An intelligent elevator inspection method

Abstract

The application relates to the field of elevator safety technology and discloses an intelligent elevator inspection method, which comprises the following steps: collecting operation data, state data and fault data of an elevator through a sensor to obtain an elevator data set; analyzing the data in the elevator data set to generate service data; obtaining a maintenance task of the elevator; forming a function library of an inspection robot according to the maintenance task; obtaining a health state analysis result of the elevator according to the service data and determining an inspection task of the elevator; issuing the inspection task of the elevator to the inspection robot; integrating task parameters and task conditions to obtain task data; analyzing the task data to predict the time required for completing each task; combining the elevator position and the action track of the inspection robot to plan the inspection route of each inspection robot, decompose the inspection route into individual tasks and issue the individual tasks to the inspection robot; and outputting an inspection report through the inspection robot. The application improves the efficiency, nature and effectiveness of elevator inspection and realizes on-demand inspection.

Description

Technical Field

[0001] This invention relates to the field of elevator safety technology, and specifically to an intelligent elevator inspection method. Background Technology

[0002] With the increasing number of high-rise buildings, elevators are being used more and more widely. The safe operation of elevators is of paramount importance. Traditional elevator inspections mainly rely on regular manual checks, which suffers from low efficiency, high risk of missed inspections, and difficulty in real-time monitoring. Therefore, a more intelligent and efficient elevator inspection method is needed to ensure the safe operation of elevators. Summary of the Invention

[0003] The purpose of this invention is to solve the above-mentioned problems by designing an intelligent elevator inspection method to address the issues existing in current elevator inspection methods.

[0004] This invention provides an intelligent elevator inspection method, which includes the following steps:

[0005] Elevator operation data, status data, and fault data are collected by sensors to obtain an elevator data set. Service data is generated by analyzing the data in the elevator data set, wherein the service data includes at least maintenance items and automatic detection results of elevator components.

[0006] The elevator maintenance tasks are obtained, a function library for the inspection robot is formed for the maintenance tasks, the health status analysis results of the elevator are obtained based on the service data, and the elevator inspection tasks are determined.

[0007] The elevator inspection task is assigned to the inspection robot, the task parameters and task status are integrated to obtain task data, the task data is analyzed, and the time required to complete each task is predicted.

[0008] By combining the elevator location and the movement trajectory of the inspection robots, the inspection routes of each inspection robot are planned, and the tasks are broken down into individual tasks and assigned to the inspection robots. After the inspection robots conduct their inspections, they output inspection reports.

[0009] Optionally, in the first implementation of the present invention, the step of collecting elevator operation data, status data, and fault data through sensors to obtain an elevator data set, and generating service data by analyzing the data in the elevator data set, includes:

[0010] The median filtering method is used to smooth the data in the elevator dataset, removing outliers and noise data. The smoothed data is then normalized to obtain a new elevator dataset.

[0011] Z-Score standardization was used to process the data in the new elevator dataset to obtain standardized data.

[0012] Calculate the covariance matrix of the standardized data to obtain the eigenvalues ​​and eigenvectors of the covariance matrix. Sort the eigenvectors according to the magnitude of the eigenvalues ​​and select the eigenvectors corresponding to the first n eigenvalues ​​to form the principal components.

[0013] Projecting the data from the new elevator dataset onto the principal components yields new data after dimensionality reduction. Based on the results of the principal component analysis, the focus of maintenance items is determined.

[0014] Clustering algorithms are used to classify the data in the new elevator dataset after dimensionality reduction, identify different types of elevator operating status and fault modes, and generate automatic detection results for elevator components based on the clustering results.

[0015] Based on the combined results of principal component analysis and cluster analysis, elevator maintenance items are identified, the importance and urgency of each maintenance item are assessed, priorities are determined, maintenance items are sorted according to priority, and service data is generated.

[0016] Optionally, in the second implementation of the present invention, the step of using a clustering algorithm to classify the data in the new elevator dataset after dimensionality reduction, identifying different types of elevator operating states and fault modes, and generating automatic detection results for elevator components based on the clustering results includes:

[0017] The first cluster center is randomly selected from the new elevator data set after dimensionality reduction, and the shortest distance between the first cluster center and the data in the new elevator data set is obtained through iterative calculation to establish a distance set;

[0018] Normalize the distance set, calculate the sum of all distances in the distance set, add Laplace noise, and obtain the shortest distance sum after adding noise;

[0019] In the new elevator dataset after dimensionality reduction, seed cluster centers are selected within the normalized range of the data, and the shortest distance between the data and the cluster centers is calculated iteratively until the selection of cluster centers is completed.

[0020] Based on the obtained shortest distance set and the selection of cluster centers, the data corresponding to the shortest distance set are divided into cluster centers to obtain data categories;

[0021] The mean of each data category is calculated, noise is added to the calculation result, and the result is used as a new data center point. This process is repeated until the cluster center no longer changes or the preset number of iterations is reached, thus completing the data clustering.

[0022] Based on the clustering results, identify different types of elevator operating states and fault modes, including at least normal operation, minor fault, and serious fault.

[0023] For each cluster, the elevator data in that cluster is analyzed to determine the operating status of each component and generate automatic detection results for the elevator components.

[0024] Optionally, in a third implementation of the present invention, the step of obtaining the elevator maintenance task, forming a function library for the inspection robot based on the maintenance task, obtaining the elevator health status analysis result based on the service data, and determining the elevator inspection task includes:

[0025] Elevator maintenance tasks are categorized and coded, wherein the task categories include at least mechanical component inspection, electrical system testing, and safety device testing.

[0026] Based on the classification and coding of maintenance tasks, a function library is built for the inspection robot, wherein the function library contains robot operation instructions corresponding to each maintenance task;

[0027] An elevator health status assessment model is established. Based on the service data, the health status of the elevator is assessed through the elevator health status assessment model to obtain the elevator health status analysis results. The elevator health status assessment model is a BP neural network optimized based on the GA algorithm.

[0028] Optionally, in the fourth implementation of the present invention, the BP neural network optimized based on the GA algorithm includes the following steps:

[0029] Choose the population size and chromosomes, determine the fitness function, and randomly generate an initial population;

[0030] The cumulative probability and selection probability of each string are calculated using the fitness ratio method, and individuals with high fitness are returned to the population.

[0031] The crossover probability determines whether to perform a crossover operation on the two selected individuals. If a crossover operation is performed, one or more positions on the chromosome are randomly selected as crossover points. Parts of the chromosomes of the two individuals are exchanged at the crossover points to generate two new individuals.

[0032] The mutation probability determines whether to perform mutation operation on the selected individual. If mutation operation is performed, a random location on the chromosome is selected as the mutation point, and the gene value at the mutation point is randomly changed.

[0033] The new population obtained after selection, crossover, and mutation operations is used as the next generation population. A new round of fitness evaluation, selection, crossover, and mutation operations are carried out until the preset number of iterations is reached. After the iteration ends, the individual with the highest fitness is selected from the last generation population as the weights and thresholds of the optimized BP neural network.

[0034] Optionally, in a fifth implementation of the present invention, the process of issuing elevator inspection tasks to the inspection robot, integrating task parameters and task status to obtain task data, analyzing the task data, and predicting the time required to complete each task includes:

[0035] Collect data from historical inspection tasks of the inspection robot, organize the collected data in chronological order to form a time series dataset, and divide the time series dataset into a training set and a validation set according to a certain ratio.

[0036] The stationarity of the prepared time series data is tested, and a time series analysis model is selected based on the stationarity test results.

[0037] The time series analysis model is trained using the training set, the model's fit is evaluated using the root mean square error, and the trained model is validated on the validation set to obtain the trained time series model.

[0038] The task data is input into the trained time series model, which predicts the time required to complete each task based on the patterns and trends of historical data.

[0039] Optionally, in the sixth implementation of the present invention, the step of combining the elevator location and the movement trajectory of the inspection robots to plan the inspection routes of each inspection robot, decomposing them into individual tasks and issuing them to the inspection robots, and the inspection robots outputting inspection reports after inspection, including:

[0040] Based on the A* algorithm, a graph is constructed between the elevator location and the area traversed by the inspection robot. Nodes represent location points, and edges represent connections between two location points. Each edge is assigned an indicator weight. The shortest path from the current inspection robot location to the target elevator location is searched in the graph. The indicators include at least distance, time, and energy consumption.

[0041] According to the planned inspection route, the entire inspection task is broken down into multiple sub-tasks, each sub-task corresponding to the inspection task of an elevator, including checking various indicators of the elevator.

[0042] The decomposed sub-tasks are sent to the corresponding inspection robots via wireless communication. After receiving the task, the inspection robot performs the inspection according to the task. During the inspection, the inspection robot collects various data of the elevator, records and stores the collected data, and generates an inspection report.

[0043] A second aspect of the present invention provides an intelligent elevator inspection device, which includes a first data analysis module, a health status analysis module, a second data analysis module, and an inspection route planning module, wherein...

[0044] The first data analysis module is used to collect elevator operation data, status data and fault data through sensors to obtain an elevator data set. By analyzing the data in the elevator data set, service data is generated, wherein the service data includes at least maintenance items and automatic detection results of elevator components.

[0045] The health status analysis module is used to obtain elevator maintenance tasks, form a function library for inspection robots based on the maintenance tasks, obtain the elevator health status analysis results based on the service data, and determine the elevator inspection tasks.

[0046] The second data analysis module is used to send elevator inspection tasks to the inspection robot, integrate task parameters and task status to obtain task data, analyze the task data, and predict the time required to complete each task.

[0047] The inspection route planning module is used to combine the elevator location and the movement trajectory of the inspection robots to plan the inspection routes of each inspection robot, break them down into individual tasks and distribute them to the inspection robots, and output inspection reports after the inspection robots have completed their inspections.

[0048] A third aspect of the present invention provides an intelligent elevator inspection device, the intelligent elevator inspection device comprising: a memory and at least one processor, the memory storing instructions; the at least one processor calling the instructions in the memory to cause the intelligent elevator inspection device to perform each step of the intelligent elevator inspection method described in any of the preceding claims.

[0049] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the elevator intelligent inspection method described in any of the preceding claims.

[0050] In the technical solution provided by this invention, elevator operation data, status data, and fault data are collected by sensors to obtain an elevator data set. Service data is generated by analyzing the data in the elevator data set. Elevator maintenance tasks are obtained, and a function library for inspection robots is formed based on these tasks. The elevator's health status analysis results are obtained based on the service data, and elevator inspection tasks are determined. Inspection tasks are issued to the inspection robots, and task parameters and task status are integrated to obtain task data. This task data is analyzed to predict the time required to complete each task. Finally, the inspection routes of each inspection robot are planned based on the elevator's location and the inspection robot's movement trajectory, and the inspection is decomposed into... Individual tasks are assigned to the inspection robot, which then performs the inspection and outputs an inspection report. This invention combines big data to automatically generate inspection tasks and plan group tasks, improving the efficiency of elevator inspections and reducing the workload of manual inspections. Through big data analysis and the application of intelligent devices, elevator malfunctions can be diagnosed more accurately, improving the accuracy of fault diagnosis. Based on the elevator's health status and preset task requirements, personalized inspection tasks are formulated for each elevator, achieving on-demand inspection, improving the targeting and effectiveness of inspections, enhancing the robot's intelligence level, and continuously improving the robot's inspection capabilities through collaboration and training. It also enhances data security and privacy protection, ensuring the safety and reliability of elevator inspection data. Attached Figure Description

[0051] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention.

[0052] Figure 1 This is a schematic diagram of the first embodiment of the intelligent elevator inspection method provided by the present invention;

[0053] Figure 2 A schematic diagram of a second embodiment of the intelligent elevator inspection method provided by the present invention;

[0054] Figure 3 This is a schematic diagram of the structure of an intelligent elevator inspection device provided in an embodiment of the present invention;

[0055] Figure 4 This is a structural schematic diagram of the intelligent elevator inspection device provided in an embodiment of the present invention. Detailed Implementation

[0056] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0057] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 A schematic diagram of the first embodiment of the intelligent elevator inspection method provided by this invention is shown. The method specifically includes the following steps:

[0058] Step 101: Collect elevator operation data, status data and fault data through sensors to obtain elevator data set. Analyze the data in the elevator data set to generate service data, which includes at least maintenance items and automatic detection results of elevator components.

[0059] Step 102: Obtain elevator maintenance tasks, create a function library for the inspection robot based on the maintenance tasks, obtain the elevator health status analysis results based on the service data, and determine the elevator inspection tasks.

[0060] Step 103: Issue elevator inspection tasks to the inspection robot, integrate task parameters and task status to obtain task data, analyze the task data, and predict the time required to complete each task.

[0061] Step 104: Based on the elevator location and the movement trajectory of the inspection robots, plan the inspection routes of each inspection robot, break them down into individual tasks and assign them to the inspection robots. After the inspection robots perform their inspections, they will output inspection reports.

[0062] In this embodiment, the elevator data set includes elevator door opening and closing status, control panel and button operation status, elevator start and stop status, number of trips, waiting time, elevator interior lighting, elevator interior fan, elevator interior alarm device, elevator interior intercom system, elevator interior maintenance switch, elevator interior stop device, elevator interior fire switch, elevator exterior door, elevator exterior pit environment, elevator exterior shaft lighting, elevator exterior machine room environment, overload status, and uncivilized elevator riding behavior.

[0063] In this embodiment, elevator maintenance tasks are classified and coded. The task classification includes at least mechanical component inspection, electrical system testing, and safety device testing. Based on the classification and coding of maintenance tasks, a function library is built for the inspection robot, which contains robot operation instructions corresponding to each maintenance task. An elevator health status assessment model is established, and the elevator health status is assessed based on service data using the elevator health status assessment model to obtain the elevator health status analysis results. The elevator health status assessment model is a BP neural network optimized based on the GA algorithm.

[0064] In this embodiment, the population size and chromosomes are selected, the fitness function is determined, and an initial population is randomly generated. The cumulative probability and selection probability of each string are calculated using the fitness ratio method, and individuals with high fitness are returned to the population. Based on the crossover probability, it is determined whether to perform a crossover operation on the two selected individuals. If a crossover operation is performed, one or more positions on the chromosome are randomly selected as crossover points, and parts of the chromosomes of the two individuals are exchanged at the crossover points to generate two new individuals. Based on the mutation probability, it is determined whether to perform a mutation operation on the selected individuals. If a mutation operation is performed, a position on the chromosome is randomly selected as a mutation point, and the gene value at the mutation point is randomly changed. The new population obtained after selection, crossover, and mutation operations is used as the next generation population, and a new round of fitness evaluation, selection, crossover, and mutation operations is performed until a preset number of iterations is reached. After the iteration ends, the individual with the highest fitness is selected from the last generation population as the weights and thresholds of the optimized BP neural network.

[0065] In this embodiment, historical inspection task data of the inspection robot are collected, and the collected data is organized in chronological order to form a time series dataset. The time series dataset is then divided into a training set and a validation set according to a certain ratio. The stationarity of the organized time series data is tested, and a time series analysis model is selected based on the stationarity test results. The time series analysis model is trained using the training set, and the root mean square error is used to evaluate the model's fitting effect. The trained model is then validated on the validation set to obtain a trained time series model. The task data is input into the trained time series model, and the model predicts the time required to complete each task based on the patterns and trends of historical data.

[0066] In this embodiment, the elevator location and the area traversed by the inspection robot are constructed into a graph based on the A* algorithm. Nodes represent location points, and edges represent connections between two location points. Each edge is assigned a weight, and the shortest path from the current inspection robot location to the target elevator location is searched in the graph. The weights include at least distance, time, and energy consumption. According to the planned inspection route, the entire inspection task is decomposed into multiple sub-tasks, each corresponding to the inspection task of one elevator, including checking various indicators of the elevator. The decomposed sub-tasks are sent to the corresponding inspection robots via wireless communication. After receiving the task, the inspection robot performs the inspection according to the task. During the inspection process, the inspection robot collects various data of the elevator, records and stores the collected data, and generates an inspection report.

[0067] Please see Figure 2 A schematic diagram of the second embodiment of the intelligent elevator inspection method provided by the present invention is shown. The method includes:

[0068] Step 201: Use median filtering to smooth the data in the elevator dataset, remove outliers and noise, and normalize the smoothed data to obtain a new elevator dataset.

[0069] Step 202: Perform principal component analysis on the data in the new elevator dataset to extract the main features, reduce the data dimensionality, and determine the focus of maintenance items based on the results of the principal component analysis.

[0070] Step 203: Use a clustering algorithm to classify the data in the new elevator dataset after dimensionality reduction, identify different types of elevator operating status and fault modes, and generate automatic detection results for elevator components based on the clustering results.

[0071] Step 204: Based on the results of principal component analysis and cluster analysis, determine the maintenance items for the elevator, assess the importance and urgency of each maintenance item, determine the priority, sort the maintenance items according to the priority, and generate service data.

[0072] In this embodiment, Z-Score standardization is used to process the data in the new elevator dataset to obtain standardized data; the covariance matrix of the standardized data is calculated to obtain the eigenvalues ​​and eigenvectors of the covariance matrix; the eigenvectors are sorted according to the size of the eigenvalues, and the eigenvectors corresponding to the first n eigenvalues ​​are selected to form the principal components; the data in the new elevator dataset is projected onto the principal components to obtain new data after dimensionality reduction; based on the results of principal component analysis, the focus of maintenance items is determined.

[0073] In this embodiment, a first cluster center is randomly selected from the data in the new elevator dataset after dimensionality reduction. The shortest distance between the first cluster center and the data in the new elevator dataset is obtained through iterative calculation, and a distance set is established. The distance set is normalized, all distance sums in the distance set are calculated, and Laplace noise is added to obtain the shortest distance sum after noise addition. Seed cluster centers are selected within the normalized range of the data in the new elevator dataset after dimensionality reduction, and the shortest distance between the data and the cluster centers is calculated iteratively until the selection of cluster centers is completed. Based on the obtained shortest distance set and the selection of cluster centers, the data corresponding to the shortest distance set is divided into cluster centers to obtain data categories. The mean of the data categories is expanded and calculated, and noise is added to the calculation results. The results are used as new data data points, and the iteration is repeated until the cluster centers no longer change or the preset number of iterations is reached to complete the data clustering. According to the clustering results, different types of elevator operating states and fault modes are identified, including at least normal operation, minor fault, and serious fault categories. For each cluster, the elevator data in that category is analyzed to determine the operating status of each component and generate automatic detection results for the elevator components.

[0074] This invention collects usage data and electrical status data of each component, analyzes the health status of each component, and combines relevant external devices, such as cameras, to obtain the internal and external conditions of the elevator. After system analysis, it issues tasks to the inspection robot, which inspects the elevator according to the preset task methods and trajectories, and outputs the final inspection report in the system. It also issues automatic inspection tasks based on the detection results of big data and the user's tasks.

[0075] Based on big data technology, the system automatically inspects the elevator: the system collects elevator operation data, status data and fault data, performs in-depth analysis and calculation through the big data platform, and generates maintenance items, elevator component automatic detection results, etc., to provide data services for robot inspection;

[0076] 1) Generate elevator equipment operating status, including elevator door opening and closing status, control panel and button operation status, elevator start and stop status, number of trips, and waiting time;

[0077] 2) Elevator interior safety conditions, including car lighting, fans, alarm devices, intercom system, inspection switches, stop devices, and fire switches;

[0078] 3) External safety conditions of the elevator, including elevator doors, pit environment, shaft lighting, machine room environment, overloading, uncivilized elevator riding, etc.

[0079] Each maintenance task is assigned a corresponding function, forming a function library for the inspection robot. For example, in the task of detecting lighting anomalies, the robot moves from point A to point B, enters the elevator, and uses a camera to check the lights on the car ceiling to see if the lighting is normal.

[0080] Based on the system's analysis of the elevator's health status and the preset task requirements, each elevator has a different inspection task and its own combination of tasks.

[0081] The intelligent edge devices in elevators can share computing power and computing models, deploying part of the robot's intelligence at the edge. Through collaboration and continuous training, the robot's intelligence can be continuously improved. At the same time, edge computing can also support real-time multi-machine collaboration, enabling real-time knowledge graph extraction, understanding, and decision-making, thereby improving robot intelligence and enhancing data security and privacy protection.

[0082] Group task planning:

[0083] To improve inspection efficiency, multiple robots need to be deployed for the inspection of multiple elevators.

[0084] Based on the results of big data inspection, the condition of various elevator components, such as the buttons in the car and the buttons on each floor, can be known. The robot can then be assigned to floors that may have problems for inspection, thus achieving on-demand inspection.

[0085] After the robot completes each task module, it will save the parameters and completion status to the large model system for analysis, in order to predict the time required to complete each task.

[0086] Based on 1) and 2), the task objectives that need to be inspected for each elevator and each waiting hall can be obtained. The overall task time can be calculated. Then, combined with the location of the elevator and the movement trajectory of the robot, the inspection route of each robot can be planned and decomposed into individual tasks and assigned to the robot.

[0087] Related collaborative tasks are also simultaneously distributed to various intelligent edge devices to assist robots in completing inspection tasks.

[0088] Based on environmental and historical data analysis, the large model is guided to generate environment-matched task plans. Then, through a multi-level decomposition task planning method, tasks are broken down into goal, task, and behavior levels, enhancing the large model's task planning capabilities. Through continuous learning, the task planning is further improved and optimized.

[0089] Inspection task function:

[0090] Check for any abnormal vibrations or noises during the operation of the drive unit;

[0091] Check the main unit's trajectory information and electrical condition for any abnormalities. Simultaneously, the robot carries vibration sensors to collect and analyze data to detect any abnormal vibrations or sounds during elevator operation.

[0092] Are the in-car alarm device and intercom system working properly?

[0093] The system analyzes the communication status and self-test status of the intercom equipment for the day to check for any abnormalities. The robot activates the alarm button and communicates with the monitoring center via recording to check if the alarm function is working properly.

[0094] Are the anti-collision protection devices for the car doors effective? (Safety touch panel, light curtain, photoelectric, etc.)

[0095] The robot periodically checks the button's operational status. If the number of actions of the light curtain, safety touch panel, photoelectric sensors, etc., exceeds a preset value, it is considered normal. The robot uses its tentacles to obstruct the door from closing and records the status of the car door's anti-collision protection device on video.

[0096] Elevator leveling inspection:

[0097] By collecting records of the elevator's own "leveling faults" and thresholds for minor anomalies, a task is assigned to the robot for floors with leveling anomalies. The robot, equipped with an RGBD camera on its extendable robotic arm, captures a close-up depth map when the elevator is leveling, analyzing whether there is a height difference between the car floor and the landing sill. Using the car floor as a reference, points are taken to fit a plane, and then points are taken at the landing sill. The maximum vertical distance from these points to the car floor is used as the leveling accuracy.

[0098] By implementing the above solution, and combining big data to automatically generate inspection tasks and group task planning, the efficiency of elevator inspection has been improved, the workload of manual inspection has been reduced, and the application of big data analysis and intelligent equipment can more accurately diagnose elevator faults, improving the accuracy of fault diagnosis. Based on the health status of the elevator and preset task requirements, personalized inspection tasks are formulated for each elevator, realizing on-demand inspection, improving the pertinence and effectiveness of inspection, enhancing the intelligence level of the robot, and continuously improving the robot's inspection capabilities through collaboration and training, while strengthening data security and privacy protection, ensuring the safety and reliability of elevator inspection data.

[0099] The elevator intelligent inspection method in the embodiments of the present invention has been described above. The elevator intelligent inspection device in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 3 One embodiment of the intelligent elevator inspection device in this invention includes:

[0100] The first data analysis module 301 is used to collect elevator operation data, status data and fault data through sensors to obtain an elevator data set. By analyzing the data in the elevator data set, service data is generated, which includes at least maintenance items and automatic detection results of elevator components.

[0101] The health status analysis module 302 is used to obtain the elevator maintenance tasks, form a function library for the inspection robot based on the maintenance tasks, obtain the elevator health status analysis results based on the service data, and determine the elevator inspection tasks.

[0102] The second data analysis module 303 is used to send elevator inspection tasks to the inspection robot, integrate task parameters and task status to obtain task data, analyze the task data, and predict the time required to complete each task.

[0103] The inspection route planning module 304 is used to combine the elevator location and the movement trajectory of the inspection robot to plan the inspection route of each inspection robot, decompose it into individual tasks and issue them to the inspection robots, and output inspection reports after the inspection robots have carried out the inspection.

[0104] The elevator intelligent inspection device in this embodiment of the invention will be described in detail below from the perspective of hardware processing.

[0105] Figure 4 This is a schematic diagram of the structure of an intelligent elevator inspection device 600 provided in an embodiment of the present invention. The intelligent elevator inspection device 600 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 610 (e.g., one or more processors) and a memory 620, and one or more storage media 630 (e.g., one or more mass storage devices) for storing application programs 633 or data 632. The memory 620 and storage media 630 can be temporary or persistent storage. The program stored in the storage media 630 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the intelligent elevator inspection device 600. Furthermore, the processor 610 may be configured to communicate with the storage media 630 and execute the series of instruction operations in the storage media 630 on the intelligent elevator inspection device 600 to implement the steps of the intelligent elevator inspection method provided in the above-described method embodiments.

[0106] The elevator intelligent inspection device 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input / output interfaces 660, and / or one or more operating systems 631, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 4 The structure of the elevator intelligent inspection equipment shown does not constitute a limitation on elevator intelligent inspection equipment. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0107] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the elevator intelligent inspection method.

[0108] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0109] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0110] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent elevator inspection method, characterized by, The elevator intelligent inspection method comprises the following steps: Collecting operation data, state data and fault data of the elevator through sensors to obtain an elevator data set, analyzing the data in the elevator data set to generate service data, wherein the service data at least includes maintenance items and automatic detection results of elevator components; Obtaining maintenance tasks of the elevator, forming a function library of the inspection robot for the maintenance tasks, obtaining health state analysis results of the elevator according to the service data, and determining inspection tasks of the elevator; specifically comprising: Classifying the elevator maintenance tasks and coding each maintenance task, wherein the task classification at least includes mechanical component inspection, electrical system detection and safety device testing; Constructing a function library for the inspection robot according to the classification and coding of the maintenance tasks, wherein the function library contains robot operation instructions corresponding to each maintenance task; Establishing an elevator health state evaluation model, evaluating the health state of the elevator through the elevator health state evaluation model based on the service data to obtain health state analysis results of the elevator, wherein the elevator health state evaluation model is a BP neural network optimized based on a GA algorithm; specifically, the BP neural network optimized based on the GA algorithm comprises the following steps: Selecting a population size and a chromosome, determining a fitness function, and randomly generating an initial population; Calculate the cumulative probability and selection probability of each string by the fitness proportion method, and put the individuals with high fitness back into the population; Determine whether to perform a crossover operation on the selected two individuals according to a crossover probability, if the crossover operation is performed, randomly select one or more positions on the chromosome as crossover points, exchange part of the chromosomes of the two individuals at the crossover points to generate two new individuals; Determine whether to perform a mutation operation on the selected individual according to a mutation probability, if the mutation operation is performed, randomly select a position on the chromosome as a mutation point, and randomly change the gene value at the mutation point; Take the new population obtained after the selection, crossover and mutation operations as the next generation population, and perform a new round of fitness evaluation, selection, crossover and mutation operations until a preset iteration number is reached, and after the iteration ends, select the individual with the highest fitness from the last generation population as the weight and threshold of the optimized BP neural network; Issue the inspection tasks of the elevator to the inspection robot, integrate the task parameters and task conditions to obtain task data, analyze the task data to predict the time required for completing each task; Plan the inspection routes of each inspection robot in combination with the elevator positions and the action trajectories of the inspection robots, decompose the inspection routes into individual tasks and issue the individual tasks to the inspection robots, and output an inspection report through the inspection robots.

2. The intelligent elevator inspection method of claim 1, wherein, The method for collecting operation data, state data and fault data of the elevator through sensors to obtain an elevator data set, and analyzing the data in the elevator data set to generate service data comprises: Smoothing the data in the elevator data set by a median filtering method to remove abnormal values and noise data, and normalizing the smoothed data to obtain a new elevator data set; The data in the new elevator data set is processed by Z-Score standardization to obtain standardized data; The covariance matrix of the standardized data is calculated, the eigenvalues and eigenvectors of the covariance matrix are obtained, the eigenvectors are sorted according to the size of the eigenvalues, and the first n eigenvectors are selected to form the principal components; The data in the new elevator data set is projected onto the principal components to obtain the new data after dimensionality reduction, and the attention direction of the maintenance item is determined according to the result of the principal component analysis; The data in the new elevator data set after dimensionality reduction is classified by using a clustering algorithm, different types of elevator operating states and fault modes are identified, and an automatic detection result of the elevator components is generated according to the clustering result; The results of principal component analysis and clustering analysis are combined to determine the maintenance items of the elevator, the importance and urgency of each maintenance item are evaluated, the priority is determined, the maintenance items are sorted according to the priority, and service data is generated.

3. The method of claim 2, wherein, The data in the new elevator data set after dimensionality reduction is classified by using a clustering algorithm, different types of elevator operating states and fault modes are identified, and an automatic detection result of the elevator components is generated according to the clustering result, including: A first cluster center is randomly selected from the data in the new elevator data set after dimensionality reduction, and the shortest distance sum between the first cluster center and the data in the new elevator data set is calculated by iteration to establish a distance set; The distance set is normalized, the distance sum in the distance set is calculated, and Laplace noise is added to obtain the shortest distance sum after adding noise; Within the normalized range of the data in the new elevator data set after dimensionality reduction, a seed cluster center is selected, and the shortest distance between the data and the cluster center is calculated in a loop until the selection of the cluster center is completed; Based on the obtained shortest distance set and the selection of the cluster center, the data corresponding to the shortest distance set is divided into the cluster center to obtain the data category; The mean value of the data category is calculated, noise is added to the calculation result, and the result is used as a new data center point. The iteration is repeated until the cluster center no longer changes or the preset number of iterations is reached, and the data clustering is completed; According to the clustering result, different types of elevator operating states and fault modes are identified, including at least normal operation class, minor fault class and serious fault class; For each cluster, the operating state of each component is determined by analyzing the elevator data in the cluster, and an automatic detection result of the elevator components is generated.

4. The method of claim 1, wherein, The task data is obtained by integrating the task parameters and task conditions, the task data is analyzed, and the time required for completing each task is predicted, including: Collecting data of historical inspection tasks of the inspection robot, arranging the collected data in chronological order to form a time series data set, dividing the time series data set into a training set and a validation set according to a certain proportion; Performing stationarity test on the arranged time series data, and selecting a time series analysis model based on the stationarity test result; The time series analysis model is trained using the training set, the fitting effect of the model is evaluated using the root mean square error, the trained model is verified on the verification set, and a trained time series model is obtained; The task data is input into the trained time series model, and the model predicts the time required for each task to be completed according to the patterns and trends of historical data.

5. The method of claim 1, wherein, The elevator positions and the action trajectories of the inspection robots are combined to plan inspection routes for each inspection robot, which are decomposed into individual tasks and issued to the inspection robots, and inspection reports are output by the inspection robots after inspection, including: The elevator positions and the areas passed by the inspection robots are constructed into a graph based on the A* algorithm, the nodes represent position points, the edges represent the connection between two position points, and each edge is assigned a weight, and the shortest path from the current inspection robot position to the target elevator position is searched in the graph, wherein the index at least includes distance, time and energy consumption; According to the planned inspection route, the entire inspection task is decomposed into multiple subtasks, each corresponding to an inspection task of an elevator, including checking various indicators of the elevator; The decomposed subtasks are issued to the corresponding inspection robots through wireless communication, and the inspection robots perform inspection according to the tasks after receiving the tasks. During the inspection process, the inspection robots collect various data of the elevator, record and store the collected data, and generate inspection reports.

6. An intelligent elevator inspection device, characterized by, The elevator intelligent inspection method according to any one of claims 1-5, the elevator intelligent inspection device comprises a first data analysis module, a health state analysis module, a second data analysis module and an inspection route planning module, wherein, The first data analysis module is used to collect the operation data, state data and fault data of the elevator through sensors to obtain an elevator data set, and generate service data by analyzing the data in the elevator data set, wherein the service data at least includes maintenance items and automatic detection results of elevator components; The health state analysis module is used to obtain maintenance tasks of the elevator, form a function library of the inspection robot for the maintenance tasks, obtain health state analysis results of the elevator according to the service data, and determine inspection tasks of the elevator; The second data analysis module is used to issue the inspection tasks of the elevator to the inspection robot, integrate task parameters and task conditions to obtain task data, analyze the task data, and predict the time required for each task to be completed; The inspection route planning module is used to combine the elevator positions and the action trajectories of the inspection robots to plan inspection routes for each inspection robot, which are decomposed into individual tasks and issued to the inspection robots, and inspection reports are output by the inspection robots after inspection.

7. An elevator intelligent patrol device, characterized by, The elevator intelligent inspection device comprises a memory and at least one processor, and the memory stores instructions; At least one processor calls the instructions in the memory to enable the elevator intelligent inspection device to perform each step of the elevator intelligent inspection method according to any one of claims 1-5.

8. A computer-readable storage medium having stored thereon instructions, the computer-readable storage medium comprising: The instructions are executed by the processor to implement each step of the elevator intelligent inspection method according to any one of claims 1-5.

Images