step S1145: Based on the collaborative node screening results, define the participating nodes corresponding to each device operation part that needs collaborative sensing, the sensing responsibilities of each node and the collaboration priority, and obtain the collaborative sensing division of labor results.

[0028] For example, step S11451: parse the collaborative node screening results, extract the candidate collaborative nodes corresponding to the operating parts of each device that needs collaborative sensing, record the sensing capabilities, deployment locations and current task status of each candidate node, mark the core capabilities and task load of each node, and obtain the candidate node details.

[0029] From the collaborative node screening results, for each piece of equipment requiring collaborative sensing (such as a hydraulic pump), all candidate collaborative nodes (e.g., N05, N06, N07) are extracted. The specific sensing capabilities of each candidate node are recorded, including sensor type, measurement range, and accuracy level; deployment location information, such as the straight-line distance from the target equipment's operating part and installation orientation; and the current sensing tasks it is undertaking, such as the names of tasks like "oil temperature monitoring" and "pressure inspection," and the proportion of resources used. Based on this information, the core capabilities of the nodes are labeled, such as "high-precision pressure measurement" for N05 and "real-time flow monitoring" for N06; the task load is also labeled, such as N05's current load being 1% and N06's being J%, forming a detailed list of candidate nodes.

[0030] Step S11452: Evaluate the matching relationship between the sensing capabilities of each node in the candidate node list and the operating parts of the sensing equipment that need to be coordinated, select the nodes whose sensing capabilities match the sensing requirements as core participating nodes, mark the selection criteria and core capabilities of the core nodes, and obtain the core node determination results.

[0031] Based on the key sensing requirements of the equipment operation that needs to be sensed collaboratively (such as the pressure parameter of the hydraulic pump as a core monitoring indicator), the matching degree of the sensing capabilities of candidate nodes is evaluated. The pressure sensor module of node N05 has an accuracy level of K, its measurement range completely covers the working pressure range of the hydraulic pump, and its deployment location is closest to the hydraulic pump outlet (L meters), resulting in minimal signal transmission attenuation. Therefore, N05 is selected as the core participating node, with its core capability being "high-precision, real-time acquisition of hydraulic pump outlet pressure." The selection criterion is that "pressure parameters are the core indicator for evaluating the operating status of the hydraulic pump, and node N05 has the best pressure sensing capability (accuracy and distance)." This is recorded as the result of determining the core node.

[0032] Step S11453: Based on the core node determination results, assign the main sensing responsibilities to the core participating nodes, covering the key parameter collection, preliminary data processing and collaborative coordination of the equipment operation part, and mark the specific content and execution standards of each responsibility to obtain the core responsibility allocation results.

[0033] Assign key sensing responsibilities to core node N05: For key parameter acquisition, responsible for hydraulic pump outlet pressure (collected every M milliseconds, data accuracy must reach ±N%FS) and pump body temperature (collected every 0 seconds); for preliminary data processing, filtering pressure data (using a P-type filter) and calculating the pressure peak and average values ​​within 1 minute; for collaborative work, as the master node, receiving supplementary data from N06 and N07, synchronizing timestamps (error controlled within Q milliseconds), and aggregating data to the cloud. Define the execution standards for each responsibility, such as "filtering delay must not exceed R milliseconds," thus forming the core responsibility allocation result.

[0034] Step S11454: Assign auxiliary sensing responsibilities to the remaining candidate nodes, including secondary parameter collection, data backup, and emergency alternative sensing in case of core node failure. Mark the specific content and switching conditions of the auxiliary responsibilities to obtain the auxiliary responsibility assignment results.

[0035] Assign auxiliary responsibilities to candidate node N06 (non-core node): secondary parameter acquisition (hydraulic pump outlet flow rate, acquired every s milliseconds) and data backup (backing up its own acquired flow data and pressure data sent by N05 to local storage for a duration of T hours). Assign auxiliary responsibilities to candidate node N07: secondary parameter acquisition (hydraulic pump housing vibration, sampling frequency UHz) and emergency replacement sensing (automatically taking over N05's pressure and temperature acquisition responsibilities when N05's heartbeat packets are interrupted V times or no data is received for W seconds). Label the switching conditions, such as "Emergency replacement trigger condition: N05 heartbeat interruption ≥ V times or data interruption ≥ W seconds," to form the auxiliary responsibility allocation results.

[0036] Step S11455: Analyze the deployment location of each participating node and the physical distance to the operating parts of the sensing equipment that need to be coordinated, set the main link and backup link for data interaction, determine the information transmission order between each node, mark the link selection basis and switching process, and obtain the interaction link planning results.

[0037] The core node N05 is X meters away from the auxiliary node N06, and N07 is Y meters away. The primary links are set as follows: N06—N05 (traffic data) and N07—N05 (vibration data), both using the ZigBee protocol (channel A, baud rate Bbps). The backup link is N06—N07—N05 (activated when the signal strength of the direct link between N06 and N05 is < CdBm). The information transmission order is: N07 sends vibration data first, followed by N06 sending traffic data, with an interval of D milliseconds to avoid data conflicts. The link selection criterion is "primary link signal strength priority, switching to the backup link when the signal strength is below the threshold." The switching process is "detect signal strength—send link switching request—receive confirmation—switch and send data," and the results are recorded as interactive link planning results.

[0038] Step S11456: Based on the correlation strength of the equipment operation parts and the urgency of the sensing tasks, set the collaboration priority. The core parameter collection and collaborative coordination work has a higher priority than the secondary parameter collection and data backup. Mark the basis for priority setting and execution order to obtain the priority setting result.

[0039] According to the equipment association details, the association strength level between the hydraulic pump and the downstream hydraulic valve group is Level E (highest), and the urgency of its pressure parameter acquisition task is "high". The collaboration priority is set as follows: Pressure acquisition of core node N05 (priority 1) > Coordination of N05 (priority 2) > Vibration acquisition of N07 (priority 3) > Flow acquisition of N06 (priority 4) > Data backup (priority 5). The priority is based on "parameter impact on equipment failure weight: pressure (F%) > vibration (G%) > flow (H%)", and the execution order is "priority 1—2—3—4—5, with lower priority tasks delayed in case of resource conflicts," thus forming the priority setting result.

[0040] Step S11457: Associate the core responsibility allocation results, auxiliary responsibility allocation results, interaction link planning results and priority setting results, classify and organize the equipment operation parts that need to be collaboratively perceived, determine the list of participating nodes, responsibilities and collaboration requirements of each node for each equipment part, supplement the division of labor basis, and obtain the initial collaborative perception division of labor information.

[0041] The above results are integrated according to the "hydraulic pump" equipment operation section: the participating nodes are N05 (core), N06 (auxiliary), and N07 (auxiliary); the responsibilities of each node are as shown in the core responsibility allocation results and auxiliary responsibility allocation results; the collaboration requirements include the interaction link (primary / backup), information transmission order, and priority order. Supplement the division of labor basis, such as "N05 is the core node because of its optimal pressure sensing capability, and N06 is responsible for flow because it has a flow sensor", forming the initial collaborative sensing division of labor information.

[0042] Step S11458: Detect whether there are duplicate assignments in the initial collaborative sensing task allocation information where multiple nodes are assigned the same sensing responsibility, or whether there are omissions in the allocation where no node in the equipment operation part undertakes sensing responsibility; merge the nodes for duplicate assignments, supplement the necessary node sensing responsibilities for omissions, adjust the responsibility list of each node, and obtain the task allocation optimization result.

[0043] The inspection revealed that in the initial task allocation, both N05 and N07 were assigned the "pump body temperature acquisition" responsibility (duplicate allocation), and the "hydraulic pump motor current" parameter was not covered by any node (omission in allocation). The duplicate allocation was resolved by retaining N05's temperature acquisition responsibility and removing it from N07's allocation. The omission in allocation was addressed by selecting N08 (equipped with a current sensor) from the candidate nodes, located near the hydraulic pump motor, and assigning it the "motor current acquisition" responsibility. After the adjustment, the task allocation list for each node was updated, resulting in an optimized task allocation.

[0044] Step S11459: Evaluate the task capacity of each participating node in the division of labor optimization results, adjust the responsibility allocation ratio, balance the workload of each node, mark the basis and specific results of the load balancing adjustment, and obtain the load balancing result.

[0045] Assess the task capacity of each node: N05 current load rate I% (core responsibility) + J% (coordination work) = K%; N06 load rate L% (traffic acquisition) + M% (backup) = N%; N07 load rate 0% (vibration acquisition) = P%; N08 load rate Q% (current acquisition) = R%. It was found that N05's load rate K% is close to its maximum capacity threshold S%, requiring adjustment. The responsibility of "1-minute peak pressure data calculation" (accounting for T% of load rate) of N05 is transferred to N07, which has a lower load rate (P% + T% ≤ S%). The adjustment is based on the principle that "N05's load rate is close to the threshold, and N07 has sufficient remaining capacity." Record the changes in the load rate of each node before and after the adjustment to form a load balancing result.

[0046] Step S114510: Integrate the division of labor optimization results with the load balancing results, mark the participating nodes, responsibilities and collaboration priorities of each part of the equipment operation that needs to be coordinated, supplement the core basis and execution order of the division of labor, and generate the final collaborative sensing division of labor results.

[0047] After integration, the collaborative sensing task allocation results for the "hydraulic pump" are as follows: participating nodes N05 (core), N06, N07, and N08; the allocation of responsibilities (including adjustments), collaboration priorities (such as priorities 1-5), and execution order are clearly defined. Supplementary core criteria for task allocation (such as sensing capability matching and load balancing) are added to form the final collaborative sensing task allocation result.

[0048] Step S1146: Based on the collaborative sensing requirement list, each edge node negotiates the sensing task splitting method through the sensing status interaction link, and allocates the collaborative sensing tasks to the corresponding edge nodes according to the node's sensing capabilities and deployment location. The task responsibilities and coverage of each node are defined, and the correspondence between the specific sensing content and the device operation part is marked to obtain the task splitting and allocation results.

[0049] The collaborative sensing requirements list clarifies the collaborative sensing tasks for each part of the equipment operation. For example, the "belt conveyor drive unit" needs to monitor motor vibration, bearing temperature, reducer temperature, and conveyor belt tension. Relevant edge nodes (motor-side node D01, reducer-side node D02, and tension sensor node D03) negotiate task splitting through a sensing status interaction link (the established ZigBee link). Based on the node's sensing capabilities, D01 (with vibration and temperature sensors) is responsible for motor vibration (X-axis and Y-axis directions) and motor bearing temperature; D02 (with temperature sensor) is responsible for reducer input and output shaft temperatures; and D03 (with tension sensor) is responsible for conveyor belt tension. According to deployment location, the vibration sensor of D01 is installed in the bearing housing at the non-drive end of the motor (covering the motor's running part), and the sensor of D02 is close to the reducer housing (covering the reducer's running part). Responsibilities are defined: D01 is the main data aggregation node, and D02 and D03 are data providing nodes. Correspondence is labeled, such as "D01 vibration data—motor running part," forming the task splitting and allocation result.

[0050] Step S1147: Combine the physical deployment locations between nodes and the operating rhythm of the equipment to optimize the execution sequence of the sensing tasks, adjust the node sensing start time difference, so that adjacent nodes participating in the same collaborative sensing task can start sensing operations synchronously, reduce data interaction delay and task connection gap, and obtain timing optimization results.

[0051] The physical distances between nodes D01, D02, and D03 are A meters and B meters, respectively. Equipment operation rhythm: When the belt conveyor starts, the motor starts first (at time 0), and the conveyor belt runs after a C-second reducer. Tension adjustment begins D seconds after startup. Optimized timing: D01 starts sensing when the motor start signal is issued (at time 0); D02 starts at time 0 + E seconds (considering signal transmission delay, advancing the reducer's actual operation by F seconds); D03 starts at time 0 + G seconds (advancing tension adjustment by H seconds). Adjusting the startup time difference ensures that the "startup phase data" collected by D01, D02, and D03 is continuous on the timeline, data interaction delay is controlled within I milliseconds, and task connection gaps (such as the time D01 waits for D02 data) are reduced to within J seconds, resulting in optimized timing.

[0052] Step S1148: Define the collaboration mode between nodes, set the main data interaction node, transmission path selection criteria and information sharing scope in the collaborative perception task, set the trigger conditions and feedback mechanism for data interaction, mark the specific content and format requirements of data interaction, and obtain the collaboration mode definition result.

[0053] The collaborative sensing mechanism for the "belt conveyor drive unit" is defined as follows: The master node is D01, responsible for aggregating data from D02 and D03 and uploading it to the cloud. Transmission path selection criteria: Prioritize links with signal strength > KdBm; if L consecutive transmission failures occur, switch to a backup link (signal strength > MdBm). Information sharing is limited to "drive unit operating status data" (vibration, temperature, tension), excluding node internal logs. Triggering conditions: Timed triggering (data upload every N seconds) and event triggering (immediate upload when vibration value > 0 mm / s). Feedback mechanism: After receiving data, the master node D01 returns an ACK frame containing "reception status (success / failure)" and "data checksum" within P milliseconds. If the sending node does not receive an ACK, it retransmits Q times and then switches links. The data format is JSON, containing fields: device ID, parameter name, value, timestamp (accurate to microseconds), and node ID. These fields are labeled as part of the collaborative mechanism definition.

[0054] Step S1149: Combine the task splitting and allocation results, timing optimization results, and collaboration method definition results, classify and store them according to edge node identifiers, and record the perception task content, execution time, collaboration objects, and data interaction requirements of each edge node in the combined data to obtain the initial perception task self-organization results.

[0055] The combined task breakdown and allocation results (tasks for each node), timing optimization results (startup time difference), and collaboration mode definition results (master node, triggering conditions, etc.) of the "belt conveyor drive unit" are stored in categories according to node identifiers D01, D02, and D03: D01 records "vibration acquisition (started at time 0, continuous sampling), temperature acquisition (started at time 0, interval R seconds), collaborating objects D02 / D03, data uploaded to the cloud every N seconds"; D02 records "reducer temperature acquisition (started at 0+E seconds, interval S seconds), collaborating object D01, data sent to D01," etc. The combined data is stored in the node's local task table, forming the initial sensing task self-organization results.

[0056] Step S11410: Retrieve historical data on the self-organization effect of perception tasks, identify historical cases where node load exceeds the threshold due to task allocation, or where data collection time windows overlap due to timing issues; based on the identified historical cases, adjust the task allocation ratio and timing arrangement in the initial self-organization result of perception tasks, update the details of the collaboration method, and obtain the corrected self-organization result of perception tasks.

[0057] Historical data was retrieved, identifying Case 1: Node D01 experienced data loss due to simultaneously undertaking three high-frequency data acquisition tasks (vibration, temperature, and data aggregation), exceeding the load rate threshold T%. Case 2: The startup time difference between D02 and D03 was too small (only U seconds), causing overlapping data acquisition time windows (V time - W time), resulting in data conflicts. Based on Case 1, the task allocation ratio of D01 was adjusted, transferring some preprocessing in "data aggregation" (such as temperature average calculation) to D02. Based on Case 2, the startup time of D03 was adjusted to 0 time + G + X seconds to avoid window overlap. The collaboration method details were updated, adding "real-time node load monitoring (checked every Y seconds), automatic task migration request when load rate > Z%", forming a corrected self-organizing result for the perceived tasks.

[0058] Step S11411: Standardize the terminology and data format of the corrected perception task self-organization result, add metadata fields to the data structure to describe the input source and output target of each part, and add identifiers to indicate the interface with the cloud dynamic collaboration rules to generate the final perception task self-organization result.

[0059] Terminology standardization: "Vibration value" is standardized as "effective value of vibration acceleration," and "temperature" as "surface temperature." Data format standardization: Numerical values ​​are retained to A decimal places, timestamps use UTC format, and Boolean values ​​use "True / False." Metadata fields are added: "Task allocation result input source: Inter-node perception capability interaction set V1.0," and "Time series optimization result output target: Collaborative perception execution module." Cloud interface identifiers are added: "Interface version: DR-2023-XX," and "Rule type: Perception task adjustment rule," ultimately generating the perception task self-organization result.

[0060] Step S115: Combining the physical deployment locations between nodes and the operating rhythm of the equipment, optimize the execution sequence of sensing tasks, adjust the node sensing start-up time difference, so that adjacent nodes participating in the same collaborative sensing task can start sensing operations synchronously, reduce data interaction delay and task connection gap, and obtain timing optimization results.

[0061] In the collaborative sensing task of the belt conveyor drive system, the physical deployment locations of each node are different. The nodes near the motor and the nodes near the reducer are relatively close, while the nodes near the current sensor in the control cabinet are relatively far away. The equipment operation rhythm is as follows: when the belt conveyor starts, the motor starts first, and after a specific time, the reducer drives the belt to run. The tension adjustment system starts working at a specific time after startup. When optimizing the execution timing of the sensing task, the nodes near the motor and the current sensor start sensing synchronously when the motor start signal is issued; the nodes near the reducer start sensing at a specific time after the motor starts; and the tension sensor node starts sensing at a specific time after startup. By adjusting the timing difference of sensing start-up of each node, it is ensured that nodes participating in the same collaborative sensing task begin data acquisition the instant the relevant components of the equipment start operating, reducing data interaction delays and task connection gaps, thus obtaining the timing optimization result.

[0062] Step S116: Define the collaboration mode between nodes, set the main data interaction node, transmission path selection criteria and information sharing scope in the collaborative perception task, set the trigger conditions and feedback mechanism for data interaction, mark the specific content and format requirements of data interaction, and obtain the collaboration mode definition result.

[0063] In the collaborative sensing task of the telescopic chute control system, the edge node within the PLC control cabinet is defined as the data interaction master node, responsible for receiving data from each collaborative node and summarizing and forwarding it. The transmission path selection criteria are: priority is given to wireless links with signal strength meeting specific requirements; when the signal strength falls below a specific threshold, the system switches to a backup link. Information sharing is limited to parameters directly related to the operation of the telescopic chute, such as cylinder telescopic displacement, hydraulic valve group working status, and chute position feedback signals. Data interaction triggering conditions include: timed triggering (data upload once every specific time period) and event triggering (immediate upload when the chute position deviation exceeds a specific value). The feedback mechanism is as follows: after receiving data, the master node returns an acknowledgment frame containing the reception status and data verification results within a specific time period; if the sending node does not receive the acknowledgment frame, it retransmits after a specific time period, and if retransmission fails after a specific number of attempts, it switches to the backup path. The specific content of the data interaction includes the equipment operating parameter name, value, timestamp, and node ID, formatted as structured text. Labeling this content yields the defined collaboration method.

[0064] Step S117: Combine the task splitting and allocation results, timing optimization results, and collaboration method definition results, classify and store them according to edge node identifiers, and record the perception task content, execution time, collaboration objects, and data interaction requirements of each edge node in the combined data to obtain the initial perception task self-organization results.

[0065] The task breakdown and allocation results, timing optimization results, and collaboration method definition results for each device's operational components are combined. The data is stored categorized by edge node ID, with each node's record including: sensing task content, execution time, collaborating partners, and data interaction requirements. The combined data is stored in the local task list of the edge node, forming the initial self-organizing result of the sensing task.

[0066] Step S118: Retrieve historical data on the self-organization effect of perception tasks, identify historical cases where node load exceeds the threshold due to task allocation, or where data collection time windows overlap due to timing issues; based on the identified historical cases, adjust the task allocation ratio and timing arrangement in the initial self-organization result of perception tasks, update the details of the collaboration method, and obtain the corrected self-organization result of perception tasks.

[0067] Historical data on the self-organization effects of sensing tasks is stored in a cloud database, and edge nodes acquire it through periodic synchronization. Data from the most recent specific time period is retrieved to identify historical cases where task allocation caused node load to exceed thresholds, or where data collection time windows overlapped due to timing issues. Based on these identified historical cases, the task allocation ratios and timing arrangements in the initial sensing task self-organization results are adjusted, and collaboration details are updated, such as adding a real-time node load monitoring mechanism. When the load exceeds a specific threshold, a task migration request is automatically triggered, resulting in a corrected sensing task self-organization result.

[0068] Step S119: Standardize the terminology and data format of the corrected perception task self-organization result, add metadata fields to the data structure to describe the input source and output target of each part, and add identifiers to indicate the interface with the cloud dynamic collaboration rules to generate the final perception task self-organization result.

[0069] Terminology standardization unifies "vibration signal" as "mechanical vibration acceleration signal," and "temperature" as "component surface temperature," etc. Data format standardization ensures that all numerical data retains a specific number of decimal places, timestamps use a specific format, and Boolean values ​​are represented by specific characters. Metadata fields are added to the data structure, such as "Task allocation result input source: inter-node perception capability interaction set" and "Time series optimization result output target: collaborative perception execution module." Cloud-based dynamic collaborative rule interface identifiers are added, such as "Interface version: DR-2023-001" and "Rule type: perception task adjustment rule," ultimately generating a self-organizing result for perception tasks containing standardized content, metadata, and interface identifiers.

[0070] Step S120: The cloud receives the self-organization results of the sensing tasks uploaded by each edge node, and generates dynamic coordination rules in combination with the operating characteristics of the loading equipment. The dynamic coordination rules define the adjustment boundary of the edge node sensing tasks, the data interaction method and the abnormal response process.

[0071] The monitoring center server, deployed in the cloud at the port terminal, possesses powerful data processing and storage capabilities. Each edge node uploads the self-organized results of its sensing tasks to the cloud via a wireless network. The cloud centrally processes these results and, combined with the inherent operating characteristics of the loading equipment, formulates dynamic collaborative rules to guide the subsequent sensing tasks executed by the edge nodes.

[0072] Step S121: Build a cloud-based platform for receiving the self-organized results of the sensing task, configure multi-channel data receiving interfaces, classify and receive the self-organized results of the sensing task uploaded by each edge node according to the edge node identifier and time sequence, record the data receiving time, transmission status and integrity information, and generate a receiving ledger.

[0073] The cloud-based receiving platform adopts a distributed architecture, deploying multiple data receiving service instances, each responsible for edge node data in a specific region. It is configured with multi-channel data receiving interfaces, including HTTP, MQTT, and WebSocket interfaces. Each interface is configured with its corresponding port number, IP address, and authentication method. When receiving data, it categorizes data by edge node identifier and sorts the data in chronological order according to the timestamp of data upload. It records the data reception time, transmission status, and integrity information, storing this information in a database table in the order of reception to generate a receiving ledger.

[0074] Step S122: Analyze the self-organization results of sensing tasks in the receiving ledger, extract the sensing task allocation of each edge node, the cooperation mode of adjacent nodes, the task execution sequence and the operating part of the covered equipment, associate the cross content of the self-organization results of sensing tasks of different edge nodes, extract common associations and differential features, and obtain the self-organization result parsing data.

[0075] The self-organization results of sensing tasks from each edge node are read from the receiving ledger, and the sensing task allocation, the cooperation mode of adjacent nodes, the task execution sequence, and the operating parts of the covered equipment are parsed out. The cross-content of the results from different nodes is correlated, common correlations and differential features are extracted, and the parsed information is summarized to obtain the self-organization result parsing data.

[0076] Step S123: Retrieve the operating characteristic data of the loading equipment, including the transmission ratio range of the mechanical transmission part, the pressure regulation characteristics of the hydraulic control part, the power output law of the electrical drive part, and the fault propagation threshold of each part, and generate the equipment operating characteristic dataset.

[0077] The operational characteristics data of the loading equipment are stored in a cloud-based equipment model database and retrieved by equipment model. The transmission ratio range of the mechanical transmission section is as follows: the reducer transmission ratio of the belt conveyor drive system is within a specific range; the gear and rack transmission ratio of the telescopic chute is within a specific range. The pressure regulation characteristics of the hydraulic control section are as follows: the working pressure regulation range of the hydraulic pump is within a specific range, and the pressure regulation response time does not exceed a specific time; the speed regulation range of the hydraulic cylinder is within a specific range. The power output pattern of the electrical drive section is as follows: the rated power of the drive motor is a specific value; the power at startup is a specific multiple of the rated power, and returns to the rated power after a specific time; the output frequency range of the frequency converter is within a specific range, corresponding to a specific range of motor speed. Fault propagation thresholds for each part are also defined: when the motor bearing temperature exceeds a specific value, it may cause the reducer temperature to rise within a specific time; when the hydraulic oil contamination level exceeds a specific level, it may cause the hydraulic valve to stick within a specific time. This data is structured and stored to generate an equipment operational characteristics dataset.

[0078] Step S124: Associate the self-organization result parsing data with the equipment operation characteristic dataset, compare the matching relationship between the coverage of the sensing task and the key parts of the equipment operation, analyze the fit relationship between the execution sequence of the sensing task and the rhythm of the equipment operation, identify the parts in the self-organization result of the sensing task that do not match the equipment operation characteristics, and obtain the adaptation deviation analysis results.

[0079] Step S1241: Extract the sensing task coverage, execution sequence and collaboration method from the self-organization result parsing data, classify and organize them according to the equipment operation part, mark the sensing task details and timing arrangement corresponding to each part, and obtain the sensing task deployment details.

[0080] The sensing tasks covered in the self-organizing results analysis data include the operating parts of equipment such as motors, reducers, hydraulic pumps, cylinders, and PLC controllers. The data is categorized by operating part, and the details and timing of the sensing tasks for each part are labeled to obtain a detailed deployment list of sensing tasks.

[0081] Step S1242: Extract the core content of the equipment operation characteristic dataset, including the key operating parameter ranges, operating status change cycles, and fault-sensitive periods of each equipment operation part, and mark the specific performance of each parameter range and cycle to obtain the core data of equipment characteristics.

[0082] Core data was extracted from the equipment operating characteristic dataset: For the motor operating section, key operating parameters ranged within specific speed and temperature ranges; the operating state change cycle consisted of the startup phase and the stable operation phase; fault-sensitive periods were a specific time after startup and continuous operation exceeding a specific time. For the hydraulic pump operating section, key parameters ranged within specific pressure and flow ranges; the operating state change cycle consisted of the loading phase and the stable operation phase; the fault-sensitive period was the loading phase. The upper and lower limits of these parameter ranges, the time nodes of the cycles, and the specific duration of the fault-sensitive periods were labeled to obtain the core data of the equipment characteristics.

[0083] Step S1243: Associate the deployment details of the sensing tasks with the core data of the equipment characteristics one by one according to the equipment operation parts, compare the correspondence between the coverage of the sensing tasks and the key operating parameters of the equipment, identify the key parameters that are not covered, mark the specific content and impact of the uncovered parameters, and obtain the coverage deviation analysis.

[0084] Taking the hydraulic pump operation section as an example, the parameter covered in the sensing task deployment details is pressure, while the key operating parameters of the hydraulic pump in the core equipment characteristic data also include flow rate. Therefore, the key parameter that is not covered is flow rate, which means that changes in the volumetric efficiency of the hydraulic pump cannot be monitored, potentially leading to missed detections of pump wear faults. This situation will be recorded in the coverage deviation analysis.

[0085] Step S1244: Analyze the fit between the execution timing of the sensing task and the cycle of equipment operating status changes, investigate cases where the start time of the sensing task misses the critical period of equipment operating status changes, mark the specific manifestations and impacts of the missed period, and obtain the timing fit deviation analysis.

[0086] The startup phase is a critical period in the operating state change cycle of the motor. In the sensing task deployment details, the motor temperature sampling startup time is the exact startup moment, and the sampling interval is a specific time interval. This results in only a limited number of temperature data points being collected during the startup phase, missing the rapid temperature changes within this phase. Specifically, this manifests as an inability to capture the dynamic temperature characteristics during startup, affecting the evaluation of motor startup performance. This situation will be recorded in the timing fit deviation analysis.

[0087] Step S1245: Evaluate the compatibility between the perception task collaboration method and the correlation strength between the equipment operation parts, investigate the mismatch between the number of collaborative nodes and the collaboration frequency and the degree of influence between the equipment parts, mark the specific manifestations and reasons of the mismatch, and obtain the collaboration method deviation analysis.

[0088] The telescopic chute's operating section is highly interconnected with the hydraulic and electrical systems. The number of collaborative nodes and the collaboration frequency in the sensing task collaboration mode are specific. However, based on core equipment characteristics, the operating status of this section changes rapidly, and the current collaboration frequency cannot meet data synchronization requirements, leading to delays in collaborative decision-making. The mismatch is due to the collaboration frequency setting not considering the interconnectedness and response speed of the equipment's operating section; this will be recorded in the collaboration mode deviation analysis.

[0089] Step S1246: Integrate coverage deviation analysis, timing matching deviation analysis, and collaboration mode deviation analysis, label the equipment operation part, sensing task, and specific performance corresponding to each deviation, supplement the scenario conditions that caused the deviation, and obtain preliminary deviation analysis results.

[0090] This paper integrates the findings regarding issues such as incomplete hydraulic pump flow parameters, insufficient temperature sampling points during motor startup, and insufficient coordination frequency of the telescopic chute. It labels the corresponding equipment operating components, sensing tasks, and specific manifestations for each deviation, supplements the scenario conditions that caused the deviation, and obtains preliminary deviation analysis results.

[0091] Step S1247: Analyze the causes of various deviations in the preliminary deviation analysis results, distinguish between deviations caused by unreasonable self-organizing processes of perception tasks and deviations caused by missing equipment operating characteristic data, mark the specific scenarios corresponding to the causes, and obtain the deviation cause analysis results.

[0092] The reasons for the lack of coverage of hydraulic pump flow parameters: Inspection of the self-organizing process of the sensing task revealed that the hydraulic pump node was not equipped with a flow sensor, indicating an unreasonable process. The reason for insufficient temperature sampling points during the motor startup phase: The equipment operating characteristic data did not explicitly indicate the need for increased sampling during the startup phase, leading to an unreasonable self-organizing timing arrangement. The reason for insufficient telescopic chute collaboration frequency: The collaboration frequency calculation formula in the self-organizing process did not include the correlation strength parameter. The specific scenarios corresponding to the identified reasons were used to obtain the deviation cause analysis results.

[0093] Step S1248: Based on the deviation cause analysis results, filter out deviations caused by unreasonable self-organization results of perception tasks, exclude non-substantive deviations caused by missing data, mark the specific range and impact of effective deviations, and obtain the effective deviation screening results.

[0094] From the analysis of deviation causes, the following were identified as valid deviations: incomplete hydraulic pump flow coverage and insufficient coordination frequency of the telescopic chute. Insufficient temperature sampling points during motor startup, due to missing equipment operating characteristic data, were excluded as non-substantial deviations. The specific range and impact of the valid deviations were then labeled, resulting in the valid deviation screening results.

[0095] Step S1249: Sort the effective deviation screening results according to the degree of impact of the deviation on the equipment operation status monitoring, highlight the key deviations that affect the completeness and accuracy of monitoring, mark the sorting basis and the core impact of the key deviations, and obtain the deviation priority sorting results.

[0096] Assessing the impact of effective deviations: Incomplete coverage of hydraulic pump flow rates leads to the inability to monitor pump volumetric efficiency, affecting the overall performance assessment of the hydraulic system; insufficient coordination frequency of the telescopic chute results in inaccurate positioning, affecting loading accuracy. Deviations are ranked from highest to lowest impact based on the probability of missed fault detection and the severity of the consequences. The critical deviation is the incomplete coverage of hydraulic pump flow rates, and its core impact is the lag in hydraulic system fault warnings, resulting in a deviation priority ranking.

[0097] Step S12410: Integrate the effective deviation screening results and deviation priority ranking results, determine the corresponding equipment part, perception task related information, specific performance and impact of each deviation, supplement the core basis of deviation analysis, and generate the final adaptation deviation analysis results.

[0098] The deviation in the integrated hydraulic pump operation section is due to the lack of coverage of flow parameters. The associated sensing task is the hydraulic pump pressure acquisition task, which manifests as the inability to calculate volumetric efficiency, resulting in a high degree of impact. The deviation in the telescopic chute operation section is due to insufficient coordination frequency. The associated sensing task is the chute position coordination control task, which manifests as positioning delay, resulting in a medium degree of impact. The core basis includes deviation cause analysis and impact assessment, forming the final adaptation deviation analysis results.

[0099] Step S125: Based on the adaptation deviation analysis results, define the edge node perception task adjustment boundary, divide the perception task types, scope and adjustment range that each edge node can adjust independently, divide the perception task categories that can only be adjusted after submitting an application to the cloud, mark the specific division criteria and adjustment process of the two types of tasks, and obtain the perception task adjustment boundary rules.

[0100] Based on the adaptation deviation analysis results, the adjustment boundaries for sensing tasks are defined as follows: Sensing task types that can be adjusted autonomously include sampling frequency, data upload interval, and local data storage cycle; the adjustment range is a specific proportion of the original parameters; the adjustment magnitude is such as adjusting the sampling frequency within a specific range. Sensing task categories requiring cloud-based adjustments include adding / deleting sensing parameters, changing the main collaborating node, and adjusting the operational aspects of covered devices. The classification criteria are: adjustments affecting the load of a single node or local data interaction are considered autonomous adjustments; adjustments affecting multi-node collaboration, the overall monitoring range, or involving hardware changes are considered requests for adjustment. The adjustment process is as follows: During autonomous adjustments, the node records the parameters before and after the adjustment and synchronizes them to adjacent nodes; during requests for adjustment, the node sends an application message containing the reason for the adjustment, the target parameters, and the expected impact to the cloud, and executes the adjustment after receiving approval instructions from the cloud. These elements are labeled to obtain the sensing task adjustment boundary rules.

[0101] Step S126: Design the data interaction rules for edge nodes. Based on the priority of collaborative sensing tasks and data transmission requirements, set the frequency of data transmission between nodes, data packet format, transmission link selection criteria, and data loss response measures. Mark the differences in transmission parameters corresponding to tasks with different priorities to obtain the data interaction rules.

[0102] Collaborative sensing tasks are prioritized into high, medium, and low. High-priority tasks transmit data at a specific frequency, medium-priority tasks at a specific frequency, and low-priority tasks at a specific frequency. Data packets are uniformly formatted as structured text containing task ID, parameter name, value, timestamp, and checksum; high-priority packets have an additional priority identifier field. Transmission link selection criteria: high-priority tasks preferentially use wired Ethernet links, and select specific types of wireless links; medium- and low-priority tasks use specific types of links. Data loss handling measures: high-priority tasks employ retransmission mechanisms and backup link switching; medium-priority tasks employ retransmission mechanisms; low-priority tasks do not retransmit and wait for the next cycle. The differences in parameters such as transmission frequency, link type, and retransmission count corresponding to different priorities are marked to obtain the data interaction rules.

[0103] Step S127: Construct anomaly response process rules. For situations such as interruption of perception task execution, failure of data interaction, and sudden changes in device operating status, set the autonomous processing process of edge nodes, the content and time limit requirements for anomaly information reporting, define the remote collaborative handling method in the cloud, and mark the differences in the processing process corresponding to different anomaly types to obtain anomaly response process rules.

[0104] Anomaly types include sensing task execution interruption, data interaction failure, and sudden changes in device operating status. The autonomous handling process for sensing task execution interruption is as follows: immediately activate the backup sensor module and switch to degraded sampling mode; the reported anomaly information includes the fault module ID, fault time, and attempted recovery measures; the reporting timeframe is a specific time after the fault occurs. The autonomous handling process for data interaction failure is as follows: initiate link switching and cache local data; the reported information includes the failed link ID, number of failures, and cached data volume; the reporting timeframe is a specific time after a specific number of consecutive failures. The autonomous handling process for sudden changes in device operating status is as follows: trigger emergency sampling and initiate local anomaly diagnosis; the reported information includes the mutation parameters, mutation magnitude, and diagnostic results; the reporting timeframe is a specific time after the mutation occurs. The cloud-based remote collaborative handling method is as follows: after receiving anomaly information, high-priority anomalies are handled with instructions within a specific timeframe, and medium- and low-priority anomalies are handled with instructions within a specific timeframe. The differences in handling processes corresponding to different anomaly types are marked to obtain anomaly response process rules.

[0105] Step S128: Merge the boundary adjustment rules for perception tasks, data interaction method rules, and abnormal response process rules, classify and store them according to the applicable scenarios of the rules, define the trigger conditions and execution priority order between each rule in the rule base, detect and parse rule conflicts, and add rule chain trigger clauses to obtain the initial dynamic collaborative rules.

[0106] The rules for adjusting the boundary of perception tasks, the rules for data interaction methods, and the rules for abnormal response processes are merged and stored according to the applicable scenarios of the rules. The trigger conditions and execution priority order between each rule are defined in the rule base. Rule conflicts are detected and resolved, and rule chain trigger clauses are added to obtain the initial dynamic collaborative rules.

[0107] Step S129: Retrieve historical dynamic collaborative rule execution effect data, calculate the completeness rate of perception data collection and the success rate of collaborative tasks under different device operating scenarios; filter out rule clauses that cause the completeness rate of collection or the success rate of collaboration to be lower than the preset threshold, correct the corresponding clause content in the initial dynamic collaborative rule, update the connection logic between rules, adjust the rule parameter settings, and obtain the corrected dynamic collaborative rule.

[0108] Retrieve historical data on the execution effectiveness of dynamic collaboration rules, and calculate the completeness rate of perception data collection and the success rate of collaboration tasks under different device operating scenarios. Identify rule clauses that cause the completeness rate or success rate of collaboration to fall below a preset threshold, correct the corresponding clauses in the initial dynamic collaboration rules, update the connection logic between rules, and adjust rule parameter settings to obtain the corrected dynamic collaboration rules.

[0109] Step S1210: According to the deployment area of ​​the edge node and the type of perception task it carries, classify and organize the revised dynamic collaboration rules, mark the rule content and execution requirements corresponding to each area and type of node, supplement the operation guide for rule execution, and generate the final dynamic collaboration rules.

[0110] Based on the deployment area and the type of sensing task carried by the edge nodes, the dynamic collaboration rules are categorized, organized, and revised. The rule content and execution requirements corresponding to each area and type of node are marked, and the operation guidelines for rule execution are supplemented to generate the final dynamic collaboration rules.

[0111] Step S130: The edge node optimizes the self-organizing process of the sensing task according to the dynamic collaboration rules, collects the feedback data of the loading equipment operation status and uploads it to the cloud. The feedback data of the loading equipment operation status records the execution effect of the sensing task and the changes in the equipment operation status.

[0112] Step S131: The edge node receives the dynamic collaboration rules issued by the cloud, extracts the boundary adjustment rules for perception tasks, data interaction method rules, and abnormal response process rules, and generates rule parsing results by combining its own perception module characteristics and the status of adjacent nodes.

[0113] Edge nodes receive dynamic collaboration rules from the cloud via wireless network, parse out the boundary adjustment rules for perception tasks, data interaction rules, and anomaly response process rules, and generate rule parsing results by combining the type and working mode of their own perception modules and the perception capability data of neighboring nodes.

[0114] Step S132: Adjust the boundary rules according to the sensing task, analyze the task allocation, execution sequence and collaboration method in the current sensing task self-organization process, adjust the sensing task content that exceeds the autonomous adjustment boundary, supplement the sensing tasks of the key operating parts of the equipment that are not covered, optimize the task allocation ratio, and obtain the task process optimization results.

[0115] Based on the boundary rules for sensing tasks, the task allocation, execution sequence, and collaboration methods in the current self-organizing process of sensing tasks are analyzed. Sensing task content that exceeds the autonomous adjustment boundary is adjusted, sensing tasks of key operating parts of the equipment that are not covered are added, the task allocation ratio is optimized, and the task process optimization results are obtained.

[0116] Step S133: Based on the data interaction rules, reconfigure the data transmission link with adjacent nodes, adjust the data transmission frequency and data packet format, enable the link redundancy mechanism to deal with possible transmission failures, set the link switching conditions and switching process, and obtain the data interaction configuration result.

[0117] Based on the data interaction rules, the data transmission links with adjacent nodes are reconfigured, the data transmission frequency and data packet format are adjusted, the link redundancy mechanism is enabled, the link switching conditions and switching process are set, and the data interaction configuration results are obtained.

[0118] Step S134: Integrate the task flow optimization results with the data interaction configuration results, update the perception task self-organizing process, mark the optimized task execution steps, collaboration nodes, data interaction nodes and timing requirements, and generate the optimized perception task self-organizing process.

[0119] Integrate the task flow optimization results with the data interaction configuration results, update the perceived task self-organizing process, mark the optimized task execution steps, collaboration nodes, data interaction nodes and timing requirements, and generate the optimized perceived task self-organizing process.

[0120] Step S135: Start the perception task execution according to the optimized perception task self-organization process, collect the operating status data of each operating part of the loading equipment through the perception module, and synchronously record the perception task execution progress, data collection status and the cooperation status of adjacent nodes to generate the original operating status data.

[0121] Step S1351: The edge node starts the sensing module preheating process according to the timing requirements in the optimized sensing task self-organizing process, checks the power supply status, working mode and data acquisition channel of the sensing module, switches to normal working mode, records the module startup process and status changes, and obtains the module startup result.

[0122] Edge nodes initiate the sensing module preheating process a specific time in advance, based on the sensing start time specified in the optimized sensing task self-organizing process. During this process, the first step is to check whether the sensing module's power supply voltage is within the normal operating voltage range and whether the power supply current is stable. Next, it confirms whether the sensing module's current operating mode matches the task requirements, such as whether the vibration sensor module has switched to continuous sampling mode and whether the temperature sensor module is set to the preset intermittent sampling interval. Simultaneously, it checks whether the data acquisition channels are unobstructed, including whether the connection lines between the sensor probes and the module host are conductive and whether the analog-to-digital conversion channel responds normally to standard signal input. After completing the checks, if any abnormalities are found, the module reset mechanism is activated; if everything is normal, it switches to normal operating mode and logs the entire process from preheating to entering normal operating mode, including changes in status parameters at key time points, such as power supply voltage fluctuations and mode switching time, ultimately forming the module startup result.

[0123] Step S1352: According to the coverage area assigned by the sensing task, adjust the acquisition angle, sensitivity and sampling frequency of the sensing module to ensure that the acquisition range accurately covers the corresponding operating part of the device. Adjust the acquisition angle to the optimal position, set the specific parameters of sensitivity and sampling frequency, and obtain the module parameter configuration result.

[0124] Based on the coverage area of ​​the equipment operating parts corresponding to each sensing module as specified in the task breakdown and allocation results, the edge nodes fine-tune the physical installation angle of the sensing modules. For example, the probe direction of the vibration sensor module is adjusted to be perpendicular to the center line of the motor shaft and directly opposite the bearing housing, ensuring that the collected vibration signal is the main vibration direction signal of the equipment operating part. For sensitivity parameters, the sensor sensitivity level is set according to the normal vibration amplitude range or temperature change range of the equipment operating part in the equipment operating characteristic data, ensuring that the signal-to-noise ratio of the sensor output signal is within the optimal range, avoiding distortion due to excessively weak signals or saturation due to excessively strong signals. The sampling frequency is set in conjunction with the equipment operating rhythm. For equipment operating parts with rapidly changing operating states (such as pressure changes in hydraulic valve groups), a higher sampling frequency is set; for parts with slower changes (such as ambient temperature), a lower sampling frequency is set. The adjusted acquisition angle, sensitivity level, and sampling frequency parameters are recorded in the module parameter configuration results.

[0125] Step S1353: Start the sensing data acquisition process, and collect the operating status data of the mechanical transmission part, the hydraulic control part, and the electrical drive part according to the optimized timing requirements and sampling frequency. Simultaneously record the time point of each acquisition and the working parameters of the sensing module to generate the original acquisition data sequence.

[0126] After configuring the module parameters, the edge nodes strictly follow the timing requirements of the optimized sensing task self-organizing process to initiate the data acquisition process. For mechanical transmission parts, such as the reducer of a belt conveyor, the installed temperature sensor module collects the shell temperature at a set sampling frequency, and the vibration sensor module collects the vibration acceleration signal. For hydraulic control parts, such as the hydraulic pump outlet, the pressure sensor module collects the real-time pressure value, and the flow sensor module collects the instantaneous flow rate. For electrical drive parts, such as the stator winding of a motor, the current sensor module collects the three-phase current value, and the voltage sensor module collects the line voltage value. During each data acquisition action, the precise timestamp of that acquisition (accurate to the millisecond level) and the current operating parameters of the sensing module, such as the actual sampling frequency, the current sensitivity level, and the power supply voltage, are recorded synchronously. The acquired equipment operating parameters and auxiliary recorded information are arranged in chronological order to form the original data acquisition sequence.

[0127] Step S1354: Define the result according to the collaboration method, send the acquisition start synchronization signal to the neighboring nodes participating in the collaborative sensing task, receive the acquisition start confirmation information from the neighboring nodes, adjust the acquisition start time of the self to be consistent with the neighboring nodes, and obtain the collaborative synchronization result.

[0128] As one of the participating nodes in the collaborative sensing task, an edge node, before initiating its own data acquisition process, sends a data acquisition start synchronization signal to all adjacent edge nodes participating in the same collaborative sensing task through the established data interaction link, according to the collaborative synchronization mechanism defined in the collaboration method definition. This signal includes the task ID for this collaborative acquisition, the precise timestamp of the planned acquisition start, and the identifier of the data interaction master node. After sending, the edge node enters a waiting state, receiving acquisition start confirmation information from each adjacent node. This confirmation information should include the adjacent nodes' confirmation status of receiving the synchronization time and their own expected acquisition start time deviation. The edge node summarizes the confirmation information from all adjacent nodes. If any node has a time deviation within the allowable range, it adjusts its own acquisition start time to match the start time of the majority of nodes. If any node has a deviation exceeding the allowable range, it resends the synchronization signal through the interaction link for calibration until the start time deviation of all participating nodes is controlled within a set threshold, ultimately forming a collaborative synchronization result.

[0129] Step S1355: During the data collection process, receive real-time data collection progress feedback information from adjacent nodes, synchronously share some of the collected running status data, supplement the data collection content that is not covered by itself, verify the consistency between the shared data and its own collected data, and obtain collaborative supplementary data.

[0130] During the collaborative sensing task execution, edge nodes receive real-time acquisition progress feedback information from neighboring nodes according to the collaboration frequency set in the data interaction rules. This progress feedback information includes the number of acquisitions completed by the neighboring node, the time range of the acquired data, and whether there are any missing or abnormal data. Simultaneously, edge nodes also send some of their acquired operational status data (such as vibration signal data from the first N sampling cycles) to neighboring nodes via a shared link to supplement any acquisition content not covered by neighboring nodes. For example, if a neighboring node experiences data loss for a certain period due to sensor failure, it can obtain shared data for that period from other nodes to supplement it. For the received shared data, edge nodes verify its consistency with their own acquired data (if there is overlap in acquisition range) by comparing data timestamps and data change trends. If inconsistencies are found, the data is marked as suspicious and the degree of difference is recorded. The supplemented missing data and the verified shared data are then integrated to form collaborative supplementary data.

[0131] Step S1356: Integrate the original collected data sequence with the collaborative supplementary data, classify and organize them according to the equipment operation section, remove duplicate data entries, retain complete operation status data records, and label the classification basis and data source to obtain the classified collection dataset.

[0132] Edge nodes integrate locally generated raw data sequences with supplementary data acquired from neighboring nodes. First, all data is categorized according to the operating components of the equipment (e.g., motors, reducers, hydraulic pumps, cylinders, etc.), grouping data of different types of parameters such as temperature, pressure, and vibration belonging to the same operating component together. Then, duplicate entries are checked after categorization. By comparing data timestamps and parameter values, duplicate data entries with the same parameters collected by different nodes at the same time are removed, retaining the earliest collected data or data with the best signal quality. For data sources, each data record is labeled whether the data was collected by the local sensing module or shared from a neighboring node. The categorization criteria are labeled with the name and number of the corresponding operating component, ultimately forming a clearly structured, non-duplicated categorized dataset.

[0133] Step S1357: Monitor sudden changes in the equipment's operating status during the data acquisition process. When an abnormal response condition is triggered, pause regular data acquisition according to the abnormal response procedure rules, start the special data acquisition mode, record the operating status data before and after the change, mark the start conditions and data range of the special data acquisition, and obtain the special data acquisition data.

[0134] During data acquisition, edge nodes monitor the collected operational status data in real time and compare it with the normal fluctuation range set in the equipment's operational characteristic data to determine whether a sudden change in the equipment's operational status has occurred. For example, when the motor current value exceeds a specific proportion of the normal range within a short period of time, or when the hydraulic system pressure experiences a sudden rise or fall, an abnormal response condition is triggered. At this time, the edge node immediately suspends the current conventional acquisition mode and starts a special acquisition mode according to the abnormal response process rules. In the special acquisition mode, the sampling frequency of relevant parameters of the equipment's operation is increased to a specific multiple of the conventional mode, expanding the data acquisition range. For example, in addition to collecting the currently mutated parameter, other parameters related to it are collected simultaneously (such as collecting motor temperature, vibration, and other parameters simultaneously when the motor current changes suddenly). Detailed operational status data for a specific time period before the mutation, the moment the mutation occurs, and a specific time period after the mutation are recorded. Simultaneously, the special acquisition data clearly indicates the start conditions for special acquisition (such as which specific parameter exceeds the threshold, the value exceeding it, etc.) as well as the range of data parameters and time range covered by the special acquisition.

[0135] Step S1358: Associate the special collection data with the regular classification collection dataset, mark the time point of the sudden change in the equipment operating status, the parameter changes before and after the change, and the special collection content, supplement the descriptive information of the change, and obtain the integrated collection dataset.

[0136] The specialized data collection data is aligned and correlated with the regular data collection data in the categorized data collection dataset along the time axis. This accurately marks the specific time points in the time series of the regular data collection data where abrupt changes in equipment operating status occur. For parameter changes before and after this time point, the magnitude of parameter change is quantified by calculating the difference between the average parameter value over a specific time period before the change and the average parameter value over a specific time period after the change. Information such as the magnitude of this change and the duration of the change is then added to the data records. Simultaneously, detailed data on the abrupt change process recorded in the specialized data collection data is attached as an appendix to the regular data record at the corresponding time point, along with a textual description of the change, such as "The motor A-phase current abruptly changes at time T; the average current before the change is X, and the average current after the change is Y, accompanied by an increase in the vibration amplitude of the motor casing." This ultimately forms an integrated data collection dataset containing both regular and specialized data.

[0137] Step S1359: Associate and store the perception module running status log, collaborative node interaction record, and acquisition environment parameters with the integrated acquisition dataset during the acquisition process, and record the acquisition timestamp and associated data entry identifier corresponding to each piece of information to obtain the complete acquisition dataset.

[0138] During data acquisition, edge nodes synchronously generate sensing module operation status logs, including records of power supply voltage fluctuations, operating temperature, and fault codes. Collaborative node interaction records include synchronization signal transmission and reception records, data sharing transmission records, and link status switching records with adjacent nodes. Environmental parameters include data on external factors that may affect equipment operation and data acquisition quality, such as ambient temperature, humidity, and dust concentration at the acquisition site. This information is then linked to the integrated acquisition dataset. By adding a corresponding acquisition timestamp to each piece of information and establishing an identifier association with data entries in the integrated acquisition dataset (e.g., through data record IDs), when querying equipment operation status data for a specific time period, the operating status of sensing modules, collaborative interactions, and environmental parameters for that period can be simultaneously accessed, forming a complete acquisition dataset containing multi-dimensional information.

[0139] Step S13510: Standardize the complete collected dataset according to the format requirements defined in the data interaction method rules, unify the data representation method and storage format, mark the specific basis for standardization, and generate the final operating status data of each operating part of the loading equipment.

[0140] Following the data format standards specified in the data interaction rules, the edge nodes standardize the complete collected dataset. First, the data representation is standardized, for example, by standardizing units of physical quantities: pressure is standardized to Pascals, temperature to degrees Celsius, and vibration acceleration to meters per second squared. Data precision is also standardized, with the number of decimal places retained according to the rules. Second, the storage format is standardized, organizing the data into a prescribed structured text format. Each data record includes fields such as equipment operating part ID, parameter type, acquisition timestamp, parameter value, data source identifier, and data quality identifier. After standardization, the header of the data file is marked with the version number and specific clause number of the data interaction rules upon which this standardization process was based, ensuring that the data recipient clearly understands the basis for the data format and ultimately generates compliant operating status data for each operating part of the loading equipment.

[0141] Step S136: Associate the exception response process rules, monitor and detect exceptions during the execution of the perception task, initiate the autonomous processing process according to the rule requirements, record the exception occurrence time, processing process and processing result, and generate exception processing records.

[0142] Associated exception response process rules, monitor and perceive exceptions during task execution, initiate autonomous processing according to rule requirements, record the exception occurrence time, processing process and results, and generate exception handling records.

[0143] Step S137: Associate the original operating status data with the anomaly handling records, remove invalid data entries based on data validity verification rules, add fields describing the changes in equipment operating parameters before and after the anomaly handling event to the data records, and store them according to the equipment operating parts to obtain standardized operating status data.

[0144] The original operating status data is associated with the anomaly handling records. Invalid data entries are removed based on data validity verification rules. Fields describing changes in equipment operating parameters before and after the anomaly handling event are added. The data is then stored according to the equipment operating parts to obtain standardized operating status data.

[0145] Step S138: According to the data interaction method rules, upload the standardized operation status data to the cloud in batches, and record the data upload progress, transmission status and cloud reception confirmation information simultaneously to generate a data upload log.

[0146] According to the data interaction rules, standardized operational status data is uploaded to the cloud in batches, and the data upload progress, transmission status and cloud reception confirmation information are recorded synchronously to generate a data upload log.

[0147] Step S139: Package the configuration information of the optimized sensing task self-organizing process, standardized operation status data and data upload logs to form the initial loading equipment operation status feedback data; add an index to the feedback data to record the correspondence between the sensing task version and the equipment operation status data, and add metadata to trace the data source to obtain the initial loading equipment operation status feedback data.

[0148] The configuration information of the optimized sensing task self-organizing process, standardized operating status data and data upload logs are packaged together, and an index is added to record the correspondence between the sensing task version and the equipment operating status data, as well as metadata to trace the data source, to obtain the initial equipment operating status feedback data.

[0149] Step S1310: Supplement the edge node identifier, dynamic collaboration rule version information, and time sequence mark to the initial loading equipment operation status feedback data, unify the data format and expression logic, generate the final loading equipment operation status feedback data, and complete the upload.

[0150] Supplement edge node identifiers, dynamic collaboration rule version information, and timing tags to the initial loading equipment operation status feedback data, unify data format and expression logic, generate the final loading equipment operation status feedback data, and upload it to the cloud.

[0151] Step S140: The cloud integrates the loading equipment operation status feedback data of all edge nodes, updates the dynamic collaborative rule generation logic, and outputs the rule update command.

[0152] Step S141: The cloud receives the loading equipment operation status feedback data uploaded by each edge node, stores it according to edge node identifier and data upload time, records data integrity and transmission quality information, and generates a feedback data reception ledger.

[0153] The cloud receives feedback data on the operating status of loading equipment uploaded by each edge node, stores it according to edge node identification and data upload time, records data integrity and transmission quality information, and generates a feedback data reception ledger.

[0154] Step S142: Analyze the feedback data on the operating status of loading equipment in the feedback data receiving ledger, extract the perception task optimization effect, standardized operating status data, anomaly handling records and data upload logs of each edge node, associate the commonalities and differences of feedback data from different nodes, extract core correlation features, and obtain the feedback data parsing results.

[0155] The system analyzes the feedback data on the operating status of loading equipment in the data receiving ledger, extracts the optimization effect of perception tasks, standardized operating status data, anomaly handling records and data upload logs of each edge node, correlates the commonalities and differences of feedback data from different nodes, extracts core correlation features, and obtains the feedback data analysis results.

[0156] Step S143: Integrate the feedback data parsing results of all edge nodes, analyze the execution effect of dynamic collaboration rules in different device operating scenarios and different node deployment areas, identify clauses in the rules that lead to insufficient perception efficiency, poor collaboration, or untimely handling of anomalies, mark the specific problem manifestations and impact scope, and obtain the rule execution effect analysis results.

[0157] By integrating the feedback data analysis results from all edge nodes, the execution effect of dynamic collaboration rules in different device operating scenarios and different node deployment areas is analyzed. Clauses in the rules that lead to insufficient perception efficiency, poor collaboration, or untimely handling of anomalies are identified, and specific problem manifestations and impact ranges are marked to obtain the analysis results of rule execution effect.

[0158] Step S144: Based on the analysis results of rule execution effect, locate the defects in the dynamic collaborative rule generation logic, including incomplete coverage of rule adaptation scenarios, unreasonable parameter settings, and loose connection logic between rules. Mark the specific location and degree of impact of each defect to obtain the defect location results of the generation logic.

[0159] Based on the analysis results of rule execution effect, defects in the dynamic collaborative rule generation logic are located, and the specific location and degree of impact of each defect are marked to obtain the defect location results of the generation logic.

[0160] Step S145: Based on the result of generating logic defect localization, adjust the generation parameters of dynamic collaborative rules, optimize the classification criteria of rule adaptation scenarios, improve the connection mechanism between rules, supplement the rule generation logic of uncovered device operation scenarios, set the parameter adjustment range and scenario classification refinement criteria, and obtain the generation logic adjustment result.

[0161] Step S1451: Analyze the result of generating logic defect location, extract the defect types existing in the dynamic collaborative rule generation logic, including unreasonable parameter settings, overly coarse scene division, and loose rule connection, determine the scope of impact and specific manifestations of each defect, and obtain a detailed list of defect types.

[0162] The cloud system performs in-depth analysis of the generated logic defect location results. By retrieving defect description information, it identifies different defect types in the dynamic collaboration rule generation logic. For defects related to unreasonable parameter settings, the specific manifestations include: the threshold parameters of the perception task adjustment boundary not matching the actual edge node carrying capacity, leading to frequent task adjustment requests from some nodes; overly coarse scenario segmentation, where device operating states under different load levels are grouped into the same adaptation scenario, resulting in poor applicability of some clauses in the generated collaboration rules under both light and heavy load scenarios; and loose rule connections, where the execution order of perception task adjustment rules and data interaction method rules conflicts under certain conditions, such as when a node simultaneously triggers task adjustment and link switching, the rules do not clearly specify which operation to prioritize. For each defect type, its impact scope is determined. For example, unreasonable parameter settings mainly affect the task execution stability of edge nodes, overly coarse scenario segmentation affects the scenario adaptation accuracy of rules, and loose rule connections affect the timeliness of anomaly handling. This information is then compiled into a detailed defect type list.

[0163] Step S1452: To address the defect of mismatched parameter settings, retrieve historical dynamic collaborative rule generation parameters and corresponding execution effect data, establish a mapping relationship between parameters and execution effect indicators, determine the direction and numerical range of parameter adjustment based on the mapping relationship, and obtain a parameter adjustment plan.

[0164] The cloud system retrieves dynamic collaborative rule generation parameters from the database across multiple rule iteration cycles, such as the CPU load threshold for autonomous boundary adjustment of perception tasks, the signal strength threshold for data interaction link switching, and the trigger latency of the anomaly response process. Simultaneously, it extracts the corresponding execution performance metrics for these parameters, including edge node task completion rate, data transmission success rate, and anomaly handling timeliness rate. Through data analysis methods, a mapping model is established between each generation parameter and various execution performance metrics to analyze the changing trends of these metrics when a parameter takes different values. For example, if the CPU load threshold is set too high, nodes may experience increased data processing latency and a decreased task completion rate due to excessive load; if it is set too low, it will lead to overly frequent task adjustments and increased network interaction overhead. Based on the above mapping relationship, the optimal adjustment direction (e.g., increasing or decreasing) and a reasonable numerical adjustment range for each parameter are determined to ensure that the adjusted parameters achieve optimal levels for various execution performance metrics, thus forming a parameter adjustment scheme.

[0165] Step S1453: Based on the parameter adjustment scheme, adjust the definition parameters of the sensing task adjustment boundary, the configuration parameters of the data interaction method, and the trigger parameters of the abnormal response process to make the parameter settings fit the device operating characteristics and edge node capabilities, record the changes before and after parameter adjustment, and obtain the generated parameter update results.

[0166] Based on the parameter adjustment scheme, the cloud system modifies specific parameters in the dynamic collaborative rule generation logic. For parameters defining the boundary of the perception task adjustment, such as expanding or narrowing the sampling frequency range that edge nodes can autonomously adjust from a specific interval to a new interval, making it more suitable for the current hardware processing capabilities of the edge nodes; for configuration parameters of data interaction methods, such as adjusting the signal strength threshold for switching transmission links for high-priority data from a specific value to a new value to adapt to the attenuation characteristics of wireless signals in different environments; for triggering parameters of the abnormal response process, such as changing the threshold for judging sudden changes in device operating status from a fixed value to a dynamic value related to the current operating load of the device—the higher the load, the more lenient the threshold setting—to avoid false triggers. During the adjustment process, the original value before adjustment, the target value after adjustment, and the specific time and basis for adjustment of each parameter are recorded in detail, forming the parameter update results.

[0167] Step S1454: To address the shortcomings of overly coarse scenario classification, analyze the operating scenario types of loading equipment, and refine the scenario classification criteria by combining factors such as equipment load changes, differences in working environment, and switching of operating modes. Add subdivided scenario types, label the classification basis and characteristics of subdivided scenarios, and obtain the scenario classification optimization results.

[0168] The cloud-based system comprehensively analyzes historical operating data of the loading equipment to identify key factors affecting its operating characteristics, including changes in equipment load (e.g., no load, half load, full load), differences in working environment (e.g., high temperature, high humidity, dust concentration), and switching of operating modes (e.g., normal loading, rapid unloading, maintenance mode). Based on these key factors, the original rule-based adaptation scenarios are further refined. For example, the original "normal operation scenario" is subdivided into multiple sub-scenarios such as "full load normal temperature normal operation scenario" and "half load high temperature normal operation scenario." Clear criteria are established for each sub-scenarios. For instance, when the actual load rate of the equipment is within a specific range and the ambient temperature is within a specific range, it is determined to be a "half load high temperature normal operation scenario." The characteristic parameter ranges for each sub-scenarios are also labeled, such as the load rate range, environmental parameter range, and typical operating mode, resulting in optimized scenario segmentation.

[0169] Step S1455: Define exclusive collaborative rule generation logic for each sub-scenarios, determine the differentiated requirements for perception task adjustment boundaries, data interaction methods and abnormal response processes under different sub-scenarios, mark the basis for differentiated settings, and obtain scenario-specific generation logic.

[0170] For each specific scenario type, the cloud system formulates exclusive collaborative rule generation logic based on the device operating characteristics and monitoring requirements of that scenario. Regarding the boundary adjustment for perception tasks, to address the high node load in full-load scenarios, the range of autonomous adjustment boundaries is appropriately narrowed, reducing the frequency of node autonomous adjustments. Regarding data interaction methods, to address the poor wireless link stability in high-temperature and high-humidity scenarios, link redundancy and data retransmission frequency are increased. Regarding anomaly response procedures, for maintenance mode scenarios, the response process for some common anomalies is simplified, prioritizing the transmission of maintenance data. Simultaneously, the basis for differentiated settings is clearly marked in the scenario-specific generation logic, such as "the basis for narrowing the adjustment boundary in full-load scenarios is that historical data statistics show that the average CPU load rate of nodes under full load is a specific percentage higher than that under idle load," ensuring the interpretability of the rules.

[0171] Step S1456: To address the shortcomings of loose rule connections, the logical relationships between the clauses of the dynamic collaborative rules are correlated, missing connecting clauses are supplemented, the execution order and triggering conditions between clauses are determined, rule conflicts or execution gaps are eliminated, and the rule connection optimization results are obtained.

[0172] The cloud system analyzes all clauses of the dynamic collaboration rules, draws a logical relationship diagram between the clauses, and identifies clause combinations with missing connections or conflicts. For example, when the "task adjustment request" clause and the "link failure switchover" clause might trigger at the same time, the original rules did not specify the execution order, leading to confusion in node processing. To address this, supplementary clauses were added, clearly stipulating that when both situations occur simultaneously, the "link failure switchover" clause should be executed first, and the "task adjustment request" should be processed only after the link has stabilized. For clauses with execution gaps, such as the "abnormal data discard" clause which lacked a subsequent data re-collection mechanism, supplementary provisions were added stipulating that after discarding abnormal data, the node should immediately send a data re-collection request to the collaborating node. Through these measures, the execution order and triggering conditions of each clause are clarified, rule conflicts and execution gaps are eliminated, resulting in optimized rule connections.

[0173] Step S1457: Integrate the generation parameter update results, scene division optimization results, scene-specific generation logic and rule connection optimization results to form a dynamic collaborative rule generation logic adjustment plan, determine the core content and basis of the adjustment, mark the execution steps of the plan, and obtain the generation logic adjustment draft.

[0174] The cloud system integrates the above optimization results, first clarifying the core content of the dynamic collaborative rule generation logic adjustment scheme, including the specific update values ​​of generation parameters, the criteria and number of subdivided scenarios, the key clauses of the scenario-specific generation logic, and supplementary rule connection clauses. Detailed basis is provided for each adjustment, such as the parameter update basis being the mapping relationship analysis results in the parameter adjustment scheme, and the scenario division optimization basis being the analysis results of key influencing factors of device operation scenarios. Then, the execution steps of the scheme are formulated, such as first, updating the generation parameter configuration table; second, reconstructing the scenario division algorithm module; third, embedding scenario-specific generation logic code; and fourth, adding rule connection clause judgment logic. The draft of the generation logic adjustment is formed in this step-by-step sequence to ensure the orderly progress of the adjustment process.

[0175] Step S1458: Apply the generated logic adjustment draft to the historical feedback data, run the rule generation logic to obtain simulated dynamic collaborative rules, calculate the performance indicators of the simulated rules under different historical scenarios, record the scenarios and rule clauses where the indicators did not meet the standards, and obtain the simulation verification results.

[0176] The cloud system selects historical equipment operation status feedback data from multiple typical equipment operation scenarios, loads the generated logic adjustment draft into the rule generation engine, runs the rule generation process, and generates simulated dynamic collaborative rules. Then, these simulated rules are applied to the corresponding historical scenario data to simulate the task execution process of edge nodes, calculating various performance indicators, such as the balance of perceived task allocation, data interaction latency, and anomaly response accuracy. The calculated performance indicators are compared with preset threshold values, and historical scenario information and corresponding simulated rule clauses are recorded for performance indicators that fail to meet the standards. For example, in a "heavy load high temperature scenario," the data interaction latency exceeds the threshold, and the corresponding rule clause is "data transmission frequency setting in high temperature scenarios." This information is then compiled into simulation verification results.

[0177] Step S1459: Based on the simulation verification results, fine-tune the parameter settings and scenario division standards in the draft of the generation logic adjustment, correct the unreasonable content found in the simulation verification, optimize the rule connection clauses, and obtain the final draft of the generation logic adjustment.

[0178] Based on the simulation results, the cloud system made targeted fine-tuning to the draft of the generation logic. Regarding parameter settings, if a task completion rate failed to meet the target in a historical scenario due to an excessively high CPU load threshold, the CPU load threshold parameter for that scenario was further reduced. Regarding scenario division criteria, if the performance metrics of two sub-scenarios were found to be insignificantly different, they were merged into one scenario to reduce rule complexity. Regarding rule connection clauses, if new execution conflicts occurred during simulation, more detailed trigger condition judgment logic was added. After multiple rounds of fine-tuning and simulation verification, until all performance metrics met the targets in all historical scenarios, the final draft of the generation logic adjustment was completed.

[0179] Step S14510: Integrate the final draft of the generation logic adjustment into the dynamic collaborative rule generation logic, update the algorithm flow, parameter configuration and scenario adaptation mechanism of rule generation, record the complete content and basis of the logic update, and complete the optimization of the generation logic.

[0180] The cloud system adjusts the final draft according to the generation logic, refactoring the underlying algorithm flow of the dynamic collaborative rule generation logic, modifying corresponding code modules, such as updating parameter values ​​in the parameter configuration module, replacing the partitioning algorithm in the scene partitioning module, embedding conditional statements for scene-specific generation logic, and adding execution order control code for rule connection clauses. Simultaneously, the documentation for the rule generation logic is updated, detailing each part of the logic update, including comparisons before and after modification, the basis for modification (such as specific scenarios and metrics in simulation verification results), and the expected impact on the overall rule performance. After completing all code modifications and documentation updates, unit tests and integration tests are used to verify the correctness of the generation logic, ensuring that the optimized generation logic can run stably and generate dynamic collaborative rules that meet the requirements.

[0181] Step S146: Integrate the results of the generation logic adjustment, update the dynamic collaborative rule generation logic, mark the new rule generation process, parameter setting requirements and scenario adaptation standards, supplement the logic adjustment basis and execution details, and obtain the updated dynamic collaborative rule generation logic.

[0182] The results of the integrated generation logic adjustment are used to update the dynamic collaborative rule generation logic. The new rule generation process, parameter setting requirements and scenario adaptation standards are marked, and the basis for logic adjustment and execution details are supplemented to obtain the updated dynamic collaborative rule generation logic.

[0183] Step S147: Based on the updated dynamic collaborative rule generation logic, generate a rule update instruction framework, mark the core content of the rule update, execution steps and time limit requirements, determine the relationship between the updates of each part, and obtain the initial rule update instruction.

[0184] Based on the updated dynamic collaborative rule generation logic, a rule update instruction framework is generated, marking the core content of the rule update, execution steps and time limit requirements, determining the relationship between the updates of each part, and obtaining the initial rule update instruction.

[0185] Step S148: Based on the edge node identifier and the perception task type it carries, split the initial rule update instructions after correction, supplement the rule adjustment details and operation instructions corresponding to each node, mark the differences between the node-specific adjustment content and the general adjustment content, and obtain the node-specific rule update instructions.

[0186] Based on the edge node identifier and the perception task type it carries, the initial rule update instructions after correction are split, the rule adjustment details and operation instructions corresponding to each node are supplemented, the differences between the node-specific adjustment content and the general adjustment content are marked, and the node-specific rule update instructions are obtained.

[0187] Step S149: Verify the feasibility of the node-specific rule update command, compare it with the hardware capacity of the edge node and the requirements for the execution of the perception task, revise the command clauses with execution obstacles, adjust the operation steps and timing requirements, and obtain the final rule update command.

[0188] The feasibility of the node-specific rule update command was verified. By comparing the hardware capacity of the edge node with the requirements of the perception task execution, the command clauses with execution obstacles were corrected, the operation steps and timing requirements were adjusted, and the final rule update command was obtained.

[0189] Step S1410: Send the final rule update command to the corresponding edge node, synchronously record the command sending time and content, update the iteration log of the cloud dynamic collaborative rule generation logic, mark the rule version and adjustment content corresponding to the iteration log, and complete the rule update process.

[0190] The final rule update command is sent to the corresponding edge node, the command sending time and content are recorded synchronously, the iteration log of the cloud dynamic collaborative rule generation logic is updated, the rule version and adjustment content corresponding to the iteration log are marked, and the rule update process is completed.

[0191] Step S150: The edge node receives the rule update instruction, iteratively perceives the task self-organization process, generates the self-organization process iteration results and uploads them to the cloud, forming a two-way driving loop of edge self-organization and cloud rule generation.

[0192] Edge nodes receive rule update instructions from the cloud, iterate the self-organizing process of perception tasks according to the instructions, generate the self-organizing process iteration results and upload them to the cloud, forming a two-way driving loop of edge self-organization and cloud rule generation, continuously optimizing the monitoring effect of loading equipment operation status.

[0193] During the data collection process, non-privacy-sensitive data such as equipment operating status is involved, and no special privacy protection measures are required. If the collection of data related to operators is expanded in the future, data anonymization, access control, and encrypted transmission will be used to protect privacy and ensure data security.

[0194] Figure 2 This embodiment illustrates a cloud-edge-device collaborative loading equipment operation status monitoring system 100, including a processor 1001, a memory 1003, and program code stored in the memory 1003. The processor 1001 executes the program code to implement the steps of the cloud-edge-device collaborative loading equipment operation status monitoring method. The processor 1001 and the memory 1003 are connected, for example, via a bus 1002. Optionally, the cloud-edge-device collaborative loading equipment operation status monitoring system 100 may further include a transceiver 1004, which can be used for data interaction between this cloud-edge-device collaborative loading equipment operation status monitoring system and other cloud-edge-device collaborative loading equipment operation status monitoring systems, such as sending and / or receiving data. It should be noted that in actual scheduling, the transceiver 1004 is not limited to one, and the structure of this cloud-edge-device collaborative loading equipment operation status monitoring system 100 does not constitute a limitation on the embodiments of this application.

[0195] The memory 1003 is used to store program code for executing the embodiments of this application, and its execution is controlled by the processor 1001. The processor 1001 is used to execute the program code stored in the memory 1003 to implement the steps shown in the foregoing method embodiments.

[0196] The above description is only an optional implementation method for some implementation scenarios of this application. It should be noted that for those skilled in the art, other similar implementation methods based on the technical concept of this application, without departing from the technical concept of this application, also fall within the protection scope of the embodiments of this application.

Claims

1. A method for monitoring the operational status of loading equipment based on cloud-edge-device collaboration, characterized in that, The method includes: The edge nodes, based on the perception status of neighboring nodes and the information related to the operation of loading equipment, self-organize perception tasks and generate self-organized perception task results. The self-organized perception task results include the perception task allocation of each edge node, the cooperation mode of neighboring nodes, and the task execution sequence. The cloud receives the self-organization results of the sensing tasks uploaded by each edge node, and generates dynamic coordination rules by combining them with the operating characteristics of the loading equipment. The dynamic coordination rules define the adjustment boundaries of the edge node's sensing tasks, the data interaction methods, and the abnormal response process. The edge nodes optimize the self-organizing process of the sensing task according to the dynamic collaboration rules, collect the feedback data of the loading equipment's operating status and upload it to the cloud. The feedback data of the loading equipment's operating status records the execution effect of the sensing task and the changes in the equipment's operating status. The cloud integrates the operational status feedback data of loading equipment from all edge nodes, updates the dynamic collaborative rule generation logic, and outputs rule update instructions; Edge nodes receive rule update instructions, iteratively perceive the task self-organization process, generate self-organization process iteration results and upload them to the cloud, forming a two-way driving loop of edge self-organization and cloud rule generation.

2. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 1, characterized in that, The edge node, based on the perception status of neighboring nodes and the information related to the operation of the loading equipment, self-organizes perception tasks and generates self-organized perception task results, including: Each edge node collects information on the operating status of its own sensing module and the current sensing tasks it is carrying, including the working mode of the sensing module, the progress of task execution, and the remaining carrying capacity, and generates its own sensing capability data. Edge nodes establish a perception state interaction link with adjacent edge nodes, configure the link transmission protocol and data interaction format, transmit their own perception capability data and receive the perception capability data of adjacent edge nodes, forming a perception capability interaction set between nodes. Retrieve the operation-related information of the loading equipment, covering the operation dependencies and mutual influence paths of the mechanical transmission part, hydraulic control part, and electrical drive part, and generate an equipment operation-related dataset; The interaction set of sensing capabilities between associated nodes and the associated dataset of device operation are analyzed to determine the matching relationship between the sensing requirements of each part of the device operation and the sensing capabilities of the nodes, identify the parts of the device operation that require multi-node collaborative sensing, and obtain a list of collaborative sensing requirements. Based on the collaborative sensing requirement list, each edge node negotiates the sensing task splitting method through the sensing status interaction link, and allocates the collaborative sensing tasks to the corresponding edge nodes according to the node's sensing capabilities and deployment location. The task responsibilities and coverage of each node are defined, and the correspondence between the specific sensing content and the device operation part is marked to obtain the task splitting and allocation results. By combining the physical deployment locations of nodes with the operating rhythm of equipment, the execution sequence of sensing tasks is optimized, and the node sensing start-up time difference is adjusted so that adjacent nodes participating in the same collaborative sensing task can start sensing operations synchronously, thereby reducing data interaction delays and task connection gaps and obtaining timing optimization results. Define the collaboration mode between nodes, set the main data interaction node, transmission path selection criteria and information sharing scope in the collaborative perception task, set the trigger conditions and feedback mechanism for data interaction, and mark the specific content and format requirements of data interaction to obtain the collaboration mode definition result; The task splitting and allocation results, timing optimization results, and collaboration method definition results are combined and stored according to the edge node identifier. The perception task content, execution time, collaboration object, and data interaction requirements of each edge node are recorded in the combined data to obtain the initial perception task self-organization result. Retrieve historical data on the self-organization effect of perception tasks, identify historical cases where node load exceeds threshold due to task allocation, or where data collection time windows overlap due to timing issues; based on the identified historical cases, adjust the task allocation ratio and timing arrangement in the initial self-organization result of perception tasks, update the details of collaboration methods, and obtain the corrected self-organization result of perception tasks. The terminology and data format of the corrected perception task self-organization results are standardized. Metadata fields are added to the data structure to describe the input sources and output targets of each part, and identifiers are added to indicate the interface with the cloud dynamic collaboration rules, thus generating the final perception task self-organization results.

3. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 1, characterized in that, The cloud receives the self-organization results of the sensing tasks uploaded by each edge node, and generates dynamic collaborative rules based on the operating characteristics of the loading equipment, including: A cloud-based platform for receiving the self-organized results of perception tasks is built, configured with multi-channel data receiving interfaces, and the self-organized results of perception tasks uploaded by each edge node are received in a classified manner according to the edge node identifier and time sequence. The data receiving time, transmission status and integrity information are recorded, and a receiving ledger is generated. The self-organization results of sensing tasks in the receiving ledger are analyzed to extract the sensing task allocation of each edge node, the cooperation mode of adjacent nodes, the task execution sequence and the operating part of the covered equipment. The cross content of the self-organization results of sensing tasks of different edge nodes is associated to extract common correlations and differential features to obtain the self-organization result analysis data. Retrieve the operating characteristic data of the loading equipment, including the transmission ratio range of the mechanical transmission part, the pressure regulation characteristics of the hydraulic control part, the power output law of the electrical drive part, and the fault propagation threshold of each part, and generate a dataset of equipment operating characteristics. By correlating the self-organization results parsing data with the equipment operation characteristic dataset, comparing the matching relationship between the coverage of the sensing tasks and the key parts of equipment operation, analyzing the fit between the execution sequence of the sensing tasks and the rhythm of equipment operation, identifying the parts in the self-organization results of the sensing tasks that do not match the equipment operation characteristics, and obtaining the adaptation deviation analysis results. Based on the adaptation deviation analysis results, the edge node perception task adjustment boundary is defined, the perception task types, ranges and adjustment ranges that each edge node can adjust independently are divided, the perception task categories that need to be applied to the cloud before adjustment are divided, the specific division criteria and adjustment process of the two types of tasks are marked, and the perception task adjustment boundary rules are obtained. Design rules for data interaction between edge nodes. Based on the priority of collaborative sensing tasks and data transmission requirements, set the frequency of data transmission between nodes, data packet format, transmission link selection criteria, and data loss response measures. Mark the differences in transmission parameters corresponding to tasks with different priorities to obtain the rules for data interaction. Anomaly response process rules are constructed. For situations such as interruption of sensing task execution, failure of data interaction, and sudden changes in device operating status, the autonomous processing process of edge nodes, the content and time limit requirements for anomaly information reporting are set, the remote collaborative handling method in the cloud is defined, and the differences in the processing process corresponding to different anomaly types are marked to obtain anomaly response process rules. The rules for adjusting the boundary of perception tasks, the rules for data interaction methods, and the rules for abnormal response processes are merged, stored according to the applicable scenarios of the rules, and the triggering conditions and execution priority order between each rule are defined in the rule base. Rule conflicts are detected and resolved, and rule chain trigger clauses are added to obtain the initial dynamic collaborative rules. Retrieve historical dynamic collaboration rule execution effect data, calculate the perception data collection completeness rate and collaboration task success rate of historical rules under different device operating scenarios; filter out rule clauses that cause the collection completeness rate or collaboration success rate to be lower than the preset threshold, correct the corresponding clause content in the initial dynamic collaboration rule, update the connection logic between rules, adjust the rule parameter settings, and obtain the corrected dynamic collaboration rule. Based on the deployment area and the type of sensing task carried by the edge nodes, the dynamic collaboration rules are categorized, organized, and revised. The rule content and execution requirements corresponding to each area and type of node are marked, and the operation guidelines for rule execution are supplemented to generate the final dynamic collaboration rules.

4. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 1, characterized in that, The edge nodes optimize the self-organizing process of sensing tasks according to dynamic coordination rules, collect feedback data on the operating status of loading equipment, and upload it to the cloud, including: Edge nodes receive dynamic collaboration rules from the cloud, extract boundary adjustment rules for perception tasks, data interaction rules, and anomaly response process rules, and generate rule parsing results by combining their own perception module characteristics and the status of adjacent nodes. Based on the boundary rules of the sensing task adjustment, analyze the task allocation, execution sequence and collaboration method in the current self-organizing process of the sensing task, adjust the sensing task content that exceeds the autonomous adjustment boundary, supplement the sensing tasks of the key operating parts of the equipment that are not covered, optimize the task allocation ratio, and obtain the task process optimization results. Based on the data interaction rules, the data transmission links with adjacent nodes are reconfigured, the data transmission frequency and data packet format are adjusted, the link redundancy mechanism is enabled to deal with possible transmission failures, the link switching conditions and switching process are set, and the data interaction configuration results are obtained. Integrate the task process optimization results with the data interaction configuration results, update the self-organizing process of the perceived task, mark the optimized task execution steps, collaboration nodes, data interaction nodes and timing requirements, and generate the optimized self-organizing process of the perceived task. The sensing task is initiated according to the optimized sensing task self-organizing process. The sensing module collects the operating status data of each operating part of the loading equipment, and synchronously records the sensing task execution progress, data collection status and the cooperation status of adjacent nodes to generate the original operating status data. Associate the abnormal response process rules, monitor abnormal situations during the execution of the sensing task, initiate the autonomous processing process according to the rule requirements, record the time of abnormal occurrence, processing process and processing result, and generate abnormal processing records; The original operating status data is associated with the anomaly handling records. Invalid data entries are removed based on data validity verification rules. Fields describing changes in equipment operating parameters before and after an anomaly handling event are added to the data records. The data is then stored according to the equipment operating parts to obtain standardized operating status data. According to the data interaction rules, standardized operational status data is uploaded to the cloud in batches, and the data upload progress, transmission status and cloud reception confirmation information are recorded synchronously to generate data upload logs; The configuration information of the optimized sensing task self-organizing process, standardized operation status data and data upload logs are packaged to form the initial loading equipment operation status feedback data; an index is added to the feedback data to record the correspondence between the sensing task version and the equipment operation status data, and metadata is added to trace the data source to obtain the initial loading equipment operation status feedback data. Supplement edge node identifiers, dynamic collaboration rule version information, and timing tags to the initial loading equipment operation status feedback data, unify data format and expression logic, generate the final loading equipment operation status feedback data, and complete the upload.

5. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 1, characterized in that, The cloud platform integrates the operational status feedback data of loading equipment from all edge nodes, updates the dynamic collaborative rule generation logic, and outputs rule update instructions, including: The cloud receives the loading equipment operation status feedback data uploaded by each edge node, stores it according to edge node identification and data upload time, records data integrity and transmission quality information, and generates a feedback data reception ledger. The system analyzes the feedback data on the operating status of loading equipment in the data receiving ledger, extracts the optimization effect of perception tasks, standardized operating status data, anomaly handling records and data upload logs of each edge node, associates the commonalities and differences of feedback data from different nodes, extracts core correlation features, and obtains the feedback data analysis results. By integrating the feedback data analysis results from all edge nodes, the execution effect of dynamic collaboration rules in different device operating scenarios and different node deployment areas is analyzed. Clauses in the rules that lead to insufficient perception efficiency, poor collaboration, or untimely handling of anomalies are identified, and specific problem manifestations and impact ranges are marked to obtain the analysis results of rule execution effect. Based on the analysis results of rule execution effect, the defects in the dynamic collaborative rule generation logic are located, including incomplete coverage of rule adaptation scenarios, unreasonable parameter settings, and loose connection logic between rules. The specific location and impact of each defect are marked to obtain the defect location results of the generation logic. Based on the results of generating logic defect localization, the generation parameters of dynamic collaborative rules are adjusted, the classification criteria for rule adaptation scenarios are optimized, the connection mechanism between rules is improved, the rule generation logic for uncovered device operation scenarios is supplemented, the parameter adjustment range and scenario classification refinement criteria are set, and the generation logic adjustment results are obtained. Integrate the results of the generation logic adjustment, update the dynamic collaborative rule generation logic, mark the new rule generation process, parameter setting requirements and scenario adaptation standards, supplement the logic adjustment basis and execution details, and obtain the updated dynamic collaborative rule generation logic. Based on the updated dynamic collaborative rule generation logic, a rule update instruction framework is generated, the core content of the rule update, the execution steps and time limit requirements are marked, the relationship between the updates of each part is determined, and the initial rule update instruction is obtained. Based on the edge node identifier and the type of perception task it carries, the initial rule update instructions after correction are split, the rule adjustment details and operation instructions corresponding to each node are supplemented, the differences between the node-specific adjustment content and the general adjustment content are marked, and the node-specific rule update instructions are obtained. The feasibility of the node-specific rule update command was verified. The command terms with execution obstacles were corrected according to the hardware capacity of the edge node and the execution requirements of the perception task. The operation steps and timing requirements were adjusted to obtain the final rule update command. The final rule update command is sent to the corresponding edge node, and the command sending time and content are recorded synchronously. The iteration log of the cloud dynamic collaborative rule generation logic is updated, and the rule version and adjustment content corresponding to the iteration log are marked to complete the rule update process.

6. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 2, characterized in that, The set of sensing capability interactions between associated nodes and the data set associated with device operation are used to identify the parts of device operation that require multi-node collaborative sensing, resulting in a list of collaborative sensing requirements, including: The interaction set of perception capabilities between nodes is analyzed, and the perception module type, perception range, data processing capability and remaining carrying capacity of each edge node are extracted. The nodes are classified and organized according to node identifiers, and the specific performance of each capability indicator is marked to obtain the node perception capability details. The equipment operation association dataset is analyzed to extract the operation dependencies of the mechanical transmission part, hydraulic control part, and electrical drive part. The impact path and correlation strength of any part's operation status change on other parts are determined, the specific links of the impact path and the correlation performance are marked, and the equipment part association details are obtained. The node perception capability details and the device part association details are associated according to the device operation part. The degree of coverage of the perception capability of a single edge node to the perception requirements of the corresponding device operation part is analyzed, and the specific perception content with insufficient coverage is marked to obtain the single node coverage evaluation result. For the equipment operation parts with insufficient sensing capabilities in the single-node coverage assessment results, analyze the sensing capabilities of other related equipment operation parts and corresponding edge nodes, screen adjacent nodes with supplementary sensing capabilities, label the screened nodes and supplementary sensing content, and obtain the collaborative node screening results. Based on the results of the collaborative node screening, the participating nodes, sensing responsibilities and collaboration priorities of each node are defined for each part of the equipment operation that needs to be sensed collaboratively. The division of responsibilities and priority ranking criteria of core nodes and auxiliary nodes are marked to obtain the collaborative sensing division of labor results. Analyze the operating characteristics of the equipment operating parts that require collaborative sensing, set the start conditions, duration and data interaction frequency of collaborative sensing, mark the equipment operating parameters corresponding to the start conditions and the adjustment basis of the data interaction frequency, and obtain the collaborative sensing parameter configuration results. The results of the collaborative sensing division of labor and the collaborative sensing parameter configuration are combined and organized according to the equipment operation section. The participating nodes, sensing responsibilities, start conditions and data interaction requirements of each collaborative sensing task are marked. The basis for task execution is supplemented to obtain the initial collaborative sensing requirement list. The initial collaborative sensing requirement list was checked for cases where there were too many collaborative nodes or overlapping responsibilities. Duplicate collaborative sensing tasks were merged, collaborative node configurations were optimized, responsibilities were adjusted, and resource consumption was reduced to obtain the optimized list. The importance of the equipment operation parts that need to be sensed collaboratively and their relationship with other collaborative tasks are supplemented; the execution priority of collaborative sensing tasks is set; the basis for setting the priority and the connection requirements between tasks are marked; and the priority marking results are obtained. The results of the integrated list optimization and priority labeling are presented in order of execution priority, and the core information and related content of each task are labeled to generate the final collaborative perception requirement list.

7. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 3, characterized in that, The self-organization result parsing data and the equipment operation characteristic dataset are used to identify the parts in the perception task self-organization results that do not match the equipment operation characteristics, and the adaptation deviation analysis results are obtained, including: Extract the sensing task coverage, execution sequence and collaboration method from the self-organization result analysis data, classify and organize them according to the equipment operation part, and mark the sensing task details and timing arrangement corresponding to each part to obtain the sensing task deployment details; Extract the core content of the equipment operation characteristic dataset, including the key operating parameter ranges, operating status change cycles and fault-sensitive periods of each equipment operation part, and label the specific performance of each parameter range and cycle to obtain the core equipment characteristic data; The deployment details of sensing tasks are associated with the core data of equipment characteristics one by one according to the equipment operation parts. The correspondence between the coverage of sensing tasks and the key operating parameters of the equipment is compared. The key parameters that are not covered are identified, and the specific content and impact of the uncovered parameters are marked to obtain the coverage deviation analysis. Analyze the fit between the execution timing of sensing tasks and the cycle of equipment operating status changes, identify cases where the start time of sensing tasks misses the critical period of equipment operating status changes, mark the specific manifestations and impacts of the missed period, and obtain the timing fit deviation analysis. Assess the compatibility between the collaborative methods of perception tasks and the correlation strength between the equipment operation parts, investigate the mismatch between the number of collaborative nodes and the frequency of collaboration and the degree of influence between the equipment parts, mark the specific manifestations and causes of the mismatch, and obtain the collaboration method deviation analysis. By integrating coverage deviation analysis, timing fit deviation analysis, and collaboration mode deviation analysis, the corresponding equipment operation parts, sensing tasks, and specific performances for each deviation are labeled, and the scenario conditions that caused the deviation are supplemented to obtain preliminary deviation analysis results. Analyze the causes of various deviations in the preliminary deviation analysis results, distinguish between deviations caused by unreasonable self-organizing processes of perception tasks and deviations caused by missing equipment operating characteristic data, label the specific scenarios corresponding to the causes, and obtain the deviation cause analysis results; Based on the results of the deviation cause analysis, deviations caused by unreasonable self-organization results of perception tasks are screened out, non-substantive deviations caused by missing data are excluded, and the specific range and impact of effective deviations are marked to obtain the effective deviation screening results. The effective deviation screening results are sorted according to the degree of impact of deviations on equipment operation status monitoring, highlighting the key deviations that affect the completeness and accuracy of monitoring, and marking the sorting basis and the core impact of the key deviations to obtain the deviation priority ranking results. By integrating the effective deviation screening results and the deviation priority ranking results, the corresponding equipment part, perception task related information, specific performance and impact of each deviation are determined, supplementing the core basis of deviation analysis and generating the final adaptation deviation analysis results.

8. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 4, characterized in that, The optimized sensing task self-organizing process is used to initiate the sensing task execution, and the sensing module collects the operating status data of each operating part of the loading equipment, including: According to the timing requirements in the optimized sensing task self-organizing process, the edge node starts the sensing module preheating process, checks the power supply status, working mode and data acquisition channel of the sensing module, switches to normal working mode, records the module startup process and status changes, and obtains the module startup result. According to the coverage area assigned by the sensing task, adjust the acquisition angle, sensitivity and sampling frequency of the sensing module to ensure that the acquisition range accurately covers the corresponding operating part of the device. Adjust the acquisition angle to the optimal position, set the specific parameters of sensitivity and sampling frequency, and obtain the module parameter configuration results. The sensing data acquisition process is initiated. According to the optimized timing requirements and sampling frequency, the operating status data of the mechanical transmission part, the hydraulic control part, and the electrical drive part are collected. The time point of each acquisition and the working parameters of the sensing module are recorded synchronously to generate the original acquisition data sequence. Define the result according to the collaboration method, send the acquisition start synchronization signal to the neighboring nodes participating in the collaborative sensing task, receive the acquisition start confirmation information from the neighboring nodes, adjust its own acquisition start time to be consistent with the neighboring nodes, and obtain the collaborative synchronization result; During the data collection process, the system receives real-time progress feedback from neighboring nodes, synchronously shares some collected operational status data, supplements the data collection content that it does not cover, verifies the consistency between the shared data and its own collected data, and obtains collaborative supplementary data. The original collected data sequence and the collaborative supplementary data are integrated, classified and organized according to the equipment operation, duplicate data entries are removed, complete operation status data records are retained, and the classification basis and data source are labeled to obtain the classified collection dataset. During the monitoring and data collection process, if any sudden changes in the equipment's operating status are triggered, and an abnormal response condition is established, the regular data collection process is paused according to the abnormal response procedure rules. The special data collection mode is then activated, and the operating status data before and after the sudden change is recorded. The activation conditions and data range of the special data collection are marked, and the special data collection data is obtained. By linking the special collection data with the regular classification collection dataset, the time point of the sudden change in the equipment operating status, the parameter changes before and after the change, and the special collection content are marked, and the descriptive information of the change is supplemented to obtain the integrated collection dataset; The sensing module's running status log, collaborative node interaction records, and collection environment parameters during the collection process are associated and stored with the integrated collection dataset. The collection timestamp and associated data entry identifier corresponding to each piece of information are also recorded to obtain the complete collection dataset. According to the format requirements defined in the data interaction rules, the complete collected dataset is standardized, the data representation and storage format are unified, the specific basis for the standardization process is marked, and the final operating status data of each operating part of the loading equipment is generated.

9. The method for monitoring the operating status of loading equipment based on cloud-edge-device collaboration according to claim 5, characterized in that, The process of adjusting the generation parameters of dynamic collaborative rules and optimizing the classification criteria for rule adaptation scenarios based on the generated logical defect localization results includes: The results of the logic defect localization are analyzed, and the defect types in the dynamic collaborative rule generation logic are extracted, including unreasonable parameter settings, overly coarse scenario division, and loose rule connection. The scope of impact and specific manifestations of each defect are determined, and a detailed list of defect types is obtained. To address the issue of mismatched parameter settings, historical dynamic collaborative rules are retrieved to generate parameters and corresponding execution effect data. A mapping relationship between parameters and execution effect indicators is established, and the direction and range of parameter adjustment are determined based on the mapping relationship to obtain a parameter adjustment plan. Based on the parameter adjustment scheme, the definition parameters of the sensing task adjustment boundary, the configuration parameters of the data interaction method, and the trigger parameters of the abnormal response process are adjusted to make the parameter settings fit the device operating characteristics and edge node capabilities. The changes before and after the parameter adjustment are recorded to obtain the parameter update results. To address the shortcomings of overly coarse scenario classification, this study analyzes the operating scenario types of loading equipment, and refines the scenario classification criteria by combining factors such as equipment load changes, differences in working environment, and switching of operating modes. It also adds subdivided scenario types, marks the basis and characteristics of subdivided scenario classification, and obtains the optimized scenario classification results. Define exclusive collaborative rule generation logic for each sub-scenario type, determine the differentiated requirements for perception task adjustment boundaries, data interaction methods and abnormal response processes under different sub-scenarios, mark the basis for differentiated settings, and obtain scenario-specific generation logic; To address the shortcomings of loose rule connections, the logical relationships between the clauses of the dynamic collaborative rules are linked, missing connecting clauses are supplemented, the execution order and triggering conditions between clauses are determined, rule conflicts or execution gaps are eliminated, and the rule connection optimization results are obtained. By integrating the results of parameter updates, scene segmentation optimization, scene-specific generation logic, and rule connection optimization, a dynamic collaborative rule generation logic adjustment plan is formed. The core content and basis of the adjustment are determined, the execution steps of the plan are marked, and a draft of the generation logic adjustment is obtained. The generated logic adjustment draft is applied to historical feedback data, the rule generation logic is run to obtain simulated dynamic collaborative rules, the performance indicators of the simulated rules under different historical scenarios are calculated, the scenarios and rule clauses where the indicators are not met are recorded, and the simulation verification results are obtained. Based on the simulation verification results, the parameter settings and scenario division standards in the draft of the generation logic adjustment were fine-tuned, unreasonable content found in the simulation verification was corrected, and the rule connection clauses were optimized to obtain the final draft of the generation logic adjustment. The final draft of the generation logic is integrated into the dynamic collaborative rule generation logic. The algorithm flow, parameter configuration, and scenario adaptation mechanism for rule generation are updated. The complete content and basis of the logic update are recorded to complete the optimization of the generation logic.

10. A cloud-edge-device collaborative system for monitoring the operational status of loading equipment, characterized in that, The device includes a processor and a computer-readable storage medium storing machine-executable instructions, which, when executed by the processor, implement the cloud-edge-device collaborative method for monitoring the operating status of loading equipment as described in any one of claims 1-9.

Images