Elevator health state evaluation method and system based on big data

Abstract

The invention relates to the technical field of elevator supervision in special equipment, in particular to an elevator health state evaluation method and system based on big data, and the method comprises the steps: obtaining multi-source heterogeneous data related to elevator operation; the elevator registration code serves as a unique identification code, data belonging to the same elevator are associated, and an elevator theme library is formed; building an elevator health evaluation model, and based on the data record of each elevator in the elevator theme library, performing real-time dynamic matching calculation through the elevator health evaluation model to obtain an elevator health evaluation result library; and an evaluation threshold value is set, the elevators with the evaluation results smaller than the evaluation threshold value are obtained, N elevators are selected from the elevators for early warning reminding, related supervisors go to the site for supervision and inspection according to a preset inspection period, and the inspection results are synchronized into the elevator theme library. According to the method, the accuracy and efficiency of elevator health state evaluation are effectively improved, and a powerful guarantee is provided for safe operation of the elevator.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator supervision in special equipment, and in particular to an elevator health status evaluation method and system based on big data. Background Art

[0002] With the rapid development of social economy and the accelerated advancement of urbanization, the total number of elevators is currently huge, and the proportion of medium and high-rise elevators in cities continues to rise. The number of old elevators with an age of more than 10 years has increased year by year. Coupled with the surge in the number of super-high-rise buildings, the safe operation and maintenance supervision of elevators are facing unprecedented pressure. It is urgent to introduce new technologies to achieve intelligent supervision to improve the overall quality and safety management efficiency of urban elevators.

[0003] The rapid increase in the number of elevator equipment, the prominent problem of equipment aging, as well as improper use, lack of management, inadequate maintenance and other multiple factors have jointly led to the annual increase in the failure rate of elevators. At present, elevator supervision still uses the traditional manual spot check mode, which undoubtedly increases the difficulty of supervising the operation status and health monitoring of elevators. With the continuous increase in the number of high-rise elevators in cities and the continuous accumulation of elevator age, how to use more cutting-edge technologies to achieve refined supervision of existing operating elevators by classification has become a major issue that needs to be solved urgently.

[0004] In view of this, there is an urgent need for an elevator health status evaluation method based on big data technology, which aims to achieve early warning and precise positioning of elevator failures, thereby effectively improving the intelligence level and efficiency of elevator supervision. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an elevator health status evaluation method and system based on big data, and the evaluation method comprises the following steps: S1: Acquire multi-source heterogeneous data related to elevator operation; S2: Using the elevator registration code as a unique identification code, associating the data belonging to the same elevator in the multi-source heterogeneous data to form an elevator theme library; S3: constructing an elevator health evaluation model, based on the data records of each elevator in the elevator subject library, performing real-time dynamic matching calculation through the elevator health evaluation model to obtain an elevator health evaluation result library; S4: setting an evaluation threshold, randomly selecting N elevators from the elevator health evaluation result database whose evaluation results are less than the evaluation threshold, and issuing early warning reminders. According to the preset inspection cycle, relevant supervisors conduct on-site supervision and inspection, and synchronize the inspection results to the elevator theme database; Among them, in step S2, the method of forming the elevator theme library is as follows: Collect data sources from the multi-source heterogeneous data at a pre-set time interval, combine the elevator registration code and the timestamp of data generation into a business primary key, and remove duplicate data records by comparing the business primary key of each data source to obtain deduplicated multi-source data; Constructing a field mapping table, and performing format conversion on field data of the same type but different names in the deduplicated multi-source data according to the field mapping table to obtain multi-source data with a unified format; For the newly collected data source, the multi-source data in the unified format is updated according to the business primary key thereof to obtain the updated multi-source data; Based on the updated multi-source data, data records with different data sources but belonging to the same elevator are associated through the elevator registration code to obtain a final elevator theme library.

[0006] In one embodiment of the present invention, in step S3, the method for obtaining the elevator health evaluation result library is as follows: S31: The elevator health evaluation model includes sub-models corresponding to eight dimensions: entity information, status information, operation information, IoT information, maintenance information, fault information, trapped person information, and feedback information. The sub-models are used to calculate the calculation results of each dimension, and weights are assigned to the calculation results of each dimension and then added together to obtain the elevator health evaluation result R of each elevator: R=W 1 ×R 1 +W 2 ×R 2 +W 3 ×R 3 +W 4 ×R 4 +W 5 ×R 5 +W 6 ×R 6 +W 7 ×R 7 +W 8 ×R 8 (1), Among them, R 1 is the elevator health evaluation score in the ontology information dimension, R 2 is the elevator health evaluation score in the condition information dimension, R 3 is the elevator health evaluation score in the operation information dimension, R 4 is the elevator health evaluation score in the IoT information dimension, R 5 is the elevator health evaluation score in the maintenance information dimension, R 6 is the elevator health evaluation score in the fault information dimension, R 7 is the elevator health evaluation score in the dimension of trapped person information, R 8is the elevator health evaluation score in the feedback information dimension; W 1 , W 2 , W 3 , W 4 , W 5 , W 6 , W 7 and W 8 They are all weight parameters; S32: Summarize the elevator health evaluation results of all elevators to obtain an elevator health evaluation result library.

[0007] In one embodiment of the present invention, the ontology information dimension elevator health evaluation score R 1 The method to obtain is as follows: For each elevator, the number of elevators that failed at the manufacturer is compared with the total number of elevators produced by the manufacturer to obtain the failure rate of the elevator manufacturer. , the failure rate of the elevator manufacturer Mapped to the corresponding elevator, the health score of the manufacturer's failure rate indicator item is obtained through formula (2): : (2); Aggregate all the places where elevators are used, count the number of elevator failures in each place, and the total number of elevator failures in all places, and take the ratio of the two statistical results as the place of use failure rate , the failure rate of the place of use Mapped to the corresponding elevator, the health score of the failure rate index item of the use place is obtained through formula (3): : (3); The number of elevator failures within n years after installation is obtained by the time the elevator is put into use. The number of elevators is aggregated according to the construction unit, and the ratio of the number of elevator failures within n years of installation by the construction unit to the total number of elevators installed by the construction unit is taken as the construction unit failure rate. , the construction unit failure rate Mapped to the corresponding elevator, the health score of the construction unit failure rate index item is obtained through formula (4): : (4); The number of failures is aggregated according to the elevator manufacturing unit and elevator model, and the number of elevator failures for each elevator manufacturing unit and elevator model is obtained. The total number of failed elevators is counted according to the elevator manufacturing unit and elevator model, and the ratio of the two statistical results is used as the equipment model failure rate. and the failure rate of the equipment model Mapped to the corresponding elevator, the health score of the equipment model failure rate indicator item is obtained through formula (5) : (5); The health score of the manufacturer's failure rate indicator , the health score of the failure rate indicator item of the place of use , the health score of the construction unit failure rate indicator and the health score of the failure rate indicator of the device model Perform weighted calculation to obtain the elevator health evaluation score R in the ontology information dimension 1 : (6) in, , , , are weight parameters, .

[0008] In one embodiment of the present invention, the status information dimension elevator health evaluation score The method to obtain is as follows: Different scores are deducted based on the number of deaths, entrapments, and malfunctions that occur in the elevator within a specified time period. If there are no deaths, entrapments, or malfunctions, the score is assigned 100 points to obtain the health score of the severity index item. ; Calculate the ratio of the total number of failures of all elevators to the age of the elevators, then calculate the mean and standard deviation of all ratios, set the threshold t=mean+3*std, compare each ratio with the threshold t one by one, if any ratio is greater than t, it is regarded as an outlier and replaced with t to obtain a feature column, normalize the feature column, and obtain any element in the normalized feature column , where y represents any value in the feature column Y, represents the minimum value in the feature column Y, Represents the maximum value in the feature column Y; Through formula (7), the health score of the historical annual average total failure index item is obtained : (7); Calculate the number of days since the last failure of each elevator. If there is no failure, the number of days is 365 days. Get the failure sequence. Normalize the failure sequence to get any element in the normalized failure sequence. , where y represents the number of faults Any value in Indicates the number of faults The minimum value in Indicates the number of faults The maximum value in ; Through formula (8), the health score of the elevator failure time interval index item in the past year is obtained : (8); The health score of the severity indicator item , the health score of the historical annual average total failure volume indicator and the health score of the elevator failure time interval indicator in the past year Perform weighted calculation to obtain the elevator health evaluation score in the condition information dimension : (9) in, , , are weight parameters, .

[0009] In one embodiment of the present invention, the operation information dimension elevator health evaluation score The method to obtain is as follows: The operation information data is obtained from the elevator subject database, and is cleaned and merged, and the elevator age of each elevator is normalized to obtain the normalized elevator age. ,in z represents the age of any elevator, min(Z) represents the minimum age of all elevators, max(Z) represents the maximum age of all elevators, and Z represents the sequence consisting of the ages of all elevators; The normalized ladder age Mapped to the corresponding elevator, the health score of the elevator age indicator item is obtained through formula (10): : (10); According to the elevator age level dictionary, the elevator age is mapped to a range between 0 and 1, and the mapping result is multiplied by 100 to obtain the health score of the elevator age level indicator item. ; The health score of the age indicator item and the health score of the age level indicator item Perform weighted calculation to obtain the elevator health evaluation score in the operation information dimension .

[0010] In one embodiment of the present invention, the IoT information dimension elevator health evaluation score The method to obtain is as follows: The IoT information is obtained from the elevator theme library, and is cleaned and merged, and the five indicator data, namely, the cumulative number of operations, the cumulative operation time, the number of wire rope bending times, the number of elevator outages due to faults, and the cumulative number of car door openings, are counted. According to the five statistical indicator data, segmented discretization processing is performed respectively, and the data is mapped to a score range of 0 to 100, and health scores for the cumulative number of operations, the cumulative operation time, the number of wire rope bending times, the number of elevator outages due to faults, and the cumulative number of car door openings are obtained respectively; The calculated health evaluation scores of the five indicators, namely, cumulative operation times, cumulative operation time, number of wire rope bending times, number of elevator outages due to faults, and cumulative number of elevator door opening times, are weighted to obtain the elevator health evaluation score in the IoT information dimension. .

[0011] In one embodiment of the present invention, the maintenance information dimension elevator health evaluation score The method to obtain is as follows: Obtain maintenance information from the elevator subject database, clean and merge it, and determine whether the elevator is factory-maintained or not based on whether the maintenance unit is the same as the manufacturing unit. Assign 100 points to factory-maintained elevators and 0 points to non-factory-maintained elevators, and then multiply it by its weight 0.05 to obtain the health score of the factory-maintained indicator; According to the star rating of the elevator maintenance unit, different star ratings are assigned different scores, and then multiplied by its weight 0.05 to obtain the health score of the maintenance unit star rating indicator; The maintenance unit evaluation index score is the evaluation score obtained from the maintenance unit reporting system, multiplied by its weight 0.15 to obtain the health score of the maintenance unit evaluation index item; The integrity of the maintenance records in the past year is evaluated. If there are no omissions in the maintenance records, 100 points are assigned. If there is one omission in the maintenance records, 80 points are assigned. If there are two omissions in the maintenance records, 60 points are assigned. If there are three or more omissions in the maintenance records, 0 points are assigned. Multiplying the score by its weight of 0.20, the health score of whether the maintenance index item has been maintained according to the prescribed cycle in the past year is obtained. The integrity of the maintenance records in the past three months is evaluated. If there are no omissions in the maintenance records, 100 points are assigned. If there is one omission in the maintenance records, 80 points are assigned. If there are two omissions in the maintenance records, 60 points are assigned. If there are three or more omissions in the maintenance records, 0 points are assigned. Multiplying the score by its weight of 0.40, the health score of whether the maintenance index item has been maintained according to the prescribed cycle in the past three months is obtained. Assign a value based on whether the acquired data field is on-demand maintenance. If it is on-demand maintenance, it is assigned 100 points, and if it is not on-demand maintenance, it is assigned 0 points. Then multiply it by its weight 0.15 to get the health score of the on-demand maintenance indicator item; The calculated health evaluation scores of the six indicators, including whether it is original factory maintenance, maintenance unit star rating, maintenance unit evaluation, whether it is maintained according to the prescribed cycle in the past year, whether it is maintained according to the prescribed cycle in the past three months, and whether it is maintained on demand, are added together to obtain the elevator health evaluation score in the maintenance information dimension. .

[0012] In one embodiment of the present invention, the fault information dimension elevator health evaluation score The method to obtain is as follows: Fault information is obtained from the elevator subject database, and cleaned and merged. According to whether the elevator has experienced a fault within one year, a score of 100 is assigned if no fault has occurred, and a score of 0 is assigned if a fault has occurred. The score is then multiplied by a weight of 0.20 to obtain a health score for the fault indicator item in the past year. The number of elevator failures in the past three months is counted. If there is no failure, 100 points are assigned, if there is one failure, 90 points are assigned, if there is two failures, 80 points are assigned, if there is three failures, 70 points are assigned, if there is four failures, 60 points are assigned, and if there are five failures or more, 0 points are assigned. Then multiply it by its weight 0.80 to get the health score of the failure number indicator item in the past three months. The health evaluation score of the elevator in the fault information dimension is obtained by adding the calculated health evaluation scores of the two indicators, whether there has been a fault in the past year and the number of faults in the past three months. ; The elevator health evaluation score of the trapped person information dimension The method to obtain is as follows: The information on trapped people is obtained from the elevator subject database, and is cleaned and merged. According to whether there has been a trapped person incident in the elevator within one year, a score of 100 is assigned if there has been no trapped person incident, and a score of 0 is assigned if there has been a trapped person incident. The score is then multiplied by a weight of 0.20 to obtain a health score for the indicator item of whether there has been a trapped person incident in the past year. The number of incidents of people being trapped in elevators in the past three months is counted. If it happens 0 times, 100 points are assigned; if it happens once, 80 points are assigned; if it happens twice, 60 points are assigned; if it happens three times or more, 0 points are assigned. Then multiply it by the weight of 0.80 to get the health score of the indicator item of the number of incidents of people being trapped in the past three months; The health evaluation score of the elevator in the dimension of trapped person information is obtained by adding the calculated health evaluation scores of the two indicators: whether there has been any trapped person in the past year and the number of trapped persons in the past three months. ; The feedback information dimension elevator health evaluation score The method to obtain is as follows: The feedback information data is obtained from the elevator subject database, and is cleaned and merged. The complaint information related to the elevator is screened out from the feedback information data, and the corresponding elevator is located. If the corresponding elevator cannot be located, the complaint information is assigned to the elevator at the lowest level of the location. If there are 0 complaints, 100 points are assigned, if there are 1 complaint, 80 points are assigned, if there are 2 complaints, 60 points are assigned, if there are 3 complaints, 40 points are assigned, if there are 4 complaints, 20 points are assigned, and if there are 5 complaints or more, 0 points are assigned. The health score of the indicator item of the number of elevator complaints is obtained, that is, the elevator health evaluation score of the feedback information dimension. .

[0013] In one embodiment of the present invention, in step S4, an evaluation threshold is set, N elevators are randomly selected from the elevators whose evaluation results are less than the evaluation threshold in the elevator health evaluation result library, and early warning reminders are given. According to the preset inspection cycle, relevant supervisors go to the site for supervision and inspection, and the inspection results are synchronized to the elevator theme library. The method is as follows: S41: putting the first M elevators whose evaluation results are less than the evaluation threshold in the elevator health evaluation result database into an early warning push pool; S42: further screening the M elevators put into the early warning push pool, removing the elevators that have been selected in the past three years from the early warning push pool, and obtaining a deduplicated early warning push pool; S43: Determine whether the number of elevators in the warning push pool after deduplication is not less than a preset number Q: If not, if the number of elevators in the warning push pool after deduplication is less than the preset number, return to step S41 and fill in the gaps with elevators with lower scores; If yes, count the number of elevators maintained by the maintenance units to which the elevators belong in the deduplicated early warning push pool, select maintenance units with fewer maintained elevators that are consistent with the preset number, and select one elevator from each of the elevators maintained by these maintenance units to obtain a pre-processed early warning push pool; S44: Determine whether there are Q elevators in the pre-processed warning push pool that belong to the same functional area: If not, retain elevators for different functional areas; If so, the elevators in the same functional area are deduplicated, and only one elevator with a lower elevator health evaluation result in the same functional area is retained; S45: Obtain an optimized early warning push pool, and issue early warning reminders to N elevators in the optimized early warning push pool through a visual interface. At the same time, push the elevator information in the optimized early warning push pool to relevant supervisors, who conduct further on-site inspection and processing based on the pushed information, and synchronously update the elevator inspection results to the elevator theme library.

[0014] Based on the same inventive concept, the present invention also provides an elevator health status evaluation system based on big data, which is used to implement the steps of the elevator health status evaluation method based on big data. The elevator health status evaluation system based on big data includes the following modules: A data acquisition module is used to acquire multi-source heterogeneous data related to elevator operation; A data association module, used to associate data belonging to the same elevator in the multi-source heterogeneous data using the elevator registration code as a unique identification code to form an elevator subject library; An evaluation index calculation module is used to construct an elevator health evaluation model, based on the data records of each elevator in the elevator subject library, and to perform real-time dynamic matching calculations through the elevator health evaluation model to obtain an elevator health evaluation result library; and an elevator early warning module, which is used to set an evaluation threshold, select N elevators from the elevator health evaluation result library whose evaluation results are less than the evaluation threshold, issue early warning reminders, and have relevant supervisors conduct on-site supervision and inspection according to a preset inspection cycle, and synchronize the inspection results to the elevator theme library; The data association module uses the elevator registration code as a unique identification code to associate the data belonging to the same elevator in the multi-source heterogeneous data to form an elevator theme library, including: Collect data sources from the multi-source heterogeneous data at a pre-set time interval, combine the elevator registration code and the timestamp of data generation into a business primary key, and remove duplicate data records by comparing the business primary key of each data source to obtain deduplicated multi-source data; Constructing a field mapping table, and performing format conversion on field data of the same type but different names in the deduplicated multi-source data according to the field mapping table to obtain multi-source data with a unified format; For the newly collected data source, the multi-source data in the unified format is updated according to the business primary key thereof to obtain the updated multi-source data; Based on the updated multi-source data, data records with different data sources but belonging to the same elevator are associated through the elevator registration code to obtain a final elevator theme library.

[0015] The above technical solution of the present invention has the following advantages compared with the prior art: By acquiring multi-source heterogeneous elevator operation data and associating them with the elevator registration code as the unique identifier, an elevator subject library is formed, and an elevator health evaluation model containing multiple dimensions is constructed for real-time dynamic matching calculations to obtain an elevator health evaluation result library. At the same time, evaluation thresholds are set for early warning reminders, and the process of selecting elevators for on-site supervision and inspection is optimized, which effectively improves the accuracy and efficiency of elevator health status evaluation and provides a strong guarantee for the safe operation of elevators. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein: Figure 1 is a flow chart of an elevator health status evaluation method based on big data provided in an embodiment of the present invention; Figure 2 It is a flow chart of the elevator push warning reminder method provided in an embodiment of the present invention; Figure 3 1 is a schematic diagram of the structure of an elevator health status evaluation system based on big data provided in an embodiment of the present invention; Description of the accompanying drawings in the specification: 100, data acquisition module; 200, data association module; 300, evaluation index calculation module; 400, elevator warning module. DETAILED DESCRIPTION

[0017] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention. Embodiment 1

[0018] like Figure 1 As shown, the present invention provides an elevator health status evaluation method based on big data, and the method comprises the following steps: S1: Acquire multi-source heterogeneous data related to elevator operation; S2: Using the elevator registration code as a unique identification code, associating the data belonging to the same elevator in the multi-source heterogeneous data to form an elevator theme library; S3: constructing an elevator health evaluation model, based on the data records of each elevator in the elevator subject library, performing real-time dynamic matching calculation through the elevator health evaluation model to obtain an elevator health evaluation result library; S4: setting an evaluation threshold, randomly selecting N elevators from the elevator health evaluation result database whose evaluation results are less than the evaluation threshold, and issuing early warning reminders. According to the preset inspection cycle, relevant supervisors conduct on-site supervision and inspection, and synchronize the inspection results to the elevator theme database; Among them, in step S2, the method of forming the elevator theme library is as follows: Collect data sources from the multi-source heterogeneous data at a pre-set time interval, combine the elevator registration code and the timestamp of data generation into a business primary key, and remove duplicate data records by comparing the business primary key of each data source to obtain deduplicated multi-source data; Constructing a field mapping table, and performing format conversion on field data of the same type but different names in the deduplicated multi-source data according to the field mapping table to obtain multi-source data with a unified format; For the newly collected data source, the multi-source data in the unified format is updated according to the business primary key thereof to obtain the updated multi-source data; Based on the updated multi-source data, data records with different data sources but belonging to the same elevator are associated through the elevator registration code to obtain a final elevator theme library.

[0019] It can be seen from the above technical solution that the present invention can comprehensively cover various information of elevator operation such as operation logs, fault records, maintenance history, etc. by acquiring multi-source heterogeneous data related to elevator operation, and use the elevator registration code as a unique identification code to associate data from different sources to form an elevator theme library, which effectively integrates scattered data resources and improves data integrity and availability.

[0020] The data source is collected at pre-set time intervals and deduplication is performed using the business primary key (elevator registration code and the timestamp of data generation), thus avoiding repeated calculation of data and improving data processing efficiency.

[0021] By constructing an elevator health evaluation model, we can perform dynamic matching calculations based on the real-time data in the elevator subject library to obtain an elevator health evaluation result library. This real-time dynamic evaluation method can promptly detect elevator health problems and improve the timeliness and accuracy of elevator maintenance.

[0022] Setting an evaluation threshold and issuing early warnings to elevators with evaluation results below the threshold can help to promptly identify potential safety hazards. According to the preset inspection cycle, N elevators are randomly selected from elevators with evaluation results below the threshold for on-site supervision and inspection, ensuring the pertinence and effectiveness of supervision.

[0023] Synchronizing the inspection results to the elevator theme library can continuously update and improve the elevator's health status information, providing a basis for subsequent maintenance and management.

[0024] Furthermore, multi-source heterogeneous data related to elevator operation is obtained from the elevator data center, and the multi-source heterogeneous data originates from different business departments and systems, including two categories: IoT data and non-IoT data. Among them, the IoT data reflects the elevator operation status in real time, including real-time operation information, IoT statistical indicators and IoT fault information; the non-IoT data includes basic elevator information, elevator maintenance unit information, elevator user unit information, inspection report information, hidden danger information, start-up notification information, supervision and inspection information, maintenance unit star information, maintenance record information, on-demand maintenance information, test report information, first registration information, implementation of principal responsibility information, maintenance contract information, elevator trapped information, elevator rescue information, elevator maintenance information, administrative penalty information, credit evaluation information, complaint information and legal person information. These data together constitute a comprehensive and detailed elevator health file.

[0025] In step S2, for the multi-source heterogeneous data after deduplication processing, since different data sources may use different field naming methods, data formats or encoding rules for information with the same meaning, it is necessary to establish a unified field mapping table. For example, in one data source, "Elevator_Manufacturer" is used to represent the elevator manufacturer, while in another data source, "Maker_Name" is used to represent the elevator manufacturer. The field name of "Elevator Manufacturer" is uniformly converted to a standard field name, such as "Manufacturer" through the field mapping table. At the same time, for inconsistent data formats, such as date formats of "YYYYMMDD" and "DD / MM / YYYY", format standardization conversion is performed; for differences in encoding rules, such as elevator fault type encoding, conversion is performed according to a pre-established standard encoding system. According to this mechanism, multi-source data with a unified format is obtained.

[0026] For the newly collected data source, the multi-source data in a unified format is updated according to its business primary key to obtain the updated multi-source data; specifically, the data is judged to be the same elevator record according to the business primary key, and the newly acquired more accurate and complete data is updated to the corresponding record position according to the update timestamp or specific update identification field of the data source; for a brand new elevator record, it is directly inserted into the appropriate storage position of the buffer layer.

[0027] Based on the updated multi-source data, the fusion of multi-source data and the establishment of association relationships are started. With the elevator registration code as the core unique identification code, various data records from different data sources but belonging to the same elevator are associated. For example, the real-time running speed, running direction and other information of the elevator in the IoT data are associated with the elevator model, installation date, maintenance plan and other information in the non-IoT data through the registration code to form a complete elevator data chain. In this process, the accuracy and efficiency of data association are ensured by using the database association query technology and the data fusion function module in the ETL (Extract, Transform, Load) tool.

[0028] Furthermore, in step S3, the method for obtaining the elevator health evaluation result library is as follows: S31: The elevator health evaluation model includes sub-models corresponding to eight dimensions: entity information, status information, operation information, IoT information, maintenance information, fault information, trapped person information, and feedback information. The sub-models are used to calculate the calculation results of each dimension, and weights are assigned to the calculation results of each dimension and then added together to obtain the elevator health evaluation result R of each elevator: R=W 1 ×R 1 +W 2 ×R 2 +W 3 ×R 3 +W 4 ×R 4 +W 5 ×R 5 +W 6 ×R 6 +W 7 ×R 7 +W 8 ×R 8 (1), Among them, R 1 is the elevator health evaluation score in the ontology information dimension, R 2 is the elevator health evaluation score in the condition information dimension, R 3 is the elevator health evaluation score in the operation information dimension, R 4 is the elevator health evaluation score in the IoT information dimension, R 5 is the elevator health evaluation score in the maintenance information dimension, R 6 is the elevator health evaluation score in the fault information dimension, R 7 is the elevator health evaluation score in the dimension of trapped person information, R 8 is the elevator health evaluation score in the feedback information dimension; W 1 , W 2 , W 3 , W 4, W 5 , W 6 , W 7 and W 8 are weight parameters, and W 1 =W 2 =W 3 =W 4 =W 5 =W 6 =W 7 =W 8 =0.125; S32: Summarize the elevator health evaluation results of all elevators to obtain an elevator health evaluation result library.

[0029] Furthermore, the elevator health evaluation score R of the ontology information dimension 1 The method to obtain is as follows: For each elevator, the number of elevators that failed at the manufacturer is compared with the total number of elevators produced by the manufacturer to obtain the failure rate of the elevator manufacturer. , the failure rate of the elevator manufacturer Mapped to the corresponding elevator, the health score of the manufacturer's failure rate indicator item is obtained through formula (2): : (2); Aggregate all the places where elevators are used, count the number of elevator failures in each place, and the total number of elevator failures in all places, and take the ratio of the two statistical results as the place of use failure rate , the failure rate of the place of use Mapped to the corresponding elevator, the health score of the failure rate index item of the use place is obtained through formula (3): : (3); The number of elevator failures within n years after installation is obtained by the time the elevator is put into use. The number of elevators is aggregated according to the construction unit, and the ratio of the number of elevator failures within n years of installation by the construction unit to the total number of elevators installed by the construction unit is taken as the construction unit failure rate. , the construction unit failure rate Mapped to the corresponding elevator, the health score of the construction unit failure rate index item is obtained through formula (4): : (4); The number of failures is aggregated according to the elevator manufacturing unit and elevator model, and the number of elevator failures for each elevator manufacturing unit and elevator model is obtained. The total number of failed elevators is counted according to the elevator manufacturing unit and elevator model, and the ratio of the two statistical results is used as the equipment model failure rate. and the failure rate of the equipment model Mapped to the corresponding elevator, the health score of the equipment model failure rate indicator item is obtained through formula (5) : (5); The health score of the manufacturer's failure rate indicator , the health score of the failure rate indicator item of the place of use , the health score of the construction unit failure rate indicator and the health score of the failure rate indicator of the device model Perform weighted calculation to obtain the elevator health evaluation score R in the ontology information dimension 1 : (6) in, , , , are weight parameters, .

[0030] Furthermore, the status information dimension elevator health evaluation score The method to obtain is as follows: Different points are deducted based on the deaths, entrapments, and malfunctions that occurred in the elevator within multiple specified time periods. If a death, entrapment, or malfunction occurred within 1 year, 5 points, 4 points, and 2 points will be deducted respectively; if a death, entrapment, or malfunction occurred within 2 years, 4 points, 3 points, and 1 point will be deducted respectively; if a death, entrapment, or malfunction occurred within 3 years, 3 points, 2 points, and 0 points will be deducted respectively; if a death, entrapment, or malfunction occurred within 4 years, 2 points, 1 point, and 0 points will be deducted respectively; if a death, entrapment, or malfunction occurred within 5 years, 1 point, 0 point, and 0 point will be deducted respectively. If no death, entrapment, or malfunction occurred, 100 points will be assigned to obtain the health score of the severity indicator item. ; Calculate the ratio of the total number of failures of all elevators to the age of the elevators, then calculate the mean and standard deviation of all ratios, set the threshold t=mean+3*std, compare each ratio with the threshold t one by one, if any ratio is greater than t, it is regarded as an outlier and replaced by t. Through this process, a processed feature column is obtained, and the feature column is normalized to obtain any element in the normalized feature column. , where y represents any value in the feature column Y, represents the minimum value in the feature column Y, Represents the maximum value in the feature column Y; Through formula (7), the health score of the historical annual average total failure index item is obtained : (7); Calculate the number of days since the last failure of each elevator. If there is no failure, the number of days is 365 days. Get the failure sequence. Normalize the failure sequence to get any element in the normalized failure sequence. , where y represents the number of faults Any value in Indicates the number of faults The minimum value in Indicates the number of faults The maximum value in ; Through formula (8), the health score of the elevator failure time interval index item in the past year is obtained : (8); The health score of the severity indicator item , the health score of the historical annual average total failure volume indicator and the health score of the elevator failure time interval indicator in the past year Perform weighted calculation to obtain the elevator health evaluation score in the condition information dimension : (9) in, , , are weight parameters, , .

[0031] Furthermore, the elevator health evaluation score of the operation information dimension The method to obtain is as follows: The operation information data is obtained from the elevator subject database, and is cleaned and merged, and the elevator age of each elevator is normalized to obtain the normalized elevator age. ,in z represents the age of any elevator, min(Z) represents the minimum age of all elevators, max(Z) represents the maximum age of all elevators, and Z represents the sequence consisting of the ages of all elevators; The normalized ladder age Mapped to the corresponding elevator, the health score of the elevator age indicator item is obtained through formula (10): : (10); According to the elevator age level dictionary, the elevator age is mapped to a range between 0 and 1, and the mapping result is multiplied by 100 to obtain the health score of the elevator age level indicator item. ; The health score of the age indicator item and the health score of the age level indicator item Perform weighted calculation to obtain the elevator health evaluation score in the operation information dimension .

[0032] In this embodiment, the IoT information dimension elevator health evaluation score The method to obtain is as follows: The IoT information is obtained from the elevator theme library, and is cleaned and merged. The five indicator data, namely, the cumulative number of operations, the cumulative operating time, the number of wire rope bending times, the number of elevator outages caused by faults, and the cumulative number of elevator door opening times, are counted. The five indicator data are discretized using quartiles. According to the statistical data of each indicator item, the data is sorted from small to large, and then divided into different intervals according to the quantiles. Different intervals correspond to different scores. In this way, each indicator can be mapped to between 0 and 100 points, and the cumulative number of operations and the cumulative operating time are obtained respectively. , the health scores of the indicators of wire rope bending times, elevator outage times caused by faults, and cumulative door opening times; for example, the maximum cumulative running time of all elevators is 10,000 minutes, which is divided into four scoring intervals: 0-2,500 minutes (including 2,500 minutes) is 100 points, 250-5,000 minutes (including 5,000 minutes) is 80 points, 5,000-7,500 minutes (including 7,500 minutes) is 60 points, and 7,500-10,000 minutes (including 10,000 minutes) is 40 points; the same scoring rules are used for other indicators; The calculated health evaluation scores of the five indicators, namely, cumulative operation times, cumulative operation time, number of wire rope bending times, number of elevator outages due to faults, and cumulative number of elevator door opening times, are weighted to obtain the elevator health evaluation score in the IoT information dimension. .

[0033] Furthermore, the elevator health evaluation score of the maintenance information dimension The method to obtain is as follows: Obtain maintenance information from the elevator subject database, clean and merge it, and determine whether the elevator is factory-maintained or non-factory-maintained based on whether the maintenance unit is the same as the manufacturing unit. If the maintenance unit is the same as the manufacturing unit, it is factory-maintained, and if the maintenance unit is different from the manufacturing unit, it is non-factory-maintained. Elevators with factory maintenance are assigned 100 points, and elevators with non-factory maintenance are assigned 0 points. Multiply it by its weight 0.05 to obtain the health score of the factory maintenance indicator; The star ratings of maintenance units are two-star, three-star, four-star, five-star and no star rating, and there is no one-star rating; the higher the star rating, the higher the maintenance capability of the maintenance unit. Different star ratings are assigned different scores according to the star rating of the elevator maintenance unit. Five-star ratings are 100 points; four-star ratings are 80 points; three-star ratings are 60 points; two-star ratings are 40 points, and no star rating is no score. Multiply it by its weight 0.05 to get the health score of the maintenance unit star rating indicator item; The maintenance unit evaluation index score is the evaluation score obtained from the maintenance unit reporting system, multiplied by its weight 0.15 to obtain the health score of the maintenance unit evaluation index item; The integrity of the maintenance records in the past year is evaluated. If there are no omissions in the maintenance records, 100 points are assigned. If there is one omission in the maintenance records, 80 points are assigned. If there are two omissions in the maintenance records, 60 points are assigned. If there are three or more omissions in the maintenance records, 0 points are assigned. Multiplying the score by its weight of 0.20, the health score of whether the maintenance index item has been maintained according to the prescribed cycle in the past year is obtained. The integrity of the maintenance records in the past three months is evaluated. If there are no omissions in the maintenance records, 100 points are assigned. If there is one omission in the maintenance records, 80 points are assigned. If there are two omissions in the maintenance records, 60 points are assigned. If there are three or more omissions in the maintenance records, 0 points are assigned. Multiplying the score by its weight of 0.40, the health score of whether the maintenance index item has been maintained according to the prescribed cycle in the past three months is obtained. Assign a value based on whether the acquired data field is on-demand maintenance. If it is on-demand maintenance, it is assigned 100 points, and if it is not on-demand maintenance, it is assigned 0 points. Then multiply it by its weight 0.15 to get the health score of the on-demand maintenance indicator item; The calculated health evaluation scores of the six indicators, including whether it is original factory maintenance, maintenance unit star rating, maintenance unit evaluation, whether it is maintained according to the prescribed cycle in the past year, whether it is maintained according to the prescribed cycle in the past three months, and whether it is maintained on demand, are added together to obtain the elevator health evaluation score in the maintenance information dimension. .

[0034] Furthermore, the fault information dimension elevator health evaluation score The method to obtain is as follows: Fault information is obtained from the elevator subject database, and cleaned and merged. According to whether the elevator has experienced a fault within one year, a score of 100 is assigned if no fault has occurred, and a score of 0 is assigned if a fault has occurred. The score is then multiplied by a weight of 0.20 to obtain a health score for the fault indicator item in the past year. The number of elevator failures in the past three months is counted. If there is no failure, 100 points are assigned, if there is one failure, 90 points are assigned, if there is two failures, 80 points are assigned, if there is three failures, 70 points are assigned, if there is four failures, 60 points are assigned, and if there are five failures or more, 0 points are assigned. Then multiply it by its weight 0.80 to get the health score of the failure number indicator item in the past three months. The health evaluation score of the elevator in the fault information dimension is obtained by adding the calculated health evaluation scores of the two indicators, whether there has been a fault in the past year and the number of faults in the past three months. .

[0035] Furthermore, the elevator health evaluation score of the trapped person information dimension The method to obtain is as follows: The information on trapped people is obtained from the elevator subject database, and is cleaned and merged. According to whether there has been a trapped person incident in the elevator within one year, a score of 100 is assigned if there has been no trapped person incident, and a score of 0 is assigned if there has been a trapped person incident. The score is then multiplied by a weight of 0.20 to obtain a health score for the indicator item of whether there has been a trapped person incident in the past year. The number of incidents of people being trapped in elevators in the past three months is counted. If it happens 0 times, 100 points are assigned; if it happens once, 80 points are assigned; if it happens twice, 60 points are assigned; if it happens three times or more, 0 points are assigned. Then multiply it by the weight of 0.80 to get the health score of the indicator item of the number of incidents of people being trapped in the past three months; The health evaluation score of the elevator in the dimension of trapped person information is obtained by adding the calculated health evaluation scores of the two indicators: whether there has been any trapped person in the past year and the number of trapped persons in the past three months. .

[0036] Furthermore, the feedback information dimension elevator health evaluation score The method to obtain is as follows: The feedback information data from the 12345 official platform is obtained from the elevator theme library, and is cleaned and merged. The complaint information related to the elevator is screened out from the feedback information data according to the pre-trained classifier, and the corresponding elevator is located. If the corresponding elevator cannot be located, the complaint information is assigned to the smallest level of the located elevator. 100 points are assigned if there are 0 complaints, 80 points if there are 1 complaints, 60 points if there are 2 complaints, 40 points if there are 3 complaints, 20 points if there are 4 complaints, and 0 points if there are 5 complaints or more. The health score of the elevator complaint number indicator is obtained, that is, the elevator health evaluation score of the feedback information dimension. .

[0037] In addition to the above technical solutions, in the process of calculating the calculation results of each dimension through sub-models of eight dimensions, namely, entity information, status information, operation information, Internet of Things information, maintenance information, fault information, trapped information, and feedback information, effective management of the algorithm model is also realized, mainly including two aspects: model feature management and algorithm model management.

[0038] In terms of model feature management, we continuously monitor the performance of the model indicators, and promptly detect and handle abnormal situations such as overfitting and underfitting. At the same time, we iteratively optimize the model based on the latest data updates to continuously improve its accuracy and reliability.

[0039] In terms of algorithm model management, based on the model's monitoring results, accumulated business experience and new business needs, the algorithm model is processed through parameter adjustment, algorithm optimization, data enhancement, etc., in order to improve the model's performance indicators and accuracy.

[0040] In order to support the rapid response and efficient operation of the algorithm model, a model scheduling engine was introduced to uniformly manage and schedule the model, allocate resources and execute it. Specifically, the following management and scheduling methods were adopted: First, the models are managed by categories, or the algorithm features are categorized so that users can quickly find and use the required models.

[0041] Secondly, a version management function is provided for each model, which records each update and modification of the model in detail, so that users can choose to use a specific version of the model for evaluation or analysis as needed.

[0042] In addition, for complex algorithmic models, identify and manage the dependencies between multiple sub-models or preprocessing steps that may exist to ensure that the model can be executed in the correct order and under the correct conditions.

[0043] In terms of resource allocation and execution, it supports parallel and distributed execution of large-scale data processing or complex model calculations. By splitting tasks into multiple subtasks and processing them in parallel on multiple nodes, the computational efficiency of health assessment is significantly improved.

[0044] like Figure 2 As shown, in step S4, an evaluation threshold is set, N elevators are randomly selected from the elevators whose evaluation results are less than the evaluation threshold in the elevator health evaluation result library, and early warning reminders are given. According to the preset inspection cycle, relevant supervisors go to the site for supervision and inspection, and the inspection results are synchronized to the elevator theme library as follows: S41: putting the first 20 elevators in the elevator health evaluation result database whose evaluation results are less than the evaluation threshold into an early warning push pool; S42: To avoid repeated warnings for the same elevator and ensure effective use of resources, the 20 elevators in the warning push pool are further screened, and elevators that have been selected in the past three years are removed from the warning push pool to obtain a deduplicated warning push pool; S43: Determine whether the number of elevators in the warning push pool after deduplication is not less than 5: If not, if the number of elevators in the warning push pool after deduplication is less than 5, return to step S41 and complete it with elevators with low scores; If yes, count the number of elevators maintained by the maintenance units to which the elevators belong in the deduplicated early warning push pool, select five maintenance units with fewer maintained elevators, and select one elevator each from the elevators maintained by these five maintenance units to obtain a preprocessed early warning push pool; S44: Determine whether there are any elevators in the pre-processed warning push pool that belong to the same functional area: If not, retain elevators for different functional areas; If so, the elevators in the same functional area are deduplicated, and only one elevator with a lower elevator health evaluation result in the same functional area is retained; S45: Obtain an optimized early warning push pool, and issue early warning reminders to N elevators in the optimized early warning push pool through a visual interface. At the same time, push the elevator information in the optimized early warning push pool to relevant supervisors, who conduct further on-site inspection and processing based on the pushed information to ensure the safe operation of the elevators, and synchronously update the inspection results of the elevators to the elevator theme library. Embodiment 2

[0045] Based on the same inventive concept as that of the first embodiment, the present invention also provides an elevator health status evaluation system based on big data, which is used to implement the steps of the elevator health status evaluation method based on big data described in the first embodiment. Figure 3 As shown, the elevator health status evaluation system based on big data includes the following modules: The data acquisition module 100 is used to acquire multi-source heterogeneous data related to elevator operation; The data association module 200 is used to associate the data belonging to the same elevator in the multi-source heterogeneous data using the elevator registration code as a unique identification code to form an elevator theme library; An evaluation index calculation module 300 is used to construct an elevator health evaluation model, based on the data records of each elevator in the elevator subject library, and perform real-time dynamic matching calculation through the elevator health evaluation model to obtain an elevator health evaluation result library; and an elevator early warning module 400, which is used to set an evaluation threshold, select N elevators from the elevators whose evaluation results are less than the evaluation threshold in the elevator health evaluation result library, issue early warning reminders, and have relevant supervisors conduct on-site supervision and inspection according to a preset inspection cycle, and synchronize the inspection results to the elevator theme library; The data association module uses the elevator registration code as a unique identification code to associate the data belonging to the same elevator in the multi-source heterogeneous data to form an elevator theme library, including: Collect data sources from the multi-source heterogeneous data at a pre-set time interval, combine the elevator registration code and the timestamp of data generation into a business primary key, and remove duplicate data records by comparing the business primary key of each data source to obtain deduplicated multi-source data; Constructing a field mapping table, and performing format conversion on field data of the same type but different names in the deduplicated multi-source data according to the field mapping table to obtain multi-source data with a unified format; For the newly collected data source, the multi-source data in the unified format is updated according to the business primary key thereof to obtain the updated multi-source data; Based on the updated multi-source data, data records with different data sources but belonging to the same elevator are associated through the elevator registration code to obtain a final elevator theme library.

[0046] The present embodiment proposes an elevator health status evaluation system based on big data, which is used to implement the aforementioned elevator health status evaluation method based on big data. Therefore, the specific implementation method of the elevator health status evaluation system based on big data can be seen in the embodiment part of the aforementioned elevator health status evaluation method based on big data. For example, the data acquisition module 100, the data association module 200, the evaluation index calculation module 300 and the elevator early warning module 400 are respectively used to implement the steps S1, S2, S3 and S4 in the elevator health status evaluation method based on big data in the first embodiment. Therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part. In order to avoid redundancy, it will not be repeated here.

[0047] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0048] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0049] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0050] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0051] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A method for evaluating the health status of an elevator based on big data, characterized in that: The following steps are involved: S1: Acquire multi-source heterogeneous data related to elevator operation; S2: Using the elevator registration code as a unique identification code, associating the data belonging to the same elevator in the multi-source heterogeneous data to form an elevator theme library; S3: constructing an elevator health evaluation model, based on the data records of each elevator in the elevator subject library, performing real-time dynamic matching calculation through the elevator health evaluation model to obtain an elevator health evaluation result library; S4: setting an evaluation threshold, randomly selecting N elevators from the elevator health evaluation result database whose evaluation results are less than the evaluation threshold, and issuing early warning reminders. According to the preset inspection cycle, relevant supervisors conduct on-site supervision and inspection, and synchronize the inspection results to the elevator theme database; Among them, in step S2, the method of forming the elevator theme library is as follows: Collect data sources from the multi-source heterogeneous data at a pre-set time interval, combine the elevator registration code and the timestamp of data generation into a business primary key, and remove duplicate data records by comparing the business primary key of each data source to obtain deduplicated multi-source data; Constructing a field mapping table, and performing format conversion on field data of the same type but different names in the deduplicated multi-source data according to the field mapping table to obtain multi-source data with a unified format; For the newly collected data source, the multi-source data in the unified format is updated according to the business primary key thereof to obtain the updated multi-source data; Based on the updated multi-source data, data records with different data sources but belonging to the same elevator are associated through the elevator registration code to obtain a final elevator theme library.

2. The elevator health status evaluation method based on big data according to claim 1 is characterized in that: In step S3, the method for obtaining the elevator health evaluation result library is as follows: S31: The elevator health evaluation model includes sub-models corresponding to eight dimensions: entity information, status information, operation information, IoT information, maintenance information, fault information, trapped person information, and feedback information. The sub-models are used to calculate the calculation results of each dimension, and weights are assigned to the calculation results of each dimension and then added together to obtain the elevator health evaluation result R of each elevator: R=W1×R1+W2×R2+W3×R3+W4×R4+W5×R5+W6×R6+W7×R7+W8×R8 (1), Among them, R1 is the elevator health evaluation score in the entity information dimension, R2 is the elevator health evaluation score in the condition information dimension, R3 is the elevator health evaluation score in the operation information dimension, R4 is the elevator health evaluation score in the IoT information dimension, R5 is the elevator health evaluation score in the maintenance information dimension, R6 is the elevator health evaluation score in the fault information dimension, R7 is the elevator health evaluation score in the trapped person information dimension, and R8 is the elevator health evaluation score in the feedback information dimension; W1, W2, W3, W4, W5, W6, W7 and W8 are all weight parameters; S32: Summarize the elevator health evaluation results of all elevators to obtain an elevator health evaluation result library.

3. The elevator health status evaluation method based on big data according to claim 2 is characterized in that: The method for obtaining the elevator health evaluation score R1 in the ontology information dimension is as follows: For each elevator, the number of elevators that failed at the manufacturer is compared with the total number of elevators produced by the manufacturer to obtain the failure rate of the elevator manufacturer. , the failure rate of the elevator manufacturer Mapped to the corresponding elevator, the health score of the manufacturer's failure rate indicator item is obtained through formula (2): : （2）； Aggregate all the places where elevators are used, count the number of elevator failures in each place, and the total number of elevator failures in all places, and take the ratio of the two statistical results as the place of use failure rate , the failure rate of the place of use Mapped to the corresponding elevator, the health score of the failure rate index item of the use place is obtained through formula (3): : （3）； The number of elevator failures within n years after installation is obtained by the time the elevator is put into use. The number of elevators is aggregated according to the construction unit, and the ratio of the number of elevator failures within n years of installation by the construction unit to the total number of elevators installed by the construction unit is taken as the construction unit failure rate. , the construction unit failure rate Mapped to the corresponding elevator, the health score of the construction unit failure rate index item is obtained through formula (4): : (4)； The number of failures is aggregated according to the elevator manufacturing unit and elevator model, and the number of elevator failures for each elevator manufacturing unit and elevator model is obtained. The total number of failed elevators is counted according to the elevator manufacturing unit and elevator model, and the ratio of the two statistical results is used as the equipment model failure rate. and the failure rate of the equipment model Mapped to the corresponding elevator, the health score of the equipment model failure rate indicator item is obtained through formula (5) : (5)； The health score of the manufacturer's failure rate indicator , the health score of the failure rate indicator item of the place of use , the health score of the construction unit failure rate indicator and the health score of the failure rate indicator of the device model Perform weighted calculation to obtain the elevator health evaluation score R1 in the ontology information dimension: (6)， in, , , , are weight parameters, .

4. The elevator health status evaluation method based on big data according to claim 2 is characterized in that: The elevator health evaluation score of the status information dimension The method to obtain is as follows: Different scores are deducted based on the number of deaths, entrapments, and malfunctions that occur in the elevator within a specified time period. If there are no deaths, entrapments, or malfunctions, the score is assigned 100 points to obtain the health score of the severity index item. ; Calculate the ratio of the total number of failures of all elevators to the age of the elevators, then calculate the mean and standard deviation of all ratios, set the threshold t=mean+3*std, compare each ratio with the threshold t one by one, if any ratio is greater than t, it is regarded as an outlier and replaced with t to obtain a feature column, normalize the feature column, and obtain any element in the normalized feature column , where y represents any value in the feature column Y, represents the minimum value in the feature column Y, Represents the maximum value in the feature column Y; Through formula (7), the health score of the historical annual average total failure index item is obtained : (7)； Calculate the number of days since the last failure of each elevator. If there is no failure, the number of days is 365 days. Get the failure sequence. Normalize the failure sequence to get any element in the normalized failure sequence. , where y represents the number of faults Any value in Indicates the number of faults The minimum value in Indicates the number of faults The maximum value in ; Through formula (8), the health score of the elevator failure time interval index item in the past year is obtained : (8)； The health score of the severity indicator item , the health score of the historical annual average total failure volume indicator and the health score of the elevator failure time interval indicator in the past year Perform weighted calculation to obtain the elevator health evaluation score in the condition information dimension : (9)， in, , , are weight parameters, .

5. The elevator health status evaluation method based on big data according to claim 2 is characterized in that: The elevator health evaluation score of the operation information dimension The method to obtain is as follows: The operation information data is obtained from the elevator subject database, and is cleaned and merged, and the elevator age of each elevator is normalized to obtain the normalized elevator age. ,in z represents the age of any elevator, min(Z) represents the minimum age of all elevators, max(Z) represents the maximum age of all elevators, and Z represents the sequence consisting of the ages of all elevators; The normalized ladder age Mapped to the corresponding elevator, the health score of the elevator age indicator item is obtained through formula (10): : (10)； According to the elevator age level dictionary, the elevator age is mapped to a range between 0 and 1, and the mapping result is multiplied by 100 to obtain the health score of the elevator age level indicator item. ; The health score of the age indicator item and the health score of the age level indicator item Perform weighted calculation to obtain the elevator health evaluation score in the operation information dimension .

6. The elevator health status evaluation method based on big data according to claim 2 is characterized in that: Elevator health evaluation score in IoT information dimension The method to obtain is as follows: The IoT information is obtained from the elevator theme library, and is cleaned and merged, and the five indicator data, namely, the cumulative number of operations, the cumulative operation time, the number of wire rope bending times, the number of elevator outages due to faults, and the cumulative number of car door openings, are counted. According to the five statistical indicator data, segmented discretization processing is performed respectively, and the data is mapped to a score range of 0 to 100, and health scores for the cumulative number of operations, the cumulative operation time, the number of wire rope bending times, the number of elevator outages due to faults, and the cumulative number of car door openings are obtained respectively; The calculated health evaluation scores of the five indicators, namely, cumulative operation times, cumulative operation time, number of wire rope bending times, number of elevator outages due to faults, and cumulative number of elevator door opening times, are weighted to obtain the elevator health evaluation score in the IoT information dimension. .

7. The elevator health status evaluation method based on big data according to claim 2 is characterized in that: Elevator health evaluation score in the maintenance information dimension The method to obtain is as follows: Obtain maintenance information from the elevator subject database, clean and merge it, and determine whether the elevator is factory-maintained or not based on whether the maintenance unit is the same as the manufacturing unit. Assign 100 points to factory-maintained elevators and 0 points to non-factory-maintained elevators, and then multiply it by its weight 0.05 to obtain the health score of the factory-maintained indicator; According to the star rating of the elevator maintenance unit, different star ratings are assigned different scores, and then multiplied by its weight 0.05 to obtain the health score of the maintenance unit star rating indicator; The maintenance unit evaluation index score is the evaluation score obtained from the maintenance unit reporting system, multiplied by its weight 0.15 to obtain the health score of the maintenance unit evaluation index item; The integrity of the maintenance records in the past year is evaluated. If there are no omissions in the maintenance records, 100 points are assigned. If there is one omission in the maintenance records, 80 points are assigned. If there are two omissions in the maintenance records, 60 points are assigned. If there are three or more omissions in the maintenance records, 0 points are assigned. Multiplying the score by its weight of 0.20, the health score of whether the maintenance index item has been maintained according to the prescribed cycle in the past year is obtained. The integrity of the maintenance records in the past three months is evaluated. If there are no omissions in the maintenance records, 100 points are assigned. If there is one omission in the maintenance records, 80 points are assigned. If there are two omissions in the maintenance records, 60 points are assigned. If there are three or more omissions in the maintenance records, 0 points are assigned. Multiplying the score by its weight of 0.40, the health score of whether the maintenance index item has been maintained according to the prescribed cycle in the past three months is obtained. Assign a value based on whether the acquired data field is on-demand maintenance. If it is on-demand maintenance, it is assigned 100 points, and if it is not on-demand maintenance, it is assigned 0 points. Then multiply it by its weight 0.15 to get the health score of the on-demand maintenance indicator item; The calculated health evaluation scores of the six indicators, including whether it is original factory maintenance, maintenance unit star rating, maintenance unit evaluation, whether it is maintained according to the prescribed cycle in the past year, whether it is maintained according to the prescribed cycle in the past three months, and whether it is maintained on demand, are added together to obtain the elevator health evaluation score in the maintenance information dimension. .

8. The elevator health status evaluation method based on big data according to claim 2 is characterized in that: The elevator health evaluation score of the fault information dimension The method to obtain is as follows: Fault information is obtained from the elevator subject database, and cleaned and merged. According to whether the elevator has experienced a fault within one year, a score of 100 is assigned if no fault has occurred, and a score of 0 is assigned if a fault has occurred. The score is then multiplied by a weight of 0.20 to obtain a health score for the fault indicator item in the past year. The number of elevator failures in the past three months is counted. If there is no failure, 100 points are assigned, if there is one failure, 90 points are assigned, if there is two failures, 80 points are assigned, if there is three failures, 70 points are assigned, if there is four failures, 60 points are assigned, and if there are five failures or more, 0 points are assigned. Then multiply it by its weight 0.80 to get the health score of the failure number indicator item in the past three months. The health evaluation score of the elevator in the fault information dimension is obtained by adding the calculated health evaluation scores of the two indicators: whether there has been a fault in the past year and the number of faults in the past three months. ; The elevator health evaluation score of the trapped person information dimension The method to obtain is as follows: The information on trapped people is obtained from the elevator subject database, and is cleaned and merged. According to whether there has been a trapped person incident in the elevator within one year, a score of 100 is assigned if there has been no trapped person incident, and a score of 0 is assigned if there has been a trapped person incident. The score is then multiplied by a weight of 0.20 to obtain a health score for the indicator item of whether there has been a trapped person incident in the past year. The number of incidents of people being trapped in elevators in the past three months is counted. If it happens 0 times, 100 points are assigned; if it happens once, 80 points are assigned; if it happens twice, 60 points are assigned; if it happens three times or more, 0 points are assigned. Then multiply it by the weight of 0.80 to get the health score of the indicator item of the number of incidents of people being trapped in the past three months; The health evaluation score of the elevator in the dimension of trapped person information is obtained by adding the calculated health evaluation scores of the two indicators: whether there has been any trapped person in the past year and the number of trapped persons in the past three months. ; The feedback information dimension elevator health evaluation score The method to obtain is as follows: The feedback information data is obtained from the elevator subject database, and is cleaned and merged. The complaint information related to the elevator is screened out from the feedback information data, and the corresponding elevator is located. If the corresponding elevator cannot be located, the complaint information is assigned to the elevator at the lowest level of the location. If there are 0 complaints, 100 points are assigned, if there are 1 complaint, 80 points are assigned, if there are 2 complaints, 60 points are assigned, if there are 3 complaints, 40 points are assigned, if there are 4 complaints, 20 points are assigned, and if there are 5 complaints or more, 0 points are assigned. The health score of the indicator item of the number of elevator complaints is obtained, that is, the elevator health evaluation score of the feedback information dimension. .

9. The elevator health status evaluation method based on big data according to claim 1, characterized in that: In step S4, an evaluation threshold is set, and N elevators are randomly selected from the elevators whose evaluation results are less than the evaluation threshold in the elevator health evaluation result library, and early warning reminders are given. According to the preset inspection cycle, relevant supervisors go to the site for supervision and inspection, and the inspection results are synchronized to the elevator theme library as follows: S41: putting the first M elevators whose evaluation results are less than the evaluation threshold in the elevator health evaluation result database into an early warning push pool; S42: further screening the M elevators put into the early warning push pool, removing the elevators that have been selected in the past three years from the early warning push pool, and obtaining a deduplicated early warning push pool; S43: Determine whether the number of elevators in the warning push pool after deduplication is not less than a preset number Q: If not, if the number of elevators in the warning push pool after deduplication is less than the preset number, return to step S41 and fill in the gaps with elevators with lower scores; If yes, count the number of elevators maintained by the maintenance units to which the elevators belong in the deduplicated early warning push pool, select maintenance units with fewer maintained elevators that are consistent with the preset number, and select one elevator from each of the elevators maintained by these maintenance units to obtain a pre-processed early warning push pool; S44: Determine whether there are Q elevators in the pre-processed warning push pool that belong to the same functional area: If not, retain elevators for different functional areas; If so, the elevators in the same functional area are deduplicated, and only one elevator with a lower elevator health evaluation result in the same functional area is retained; S45: Obtain an optimized early warning push pool, and issue early warning reminders to N elevators in the optimized early warning push pool through a visual interface. At the same time, push the elevator information in the optimized early warning push pool to relevant supervisors, who conduct further on-site inspection and processing based on the pushed information, and synchronously update the elevator inspection results to the elevator theme library.

10. An elevator health status evaluation system based on big data, characterized in that: The steps for implementing the method for evaluating the health status of an elevator based on big data according to any one of claims 1 to 9, wherein the system for evaluating the health status of an elevator based on big data comprises the following modules: A data acquisition module is used to acquire multi-source heterogeneous data related to elevator operation; A data association module, used to associate data belonging to the same elevator in the multi-source heterogeneous data using the elevator registration code as a unique identification code to form an elevator subject library; An evaluation index calculation module is used to construct an elevator health evaluation model, based on the data records of each elevator in the elevator subject library, and to perform real-time dynamic matching calculations through the elevator health evaluation model to obtain an elevator health evaluation result library; and an elevator early warning module, which is used to set an evaluation threshold, select N elevators from the elevator health evaluation result library whose evaluation results are less than the evaluation threshold, issue early warning reminders, and have relevant supervisors conduct on-site supervision and inspection according to a preset inspection cycle, and synchronize the inspection results to the elevator theme library; The data association module uses the elevator registration code as a unique identification code to associate the data belonging to the same elevator in the multi-source heterogeneous data to form an elevator theme library, including: Collect data sources from the multi-source heterogeneous data at a pre-set time interval, combine the elevator registration code and the timestamp of data generation into a business primary key, and remove duplicate data records by comparing the business primary key of each data source to obtain deduplicated multi-source data; Constructing a field mapping table, and performing format conversion on field data of the same type but different names in the deduplicated multi-source data according to the field mapping table to obtain multi-source data with a unified format; For the newly collected data source, the multi-source data in the unified format is updated according to the business primary key thereof to obtain the updated multi-source data; Based on the updated multi-source data, data records with different data sources but belonging to the same elevator are associated through the elevator registration code to obtain a final elevator theme library.

Images