Search

CN-121998398-A - Elevator inspection procedure optimization method

CN121998398ACN 121998398 ACN121998398 ACN 121998398ACN-121998398-A

Abstract

The invention provides an elevator inspection procedure optimizing method, which relates to the field of internet intelligent manufacturing and comprises the steps of collecting real-time sensor data corresponding to a target elevator, preprocessing the real-time sensor data corresponding to each sensor, updating a digital twin elevator model corresponding to the target elevator in real time, outputting a corresponding current running state, determining the point to be detected of a virtual elevator in the digital twin elevator model at the current moment, mapping and encoding each point to be detected into a specific detection procedure, forming a plurality of candidate detection sequences according to the detection procedures, optimizing each candidate detection sequence, selecting the optimal detection sequence as a final detection sequence, feeding the final detection sequence back to an inspector, collecting current position information corresponding to the inspector and an inspection picture corresponding to a worn intelligent acquisition terminal, and judging whether the current inspection procedure executed by the inspector is correct or not according to the current position information and the inspection picture of the inspector.

Inventors

  • SHEN HAIPING

Assignees

  • 杭州奥创电梯工程有限公司

Dates

Publication Date
20260508
Application Date
20260121

Claims (8)

  1. 1. A method of optimizing an elevator inspection process, the method comprising: collecting real-time sensor data corresponding to a target elevator, and preprocessing the real-time sensor data corresponding to each sensor; According to the preprocessed real-time sensor data, updating the digital twin elevator model corresponding to the target elevator in real time, and outputting the current running state corresponding to the target elevator; Determining the point positions to be detected of the virtual elevator in the digital twin elevator model at the current moment based on the current running state, and mapping and encoding each point position to be detected into a specific detection procedure; Forming a plurality of candidate detection sequences according to the detection procedure, optimizing each candidate detection sequence based on a preset sequence optimization algorithm, selecting an optimal detection sequence as a final detection sequence, and feeding back the final detection sequence to a inspector; When the inspector receives the final detection sequence, collecting current position information corresponding to the inspector and a detection picture corresponding to the worn intelligent acquisition terminal; Judging whether the current inspection procedure carried out by the inspector is correct or not according to the current position information of the inspector and the inspection picture, and visually displaying the current position of the inspector and the inspection picture in the digital twin elevator model when the current inspection procedure is incorrect, and sending an alarm signal.
  2. 2. The elevator inspection process optimization method according to claim 1, wherein preprocessing the real-time sensor data corresponding to each sensor comprises: Extracting running state parameters corresponding to a target elevator, and dividing the elevator running process into four dynamic sections, namely a starting acceleration section, a uniform running section, a decelerating stopping section and a static section, based on the running state parameters, wherein the running state parameters comprise real-time position data, running direction and speed change rate of the elevator; aiming at different dynamic intervals, adopting a differential time alignment strategy to perform alignment operation on real-time sensor data corresponding to each sensor; Denoising the real-time sensor data after the alignment operation and filling in the missing data.
  3. 3. The elevator inspection process optimization method according to claim 2, wherein determining the point to be detected of the virtual elevator in the digital twin elevator model at the current moment based on the current running state comprises: Calculating the dynamic risk indexes of all parts in the digital twin elevator model by adopting a weighted fusion algorithm based on the historical operation data stored in a database according to the current operation state of the digital twin elevator model and the real-time sensor data of all corresponding sensors; correcting the dynamic risk index based on the risk factor weight corresponding to each predefined component to generate a comprehensive risk priority number, and sequencing all components to be detected according to the risk priority number; And selecting the components ranked at the preset ranks as high-risk components according to the risk priority ranks corresponding to the components, and determining the specific geometric feature points corresponding to the high-risk components in the digital twin elevator model as corresponding points to be detected.
  4. 4. The elevator inspection process optimization method according to claim 3, wherein the encoding each point to be detected map into a specific detection process comprises: Generating corresponding multidimensional dynamic point position codes for each point position to be detected based on the current running state and the corresponding historical running data of a digital twin elevator model corresponding to the target elevator, wherein the dynamic point position codes at least comprise a basic identity ID for representing the identity of a component, a real-time risk tag for representing the real-time risk level, and a resource and environment tag for representing the field condition; According to the dynamic point position codes, based on a preset detection procedure knowledge base and a procedure matching strategy, matching corresponding detection procedures for all the dynamic point position codes; The preset procedure matching strategy is as follows: searching a process template which is completely matched with the complete dynamic point position coding in a detection process knowledge base; If the matching fails, selecting an optimal working procedure template from a plurality of candidate templates based on the values of one or more specific labels in the dynamic point position coding; If the adaptive template is still not available, selecting a corresponding minimum operation unit from a preset procedure atomic operation library, and dynamically combining according to rules to generate a temporary detection procedure.
  5. 5. The elevator inspection process optimization method according to claim 4, wherein the forming a plurality of candidate inspection sequences according to the inspection process, optimizing each candidate inspection sequence based on a preset sequence optimization algorithm, selecting an optimal inspection sequence as a final inspection sequence, and feeding back to inspection personnel comprises: According to the detection procedure, the positions of all the sub-steps in the detection procedure are randomly changed to form a plurality of candidate detection sequences which are used as corresponding initial populations; calculating a process detection efficiency value and a process detection rationality corresponding to individuals of the population; Selecting individuals with the process detection efficiency value larger than a first preset efficiency value in the first iteration times, and storing the individuals with the process detection efficiency value larger than or equal to a second preset efficiency value and smaller than the first preset efficiency value into a standby library to form a standby population; after the first iteration times are exceeded, combining the current population with the standby population to form a new population, and selecting individuals with the process detection rationality of more than or equal to a preset rationality in the current population; obtaining a offspring population from the selected population through a hybridization variation strategy; after the offspring population is obtained, calculating and screening the process detection efficiency value and the process detection rationality are continued until the preset iteration times are met; and outputting the offspring population as a final detection sequence and feeding back to the inspector.
  6. 6. The elevator inspection process optimization method according to claim 5, wherein storing individuals with process detection efficiency values greater than or equal to a second preset efficiency value and less than a first preset efficiency value in a backup library to form a backup population comprises: selecting an individual with the process detection efficiency value being more than or equal to the second preset efficiency value and less than the first preset efficiency value; calculating the distance value among the individuals according to the selected individuals; Comparing the iteration times with the first iteration times: when the iteration times are smaller than or equal to the first percentage of the first iteration times, sequencing according to the sequence from the large distance value to the small distance value, and selecting individuals with the previous second percentage to store in a standby library; When the iteration times are larger than the first percentage of the first iteration times and smaller than or equal to the first iteration times, after the first iteration times are ordered according to the distance values, selecting individuals with the first third percentage and storing the individuals in the standby library, wherein the second percentage is larger than the third percentage.
  7. 7. The elevator inspection process optimizing method according to claim 6, wherein the step of judging whether the inspection process currently performed by the inspector is correct based on the current position information of the inspector and the inspection picture, and visually displaying the current position of the inspector and the inspection picture in the digital twin elevator model when the inspection process is incorrect, and sending out an alarm signal comprises: determining node procedure information associated with each procedure node in the final detection sequence according to the final detection sequence, wherein the node procedure information comprises a node unique sequence number, a theoretical execution area and standard point location characteristics; determining the current position coordinates of the inspector and the actual point location characteristics in the inspection picture according to the current position information of the inspector and the inspection picture; searching a last procedure node of the finished procedure nodes in the final detection sequence, and determining a next procedure node corresponding to the last procedure node as a next execution node; Checking whether the current position coordinates match with a theoretical execution area of the next execution node or not, and determining that the matching degree of the actual point location features and the standard point location features of the next execution node is larger than a preset matching threshold; If the positions are matched and the matching degree reaches the standard, judging that the currently executed checking procedure of the checking staff is correct, and if any item is not satisfied, judging that the currently executed checking procedure of the checking staff is wrong; At this time, if the positions are not matched, marking the theoretical execution position of the next execution node by using a yellow dynamic flashing icon, marking the current position of the inspector by using a red solid icon, connecting the two by using a red dotted line, and marking the linear distance; If the matching degree does not reach the standard, ejecting red text bubbles at the current position of the inspector in the digital twin elevator model, and marking the current working procedure node to be executed and the actual working procedure node to be executed.
  8. 8. The elevator inspection process optimization method of claim 7, wherein the hybridization variation strategy comprises: dividing individuals in the selected group into 3 associated procedure partitions according to a preset elevator detection space region and a pre-arranged dependency relationship of the checking procedure, wherein procedures in each associated procedure partition meet the condition that space positions are continuous and pre-arranged dependency conflicts are avoided; selecting individuals from the population to pair every two, and generating 3 independent first random numbers for three associated procedure partitions of one individual randomly, wherein the value range of the random numbers is 0 to 1; If the first random number is larger than a preset random threshold value, carrying out pairwise exchange on the association procedure partition of the first individual corresponding to the first random number and the association procedure partition of the second individual corresponding to the second random number; And if the first random number is smaller than or equal to a preset random threshold value, randomly generating a second random number from 0 to 1 for the associated process partition of the first individual corresponding to the first random number, and if the second random number is larger than the first random number, mutating the associated process partition of the first individual corresponding to the second random number.

Description

Elevator inspection procedure optimization method Technical Field The invention relates to the technical field of internet intelligent manufacturing, in particular to an elevator inspection procedure optimization method. Background With the deep penetration of the internet intelligent manufacturing technology in the field of special equipment, the elevator is used as a key infrastructure of urban traffic, and the requirements on operation safety and inspection efficiency are increasingly increased. The conventional inspection relies on a fixed flow, and a dynamic adjustment scheme of the real-time running state of the elevator is not combined, so that the problems of missed inspection of high-risk components or excessive inspection of low-risk components often occur, and the individual differential safety requirements of the elevator cannot be met. The multi-sensor data processing link has the defects that the sensor data of the elevator in different dynamic intervals such as starting acceleration, uniform speed and the like are easy to interfere, and the conditions such as asynchronous time, noise superposition, data loss and the like occur, so that the subsequent detection point location and process generation lack of reliable data support. And the optimization of the test sequence mostly adopts a general algorithm, the space relevance and the procedure dependency relationship of elevator test are not considered, and the optimized sequence has low efficiency and poor execution rationality. The inspection process lacks a real-time supervision error correction mechanism, and inspection personnel are prone to errors such as jump sequence and position deviation, and the traditional mode is difficult to identify and guide in time, so that the inspection quality is affected, and safety risks can be possibly caused. Therefore, the elevator inspection procedure optimizing method is provided, the internet intelligent manufacturing is used as a support, the high-risk point positions are updated and positioned in real time through the sensor data fine preprocessing and digital twin, an optimal detection sequence is generated by combining a multi-target algorithm, and the elevator inspection intellectualization, the precision and the safety are realized by matching with real-time checking and alarming, so that the high-quality supervision requirement of special equipment is met. Disclosure of Invention In order to solve the technical problems, the invention aims to provide an elevator inspection procedure optimizing method which takes intelligent manufacturing of the Internet as a support, generates an optimal detection sequence by means of sensor data fine preprocessing and digital twin real-time updating and positioning of high-risk points and combining a multi-target algorithm, and achieves intelligent, accurate and safe elevator inspection by matching with real-time checking alarm, thereby meeting the high-quality supervision requirement of special equipment. In order to achieve the purpose, the invention provides the following technical scheme that the elevator inspection procedure optimizing method comprises the following steps: collecting real-time sensor data corresponding to a target elevator, and preprocessing the real-time sensor data corresponding to each sensor; According to the preprocessed real-time sensor data, updating the digital twin elevator model corresponding to the target elevator in real time, and outputting the current running state corresponding to the target elevator; Determining the point positions to be detected of the virtual elevator in the digital twin elevator model at the current moment based on the current running state, and mapping and encoding each point position to be detected into a specific detection procedure; Forming a plurality of candidate detection sequences according to the detection procedure, optimizing each candidate detection sequence based on a preset sequence optimization algorithm, selecting an optimal detection sequence as a final detection sequence, and feeding back the final detection sequence to a inspector; When the inspector receives the final detection sequence, collecting current position information corresponding to the inspector and a detection picture corresponding to the worn intelligent acquisition terminal; Judging whether the current inspection procedure carried out by the inspector is correct or not according to the current position information of the inspector and the inspection picture, and visually displaying the current position of the inspector and the inspection picture in the digital twin elevator model when the current inspection procedure is incorrect, and sending an alarm signal. Preferably, the preprocessing the real-time sensor data corresponding to each sensor includes: Extracting running state parameters corresponding to a target elevator, and dividing the elevator running process into four dynamic sections, namely a starting accelerat