Industrial robot remote monitoring method and system based on Internet of Things

Abstract

The invention relates to the technical field of industrial automation, in particular to an industrial robot remote monitoring method and system based on the Internet of Things, and the method comprises the following steps: extracting a load state and transmission link delay through the operation data of an industrial robot node, comparing the communication performance with task load distribution, and weighting link relevance; and generating a node association topological structure. According to the method, the operation data of the industrial robot is deeply analyzed, the production efficiency is improved, the system response speed is increased, the load states and link delays of the nodes are extracted, communication between the nodes and task load distribution are optimized, the task execution sequence is optimized through a data-driven strategy, real-time state changes are coped through dynamic adjustment, and therefore the real-time state change of the industrial robot is optimized. The delay and failure rate are remarkably reduced, the link and node capacity is continuously monitored, the task allocation reasonability is ensured, the problem can be quickly positioned, shutdown is reduced, and the stability and predictability of the production process are improved by optimizing the task path and adjusting cloud edge cooperation.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation, and in particular to a remote monitoring method and system for industrial robots based on the Internet of Things. Background Art

[0002] The field of industrial automation technology involves the use of various automatic control devices, such as logic controllers, sensors, robots and advanced mechanical equipment, to improve production efficiency, reduce human intervention and optimize the production process. The key technologies in this field include machine vision systems, automated assembly lines, control systems and feedback mechanisms, which aim to enhance the flexibility and reliability of factory operations and ensure the safety of the production environment. The implementation of industrial automation can significantly improve product quality, reduce production costs, and improve productivity and efficiency through continuous process monitoring and precise control.

[0003] Among them, the industrial robot remote monitoring method is a technology that allows users to monitor and control industrial robots from a remote location. The core functions of this topic include real-time data transmission, performance analysis, early warning and maintenance instruction communication through the network. This method is mainly used to ensure the robot's operating efficiency and timely problem solving when there are no on-site access conditions. It is widely used in production line monitoring, quality control and maintenance management. By realizing remote monitoring, enterprises can reduce downtime, optimize resource allocation, and improve the reliability and safety of the overall production line.

[0004] Although existing technologies have made remarkable achievements in improving production efficiency and reducing human intervention, they are still limited in processing large-scale data and real-time requirements, especially in the field of remote monitoring and control. Conventional technologies rely on static system settings and lack the ability to adapt to dynamic changes in the production environment. This fixed processing logic leads to insufficient response speed and flexibility of the system in the face of rapidly changing production needs, and it is impossible to adjust the production line or optimize resource allocation in time, thereby increasing downtime and maintenance costs. The deficiencies of existing methods in data processing and priority judgment also lead to uneven task distribution and low execution efficiency, which is particularly obvious under high load conditions. The potential capabilities of each node cannot be fully utilized, affecting the output efficiency and product quality of the overall production line. Summary of the invention

[0005] In order to solve the problem that the existing technology has made remarkable achievements in improving production efficiency and reducing manual intervention, but there are still limitations in processing large-scale data and real-time requirements, especially in the field of remote monitoring and control, conventional technology relies on static system settings and lacks the ability to adapt to dynamic changes in the production environment. This fixed processing logic leads to insufficient response speed and flexibility of the system in the face of rapidly changing production needs, and it is impossible to adjust the production line or optimize resource allocation in time, thereby increasing downtime and maintenance costs. The deficiencies of existing methods in data processing and priority judgment also lead to uneven task distribution and low execution efficiency, which is particularly obvious under high load conditions. It is impossible to fully utilize the potential capabilities of each node, affecting the output efficiency and product quality of the overall production line. The embodiment of the present invention provides a remote monitoring method and system for industrial robots based on the Internet of Things. The technical solution is as follows:

[0006] On the one hand, a remote monitoring method for an industrial robot based on the Internet of Things is provided, the method comprising:

[0007] S1: Extract the load status and transmission link delay through the operation data of the industrial robot nodes, compare the communication performance and task load distribution, weight the link association, and generate the node association topology structure;

[0008] S2: Based on the node association topology, extract the task load and execution capability, combine the link transmission performance priority comparison, classify the task load and screen the key nodes to obtain the task priority allocation matrix;

[0009] S3: Based on the task priority allocation matrix, combined with the real-time status of the task and the path link information, the path execution time is analyzed and the order and distribution are optimized to generate an optimized task transmission path planning scheme;

[0010] S4: Based on the optimized task transmission path planning scheme, extract the execution capability distribution data, adjust the local and cloud task load ratios and sort the priorities, and generate a task execution priority status table;

[0011] S5: Based on the task execution priority status table, analyze the task path status and compare the completion time and feedback, locate abnormal nodes and status deviations, and generate industrial robot operation status monitoring and evaluation results.

[0012] As a further scheme of the present invention, the node association topology structure includes node priority, link weighted index, and node load information; the task priority allocation matrix includes task execution nodes, node execution capabilities, and task load levels; the optimized task transmission path planning scheme includes path optimization strategy, link load sorting, and task execution scheduling; the task execution priority status table includes task priority, task type classification, and cloud-edge collaboration ratio; the industrial robot operation status monitoring and evaluation results include node anomaly analysis, task execution deviation, and status feedback evaluation.

[0013] As a further solution of the present invention, the steps of extracting load status and transmission link delay through the operation data of industrial robot nodes, comparing communication performance and task load distribution, weighting link association, and generating node association topology structure are as follows:

[0014] S101: Based on the operation data of the industrial robot, extract the load status and transmission link delay of each robot node, analyze the numerical distribution of the load parameters and transmission time parameters in the operation data, adjust the thresholds of the node load and link delay, and obtain the robot node performance data;

[0015] S102: Based on the robot node performance data, analyze the communication performance parameters between each node, analyze the corresponding relationship between the load ratio and the delay time, perform difference calculation on the node task load distribution data, perform statistics and comparison on the transmission delay in the communication path between the nodes, and generate a matching parameter set of the node task load and the communication delay;

[0016] S103: weighting the matching parameter set of the node task load and the communication delay, calling the weight parameter to mark the link priority between the nodes, and generating a node association topology structure.

[0017] As a further solution of the present invention, based on the node association topology, extracting task load and execution capability, combining link transmission performance priority comparison, classifying task load and screening key nodes, and obtaining the task priority allocation matrix are specifically as follows:

[0018] S201: extracting the task load and execution capability of the industrial robot node based on the node association topology, analyzing the task allocation of the node connections in the topology, separating the task load data from the node execution capability parameters, comparing the matching degree of the node load and the execution capability, rearranging the corresponding relationship between the load and the capability according to the node priority, and obtaining the node task load capability matrix;

[0019] S202: Based on the node task load capacity matrix, divide the task load data into task groups according to characteristic parameters, analyze the load distribution characteristics, analyze the adaptability of tasks and node capabilities, select nodes that meet the conditions, mark the key node positions, and generate task grouping and key node tables;

[0020] S203: Analyze the task grouping and key node table, extract task group priority and node weight parameters, reorder node execution priorities, adjust the priority order, and obtain a task priority allocation matrix.

[0021] As a further solution of the present invention, based on the task priority allocation matrix, combined with the real-time status of the task and the path link information, the path execution time is analyzed and the order and distribution are optimized, and the steps of generating the optimized task transmission path planning scheme are specifically as follows:

[0022] S301: Based on the task priority allocation matrix, extract real-time task execution data and link status data, evaluate task node execution time and link delay, and obtain a task path execution time matrix;

[0023] S302: extracting link load data based on the task path execution time matrix, and comparing and analyzing with link capacity data, sorting according to load ratio, identifying and optimizing task paths, reconfiguring tasks, and obtaining a task path load sorting table;

[0024] S303: Based on the task path load sorting table, analyze the sorted load table, adjust task distribution and path allocation, optimize task transmission time and link congestion, and generate an optimized task transmission path planning solution.

[0025] As a further solution of the present invention, based on the optimized task transmission path planning scheme, the execution capability distribution data is extracted, the local and cloud task load ratios are adjusted and the priorities are sorted, and the steps of generating a task execution priority status table are specifically as follows:

[0026] S401: Based on the optimized task transmission path planning scheme, collect the execution capability data of the edge device and the industrial robot, including processing speed, memory usage, and network speed, determine the task volume of each node, and generate a node capability details table;

[0027] S402: Based on the node capability details table, classify local and cloud tasks, calculate the task load ratio of the node, sort the tasks according to resource requirements and available resources, and generate a task priority and resource matching table;

[0028] S403: Based on the task priority and resource matching table, a dynamic task scheduling algorithm is used to analyze and set the task execution ratio of the cloud and edge devices, adjust the task allocation, and generate a task execution priority status table with reference to the response time of the edge device and the processing capacity of the cloud.

[0029] As a further solution of the present invention, the formula of the dynamic task scheduling algorithm is as follows:

[0030]

[0031] Calculate the task execution priority score P i , where T ri represents the average response time of task i on the edge device, R ei represents the edge computing resources required for task i, c i represents the processing capability index of task i in the cloud, L i represents the network delay of task i, w1 is the critical weight of task response time, w2 is the critical weight of edge computing resources, w3 is the critical weight of cloud processing capacity, and w4 is the network delay adjustment coefficient.

[0032] As a further solution of the present invention, based on the task execution priority status table, the task path status is parsed and the completion time and feedback are compared, abnormal nodes and status deviations are located, and the steps of generating the industrial robot operation status monitoring evaluation results are specifically as follows:

[0033] S501: Based on the task execution priority status table, collect the path status data of the industrial robot, record the task execution start and end time of each node, mark abnormal nodes and time delays, and generate a path status and time record table;

[0034] S502: Based on the path status and the time record table, compare the task completion time, parse the time data, calculate the time difference between the task completion and the expected time, mark the abnormal nodes, count the task delay ratio, and generate a time deviation list of abnormal nodes;

[0035] S503: Based on the abnormal node time deviation list, a machine learning algorithm is used to analyze the deviation in link transmission, locate abnormal nodes, extract state change records, compare with normal node data, analyze abnormal parameter differences, and generate industrial robot operation status monitoring evaluation results.

[0036] As a further solution of the present invention, the formula of the machine learning algorithm is as follows:

[0037]

[0038] Calculate the anomaly score S, where t represents the time deviation of a single node and t avg represents the average time deviation of normal nodes, n represents the total number of nodes, and d max represents the maximum deviation value, and α, β, and γ are adjustment coefficients respectively.

[0039] On the other hand, an electric vehicle state monitoring system is provided, the electric vehicle state monitoring system is used to execute the above electric vehicle state monitoring method, the system comprises:

[0040] The node load status determination module extracts the real-time load information and link delay of each node through the operation data of the industrial robot node, calculates and corrects the delay deviation using the real-time performance data, and generates a node load and delay table;

[0041] The link priority analysis module evaluates and compares the communication performance and task load between nodes according to the node load and delay table, adjusts the priority of each link through a weighted algorithm, and establishes a node association topology map;

[0042] The task allocation matrix module analyzes the task processing capability of each node and the transmission efficiency of the link according to the node association topology diagram, prioritizes the node execution tasks, and generates a task priority matrix;

[0043] The path planning optimization module analyzes the task flow and real-time link status according to the task priority matrix, recalculates the optimal task execution path, adjusts and optimizes the path, and creates an optimized transmission path solution;

[0044] The execution node optimization module extracts key execution nodes from the optimized transmission path scheme, analyzes the load balance and execution capability between nodes, reconfigures the task load, and forms an execution node priority table;

[0045] The status monitoring and evaluation module uses the execution node priority table to monitor the task execution path and real-time status of the industrial robot, detect and diagnose anomalies in task execution in real time, and generate industrial robot operation status monitoring evaluation results.

[0046] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0047] Through in-depth analysis and processing of industrial robot operation data, production efficiency and system response speed can be significantly improved. By extracting the load status and transmission link delay of each node, and comprehensively comparing and weighting the data, the communication performance between nodes and the optimal distribution of task loads can be ensured. This data-driven decision-making process not only optimizes the execution order of tasks, but also adapts to changes in real-time status through dynamic adjustment, significantly reducing overall delays and failure rates. Continuous monitoring and analysis of link and node execution capabilities further ensure the rationality of task allocation and the efficiency of execution, which helps to quickly locate and respond to potential problems and reduce the risk of production interruptions. By optimizing task transmission paths and adjusting the collaboration ratio between cloud and edge devices, the scalability and flexibility of the method are enhanced, making the entire production process more stable and predictable. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0049] Figure 2 This is a detailed flow chart of S1 of the present invention;

[0050] Figure 3 This is a detailed flow chart of S2 of the present invention;

[0051] Figure 4 This is a detailed flow chart of S3 of the present invention;

[0052] Figure 5 This is a detailed flow chart of S4 of the present invention;

[0053] Figure 6 This is a detailed flow chart of S5 of the present invention;

[0054] Figure 7 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0056] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0057] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0058] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0059] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0060] See also Figure 1 The embodiment of the present invention provides a remote monitoring method for an industrial robot based on the Internet of Things. The processing flow of the method may include the following steps:

[0061] S1: Extract the load status and transmission link delay of each robot node through the operation data of the industrial robot nodes, perform weighted processing on the link association by comparing the communication performance and task load distribution between nodes, mark the priority relationship between the nodes of the industrial robot, and generate the node association topology structure;

[0062] S2: Based on the node association topology, the task load and execution capability of the industrial robot node are extracted, and the priority is compared in combination with the transmission performance of the link. The task load is grouped and classified, the key nodes for task execution are screened, the priority order of the execution nodes of the task is determined, and the task priority allocation matrix is ​​obtained;

[0063] S3: Based on the task priority allocation matrix, combined with the real-time status of the industrial robot task and the path link information, the execution time of the task transmission path is analyzed, the link load is analyzed and sorted, the execution order of the tasks is adjusted, the path distribution is optimized, and the optimized task transmission path planning scheme is generated;

[0064] S4: Based on the optimized task transmission path planning scheme, the execution capability distribution data of edge devices and industrial robot nodes are extracted, the load between local tasks and cloud tasks is classified and adjusted, the allocation ratio of cloud collaboration and edge tasks is determined by reordering the task priorities, and the task execution priority status table is generated;

[0065] S5: Based on the task execution priority status table, the path status of the industrial robot during the task execution process is analyzed. By comparing the task completion time and execution feedback, the deviation of abnormal nodes in the link transmission is analyzed, the task execution interruption and abnormal nodes and status differences are located, and the industrial robot operation status monitoring and evaluation results are generated.

[0066] The node association topology structure includes node priority, link weighted indicators, and node load information. The task priority allocation matrix includes task execution nodes, node execution capabilities, and task load levels. The optimized task transmission path planning scheme includes path optimization strategy, link load sorting, and task execution scheduling. The task execution priority status table includes task priority, task type classification, and cloud-edge collaboration ratio. The industrial robot operation status monitoring and evaluation results include node anomaly analysis, task execution deviation, and status feedback evaluation.

[0067] Specifically, if Figure 2 As shown in FIG, the steps of extracting the load status and transmission link delay through the operation data of the industrial robot node, comparing the communication performance and task load distribution, weighting the link association, and generating the node association topology structure are as follows:

[0068] S101: Based on the operation data of the industrial robot, extract the load status and transmission link delay of each robot node, analyze the numerical distribution of the load parameters and transmission time parameters in the operation data, adjust the thresholds of the node load and link delay, and obtain the robot node performance data;

[0069] The load status and transmission link delay of each node are analyzed. By parsing the numerical distribution of load parameters and transmission time parameters, the thresholds of node load and link delay are adjusted. The real-time performance of the robot can be evaluated more accurately through data-driven methods to ensure the optimal operation of the system. This method uses machine learning algorithms to analyze a large amount of historical and real-time data. By establishing a mathematical model to predict the performance change trend of the node, the task allocation and communication strategy between nodes can be adjusted in real time, thereby reducing the overall delay of the system and improving efficiency. By monitoring the status of the node in real time and responding quickly, the stability and response speed of the robot system can be significantly improved, making the system more reliable and effective, and finally obtaining the robot node performance data.

[0070] S102: Based on the robot node performance data, analyze the communication performance parameters between each node, analyze the corresponding relationship between the load ratio and the delay time, perform difference calculation on the node task load distribution data, perform statistics and comparison on the transmission delay in the communication path between nodes, and generate a matching parameter set of the node task load and the communication delay;

[0071] Analyze the corresponding relationship between load ratio and delay time, and perform difference calculation on node task load distribution data according to the formula:

[0072]

[0073] Calculate the variance of the load difference of each node, where A j represents the load value of the jth node, A represents the average load of all nodes, and n represents the total number of nodes;

[0074] Assume there are three nodes, whose loads are A1=70, A2=65, A3=75, and the average load is A=70. Then the sum of squares of the difference between the load of each node and the average load is (70-70). 2 +(65-70) 2 +(75-70) 2 =0+25+25=50, divided by the number of nodes minus one, that is, n-1=3-1=2, we get the variance

[0075] This variance result shows that there are certain differences in the load between nodes, and the balance of the overall network needs to be optimized.

[0076] S103: weighting the matching parameter set of the node task load and the communication delay, calling the weight parameter to mark the link priority between the nodes, and generating a node association topology structure;

[0077] The link priorities between nodes are marked by calling weight parameters. The process evaluates the data transmission and processing capabilities between nodes by using advanced algorithms and computational models to ensure the optimization of data flow and the rationality of resource allocation. This method allows the system to dynamically adjust its communication paths and load distribution based on real-time data and historical performance to adapt to changing operating conditions and needs. In addition, this strategy not only optimizes resource utilization but also improves the overall reliability and efficiency of the system by precisely controlling data flow, thereby reducing problems caused by improper configuration or node overload while maintaining high communication quality, and generating a node-associated topology.

[0078] Specifically, if Figure 3 As shown in the figure, based on the node association topology, the task load and execution capability are extracted, and the task load is classified and the key nodes are screened in combination with the link transmission performance priority comparison. The specific steps to obtain the task priority allocation matrix are as follows:

[0079] S201: based on the node association topological structure, extract the task load and execution capability of the industrial robot node, analyze the task allocation of the node connection in the topology, separate the task load data and the node execution capability parameter, compare the matching degree of the node load and the execution capability, rearrange the corresponding relationship between the load and the capability according to the node priority, and obtain the node task load capability matrix;

[0080] Combined with the node connection information in the topology, the task type and amount undertaken by each node are analyzed through the task allocation situation. The task load data and node execution capability parameters can be separated by collecting the task completion time and node operating power data respectively. Based on the obtained load and execution capability values, the matching degree of the two is calculated and the results are quantified. The matching degree can be analyzed by comparing the deviation between the actual running time of the node and the theoretical completion time. By sorting the priority of each node, the task is assigned to the optimal node, thereby constructing a node task load capacity matrix.

[0081] S202: Based on the node task load capacity matrix, divide the task load data into task groups according to characteristic parameters, analyze the load distribution characteristics, analyze the adaptability of tasks and node capabilities, select nodes that meet the conditions, mark the key node positions, and generate task grouping and key node tables;

[0082] The task load data is divided into task groups according to characteristic parameters, according to the formula

[0083]

[0084] Calculate the average feature load of each task group, where B i represents the load value of the i-th task, T i represents the execution time of the ith task, and N is the total number of tasks;

[0085] Detailed explanation of the formula and the process of formula calculation and derivation:

[0086] For a task group, there are three tasks, with load values ​​of B1 = 10, B2 = 20, and B3 = 30, respectively, and execution times of D1 = 2, D2 = 3, and D3 = 5, respectively. The execution time T i They are 2, 3, and 5 respectively;

[0087] The characteristic average load is calculated according to the formula:

[0088]

[0089] The result shows that the characteristic average load of the task group is 23, indicating that the load distribution within the task group is relatively balanced, which provides a basis for the subsequent task and node adaptation analysis.

[0090] S203: Analyze the task grouping and key node table, extract the task group priority and node weight parameters, reorder the node execution priority, adjust the priority order, and obtain the task priority allocation matrix;

[0091] The task group priority is reordered according to the criticality of task execution, and the execution priority is readjusted by calculating the ratio between the average communication delay value of each node in the current topology and the task execution time. The average communication delay can be obtained by analyzing the sending and receiving time of network traffic data packets. After reordering, a task priority allocation matrix is ​​formed. The combination of task groups and node weights in the matrix can clarify the order of task execution.

[0092] Specifically, if Figure 4 As shown in the figure, based on the task priority allocation matrix, combined with the real-time status of the task and the path link information, the path execution time is analyzed and the order and distribution are optimized. The specific steps for generating the optimized task transmission path planning scheme are as follows:

[0093] S301: Based on the task priority allocation matrix, extract the real-time task execution data and link status data, evaluate the task node execution time and link delay, and obtain the task path execution time matrix;

[0094] By monitoring the start time and completion time of each task, the execution time difference is calculated to determine the actual operation cycle of the task. The link status data is obtained by collecting performance parameters such as the packet loss rate, average transmission delay, and bandwidth occupancy rate of each link in the network transmission. The link delay is obtained by calculating the packet transmission time interval between the nodes at both ends. When parsing the execution time of the task node, it is necessary to track the transmission data of each link in real time. Combined with the actual completion status of the task operation and the link status, the overall execution time of the task path is obtained by adding the task execution time and the link transmission delay. By recording the execution time of each task path, it is summarized and organized into a task path execution time matrix. The generation of this matrix depends on the comprehensive analysis of the relationship between task allocation and link performance, which can provide an intuitive data basis for the subsequent optimization of task paths.

[0095] S302: extract link load data based on the task path execution time matrix, compare and analyze with link capacity data, sort according to load ratio, identify and optimize task paths, reconfigure tasks, and obtain a task path load sorting table;

[0096] The link load data is compared and analyzed with the link capacity data according to the formula:

[0097]

[0098] Calculate the link load ratio E, where T task represents the time required for task transmission, Cl ink indicates the bandwidth capability of the link;

[0099] The time required for task transmission on a link is T task = 10 seconds, link bandwidth capacity is C l ink=50MB / s, calculated according to the formula:

[0100]

[0101] The load ratio of this link is 0.2. The link load ratio reflects the occupation of link resources by tasks. To calculate the link load ratio, it is necessary to determine the task transmission time and link bandwidth capacity through the actual link operation data. The task transmission time can be obtained by recording the time points when the task starts and completes the transmission. The link bandwidth capacity is determined by measuring the stable transmission rate of the link under high load. The sorting of link load ratios provides a clear adjustment direction for optimizing the task path. If the link load ratio value is high, it means that the link has become a bottleneck and the task needs to be reconfigured to relieve the link pressure, which ultimately provides data basis for optimizing the task path.

[0102] S303: Based on the task path load sorting table, analyzing the sorted load table, adjusting task distribution and path allocation, optimizing task transmission time and link congestion, and generating an optimized task transmission path planning scheme;

[0103] First, the task priority and the load ratio value of the corresponding link are extracted. The task priority is determined by evaluating the criticality and urgency of the task. The link load ratio is sorted based on the actual bandwidth occupancy of the link. The sorted load table is arranged from high to low according to the task priority, and the high-load path of the link is marked. In the process of adjusting the task distribution, high-priority tasks need to be assigned to the link with the lowest load ratio. By real-time monitoring of the task execution data and the performance parameters of the link, the high-load path is adjusted in real time. The optimization of task transmission time depends on the actual matching of link delay and bandwidth. Link congestion is alleviated by reducing the number of tasks on the same link. Finally, an optimized task transmission path planning scheme is formed. By comprehensively considering the fairness of task distribution and the maximum utilization of link resources, the task transmission time is minimized, and the impact of link congestion on the overall system performance is significantly reduced, thereby completing the dynamic optimization of the task path.

[0104] Specifically, if Figure 5 As shown in the figure, based on the optimized task transmission path planning scheme, the execution capability distribution data is extracted, the local and cloud task load ratios are adjusted and the priorities are sorted, and the steps of generating the task execution priority status table are as follows:

[0105] S401: Based on the optimized task transmission path planning scheme, collect the execution capability data of edge devices and industrial robots, including processing speed, memory usage, and network speed, determine the task volume of each node, and generate a node capability details table;

[0106] By collecting the actual performance data of edge devices and industrial robots when performing tasks, the processing speed can be calculated by counting the ratio of task processing time to the number of completed tasks. The memory usage is obtained by recording the memory occupancy rate of the device during task execution. The network speed is measured by the data packet transmission rate between devices. The data is summarized by the system to reflect the comprehensive capabilities of the equipment. The task volume is determined by the number of tasks assigned by the system and the processing complexity. By combining the device execution capability with the task allocation data, a node capacity details table is generated, which associates the actual capacity of each node with the task volume. This can clearly describe the capacity range and load status of different nodes, providing a reference for subsequent task classification and priority sorting.

[0107] S402: Based on the node capability details table, classify local and cloud tasks, calculate the task load ratio of the node, sort the tasks according to resource requirements and available resources, and generate a task priority and resource matching table;

[0108] Tasks are classified according to the execution characteristics of local and cloud computing. Local tasks include small-scale tasks that require quick response, while cloud tasks include tasks with high computational complexity or large data size. The task load ratio is calculated by the ratio of local node resource consumption to task completion rate. Resource requirements are quantified according to the computing time, memory consumption, bandwidth occupancy and other characteristics of each task. A task priority and resource matching table is generated. By matching and sorting the resource requirements of the tasks with the available resources of the nodes, priority is given to the allocation method that efficiently utilizes resources. Based on the impact of node capabilities on task completion time, the priority of task allocation is adjusted to ensure optimal utilization of resources.

[0109] S403: Based on the task priority and resource matching table, a dynamic task scheduling algorithm is used to analyze and set the task execution ratio of the cloud and edge devices, adjust the task allocation, and refer to the response time of the edge device and the processing capacity of the cloud to generate a task execution priority status table;

[0110] The formula of the dynamic task scheduling algorithm is as follows:

[0111]

[0112] Calculate the task execution priority score P i , where T ri represents the average response time of task i on the edge device, R ei represents the edge computing resources required for task i, ci represents the processing capability index of task i in the cloud, L i represents the network delay of task i, w1 is the critical weight of task response time, w2 is the critical weight of edge computing resources, w3 is the critical weight of cloud processing capacity, and w4 is the network delay adjustment coefficient;

[0113] Detailed explanation of the formula and the process of formula calculation and derivation:

[0114] This formula is used to calculate the execution priority score P of task i. i , where various parameters are obtained through monitoring and data analysis. The calculation of this formula involves several key steps:

[0115] T ri (The average response time of task i on the edge device) is measured by running the corresponding task on the edge device and recording the time from start to completion. Suppose the average response time of task i is 200 milliseconds after monitoring for a period of time;

[0116] R ei (Edge computing resources required for task i) is defined based on task complexity and resource consumption evaluation of previous tests, assuming that the resource units required for task i are 50 units;

[0117] c i (The processing capacity of task i in the cloud), the indicator is derived from the data provided by the cloud service provider, including the combination of computing rate and resource capacity, and the cloud processing capacity value of task i is set to 500;

[0118] L i (Network latency of task i), measured by a real-time network monitoring tool, set to 100 milliseconds;

[0119] The weights and adjustment coefficients are set as follows:

[0120] w1=0.3, indicating the impact of response time on priority;

[0121] w2=0.5, indicating the criticality of edge computing resources to task execution;

[0122] w3=0.2, the weight of cloud processing capacity, which is lower because cloud resources are more abundant;

[0123] w4=0.1, adjustment coefficient of network delay, small because network conditions can be optimized by some methods;

[0124] Substitute specific values ​​for calculation:

[0125]

[0126] Calculate the numerator: 0.3·200=60, 0.5·50=25, then 60+25=85;

[0127] Compute the square root portion of the denominator:

[0128] Calculate the absolute value part of the denominator: |0.1·100|=10;

[0129] Complete denominator calculation: 10+10=20;

[0130] This evaluates the entire expression:

[0131] The result shows that the execution priority score of task i is 4.25. The value reflects the priority of task i in the cloud and edge devices after considering response time, resource requirements, processing power, and network latency. A higher score indicates that task i should be executed first to achieve the best performance and user experience by utilizing the current resource status.

[0132] Specifically, if Figure 6 As shown in the figure, based on the task execution priority status table, the task path status is parsed and the completion time and feedback are compared, abnormal nodes and status deviations are located, and the steps for generating the industrial robot operation status monitoring and evaluation results are as follows:

[0133] S501: Based on the task execution priority status table, collect the path status data of the industrial robot, record the task execution start and end time of each node, mark abnormal nodes and time delays, and generate a path status and time record table;

[0134] By collecting the path status data of the industrial robot, the start time and end time of the task execution of each node are recorded. The time record uses a unified timestamp format. The path status data is obtained through the node's task processing status flag. The identification of abnormal nodes is based on task execution timeout or path transmission delay exceeding the standard range. By collating the task node execution records, a path status and time record table is generated. This table uses time as the main line to display the running trajectory and execution status of the task node, and provides data support for subsequent analysis.

[0135] S502: Based on the path status and the time record table, compare the task completion time, parse the time data, calculate the time difference between the task completion and the expected time, mark the abnormal nodes, count the task delay ratio, and generate a time deviation list of the abnormal nodes;

[0136] By comparing the task completion time with the expected time item by item and parsing the time data, the time difference is obtained by subtracting the expected time from the execution time. The task delay ratio is calculated based on the ratio of the number of delayed tasks to the total number of tasks. The abnormal node marking is based on the judgment rule that the delay time exceeds the allowable range. The data of the abnormal nodes are separately sorted and counted to generate a list of abnormal node time deviations. This list takes time deviation as the core indicator, supplemented by node identification and task information, to provide a basis for subsequent link and node anomaly analysis.

[0137] S503: Based on the abnormal node time deviation list, a machine learning algorithm is used to analyze the deviation in link transmission, locate the abnormal node, extract the state change record, compare the normal node data, analyze the abnormal parameter difference, and generate the industrial robot operation status monitoring evaluation result;

[0138] The formula for the machine learning algorithm is as follows:

[0139]

[0140] Calculate the anomaly score S, where t represents the time deviation of a single node and t avg represents the average time deviation of normal nodes, n represents the total number of nodes, and d max represents the maximum deviation value, α, β, and γ are adjustment coefficients respectively;

[0141] Detailed explanation of the formula and the process of formula calculation and derivation:

[0142] This formula is designed to evaluate the degree of abnormality of each node of the industrial robot. The abnormality score S is calculated by considering the statistical characteristics of time deviation, number of nodes and maximum deviation. The parameter acquisition method includes actual measurement and statistical calculation.

[0143] The parameter t represents the time deviation of a specific node, which is obtained by recording the reaction time of the node through the sensor and comparing it with the preset operation time. The actual response time of a node is set to 150 milliseconds and the preset time is 100 milliseconds, so t = 150;

[0144] Parameter t avg is the average time deviation of all normal nodes, which is calculated as the arithmetic mean of the time deviations of all nodes measured as normal. If there are 5 normal nodes with deviations of 10, 15, 20, 15, and 10 milliseconds respectively, then t avg =(10+15+20+15+10) / 5=14;

[0145] The parameter n represents the total number of nodes involved in the calculation, which is directly obtained from the system records. If the system detects 10 nodes, then n = 10;

[0146] Parameter d maxis the maximum time deviation observed in all nodes. The maximum value is obtained from the deviation data of all nodes. If the maximum deviation is 50 milliseconds, then d max =50;

[0147] The weight parameters α, β, and γ are set according to the sensitivity and deviation tolerance of the system, and the weight coefficients are optimized and determined by the expert group based on historical data, for example:

[0148] α = 0.8 adjusts the influence of time deviation, making the results more sensitive to time differences;

[0149] β = 0.5 reduces the impact of the number of nodes and is suitable for systems with more nodes;

[0150] γ = 0.3 increases the influence of the maximum deviation and ensures the significance of extreme values;

[0151] Substitute specific values ​​into the formula for calculation:

[0152]

[0153] S≈24.35

[0154] The results show that if the anomaly score is significantly higher than a certain threshold, for example, the threshold is set to 20, the current node has obviously deviated from the normal operating state. This score can be used to identify and prioritize those nodes that show significant deviations, further investigate the causes of the anomaly and take corrective measures.

[0155] like Figure 7 As shown, a remote monitoring system for industrial robots based on the Internet of Things includes:

[0156] The node load status determination module extracts the real-time load information and link delay of each node through the operation data of the industrial robot node, calculates and corrects the delay deviation using the real-time performance data, and generates a node load and delay table;

[0157] The link priority analysis module evaluates and compares the communication performance and task load between nodes based on the node load and delay table, adjusts the priority of each link through a weighted algorithm, and establishes a node association topology map;

[0158] The task allocation matrix module analyzes the task processing capability of each node and the transmission efficiency of the link according to the node association topology, prioritizes the node execution tasks, and generates a task priority matrix;

[0159] The path planning optimization module analyzes the task flow and real-time link status according to the task priority matrix, recalculates the optimal task execution path, adjusts and optimizes the path, and creates an optimized transmission path plan;

[0160] The execution node optimization module extracts key execution nodes from the optimized transmission path scheme, analyzes the load balance and execution capabilities between nodes, reconfigures the task load, and forms an execution node priority table;

[0161] The status monitoring and evaluation module uses the execution node priority table to monitor the task execution path and real-time status of the industrial robot, detect and diagnose anomalies in task execution in real time, and generate industrial robot operation status monitoring and evaluation results.

[0162] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A remote monitoring method for industrial robots based on the Internet of Things, characterized in that: The following steps are involved: Through the operation data of industrial robot nodes, load status and transmission link delay are extracted, communication performance and task load distribution are compared, link association is weighted, and node association topology is generated; Based on the node association topology, the task load and execution capability are extracted, and the task load is classified and the key nodes are screened in combination with the link transmission performance priority comparison to obtain the task priority allocation matrix; Based on the task priority allocation matrix, combined with the real-time status of the task and the path link information, the path execution time is analyzed and the order and distribution are optimized to generate an optimized task transmission path planning scheme; Based on the optimized task transmission path planning scheme, the execution capability distribution data is extracted, the local and cloud task load ratios are adjusted and the priorities are sorted, and a task execution priority status table is generated; Based on the task execution priority status table, the task path status is analyzed and the completion time and feedback are compared to locate abnormal nodes and status deviations, and generate the industrial robot operation status monitoring evaluation results.

2. The method for remote monitoring of industrial robots based on the Internet of Things according to claim 1, characterized in that: The node association topology structure includes node priority, link weighted index, and node load information; the task priority allocation matrix includes task execution nodes, node execution capabilities, and task load levels; the optimized task transmission path planning scheme includes path optimization strategy, link load sorting, and task execution scheduling; the task execution priority status table includes task priority, task type classification, and cloud-edge collaboration ratio; the industrial robot operation status monitoring and evaluation results include node anomaly analysis, task execution deviation, and status feedback evaluation.

3. The method for remote monitoring of industrial robots based on the Internet of Things according to claim 1, characterized in that: The specific steps of extracting load status and transmission link delay through the operation data of industrial robot nodes, comparing communication performance and task load distribution, weighting link association, and generating node association topology are as follows: Based on the operation data of industrial robots, the load status and transmission link delay of each robot node are extracted, the numerical distribution of load parameters and transmission time parameters in the operation data is analyzed, the thresholds of node load and link delay are adjusted, and the performance data of robot nodes are obtained; Based on the robot node performance data, the communication performance parameters between each node are parsed, the corresponding relationship between the load ratio and the delay time is analyzed, the difference calculation of the node task load distribution data is performed, the transmission delay in the communication path between the nodes is counted and compared, and a matching parameter set of the node task load and the communication delay is generated; The matching parameter set of the node task load and the communication delay is weighted in terms of correlation, and the link priority between the nodes is marked by calling the weight parameter to generate a node correlation topology structure.

4. The method for remote monitoring of industrial robots based on the Internet of Things according to claim 1, characterized in that: Based on the node association topology, the task load and execution capability are extracted, and the task load is classified and key nodes are screened in combination with the link transmission performance priority comparison to obtain the task priority allocation matrix. Specifically, the steps are as follows: Based on the node association topological structure, the task load and execution capability of the industrial robot node are extracted, the task allocation of the node connection in the topology is analyzed, the task load data and the node execution capability parameters are separated, the matching degree of the node load and the execution capability is compared, and the corresponding relationship between the load and the capability is rearranged according to the node priority to obtain the node task load capability matrix; Based on the node task load capacity matrix, the task load data is divided into task groups according to characteristic parameters, the load distribution characteristics are analyzed, the adaptability of tasks and node capabilities is analyzed, nodes that meet the conditions are screened, key node positions are marked, and task grouping and key node tables are generated; The task grouping and key node table are analyzed, task group priorities and node weight parameters are extracted, node execution priorities are reordered, and the priority order is adjusted to obtain a task priority allocation matrix.

5. The method for remote monitoring of industrial robots based on the Internet of Things according to claim 1, characterized in that: Based on the task priority allocation matrix, combined with the real-time status of the task and the path link information, the path execution time is analyzed and the order and distribution are optimized, and the steps of generating the optimized task transmission path planning scheme are as follows: Based on the task priority allocation matrix, extracting real-time task execution data and link status data, evaluating task node execution time and link delay, and obtaining a task path execution time matrix; Based on the task path execution time matrix, link load data is extracted, and compared and analyzed with link capacity data, sorted according to load ratio, and optimized task paths are identified, tasks are reconfigured, and a task path load sorting table is obtained; Based on the task path load sorting table, the sorted load table is analyzed, task distribution and path allocation are adjusted, task transmission time and link congestion are optimized, and an optimized task transmission path planning scheme is generated.

6. The method for remote monitoring of industrial robots based on the Internet of Things according to claim 1, characterized in that: Based on the optimized task transmission path planning scheme, the execution capability distribution data is extracted, the local and cloud task load ratios are adjusted and the priorities are sorted, and the steps of generating the task execution priority status table are as follows: Based on the optimized task transmission path planning scheme, collect the execution capability data of edge devices and industrial robots, including processing speed, memory usage, and network speed, determine the task volume of each node, and generate a node capability details table; Based on the node capability details table, local and cloud tasks are classified, the task load ratio of the node is calculated, tasks are sorted according to resource requirements and available resources, and a task priority and resource matching table is generated; Based on the task priority and resource matching table, a dynamic task scheduling algorithm is used to analyze and set the task execution ratio of the cloud and edge devices, adjust the task allocation, and generate a task execution priority status table with reference to the response time of the edge device and the processing capacity of the cloud.

7. The method for remote monitoring of industrial robots based on the Internet of Things according to claim 6, characterized in that: The formula of the dynamic task scheduling algorithm is as follows: Calculate the task execution priority score P i , where T ri represents the average response time of task i on the edge device, R ei represents the edge computing resources required for task i, c i represents the processing capability index of task i in the cloud, L i represents the network delay of task i, w1 is the critical weight of task response time, w2 is the critical weight of edge computing resources, w3 is the critical weight of cloud processing capacity, and w4 is the network delay adjustment coefficient.

8. The method for remote monitoring of industrial robots based on the Internet of Things according to claim 1, characterized in that: Based on the task execution priority status table, the steps of parsing the task path status and comparing the completion time and feedback, locating abnormal nodes and status deviations, and generating the industrial robot operation status monitoring evaluation results are as follows: Based on the task execution priority status table, the path status data of the industrial robot is collected, the task execution start and end time of each node is recorded, abnormal nodes and time delays are marked, and a path status and time record table is generated; Based on the path status and time record table, compare the task completion time, parse the time data, calculate the time difference between the task completion and the expected time, mark the abnormal nodes, count the task delay ratio, and generate a time deviation list of abnormal nodes; Based on the abnormal node time deviation list, a machine learning algorithm is used to analyze the deviation in link transmission, locate abnormal nodes, extract state change records, compare with normal node data, analyze abnormal parameter differences, and generate industrial robot operation status monitoring and evaluation results.

9. The industrial robot remote monitoring system based on the Internet of Things according to claim 8, characterized in that: The formula for the machine learning algorithm is as follows: Calculate the anomaly score S, where t represents the time deviation of a single node and t avg represents the average time deviation of normal nodes, n represents the total number of nodes, and d max represents the maximum deviation value, and α, β, and γ are adjustment coefficients respectively.

10. An industrial robot remote monitoring system based on the Internet of Things, characterized in that: According to any one of claims 1 to 9, the remote monitoring method for an industrial robot based on the Internet of Things comprises: The node load status determination module extracts the real-time load information and link delay of each node through the operation data of the industrial robot node, calculates and corrects the delay deviation using the real-time performance data, and generates a node load and delay table; The link priority analysis module evaluates and compares the communication performance and task load between nodes according to the node load and delay table, adjusts the priority of each link through a weighted algorithm, and establishes a node association topology map; The task allocation matrix module analyzes the task processing capability of each node and the transmission efficiency of the link according to the node association topology diagram, prioritizes the node execution tasks, and generates a task priority matrix; The path planning optimization module analyzes the task flow and real-time link status according to the task priority matrix, recalculates the optimal task execution path, adjusts and optimizes the path, and creates an optimized transmission path solution; The execution node optimization module extracts key execution nodes from the optimized transmission path scheme, analyzes the load balance and execution capability between nodes, reconfigures the task load, and forms an execution node priority table; The status monitoring and evaluation module uses the execution node priority table to monitor the task execution path and real-time status of the industrial robot, detect and diagnose anomalies in task execution in real time, and generate industrial robot operation status monitoring evaluation results.

Images