Artificial inspection dynamic planning method, device and medium based on Internet of Things platform

Abstract

The present application discloses an artificial inspection dynamic planning method, device and medium based on an Internet of Things platform, relating to the technical field of the Internet of Things. The method includes: calibrating device quantization parameters for device deployment data to obtain device importance degree scores; monitoring multi-dimensional device anomaly thresholds for device real-time operation parameters to determine the number of device anomalies; obtaining objects to be inspected through device screening score analysis based on the device importance degree scores and the number of device anomalies; dynamically adjusting inspection sections for the objects to be inspected to determine an optimal path node sequence; obtaining the current inspection plan through planning constraint analysis of inspection scheduling according to the optimal path node sequence; and performing inspection plan analysis on the inspection parameters obtained by real-time monitoring to determine inspection plan update parameters. The present application solves the technical problem of insufficient adaptability between the formulation of artificial inspection plans in industrial environments and the current industrial environments through the above method.

Description

Technical Field

[0001] This application relates to the field of Internet of Things technology, and in particular, to a dynamic planning method, device, and medium for manual inspection based on an Internet of Things platform. Background Art

[0002] Driven by industrial Internet of Things (IIoT) technology, traditional manual inspections are gradually transforming from paper records to digital management. The existing technical system is mainly based on two core frameworks: one is the fixed-cycle inspection mode relying on enterprise-level management systems, which generates a standardized task list by presetting work orders and scheduling plans and combining equipment maintenance manuals; the other is to adopt a static path planning strategy, relying on manual experience or an offline rule library to generate inspection routes, and using technologies such as two-dimensional codes and RFID to achieve personnel signing and task verification.

[0003] In the prior art, for the manual inspection planning in an industrial environment, it usually cannot meet the personnel scheduling and real-time monitoring in a complex environment. In an industrial environment with multiple devices and multiple paths, the prior art also cannot meet the real-time object screening of the devices to be inspected and the formulation of dynamic inspection plans, and it is difficult to meet the effective analysis of manual inspection planning in a complex industrial environment with multiple devices. While the efficiency of formulating inspection plans is low, the single data analysis will also lead to insufficient adaptability of the formulated inspection plans to the current industrial environment. Summary of the Invention

[0004] The embodiments of this application provide a dynamic planning method, device, and medium for manual inspection based on an Internet of Things platform, which solve the technical problem of insufficient adaptability of manual inspection planning in an industrial environment to the current industrial environment.

[0005] In a first aspect, the embodiments of this application provide a dynamic planning method for manual inspection based on an Internet of Things platform, which is characterized in that the method includes: obtaining device deployment data of an industrial environment, and calibrating device quantization parameters for the device deployment data to obtain device importance degree scores; obtaining device real-time operation parameters, and monitoring multi-dimensional device abnormality thresholds for the device real-time operation parameters to determine the number of device abnormalities; based on the device importance degree scores and the number of device abnormalities, obtaining objects to be inspected through device screening score analysis; dynamically adjusting inspection sections for the objects to be inspected to determine an optimal path node sequence; according to the optimal path node sequence, obtaining the current inspection plan through planning constraint analysis of inspection scheduling; and performing real-time monitoring on the current inspection plan, and performing inspection plan analysis on the inspection parameters obtained by the real-time monitoring to determine inspection plan update parameters.

[0006] In an implementation manner of the present application, device quantization parameter calibration is performed on device deployment data to obtain the device importance degree score, which specifically includes: based on a preset importance level list, device importance degree configuration is performed on the device deployment data to obtain device importance data; quantization parameter calibration is performed on the device importance data to obtain the device importance degree score.

[0007] In an implementation manner of the present application, multi-dimensional device anomaly threshold monitoring is performed on device real-time operation parameters to determine the number of device anomalies, which specifically includes: dividing the abnormal state of the device real-time operation parameters to obtain the device abnormal state type; based on the device abnormal state type, the number of device anomalies is determined through abnormal state monitoring.

[0008] In an implementation manner of the present application, based on the device importance degree score and the number of device anomalies, the object to be inspected is obtained through device screening score analysis, which specifically includes: based on the device importance degree score and the number of device anomalies, the device screening score is obtained through the evaluation of the screening score of the device to be inspected; among them, the calculation formula for the evaluation of the screening score of the device to be inspected is:

[0009]

[0010] Among them, is the device screening score, is the device asset price, is the device importance degree score, is the number of device anomalies; the score group range of the device screening score is divided to obtain the set of arrays of devices to be inspected; among them, the score group range division includes: screening score sorting, sub-data group division, target data group division, and the division formula for the sub-data group division is:

[0011]

[0012] Among them, is the element after sorting the device screening scores, is 's adjacent element; based on the set of arrays of devices to be inspected, the object to be inspected is obtained through dynamic update of preference requirements.

[0013] In an implementation manner of the present application, dynamic adjustment of the inspection section of the object to be inspected is performed to determine the optimal path node sequence, which specifically includes: based on the object to be inspected, device inspection nodes are determined, and through the Dijkstra algorithm, inspection priority path analysis is performed on the device inspection nodes to determine the inspection priority queue; edge weight analysis of the device inspection nodes is performed on the inspection priority queue to obtain the edge weights of the node paths; among them, the calculation formulas for point g and point h in the device inspection nodes are:

[0014]

[0015] Among them, is the path passing time, is the road condition index, and is the path selection preference index of the inspection personnel; according to the edge weights of the node paths, the inspection path vector is determined, and the graph structure of the inspection path vector is constructed to obtain the inspection path structure data; the inspection path structure data is stored as a graph structure dictionary, and the predecessor node processing is performed on the graph structure dictionary to determine the optimal path node sequence.

[0016] In an implementation manner of the present application, according to the optimal path node sequence, through the analysis of the planning constraints of the inspection schedule, the current inspection plan is obtained, specifically including: according to the optimal path node sequence, the key parameters of the inspection group are determined, and based on the key parameters of the inspection group, through the optimization of the total inspection time, the shortest total inspection time is determined; among them, the objective function of the total inspection time optimization is:

[0017]

[0018] Among them, is the inspection group from the inspection point to the inspection point time, is the inspection point decision variable, and the value range of the decision variable is , the th inspection group needs the number of inspection points, and m is the number of types of problems to be inspected divided in the corresponding industrial scenario; based on the shortest total inspection time, through the inspection schedule time range constraint, the current inspection plan is obtained; among them, the constraint condition of the inspection schedule time range constraint is:

[0019]

[0020] Among them, , is the th inspection group arriving at the inspection point , time, is the inspection group at the inspection point inspection duration.

[0021] In an implementation manner of the present application, patrol planning analysis is performed on the patrol parameters obtained by real-time monitoring to determine the patrol planning update parameters, which specifically includes: calculating the patrol passing time of the patrol parameters to obtain the equipment patrol time of the patrol personnel; matching the processing time of the equipment exception handling type for the equipment patrol time of the patrol personnel to determine the exception handling time of the equipment exception type; and determining the patrol planning update parameters through periodic exception handling planning parameter update based on the exception handling time of the equipment exception type.

[0022] In an implementation manner of the present application, after performing patrol planning analysis on the patrol parameters obtained by real-time monitoring to determine the patrol planning update parameters, the method further includes: updating the patrol planning update parameters to the current patrol plan, and performing real-time patrol monitoring on the updated current patrol plan to obtain the patrol time period; obtaining the subsequent patrol path planning cycle update parameters through periodic patrol service time analysis based on the patrol time period; acquiring the patrol check-in data, and performing patrol completion status analysis on the patrol check-in data to obtain the patrol completion rate; and determining the dynamic adjustment parameters of the personnel patrol work through patrol completion status threshold determination according to the patrol completion rate.

[0023] In a second aspect, an artificial patrol dynamic planning device based on an Internet of Things platform is further provided in an embodiment of the present application, where the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to: obtain the equipment deployment data of the industrial environment, and perform equipment quantization parameter calibration on the equipment deployment data to obtain the equipment importance degree score; obtain the equipment real-time operation parameters, and perform multi-dimensional equipment exception threshold monitoring on the equipment real-time operation parameters to determine the equipment exception times; obtain the objects to be patrolled through equipment screening score analysis based on the equipment importance degree score and the equipment exception times; perform dynamic adjustment of the patrol sections on the objects to be patrolled to determine the optimal path node sequence; obtain the current patrol plan through planning constraint analysis of the patrol schedule according to the optimal path node sequence; perform real-time monitoring on the current patrol plan, and perform patrol planning analysis on the patrol parameters obtained by the real-time monitoring to determine the patrol planning update parameters.

[0024] In a third aspect, an embodiment of the present application further provides a non-volatile computer storage medium for dynamic planning of manual inspections based on an Internet of Things platform, storing computer-executable instructions, characterized in that the computer-executable instructions are set to: obtain device deployment data of an industrial environment, and calibrate device quantization parameters for the device deployment data to obtain a device importance degree score; obtain real-time device operation parameters, and perform multi-dimensional device anomaly threshold monitoring on the real-time device operation parameters to determine the number of device anomalies; based on the device importance degree score and the number of device anomalies, obtain inspection objects to be inspected through device screening score analysis; perform dynamic adjustment of inspection sections for the inspection objects to be inspected to determine an optimal path node sequence; according to the optimal path node sequence, obtain the current inspection plan through planning constraint analysis of inspection scheduling; perform real-time monitoring on the current inspection plan, and perform inspection plan analysis on the inspection parameters obtained by the real-time monitoring to determine inspection plan update parameters.

[0025] An embodiment of the present application provides a method, device and medium for dynamic planning of manual inspections based on an Internet of Things platform. Through inspection object management, inspection path analysis, inspection planning, inspection execution judgment and personnel management, it solves the technical problem of insufficient adaptability of manual inspection planning formulation in an industrial environment to the current industrial environment, realizes the dynamic formulation of inspection plans, improves the self-adaptability of manual inspection plans in a complex industrial environment, and optimizes the travel time of manual inspection paths in a complex industrial environment. Description of the Drawings

[0026] The drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation to the present application. In the drawings:

[0027] Figure 1 It is a flowchart of a method for dynamic planning of manual inspections based on an Internet of Things platform provided by an embodiment of the present application;

[0028] Figure 2 It is a structural diagram of a system for dynamic planning of manual inspections based on an Internet of Things platform provided by an embodiment of the present application;

[0029] Figure 3 It is a schematic internal structure diagram of a device for dynamic planning of manual inspections based on an Internet of Things platform provided by an embodiment of the present application. Detailed Embodiments

[0030] To make the objectives, technical solutions, and advantages of this application more clear, the following will clearly and completely describe the technical solutions of this application in combination with specific embodiments of this application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.

[0031] The embodiments of this application provide a method, device, and medium for dynamic planning of manual inspections based on an Internet of Things platform. Through inspection object management, inspection path analysis, inspection planning, inspection execution judgment, and personnel management, the technical problem of insufficient adaptability of manual inspection planning in an industrial environment to the current industrial environment is solved, the dynamic formulation of inspection planning is realized, the self-adaptability of manual inspection planning in a complex industrial environment is improved, and the travel time of manual inspection paths in a complex industrial environment is optimized.

[0032] The following will detail the technical solutions proposed in the embodiments of this application through the drawings.

[0033] Figure 1 It is a flowchart of a method for dynamic planning of manual inspections based on an Internet of Things platform provided by the embodiments of this application. As Figure 1 shown, a method for dynamic planning of manual inspections based on an Internet of Things platform provided by the embodiments of this application specifically includes the following steps:

[0034] Step 101: Obtain the device deployment data of the industrial environment, and perform calibration of device quantization parameters on the device deployment data to obtain the device importance degree score.

[0035] Exemplarily, calibration of device quantization parameters is performed on the device deployment data to obtain the device importance degree score, and the importance degree of device inspections is intuitively represented by the quantization score.

[0036] Specifically, based on a preset importance level list, configure the device importance degree for the device deployment data to obtain device importance data; perform calibration of quantization parameters on the device importance data to obtain the device importance degree score.

[0037] In one embodiment, first, collect the operation data of various devices in the industrial environment and the data returned by the handheld mobile terminal in real time, including but not limited to physical parameters such as the temperature, pressure, rotation speed, and vibration frequency of the device, information on the operation status of the device (such as power on, power off, failure, etc.), inspection results, personnel positions, etc., to determine the device deployment data.

[0038] Through a variety of stable and reliable communication protocols (such as MQTT, HTTP, etc.), the collected data is transmitted to the Internet of Things platform by wired or wireless communication methods. The Internet of Things platform stores this data into the data storage module for subsequent calls.

[0039] When creating information of a certain device in this sub-module, the operator needs to open the device registration page and input basic information such as device name, model, number, manufacturer, purchase date, etc. Select the importance level of the device from the preset importance level list (such as high, medium, low) to indicate the criticality of the device in industrial production. The importance level list is set in advance according to the actual needs of users.

[0040] Each importance corresponds to a quantified score. For example, the importance levels are artificially divided into [extremely high importance, high importance, medium importance, low importance, extremely low importance], and the corresponding quantified scores are [5, 4, 3, 2, 1] respectively, to indicate the criticality of the device in industrial production.

[0041] IBM Maximo is an asset management software. For different importance levels in the importance level column, they are set by users through modular functions. The setting content includes how many importance levels there are, what the name of each level is, and what color is used to represent it, etc.

[0042] Through the device asset importance configuration of IBM Maximo, the importance levels of device parameters are divided into [extremely high importance, high importance, medium importance, low importance, extremely low importance] to determine the importance sequence of device parameters and characterize the importance ranking of the operating parameters of the same device.

[0043] Furthermore, the location information can be determined by manual input or map click positioning. Manual input means manually filling in the longitude and latitude coordinates of the device location in the location information input box. Professional map services (such as Baidu Map, Amap API) are also integrated in this sub-module. You can click the "Map Location" button to pop up the map interface. By clicking on the actual location of the device on the map, the system automatically obtains and fills in the longitude and latitude information. Fill in asset-related data such as the purchase price and depreciation information of the device to provide a basis for asset accounting. After confirming that all information is filled in correctly, click the "Submit" button, and the system will save the device asset information to the database.

[0044] Finally, for the S registered devices, where the s-th device, s ∈ S, the corresponding asset price is set to According to the preset quantization relationship of the importance level, the corresponding importance level score is set to .

[0045] Step 102: Obtain the real-time operation parameters of the device, and perform multi-dimensional device anomaly threshold monitoring on the real-time operation parameters of the device to determine the number of device anomalies.

[0046] Exemplarily, in order to obtain the specific data of the device anomaly situation, multi-dimensional device anomaly threshold monitoring is performed on the real-time operation parameters of the device to determine the specific number of anomalies of industrial devices, so as to meet the acquisition of the background data for the anomaly analysis of each device in a complex industrial environment.

[0047] In an implementation manner of the present application, performing multi-dimensional device anomaly threshold monitoring on the real-time operation parameters of the device to determine the number of device anomalies specifically includes: dividing the anomaly states of the real-time operation parameters of the device to obtain the types of device anomaly states; based on the types of device anomaly states, determining the number of device anomalies through anomaly state monitoring.

[0048] In one embodiment, device information is read from the database of the device asset management sub-module, and a device information list is automatically generated. In this device information list, measuring points of the data acquisition module can be configured for specific devices, and the anomaly types of the devices and their judgment rules are set.

[0049] Then, the data collected and uploaded during the operation of the device is received in real time, such as parameters such as temperature, pressure, flow rate, and vibration. According to the preset judgment rules, etc., the collected data is analyzed to determine whether the device has an anomaly and what kind of anomaly occurs. In addition to real-time monitoring means, in order to avoid anomalies in some devices not being detected, the anomaly events found during manual inspections will also be uploaded to the Internet of Things platform through a handheld mobile terminal and stored in the data storage module.

[0050] Next, m types of anomaly situations to be inspected in a specific industrial scenario are divided. For example, in the water service scenario, 3 types of device anomalies (m = 3) and their corresponding judgment rules are as follows:

[0051] First, water quality anomaly: indicators such as the acidity and alkalinity (pH value), turbidity, and microbial content in water exceed the normal range. The judgment rule is threshold judgment. For example, when the pH value is lower than 6.5 or higher than 8.5, the turbidity is greater than 5 NTU, and the microbial content exceeds the value specified by the national standard, it is determined that the water quality is abnormal. Second, mechanical failure: wear of the water pump impeller, loose valve seal, pipeline rupture, etc. The judgment rule is to monitor the vibration frequency and amplitude of the water pump through a vibration sensor. If it exceeds the normal threshold, it is judged that the water pump may have faults such as impeller wear; for the valve, the pressure difference before and after the valve is detected through a pressure sensor. If the pressure difference exceeds the normal range and the flow rate is abnormal, it is judged that the valve may have a loose seal; for the pipeline, it is comprehensively judged through a flow sensor and a pressure sensor. If the flow rate suddenly decreases and the pressure changes abnormally, there may be a pipeline rupture. Third, electrical failure: motor overload, short circuit, aging and poor contact of the power distribution cabinet lines, etc. Judgment rule: Monitor the motor current through a current sensor. If the current exceeds a certain proportion of the rated current, it is judged that the motor is overloaded; detect the insulation resistance of the power distribution cabinet lines through an insulation resistance tester. If the insulation resistance is lower than the specified value, it is judged that the lines may be aging or have poor contact.

[0052] Finally, set a counter for each device s and each type of anomaly k. Whenever an anomaly is detected, the corresponding counter is incremented by 1. For the k-th type of anomaly, there is , and the count of anomalies for the s-th device is .

[0053] Step 103: Based on the device importance degree score and the number of device anomalies, through device screening score analysis, obtain the objects to be inspected.

[0054] Exemplarily, after determining the device importance degree score and the number of device anomalies, for the actual device inspection requirements, it is necessary to distinguish the priorities of device inspections to determine which devices need to be arranged for inspection tasks in the current situation, so as to improve the rationality and timeliness of device inspection planning. Similarly, this application can also meet the iterative update of the objects to be inspected for the periodicity of inspection tasks to meet the timeliness of inspection planning in a complex industrial environment.

[0055] Specifically, based on the device importance degree score and the number of device anomalies, through device screening score analysis, obtain the objects to be inspected, including: Based on the device importance degree score and the number of device anomalies, through the evaluation of the screening score of the device to be inspected, obtain the device screening score; among them, the calculation formula for the evaluation of the screening score of the device to be inspected is:

[0056]

[0057] Among them, is the device screening score, is the price of the equipment asset, is the score of the equipment importance level, is the number of equipment anomalies; the screening scores of the equipment are divided into score group ranges to obtain the array set of equipment to be inspected; among them, the score group range division includes: screening score sorting, sub-data group division, and target data group division. The division formula for the sub-data group division is:

[0058]

[0059] wherein, is the element after sorting the equipment screening scores, is the adjacent element of; based on the array set of equipment to be inspected, through dynamic update of preference requirements, the object to be inspected is obtained.

[0060] In one embodiment, the collected equipment asset price , the importance level score , and the anomaly count are read from the data storage module. Let the screening score . Through equipment screening score analysis, the object to be inspected is obtained. The evaluation of the screening score of the inspection equipment is explained by the following formula:

[0061] (1)

[0062] wherein, is the equipment screening score, is the equipment asset price, is the equipment importance level score, is the number of equipment anomalies.

[0063] The setting principle of this screening score is to list the equipment with high asset value, high importance, and more likely to occur a certain type of accident as the object of priority inspection.

[0064] It should be noted that the setting principle of the screening score is to list the equipment with high asset value, high importance, and more likely to occur a certain type of accident as the object of priority inspection. The setting of the equipment value and the number of abnormal occurrences is also determined according to the actual situation.

[0065] Furthermore, in the actual industrial environment, the equipment with low value and prone to anomalies has a relatively low importance index and is called low-quality and vulnerable parts. It will not be used in key production links and production lines, and the inspection significance is not great. Instead, a preventive maintenance strategy is often adopted, with regular replacement and scrapping.

[0066] The equipment that needs inspection and maintenance is often relatively important and key equipment with high equipment value. Taking the water pump commonly used in the water treatment field as an example, a 1900kw water pump usually has an economic value of more than 200,000 yuan. Assuming that it has failed 4 times in a short period, the product of the two is also 800,000 yuan * times. Similarly, for equipment of the same importance as this pump, even if it fails once, the value is generally greater than 800,000 yuan. However, the proportion of the above-mentioned equipment is very small, and it often does not fail because it is very critical equipment.

[0067] In this case, the equipment with low price and many failure times will still be preferentially selected for inspection, avoiding the situation in the prior art that for equipment with low price and many abnormal times, due to the low price, the screening score is low, and it is not inspected, so the abnormal failure problems cannot be truly eliminated. The screening scores of S pieces of equipment constitute an equipment screening score array , and according to the values in it, it is rearranged from large to small into an array , where , .

[0068] According to the principle of data mutation, the data in the array is divided and sub-arrays are formed. The specific process is as follows:

[0069] Calculate the difference between two adjacent elements and in the array B, which is explained by the following formula:

[0070] , (2)

[0071] where is the element after sorting the equipment screening scores, is 's adjacent element

[0072] There are S - 1 such differences in total, which together form a new set .

[0073] According to how many categories the user wants to divide the objects to be inspected into, assuming it is W categories, then select the largest W - 1 numbers in the set b, record the subscripts of these numbers, and in descending order of the subscripts are , , , . According to these subscripts, it can be divided into new W groups of arrays, which are , , , According to the user's requirements, select one or several of these sets as the objects for inspection, and find the devices corresponding to the elements in these sets, thus completing the selection of the devices to be inspected.

[0074] Finally, divide the objects to be inspected into [worth inspecting, inspectable, not worth inspecting], a total of three categories, i.e., W = 3. In the case of a total of 30 devices, i.e., S = 30, select the two elements with the largest numbers in set b, denoted as and , i.e., , .

[0075] Therefore, according to and 's subscripts 15 and 20, divide the B array into three arrays , , .

[0076] The above three arrays correspond to "worth inspecting", "inspectable", and "not worth inspecting" in sequence. Assume that the user only wants to select the "worth inspecting" array as the object to be inspected, and find the corresponding 15 devices in array , and they are the selected objects to be inspected.

[0077] Divide this one-dimensional array according to mutation, select the two with the largest difference, and divide the data into three groups, which are set as very important, generally important, and relatively unimportant respectively.

[0078] It should be noted that the provided objects to be inspected are not fixed, and it can change according to the user's preferences and the update of data such as the device list and inspection task completion rate in the Internet of Things platform. The principle of data mutation refers to that in a data sequence, the data value suddenly and significantly changes at a certain time point or in a certain area, and this change does not conform to the original trend or pattern of the data. The objects to be inspected are divided into [worth inspecting, inspectable, not worth inspecting], a total of three categories, i.e., W = 3, but the value of W can be adjusted according to the actual situation, and it can be divided into 2 categories, 4 categories or even more, which can be set by the user based on actual needs.

[0079] For the two elements b15 and b20 with the largest numbers, they represent the difference b15 between the 16th value and the 15th value, and the difference between the 20th and 21st elements is too large. There is a mutation phenomenon between these two pairs of data. According to the principle of data mutation, it is considered that the first 15 data are a data group with stable trend, the data between the 15th and the 20th are the second data group with stable trend, and the data between the 21st and the 30th are the third data group with stable trend. Therefore, according to the size of the device screening score, it is judged that they correspond to the three categories of worth inspecting, inspectable, and not worth inspecting divided by the user in W in sequence.

[0080] Similarly, the inspection suggestions of being worthy of inspection, being inspectable, and not being worthy of inspection are only relative inspection suggestions, which can be adjusted and renamed according to the value of W. At the same time, the analysis of the objects to be inspected based on data mutation has a more sensitive detection ability for abnormally changing data and is more applicable to complex industrial scenarios.

[0081] Finally, the dataset of inspection points to be inspected for each inspection group can be obtained. Let be the number of inspection points required for the k-th inspection group (excluding the starting point).

[0082] Step 104: Dynamically adjust the inspection sections of the objects to be inspected to determine the optimal path node sequence.

[0083] Exemplarily, since the number of inspection objects is large and the manual inspection paths for each inspection object (equipment) are not unique, it is necessary to analyze the optimal path for manual inspection so that the inspection personnel can reach the inspection object as quickly as possible in the current industrial system environment during the inspection task and perform the inspection task.

[0084] Specifically, dynamically adjusting the inspection sections of the objects to be inspected to determine the optimal path node sequence includes: based on the objects to be inspected, determining the equipment inspection nodes, and through the Dijkstra algorithm, analyzing the inspection priority paths for the equipment inspection nodes to determine the inspection priority queue; performing edge weight analysis on the equipment inspection nodes in the inspection priority queue to obtain the edge weights of the node paths. The calculation formula for the edge weight analysis from equipment inspection node g to node h is:

[0085]

[0086] where is the path passing time, is the road condition index, is the path selection preference index of the inspection personnel. According to the edge weights of the node paths, determining the inspection path vector and constructing a graph structure for the inspection path vector to obtain the inspection path structure data; storing the inspection path structure data as a graph structure dictionary and performing predecessor node processing on the graph structure dictionary to determine the optimal path node sequence.

[0087] In one embodiment, first, a high-definition image is acquired. A drone is used to conduct an aerial survey of the industrial scene at an appropriate height and angle to ensure an image with high resolution, clarity, integrity, and good lighting conditions is obtained; in the case of using remote sensing images, a regional image of the industrial environment that meets the accuracy requirements needs to be acquired. Image editing or processing software (such as Photoshop, ENVI, etc.) is used to preprocess the acquired image, including operations such as removing noise, adjusting brightness and contrast, and performing geometric correction, etc., to improve the image quality and provide a good basis for subsequent analysis. A professional Geographic Information System (GIS) software (such as ArcGIS) is used to import the processed image. Through the vectorization tool of the GIS software, path features in the image can be extracted by methods based on spectrum, texture, and shape, and the path is converted into vector data.

[0088] For the vectorization process of the path, through binary processing, the extracted road pixels are converted into a specific value (such as 1), and non-road pixels are another value (such as 0) to simplify subsequent processing; for the thinned processing of the binary image, the morphological thinning algorithm (such as the Zhang-Suen algorithm) is used to reduce the width of the road pixels to one pixel width while maintaining the connectivity of the road, obtaining the road centerline; finally, through vectorization software or algorithms, a tracking-based method is adopted to trace the centerline pixel by pixel starting from the road endpoints, record the node coordinates, and convert them into a vector format (such as Shapefile or GeoJSON) to accurately represent information such as the position, length, and shape of the road. Finally, using the path vector data generated in the GIS software, considering factors such as the actual needs of the factory, the logistics and personnel flow situations, and whether it is passable, the Dijkstra algorithm is used to calculate the shortest path between any two inspection locations to be inspected and store it as a path dataset.

[0089] The dynamic adjustment of the inspection section requires the dynamic allocation of edge weights through the Dijkstra algorithm. The Dijkstra algorithm is used to calculate the shortest path from a source node to all other nodes in a weighted graph. This algorithm is used to calculate the path with the minimum cost from a source node to all other nodes in a weighted graph. There are N points on the weighted graph, and the connections represent that there is a path between two points. The number on each edge represents the cost required to go from one point to another, including the length of the distance traveled and the time required for the inspection personnel to pass through, etc.

[0090] It should be noted that there are significant differences between nodes and between inspection points. Between one inspection point and another inspection point, there may be multiple nodes.

[0091] The goal of establishing the graph structure is to obtain the shortest path in terms of travel time between two points to be inspected, fully considering the actual needs of the factory and the logistics and personnel flow situations. The actual factors such as whether passage is allowed, the degree of logistics busyness, and the density of personnel flow are quantified as the weights of the edges between adjacent nodes on the path, as follows:

[0092] Perform edge weight analysis on the equipment inspection nodes in the inspection priority queue to obtain the edge weights of the node paths; among them, the edge weight analysis from equipment inspection node g to node h is explained by the following formula:

[0093] (3)

[0094] Among them, is the path passing time, is the road condition index, is the path selection preference index of the inspection personnel.

[0095] Furthermore, the path selection preference index of the inspection personnel is obtained based on the inspection records in the Internet of Things platform. By counting the number of times the inspection personnel have walked each edge over a period of time and performing standardized processing on these data, dividing the number of times of walking a certain edge by the total number of all edges between two nodes, the probability of them passing each edge can be calculated, and this probability is the path selection preference index of the inspection personnel . Initially, because there is no accumulation of historical inspection data, the path selection preference indices of all edges between two nodes are set to be equal.

[0096] Furthermore, the road condition index is affected by whether passage is allowed and the influence of logistics and personnel flow, and its value range is . For a convenient and unobstructed path, can be set to 1; for a path where logistics is busy and passage is slow, increase this , for example, set it to 5; for a path where passage is not allowed, set its to infinity or a very large number, such as 10000.

[0097] Furthermore, the passing time is the time for the inspection personnel to pass between two nodes. Initially, assume that the traveling speed of each path on the map is the same, and the weight of each feasible path is the value obtained by dividing the path length by the walking speed, that is, the total traveling duration . However, with the accumulation of data in the Internet of Things platform, the traveling time of each actual path is updated based on historical data.

[0098] It should be noted that the design of the preference index fully considers the real environment. In a park, there are different differences in the road conditions and passable times of different routes. Some roads will pose certain safety hazards to passing pedestrians due to construction, and the inspection personnel will also choose to bypass the dangerous areas.

[0099] By using the number of times a path is passed to reflect the preference index and incorporating the considerations of the inspection personnel for the real conditions, some adverse factors for the inspection personnel on the inspection path are largely avoided. Truly considering the needs and safety of the inspection personnel, the number of passes within a certain time range can represent the action preferences of the inspection personnel during that period. The preference index is not solely affected by the subjective feelings of the inspection personnel, but also by environmental changes such as construction and road condition changes, thus avoiding the singularity of subjective judgment.

[0100] Furthermore, during the implementation of the Dijkstra algorithm, in addition to recording the distance of the shortest path, the node sequence of the path also needs to be recorded. By introducing a `previous` dictionary in the algorithm, it is used to record the predecessor node of each node in the shortest path. After the algorithm ends, starting from the target node, the complete shortest path node sequence is gradually generated by backtracking the predecessor nodes. The calculated shortest path information is converted into a geospatial data format for storage.

[0101] Any two locations to be inspected in the industrial environment , The time-optimal feasible path between them can be abstracted as an edge, and the set of these edges is . It should be noted that with the accumulation of data by the Internet of Things platform and the real-time update of the weights between nodes, that is to say, the shortest path dataset A is in a state of continuous update.

[0102] Step 105: According to the optimal path node sequence, through the analysis of the planning constraints of the inspection schedule, obtain the current inspection plan.

[0103] Exemplarily, through the analysis of the planning constraints of the inspection schedule in this application, the path planning of each inspection group within the planned time range, the arrangement of the inspection order for each location to be inspected, and the inspection group allocation plan for each location to be inspected are solved.

[0104] It should be noted that if the planning constraint analysis stipulates the planned path for one day, the relevant parameters (such as the scheduling plan, the set of locations to be inspected, etc.) are obtained or calculated with one day as the time range.

[0105] Specifically, according to the optimal path node sequence, through the analysis of the planning constraints of the inspection schedule, the current inspection plan is obtained, which specifically includes: determining the key parameters of the inspection group according to the optimal path node sequence, and based on the key parameters of the inspection group, determining the shortest total inspection time through the optimization of the total inspection time; among them, the objective function of the total inspection time optimization is:

[0106]

[0107] Among them, is the inspection group from the inspection point to the inspection point time, is the decision variable of the inspection point, and the value range of the decision variable is , the number of inspection points required by the

[0108]

[0109] Among them, , is the time when the , arrives at the inspection point is the inspection group at the inspection point inspection duration at.

[0110] In one embodiment, by obtaining the key parameters of dynamic inspection, the following parameters are determined:

[0111] m is the total number of groupings of inspection personnel and classification of abnormal situations, ;

[0112] is the number of inspection points required by the k-th inspection group (excluding the starting point);

[0113] is the earliest start time of the inspection task at the inspection point i in the schedule of the inspection group k, generally the working time of the k-th group of personnel;

[0114] is the latest end time of the inspection point i in the schedule of the inspection group k, generally the off-duty time of the k-th group of personnel;

[0115] is the time for inspection team k to reach inspection point j from inspection point i, and this parameter will be updated by the inspection execution judgment module as the Internet of Things platform runs;

[0116] is the inspection duration of inspection team k at inspection point i. This parameter can be set to 0 when there is no relevant data. However, as the inspection records in the Internet of Things platform accumulate, it will be calculated and the parameter updated by the inspection execution judgment module.

[0117] First, based on the optimal path node sequence, determine the key parameters of the inspection team, and based on the key parameters of the inspection team, determine the shortest total inspection time through the optimization of the total inspection time; among them, the objective function for the optimization of the total inspection time is:

[0118] (4)

[0119] Among them, is the inspection team from inspection point to reach inspection point time, is the inspection point decision variable.

[0120] For the constraint conditions of the objective function, the first constraint condition is constrained by the value of the decision variable, and the value range of the decision variable is

[0121] Among them, the number of inspection points required by the th inspection team.

[0122] Then, the status of each inspection team is not dispatched or dispatched. If the status is dispatched, the inspection team must start from the origin and return to the origin. This origin is different from the inspection points to be inspected and refers to the daily work or standby location of the inspection team.

[0123] Therefore, based on the status of the inspection team, a second constraint condition can be obtained and is explained by the following formula:

[0124] (5)

[0125] Furthermore, the third constraint condition is used to ensure that each inspection point assigned to inspection team k should be visited and visited only once, and is explained by the following formula:

[0126] (6)

[0127] Further, the fourth constraint condition is used to ensure the magnitude relationship of the start service times of adjacent nodes to be inspected. Here, M is a large positive number, which is used in the Big-M method for linear programming to implement the conditional constraint. It is recommended that it be a positive integer not less than +1. Through the inspection scheduling time range constraint, the current inspection plan can be obtained. Among them, the constraint condition of the inspection scheduling time range constraint (the fourth constraint condition) is:

[0128] (7)

[0129] Among them, 、 are the times when the th inspection group arrives at the inspection point 、 . is the inspection duration of the inspection group at the inspection point .

[0130] Further, the fifth constraint condition is used to ensure that the time to reach an inspection point is within the range allowed by the schedule, which is explained by the following formula.

[0131] (8)

[0132] It should be noted that the vehicle routing problems with time windows (VRPTW) refer to a certain number of customers, each with different quantities of goods demands. The distribution center provides goods to the customers, and a fleet is responsible for distributing the goods and organizing appropriate driving routes. The goal is to meet the customers' demands and, under certain constraints, achieve purposes such as the shortest distance, the minimum cost, and the least time consumption.

[0133] The constraint conditions proposed in this application are a variant of the VRPTW model. The first constraint condition is constrained by the value of the decision variable.

[0134] The first constraint condition is constrained by the value of the decision variable;

[0135] The second and third constraint conditions are used to ensure that each inspection group k is assigned inspection points that should be visited and visited only once;

[0136] The fourth constraint condition is used to ensure the magnitude relationship of the start service times of adjacent nodes to be inspected;

[0137] The fifth constraint condition is used to ensure that the time to reach an inspection point is within the range allowed by the schedule;

[0138] Based on the above constraints, the inspection group allocation plan for each inspection point can be met, which avoids subjective analysis and obtains constraints through VRPTW model improvement, thereby improving the real-time and rationality of inspection planning.

[0139] Through the construction of the above-mentioned constraints and objective functions, this application realizes the path planning of each inspection group within the planned time range, the arrangement of the inspection order of each inspection point to be inspected, and the reasonable configuration of the inspection group allocation plan for each inspection point to be inspected, further improving the real-time and rationality of the inspection plan.

[0140] Step 106: Monitor the current inspection plan in real time, and perform inspection plan analysis on the inspection parameters obtained by real-time monitoring to determine the inspection plan update parameters.

[0141] For example, for personnel positioning, the location of personnel can be grasped in real time, so that managers can monitor in real time whether the inspection personnel are working according to the scheduled inspection plan and time nodes. By optimizing the inspection route, the actual walking route and stay time of the inspection personnel can be understood, and the existing inspection route can be evaluated whether it is reasonable. According to the analysis results, the inspection route can be optimized and adjusted to reduce unnecessary trips and improve inspection efficiency.

[0142] Specifically, an inspection planning analysis is performed on the inspection parameters obtained from real-time monitoring to determine the inspection planning update parameters, including: calculating the inspection time of the inspection parameters to obtain the equipment inspection time of the inspection personnel; matching the equipment inspection time of the inspection personnel with the processing time of the equipment exception processing type to determine the exception processing time of the equipment exception type; based on the exception processing time of the equipment exception type, the inspection planning update parameters are determined by updating the periodic exception processing planning parameters.

[0143] In one embodiment, each device to be inspected is installed with an RFID tag, which stores basic information of the device, including detailed information such as device model, number, and specifications, providing necessary device identification and basic data support for the inspection process.

[0144] First, when the inspectors arrive at each inspection point (device to be inspected), they can choose to scan the QR code posted near the device, or use the NFC function of the handheld mobile terminal and the RFID of the device to trigger and complete the point inspection and clocking in operation. At the moment of successful scanning, the mobile terminal obtains the basic information of the device and records the scanning time at the same time. This time point is used as the starting time for the inspection personnel to start inspecting the equipment.

[0145] Then, the data of inspection personnel punching in and equipment inspection completion is uploaded to the IoT platform, which calculates , the inspection time of the patrol personnel at this equipment can be accurately obtained, and the inspection times recorded within the parameter update period are averaged and denoted as the service time for type-k anomalies at point i. .

[0146] Furthermore, set the parameter update period to one month. For the fan to be patrolled in January, three patrols were conducted for motor anomaly types, and the inspection times were 1 hour, 2 hours, and 3 hours respectively. Then the average inspection time is 2 hours, which is denoted as the service time for anomaly type k at point i. The updated parameters will be used in the patrol route planning for February.

[0147] Furthermore, after analyzing the patrol parameters obtained from real-time monitoring to determine the patrol plan update parameters, the method further includes: updating the patrol plan update parameters to the current patrol plan, and performing real-time patrol monitoring on the updated current patrol plan to obtain the patrol time period; based on the patrol time period, analyzing through the periodic patrol service time to obtain the subsequent patrol route planning period update parameters; obtaining the patrol check-in data, and analyzing the patrol completion status of the patrol check-in data to obtain the patrol completion rate; according to the patrol completion rate, determining the dynamic adjustment parameters for personnel patrol work through the patrol completion status threshold judgment.

[0148] In one embodiment, the IoT platform statistically counts the number of patrolled points that each patrol group has completed checking in based on the received check-in information in real time. . Using the patrol completion rate formula, calculate the patrol completion rate of each patrol group. The patrol completion rate formula is explained by the following formula:

[0149] (9)

[0150] where is the number of patrolled points that the k-th patrol group needs to patrol (excluding the starting point). This patrol completion rate can not only let the management know the patrol situation, but also be used to adjust the arrangements during the patrol. When it is less than 1, according to the value of the patrol completion rate, the parameters can be dynamically adjusted as follows:

[0151] When , the preferences of users in the patrol object screening module should be considered first to reduce the objects to be patrolled, that is, to reduce .

[0152] When , the number of personnel in the patrol group should be considered first to increase. The upper limit here is taken as 0.9, but in actual situations, it can be solved in the following way. Assume that there are peo patrol personnel in the k-th patrol group, then the upper limit value is taken as 。

[0153] When , the working hours of workers should be appropriately increased in the patrol personnel management module, that is, adjust and . The new working hours and the old working hours should satisfy the following relationship: 。

[0154] Figure 2 FIG. 0 is a structural diagram of an artificial patrol dynamic planning system based on an Internet of Things platform provided by an embodiment of the present application.

[0155] In one embodiment, a structure of an artificial patrol dynamic planning system based on an Internet of Things platform is composed of the following modules:

[0156] Information collection module: The data collection module includes, in terms of hardware, all sensors and instruments deployed on instruments in an industrial environment, and the handheld mobile terminals of patrol personnel, etc. And it can realize the collection and preprocessing of data.

[0157] Internet of Things platform: The Internet of Things platform is the main body of the system, used to control the cooperation of each module. It has functions of device connection, management and monitoring, and includes a device asset management sub-module, a device status monitoring sub-module and a visual operation interface sub-module. It can also perform data exchange with the data collection module, the patrol object management module, the personnel management module, the personnel positioning module, the patrol execution judgment module and the stored information of data.

[0158] Device asset management sub-module: Facing the industrial environment, it aims to comprehensively and efficiently manage existing devices. By registering information such as the importance degree of devices, their locations (latitude and longitude positioning and supporting map click to determine), and asset prices, it realizes the digital management of device assets, and provides accurate device information query, analysis and decision-making support for industrial enterprises.

[0159] Device status monitoring sub-module: This module aims to monitor the real-time status of devices in an industrial environment. By reading the device information registered in the device asset management sub-module, it classifies devices according to abnormal types for subsequent allocation to different patrol groups. Combining the data uploaded by the data collection module, it uses a variety of judgment rules to achieve real-time monitoring, and at the same time counts the abnormal times of different devices, providing strong support for device management and maintenance.

[0160] Visual operation interface sub-module: It is the key entry for users to interact with the Internet of Things system, and it is the medium for users to operate the Internet of Things platform and cooperate with each module. It facilitates users to view information such as device status, data reports, and analysis results, and supports users to create personalized visual dashboards through simple operations such as dragging and configuration to meet the needs of different users. It greatly improves the usability and management efficiency of the Internet of Things system.

[0161] Inspection object management module: Reads the collected device asset prices , importance level scores , anomaly counts , and based on the calculated device screening scores, user preferences, and inspection completion rates, uses the inspection object screening algorithm to screen out the objects to be inspected.

[0162] Personnel management module: This module can assign employees to inspection groups for different types of inspection tasks according to their job responsibilities, and generate a reasonable shift schedule by considering factors such as employees' working hours and vacation plans.

[0163] Path management module: First, it can use professional geographic information system (GIS) software (such as ArcGIS) to create vector data of all paths in the industrial environment based on geographic information images. Then, based on this vector data, using the preference index , road condition index and passing time , based on the Dijkstra algorithm, generates the shortest path dataset between any two objects to be inspected and updates this dataset regularly.

[0164] Path dynamic planning module: Responsible for formulating inspection routes and selecting the most frequently used inspection routes

[0165] Inspection planning module: Built-in mathematical model, through the parameter values passed in from other modules, can finally solve the path planning of each inspection group within the planned time range, the arrangement of the inspection order for each point to be inspected, and the inspection group assignment plan for each point to be inspected.

[0166] Inspection execution judgment module: Responsible for personnel positioning and real-time monitoring of the completion status of inspection tasks

[0167] Personnel positioning module: According to the collected personnel positioning records, which should include node identifiers, personnel identifiers, timestamps, and location information at each time point, calculates and updates the preference index and passing time , stores the results in the data storage module and is called by the path management module. It can also calculate the inspection completion rate based on the inspection start and completion reports from inspection personnel. And according to the value of the inspection completion rate, the parameters , , and etc. for dynamic adjustment.

[0168] The above is the method embodiment proposed in this application. Based on the same inventive concept, the embodiments of this application also provide a dynamic planning device for manual inspection based on the Internet of Things platform, and its structure is as Figure 3 shown.

[0169] Figure 3 This is a schematic diagram of the internal structure of a dynamic planning device for manual inspection based on the Internet of Things platform provided by the embodiments of this application. As Figure 3 shown, the device includes:

[0170] At least one processor 301;

[0171] And a memory 302 communicatively connected to at least one processor;

[0172] Wherein, the memory 302 stores instructions executable by at least one processor, and the instructions are executed by at least one processor 301 so that at least one processor 301 can:

[0173] Obtain the device deployment data of the industrial environment, and calibrate the device quantization parameters for the device deployment data to obtain the device importance degree score; obtain the device real-time operation parameters, and monitor the multi-dimensional device anomaly thresholds for the device real-time operation parameters to determine the number of device anomalies; based on the device importance degree score and the number of device anomalies, obtain the objects to be inspected through device screening score analysis; perform dynamic adjustment of the inspection sections for the objects to be inspected to determine the optimal path node sequence; according to the optimal path node sequence, obtain the current inspection plan through the planning constraint analysis of the inspection schedule; perform real-time monitoring on the current inspection plan, and perform inspection plan analysis on the inspection parameters obtained from the real-time monitoring to determine the inspection plan update parameters.

[0174] Some embodiments of this application provide a Figure 1 corresponding non-volatile computer storage medium for dynamic planning of manual inspection based on the Internet of Things platform, storing computer-executable instructions, and the computer-executable instructions are set as:

[0175] Obtain the device deployment data of the industrial environment, and calibrate the device quantization parameters for the device deployment data to obtain the device importance degree score; obtain the device real-time operation parameters, and monitor the multi-dimensional device anomaly thresholds for the device real-time operation parameters to determine the number of device anomalies; based on the device importance degree score and the number of device anomalies, obtain the objects to be inspected through device screening score analysis; perform dynamic adjustment of the inspection sections for the objects to be inspected to determine the optimal path node sequence; according to the optimal path node sequence, obtain the current inspection plan through the analysis of the planning constraints of the inspection schedule; perform real-time monitoring on the current inspection plan, and perform inspection plan analysis on the inspection parameters obtained from the real-time monitoring to determine the inspection plan update parameters.

[0176] Each embodiment in this application is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the Internet of Things devices and media, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0177] The systems and media provided by the embodiments of this application correspond one-to-one with the methods. Therefore, the systems and media also have beneficial technical effects similar to the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and media will not be elaborated here.

[0178] Those skilled in the art should understand that the embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0179] This application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0180] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device that implements the functions specified in one or more of the flow Figure 1 a process or processes and / or blocks Figure 1 specified in one or more of the blocks or blocks.

[0181] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the flow Figure 1 a process or processes and / or blocks Figure 1 specified in one or more of the blocks or blocks.

[0182] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0183] The memory may include non-permanent memory in the computer-readable medium, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0184] Computer-readable media includes both permanent and non-permanent, removable and non-removable media implemented by any method or technology for storing information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0185] It should also be noted that the term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0186] The above are only embodiments of the present application and are not used to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. An artificial inspection dynamic planning method based on an Internet of Things platform, characterized in that The method includes: Obtaining the device deployment data of the industrial environment, and calibrating the device quantization parameters of the device deployment data to obtain the device importance degree score; Obtaining the device real-time operation parameters, and monitoring the multi-dimensional device anomaly thresholds of the device real-time operation parameters to determine the number of device anomalies; Based on the device importance degree score and the number of device anomalies, obtaining the objects to be inspected through device screening score analysis; Dynamically adjusting the inspection sections of the objects to be inspected to determine the optimal path node sequence; According to the optimal path node sequence, obtaining the current inspection plan through the analysis of the planning constraints of the inspection schedule; Performing real-time monitoring on the current inspection plan, and performing inspection plan analysis on the inspection parameters obtained from the real-time monitoring to determine the inspection plan update parameters; Based on the device importance degree score and the number of device anomalies, obtaining the objects to be inspected through device screening score analysis, specifically including: Based on the device importance degree score and the number of device anomalies, obtaining the device screening score through the evaluation of the screening score of the devices to be inspected; wherein, the calculation formula for the evaluation of the screening score of the devices to be inspected is: Among them, is the device screening score, is the device asset price, is the device importance degree score, is the number of device anomalies; Dividing the range of the device screening score into score group ranges to obtain a set of arrays of devices to be inspected; wherein, the division of the score group ranges includes: sorting the screening scores, dividing the sub-data groups, and dividing the target data groups, and the division formula for the sub-data group division is: Among them, is the element after sorting the screening scores of the device, is the adjacent element of; Based on the set of arrays of devices to be inspected, obtaining the objects to be inspected through dynamic update of the preference requirements; According to the optimal path node sequence, obtaining the current inspection plan through the analysis of the planning constraints of the inspection schedule, specifically including: According to the optimal path node sequence, determining the key parameters of the inspection group, and based on the key parameters of the inspection group, determining the shortest total inspection time through the optimization of the total inspection time; wherein, the objective function for the optimization of the total inspection time is: Among them, is the inspection group from the inspection point to the inspection point time, is the inspection point decision variable, and the value range of the decision variable is , the th inspection group needs the number of inspection points, and m is the number of types of problems to be inspected divided in the corresponding industrial scenario; Based on the shortest total inspection time, obtaining the current inspection plan through the time range constraint of the inspection schedule; wherein, the constraint condition for the time range constraint of the inspection schedule is: Among them, and are the times when the th inspection group arrives at the inspection points and . is the inspection duration of the inspection group at the inspection point .

2. The artificial inspection dynamic planning method based on the Internet of Things platform according to claim 1, wherein, Calibrating the device quantization parameters of the device deployment data to obtain the device importance degree score, specifically including: Based on a preset importance level list, configuring the device importance degree of the device deployment data to obtain device importance data; Calibrating the quantization parameters of the device importance data to obtain the device importance degree score.

3. The artificial inspection dynamic planning method based on the Internet of Things platform according to claim 2, wherein Monitoring the multi-dimensional device anomaly thresholds of the device real-time operation parameters to determine the number of device anomalies, specifically including: Dividing the anomaly status of the device real-time operation parameters to obtain the device anomaly status type; Based on the device anomaly status type, determining the number of device anomalies through anomaly status monitoring.

4. The artificial inspection dynamic planning method based on the Internet of Things platform according to claim 1, wherein Dynamically adjusting the inspection sections of the objects to be inspected to determine the optimal path node sequence, specifically including: Based on the objects to be inspected, determining the device inspection nodes, and through the Dijkstra algorithm, performing inspection priority path analysis on the device inspection nodes to determine the inspection priority queue; Perform edge weight analysis on the inspection priority queue for the equipment inspection nodes to obtain the edge weights of the node paths; among them, the calculation formula for the edge weight analysis of the g point and the h point in the equipment inspection nodes is: Among them, is the path passing time, is the road condition index, is the path selection preference index of the inspection personnel; Determine the inspection path vector according to the edge weights of the node paths, and construct a graph structure for the inspection path vector to obtain inspection path structure data; Store the inspection path structure data as a graph structure dictionary, and perform predecessor node processing on the graph structure dictionary to determine the optimal path node sequence.

5. The artificial inspection dynamic planning method based on the Internet of Things platform according to claim 1, wherein Perform inspection plan analysis on the inspection parameters obtained by real-time monitoring to determine inspection plan update parameters, specifically including: Calculate the inspection passing time for the inspection parameters to obtain the equipment inspection time of the inspection personnel; Match the processing time of the equipment anomaly handling type for the equipment inspection time of the inspection personnel to determine the anomaly handling time of the equipment anomaly type; Based on the anomaly handling time of the equipment anomaly type, determine the inspection plan update parameters through periodic anomaly handling plan parameter update.

6. The artificial inspection dynamic planning method based on the Internet of Things platform according to claim 1, wherein, After performing inspection plan analysis on the inspection parameters obtained by real-time monitoring to determine inspection plan update parameters, the method further includes: Update the inspection plan update parameters to the current inspection plan, and perform real-time inspection monitoring on the updated current inspection plan to obtain the inspection time period; Based on the inspection time period, obtain subsequent inspection path planning period update parameters through periodic inspection service time analysis; Obtain inspection check-in data, and perform inspection completion status analysis on the inspection check-in data to obtain the inspection completion rate; Determine the dynamic adjustment parameters of the personnel inspection work according to the inspection completion rate through inspection completion status threshold judgment.

7. An artificial inspection dynamic planning device based on an Internet of Things platform, characterized in that, The device includes: At least one processor; And a memory communicatively connected to the at least one processor; Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: Obtain the device deployment data of the industrial environment, and perform calibration of device quantization parameters on the device deployment data to obtain the device importance degree score; Obtain the device real-time operation parameters, and perform multi-dimensional device anomaly threshold monitoring on the device real-time operation parameters to determine the number of device anomalies; Based on the device importance degree score and the number of device anomalies, obtain the objects to be inspected through device screening score analysis; Perform dynamic adjustment of the inspection section on the objects to be inspected to determine the optimal path node sequence; According to the optimal path node sequence, obtain the current inspection plan through planning constraint analysis of inspection scheduling; Perform real-time monitoring on the current inspection plan, and perform inspection plan analysis on the inspection parameters obtained by the real-time monitoring to determine inspection plan update parameters; Based on the device importance degree score and the number of device anomalies, obtain the objects to be inspected through device screening score analysis, specifically including: Based on the device importance degree score and the number of device anomalies, through the evaluation of the screening score of the devices to be inspected, the device screening score is obtained; wherein, the calculation formula for the evaluation of the screening score of the devices to be inspected is: Among them, is the device screening score, is the device asset price, is the device importance degree score, is the number of device anomalies; The device screening score is divided into score group ranges to obtain a set of arrays of devices to be inspected; wherein, the score group range division includes: screening score sorting, sub-data group division, target data group division, and the division formula for the sub-data group division is: Among them, is the element after the device screening score is sorted, is the adjacent element of; Based on the set of arrays of devices to be inspected, through dynamic update of preference requirements, the object to be inspected is obtained; According to the optimal path node sequence, through the analysis of the planning constraints of the inspection schedule, the current inspection plan is obtained, specifically including: According to the optimal path node sequence, the key parameters of the inspection group are determined, and based on the key parameters of the inspection group, through the optimization of the total inspection time, the shortest total inspection time is determined; wherein, the objective function for the optimization of the total inspection time is: Among them, is the inspection group from the inspection point to the inspection point time, is the decision variable of the inspection point, and the value range of the decision variable is , the number of inspection points required by the th inspection group, and m is the number of types of problems to be inspected divided in the corresponding industrial scenario; Based on the shortest total inspection time, through the constraint of the inspection schedule time range, the current inspection plan is obtained; wherein, the constraint condition for the inspection schedule time range constraint is: Among them, , is the time when the th inspection group arrives at the inspection point , , and is the inspection duration of the inspection group at the inspection point .

8. A non-volatile computer storage medium for dynamic planning of manual inspection based on an Internet of Things platform, storing computer-executable instructions, characterized in that, The computer-executable instructions are set as: Obtain the device deployment data of the industrial environment, and calibrate the device quantization parameters for the device deployment data to obtain the device importance degree score; Obtain the real-time operation parameters of the device, and monitor the multi-dimensional device anomaly thresholds for the real-time operation parameters of the device to determine the number of device anomalies; Based on the device importance degree score and the number of device anomalies, through the analysis of the device screening score, the object to be inspected is obtained; Dynamically adjust the inspection sections of the object to be inspected to determine the optimal path node sequence; According to the optimal path node sequence, through the analysis of the planning constraints of the inspection schedule, the current inspection plan is obtained; Monitor the current inspection plan in real time, and analyze the inspection plan for the inspection parameters obtained from the real-time monitoring to determine the inspection plan update parameters; Based on the device importance degree score and the number of device anomalies, through the analysis of the device screening score, the object to be inspected is obtained, specifically including: Based on the device importance degree score and the number of device anomalies, through the evaluation of the screening score of the devices to be inspected, the device screening score is obtained; wherein, the calculation formula for the evaluation of the screening score of the devices to be inspected is: Among them, is the device screening score, is the device asset price, is the device importance degree score, is the number of device anomalies; The device screening score is divided into score group ranges to obtain a set of arrays of devices to be inspected; wherein, the score group range division includes: screening score sorting, sub-data group division, target data group division, and the division formula for the sub-data group division is: Among them, is the element after the device screening score is sorted, is the adjacent element of; Based on the set of arrays of devices to be inspected, through dynamic update of preference requirements, the object to be inspected is obtained; According to the optimal path node sequence, through the analysis of the planning constraints of the inspection schedule, the current inspection plan is obtained, specifically including: According to the optimal path node sequence, the key parameters of the inspection group are determined, and based on the key parameters of the inspection group, through the optimization of the total inspection time, the shortest total inspection time is determined; wherein, the objective function for the optimization of the total inspection time is: Among them, is the inspection group from the inspection point to the inspection point time, is the inspection point decision variable, and the value range of the decision variable is , the number of inspection points required by the th inspection group, and m is the number of types of problems to be inspected divided in the corresponding industrial scenario; Based on the shortest total inspection time, the current inspection plan is obtained through the constraint of the inspection scheduling time range; wherein, the constraint condition of the inspection scheduling time range is as follows: Among them, and are the times when the th inspection team arrives at the inspection point and . is the inspection duration of the inspection team at the inspection point .

Images