CN115730850B - Rescue unit automatic recommendation method for intelligent emergency disposal process of elevator - Google Patents
Rescue unit automatic recommendation method for intelligent emergency disposal process of elevator Download PDFInfo
- Publication number
- CN115730850B CN115730850B CN202211484455.6A CN202211484455A CN115730850B CN 115730850 B CN115730850 B CN 115730850B CN 202211484455 A CN202211484455 A CN 202211484455A CN 115730850 B CN115730850 B CN 115730850B
- Authority
- CN
- China
- Prior art keywords
- maintenance
- elevator
- rescue
- unit
- maintenance unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention discloses an automatic recommending method of rescue units for an intelligent emergency disposal process of an elevator, which comprises the steps of firstly, taking the position of the elevator with a fault as the center, taking a set distance R as the radius, forming a searching area, and obtaining the information of all elevator maintenance units in the area; 2. constructing a rescue capability evaluation system model of an elevator maintenance unit; 3. and (3) comprehensively evaluating all elevator maintenance in the first step by using a rescue capability evaluation system to obtain an optimal scheme, and rescue the faulty elevator by taking the optimal scheme as a recommended unit. According to the invention, the rescue capability of each maintenance unit is comprehensively considered through multiple indexes, rescue dispatch task recommendation is carried out, the most reasonable rescue units are allocated for the failed elevator, and professional maintenance personnel are rapidly dispatched to enable trapped masses to escape, so that the elevator failure rescue speed can be accelerated, the elevator rescue resource utilization efficiency is improved, the secondary hazard is reduced, and unnecessary casualties are avoided; the problem of maintenance unit rescue task distribution is uneven, unreasonable is solved.
Description
Technical Field
The invention belongs to the technical field of elevator safety, and particularly relates to an automatic rescue unit recommending method for an intelligent emergency disposal process of an elevator.
Background
At present, when an elevator has a trapped fault, passengers in a car dial calls to an elevator emergency disposal platform, and platform staff call maintenance staff to rescue on site. The rescue unit is basically a signing maintenance unit of the elevator, and when a signing maintenance party cannot implement rescue, a maintenance rescue station closest to the grid rescue station and the geographic position of the accident elevator is selected for emergency maintenance.
For example, the Hangzhou elevator intelligent rescue platform has all the Hangzhou elevator maintenance rescue workers incorporated therein, and the platform can automatically dispatch the service rescue workers closest to the Hangzhou elevator intelligent rescue workers on the basis of an elevator accurate electronic map. When a person is trapped in the elevator, the elevator is fed back to the Hangzhou elevator safety pass software, the nearby elevator maintenance personnel can receive the order through the software, the specific position of the trapped passenger and the route pattern of the elevator for driving the trapped passenger can be seen after receiving the order, and the trapped passenger can see the contact telephone of the rescue personnel and the real-time distance from the elevator through the software. In 2016, beijing test points are a repair mode of robbing a bill, by 2017, 4 months, and more than 500 elevator maintenance workers are trained to participate in the process, scattered maintenance workers are brought into a unified platform to be mobilized, citizens can "order" through an emergency platform, and the maintenance workers are repair nearby. In 2019, the Beijing test point nearby dispatch rescue mode is developed by combining the Beijing elevator APP with an elevator emergency treatment center and referring to the dripping taxi taking mode. When the elevator is in trouble, the emergency disposal center of the elevator dispatches the contracted maintenance unit of the elevator as a first-level rescue unit, and can directly find the rescue personnel with the nearest positioning distance on the Nanjing elevator APP to rescue. If the first-level rescue is not responded in time, the '96333' platform staff can start the second-level grid rescue, search for 3 to 5 rescue stations and people closest to each other by taking the am position coordinate of the fault elevator as the center, push emergency treatment information and send orders nearby.
However, the rescue dispatch recommendation method only considering the distance has the following problems: 1) The rescue units closest to the elevator brands, elevator characteristics and elevator environments are possibly unfamiliar, and although the rescue units can arrive at the scene earlier, the rescue units often cannot be high in efficiency in the rescue process, even the corresponding rescue tasks cannot be completed, and larger economic loss and casualties can be caused; 2) Due to the uneven elevator arrangement, the number and population pressure of elevators responsible by maintenance units in high-population residential areas, commercial areas and low-density remote suburbs are not the same, resulting in uneven rescue task allocation. Rescue tasks are sent nearby, so that maintenance teams in residential areas and business areas frequently execute tasks, and maintenance units relatively far away often cannot obtain rescue tasks, and the problems of personnel idling, low utilization rate and the like occur.
Disclosure of Invention
The invention aims to provide an automatic rescue unit recommending method for an elevator intelligent emergency treatment process, which solves the technical problems of nonuniform allocation and low utilization rate of rescue tasks and unreasonable rescue tasks caused by recommending elevator rescue tasks only considering distance in the prior art.
In order to solve the problems, the invention is realized by the following technical scheme:
the rescue unit automatic recommendation method for the intelligent emergency treatment process of the elevator comprises the following steps:
taking the position of the elevator with the fault as the center, taking the set distance R as the radius, forming a search area, and acquiring information of all elevator maintenance units in the area;
step two, constructing a rescue capability evaluation system model of an elevator maintenance unit, wherein the high system model comprises a distance L between a failed elevator and the maintenance unit, a matching relation between the maintenance unit and a failed elevator manufacturer, comprehensive evaluation star-level change conditions of the maintenance unit, maintenance rescue times of the maintenance unit in the last T years and 5 indexes of average rescue response time;
and thirdly, comprehensively evaluating all elevator maintenance in the first step by using a rescue capability evaluation system to obtain an optimal scheme, and rescue the faulty elevator by taking the optimal scheme as a recommended unit.
According to the invention, the rescue capability of each maintenance unit is comprehensively considered through multiple indexes, and the rescue dispatching task is automatically recommended, so that the most reasonable rescue unit is allocated for the failed elevator, and professional maintenance personnel are rapidly dispatched to enable trapped people to escape, so that the elevator failure rescue speed can be increased, the elevator rescue resource utilization efficiency is improved, the secondary hazard is reduced, and unnecessary casualties are avoided; in addition, the rescue capability of each maintenance unit is comprehensively considered through multiple indexes, so that the problem of uneven and unreasonable rescue task allocation of the maintenance units is better solved.
In the first step, the obtained elevator maintenance unit information comprises the name of each maintenance unit, the position coordinates of the maintenance unit and the next maintenance station, the elevator brands associated with the elevator maintenance unit, all elevator maintenance rescue data information in the last T years and the maintenance rescue capability comprehensive evaluation star-level data information of each year maintenance unit in the last T years.
The elevator maintenance unit information is obtained through a platform of an elevator emergency rescue center in the urban district or through a platform of an elevator supervision department. The data information of all elevators in the city and the information of all elevator maintenance units are stored in the database of the platform. The system comprises unique numbers and position information of each elevator, such as longitude and latitude, elevator manufacturers, signed maintenance unit information, past faults and rescue unit information; each elevator maintains the position information of the unit name, maintenance site, past rescue information, comprehensive evaluation star level of rescue capability, elevator brands related to the star level, and the like. Some maintenance units may be provided with a plurality of maintenance sites, each located in a different small or the like, and thus the position data of each maintenance site also needs to be stored in the database.
Further optimizing, after obtaining the information of all the elevator maintenance units in the area, the elevator maintenance rescue data needs to be preprocessed, and the method specifically comprises the following steps:
1) And processing the summarized elevator maintenance rescue data json file by using a python json file package, extracting required fields and values, processing some missing values and abnormal values, and storing the data for subsequent use. The data information of the elevator and the maintenance unit is manually input into the platform, and in the process, the filling errors of the longitude and latitude of the elevator can occur, such as the filling errors of the longitude and latitude position outside the city of the foreign province, the lack of maintenance site position data, the missing filling errors of the maintenance partial field of a certain fault elevator, and the like.
2) And extracting the maintenance rescue capability comprehensive evaluation star-level data information of the maintenance unit, and recording the star-level data of the past year in an excel file. Because the maintenance comprehensive evaluation star-level data of each year can be stored in different files, file formats are different, such as pdf, excel and the like, unification is needed, and the subsequent data extraction is convenient.
3) And extracting the name of the maintenance unit, the elevator brand related to the name and the maintenance rescue capability of the maintenance unit by using a python openpyxl file package, comprehensively evaluating star-level data and the position coordinates of the elevator maintenance unit, and storing the star-level data and the position coordinates.
Further optimizing, processing some missing values and abnormal values, specifically including: the file format errors are corrected, and the individual error data and the missing data are corrected and supplemented by querying the corresponding original database.
Further optimizing, setting the position coordinates of the failed elevator as (x 1 ，y 1 ) The position coordinates of the maintenance unit or the maintenance site arranged below the maintenance unit are (x) 2 ，y 2 ) The distance L between the failed elevator and the maintenance unit is calculated as follows:
the maintenance site is a certain number of sites under the maintenance unit, and when the maintenance unit needs to execute tasks, workers at the certain maintenance site are generally sent to rush repair. Such as maintenance unit a, there is one maintenance site at each of three locations a1, a2, a 3.
Although a more accurate result can be obtained by calculating the Manhattan distance (walking distance) between two points after modeling urban traffic by calling map data, the method and the device adopt Euclidean distance between two points to represent the distance between a fault elevator and a maintenance site in consideration of huge calculation amount and algorithm cost of hundreds of thousands of data sets, so that the result can be obtained more quickly, and actual operation is facilitated.
Further optimizing, judging the matching relation between the maintenance unit and the fault elevator manufacturer, comprising the following steps:
1) If a maintenance unit is the maintenance unit of the fault elevator manufacturer, namely the maintenance unit is matched with the fault elevator manufacturer and recorded as the original factory maintenance, the index is fully divided into 5 points. The original factory maintenance is obtained by counting the past elevator maintenance data, so that the original factory maintenance is very beneficial to the maintenance of the fault elevator, and the original factory is clearer to the construction and common faults of the fault elevator.
2) And calculating the relevance coefficient between each maintenance unit and the fault elevator by adopting a word frequency-inverse text frequency index weighting method for all non-original factory maintenance units, and measuring the maintenance professional degree and irreplaceability of the maintenance unit for the elevator of a specific brand, wherein the importance of the elevator of the certain brand in the maintenance unit is increased in proportion to the occurrence frequency of the elevator in the history record, and is decreased in inverse proportion to the occurrence frequency of the elevator in all maintenance units. And (3) respectively calculating the TF value of each maintenance unit and the IDF value of the brand according to the formulas (2) and (3), and multiplying to obtain the TF-IDF importance index.
For elevators of brand j, the ratio TF of the number of times maintenance is taken care of by the i-th maintenance unit in all elevator malfunctions occurring i,j ：
k is a positive integer, and k is [1, D ].
In all elevator repair and rescue completed by the ith maintenance unit, the repair and rescue of the elevator with the brand of j is biased to IDF i ：
Wherein n is i,j Representing the maintenance and rescue times of maintenance units i to elevators with brands of j in the last T years, wherein D is the number of maintenance units and D j Is maintenance history data of the brand of the elevator of the recent T years.
The values TF and IDF are obtained by the two formulas, importance indexes are obtained through TF multiplied by IDF, the score between 0 and 1 is obtained through min-max normalization processing, and then the score is multiplied by 5 to obtain corresponding assignment, namely:
brand assignment= (max-current)/(max-min) ×5.
Further optimizing, comprehensively evaluating different values corresponding to star grades of different conservation and rescue capabilities of the dimension-protection unit dimension, wherein the values are specifically as follows: one star corresponds to 1 minute, two stars corresponds to 2 minutes, three stars corresponds to 3 minutes, four stars corresponds to 4 minutes, five stars corresponds to 5 minutes, and the highest position is 5 stars.
And calculating the comprehensive evaluation star-scale change of the maintenance rescue capability of the maintenance unit according to the following algorithm:
1) If the current star level of the maintenance unit is lower than the star level of the last year, assigning a value of 0.5 score to be subtracted on the basis of the corresponding score of the current star level;
2) If the current star grade of the maintenance unit is the same as the star grade of the last year, assigning a corresponding score of the current star grade;
3) If the current star level of the maintenance unit is higher than the star level of the last year, assigning a value of 0.5 score on the basis of the corresponding score of the current star level;
4) If the maintenance unit is currently star-level vacant, assigning the current star-level vacant as an average value of scores of the maintenance unit in the last 6 years;
5) If the elevator is newly installed in the current year, the value is the average value of all current elevator maintenance unit star-class values in the urban district where the accident elevator is located.
Further optimizing, wherein the average rescue response time T comprises an average value of the sum of time T1 from receiving rescue information to a rescue site and total rescue time T2 in all the maintenance and rescue participating works of the maintenance and maintenance unit in the near-T year; as an indicator t of the response time, it is assigned a value normalized to between 0 and 1 and multiplied by 5. Namely:
wherein t is max For the maximum value of average rescue response time in the last T years of all maintenance units in the area, T min For the minimum value of average rescue response time in the last T years of all maintenance units in the area, T i Average rescue sound in nearly T years for maintenance unit iThe time should be.
Further optimizing the maintenance and rescue times of the near-T-year maintenance unit: obtaining the maintenance times of each maintenance unit in the T year according to the obtained maintenance rescue data, and calculating the historical response times in the reverse order; specifically, the historical response times of the maintenance unit with the highest response times are subtracted by the historical response times of each other maintenance unit to serve as index historical response indexes of the corresponding maintenance units, and the indexes are normalized to be between 0 and 1 and multiplied by 5, namely:
wherein k is max For the maximum value, k of maintenance and rescue times in the near-T years of all maintenance units in the area min K is the minimum value of maintenance and rescue times in the last T years of all maintenance units in the area i The maintenance rescue times are maintained for the maintenance unit i in the near T years.
Repair units and repair sites with low response times cannot represent low rescue capability, and the repair units or repair sites may be in areas where elevator faults occur less or areas where elevator numbers are small, or may be always biased to allocate faults to sites where repair tasks are frequently completed when manually dispatched according to the earlier stage. Furthermore, the index is only one of a plurality of indexes, is used as a fairness measure, is not used as a unique reference index, and can be limited by reducing the weight of the final index and the weight of other indexes when the final index and the other indexes are weighted and summed.
Further optimizing, the third step specifically comprises the following steps:
1) Calculating Euclidean distances L between all elevator maintenance units and the fault elevators in the set area according to the position information of the fault elevators and the position information of the maintenance units or the maintenance stations; according to the calculation result, sorting from low to high according to Euclidean distance values, and selecting N maintenance units with the top ranking as candidate sets for follow-up recommendation;
2) Calculating brand assignment between N maintenance units in the candidate set and the fault elevator, sorting from high to low according to the brand assignment, and selecting M maintenance units with brand assignment ranking top as a preferred set; wherein N and M are positive integers greater than or equal to 5, and N is greater than M;
3) Calculating a weighted average value of 5 index assignment of each elevator maintenance unit or maintenance station in the optimized set, and taking the elevator maintenance unit with the highest score as an elevator fault rescue recommendation result;
the assignment of the distance index is that the Euclidean distance L value from each elevator maintenance or maintenance station to the fault elevator in the preference set is normalized to be between 0 and 1 and multiplied by 5, namely:
wherein L is max To preferably concentrate the maximum value of the euclidean distance of all elevator maintenance units or maintenance stops to the failed elevator, L min To preferably concentrate the minimum value of the euclidean distance of all elevator maintenance units or maintenance stops to the failed elevator, L i The euclidean distance from the maintenance unit or maintenance station i to the fault elevator.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the rescue capability of each maintenance unit is comprehensively considered through multiple indexes, and the rescue dispatching task is automatically recommended, so that the most reasonable rescue unit is allocated for the failed elevator, and professional maintenance personnel are rapidly dispatched to enable trapped people to escape, so that the elevator failure rescue speed can be increased, the elevator rescue resource utilization efficiency is improved, the secondary hazard is reduced, and unnecessary casualties are avoided; in addition, the rescue capability of each maintenance unit is comprehensively considered through multiple indexes, so that the problem of uneven and unreasonable rescue task allocation of the maintenance units is better solved.
Drawings
Fig. 1 is a flowchart of a rescue unit automatic recommendation method for an elevator intelligent emergency treatment process.
Fig. 2 is a diagram of an automatically recommended maintenance unit scoring radar in accordance with the first embodiment.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
As shown in fig. 1, the rescue unit automatic recommendation method for the intelligent emergency treatment process of the elevator comprises the following steps:
and firstly, forming a search area by taking the position of the elevator with the fault as the center and taking the set distance R as the radius, and acquiring information of all elevator maintenance units in the area.
The elevator maintenance unit information is obtained through a platform of an elevator emergency rescue center in a city-level district, such as a 96333 platform in Nanjing city. The acquired elevator maintenance unit information comprises the name and position coordinates of each maintenance unit, elevator brands associated with the elevator maintenance unit information, all elevator maintenance rescue data information in the last T years and maintenance rescue capability comprehensive evaluation star grade data information of each year maintenance unit in the last T years.
After the information of all the elevator maintenance units in the area is acquired, the elevator maintenance rescue data is required to be preprocessed, and the method specifically comprises the following steps:
1) And processing the summarized elevator maintenance rescue data json file by using a python json file package, extracting required fields and values, processing some missing values and abnormal values, and storing the data for subsequent use. The data information of the elevator and the maintenance unit is manually input into the platform, and in the process, the filling errors of the longitude and latitude of the elevator can occur, such as the filling errors of the longitude and latitude position outside the city of the foreign province, the lack of maintenance site position data, the missing filling errors of the maintenance partial field of a certain fault elevator, and the like. Processing the missing value and the abnormal value specifically comprises the following steps: the file format errors are corrected, and the individual error data and the missing data are corrected and supplemented by querying the corresponding original database.
2) And extracting the maintenance rescue capability comprehensive evaluation star-level data information of the maintenance unit, and recording the star-level data of the past year in an excel file. Because the maintenance comprehensive evaluation star-level data of each year can be stored in different files, file formats are different, such as pdf, excel and the like, unification is needed, and the subsequent data extraction is convenient.
3) And extracting the name of the maintenance unit, the elevator brand related to the name and the maintenance rescue capability of the maintenance unit by using a python openpyxl file package, comprehensively evaluating star-level data and the position coordinates of the elevator maintenance unit, and storing the star-level data and the position coordinates.
Step two, constructing a rescue capability evaluation system model of the elevator maintenance unit, wherein the system model comprises a distance L between a failed elevator and the maintenance unit or a maintenance site, a matching relation between the maintenance unit and a failed elevator manufacturer, comprehensive evaluation star-level change conditions of the maintenance unit, maintenance rescue times of the maintenance unit in the last T years and 5 indexes of average rescue response time.
(1) Distance L index between fault elevator and maintenance unit or maintenance station
Setting the position coordinates of the faulty elevator as (x 1 ，y 1 ) The position coordinates of the maintenance unit or the maintenance site arranged below the maintenance unit are (x) 2 ，y 2 ) The distance L between the failed elevator and the maintenance unit is calculated as follows:
the maintenance site is a certain number of sites under the maintenance unit, and when the maintenance unit needs to execute tasks, workers at the certain maintenance site are generally sent to rush repair. Such as maintenance unit a, there is one maintenance site at each of three locations a1, a2, a 3.
Although a more accurate result can be obtained by calculating the Manhattan distance (walking distance) between two points after modeling urban traffic by calling map data, the method and the device adopt Euclidean distance between two points to represent the distance between a fault elevator and a maintenance site in consideration of huge calculation amount and algorithm cost of hundreds of thousands of data sets, so that the result can be obtained more quickly, and actual operation is facilitated.
(2) Index of matching relation between maintenance unit and fault elevator manufacturer
Judging the matching relation between the maintenance unit and the fault elevator manufacturer, comprising the following steps:
1) If a maintenance unit is the maintenance unit of the fault elevator manufacturer, namely the maintenance unit is matched with the fault elevator manufacturer and recorded as the original factory maintenance, the index is fully divided into 5 points. The original factory maintenance is very beneficial to the maintenance of the fault elevator through the original elevator maintenance data, because the original factory is clearer to the construction and common faults of the fault elevator.
2) And calculating the relevance coefficient between each maintenance unit and the fault elevator by adopting a word frequency-inverse text frequency index weighting method for all non-original factory maintenance units, and measuring the maintenance professional degree and irreplaceability of the maintenance unit for the elevator of a specific brand, wherein the importance of the elevator of the certain brand in the maintenance unit is increased in proportion to the occurrence frequency of the elevator in the history record, and is decreased in inverse proportion to the occurrence frequency of the elevator in all maintenance units. According to formulas (2) and (3), respectively calculating the TF value of each maintenance unit and the IDF value of the brand, and multiplying to obtain a TF-IDF importance index:
for elevators of brand j, the ratio TF of the number of times maintenance is taken care of by the i-th maintenance unit in all elevator malfunctions occurring i,j ：
k is a positive integer, k is [1, D ];
in all elevator repair and rescue completed by the ith maintenance unit, the repair and rescue of the elevator with the brand of j is biased to IDF i ：
Wherein n is i,j Representing the maintenance and rescue times of maintenance units i to elevators with brands of j in the last T years, wherein D is the number of maintenance units and D j Is an elevator with brand jMaintenance history data of recent T years.
The values TF and IDF are obtained through the two formulas, importance indexes are obtained through TF multiplied by IDF, the score between 0 and 1 is obtained through min-max normalization processing, and corresponding assignment is obtained through multiplying the score by 5.
(3) Comprehensive evaluation star-level change condition index of maintenance unit
The comprehensive evaluation star level of different conservation and rescue capabilities of the dimension-protection unit corresponds to different assignments, and the comprehensive evaluation star level is specifically as follows: one star corresponds to 1 minute, two stars corresponds to 2 minutes, three stars corresponds to 3 minutes, four stars corresponds to 4 minutes, five stars corresponds to 5 minutes, and the highest position is 5 stars.
And calculating the comprehensive evaluation star-scale change of the maintenance rescue capability of the maintenance unit according to the following algorithm:
1) If the current star level of the maintenance unit is lower than the star level of the last year, assigning a value of 0.5 score to be subtracted on the basis of the corresponding score of the current star level;
2) If the current star grade of the maintenance unit is the same as the star grade of the last year, assigning a corresponding score of the current star grade;
3) If the current star level of the maintenance unit is higher than the star level of the last year, assigning a value of 0.5 score on the basis of the corresponding score of the current star level;
4) If the maintenance unit is currently star-level vacant, assigning the current star-level vacant as an average value of scores of the maintenance unit in the last 6 years;
5) If the elevator is newly installed in the current year, the value is the average value of all current elevator maintenance unit star-class values in the urban district where the accident elevator is located.
(4) Average rescue response time index
The average rescue response time T comprises an average value of the sum of time T1 from receiving rescue information to a rescue site and total rescue time T2 in the maintenance unit participating in maintenance and rescue work in the period of nearly T years; as an index t of the response time, it is normalized to between 0 and 1 and multiplied by five.
(5) Maintenance and rescue frequency index of maintenance unit
The maintenance rescue times of the near-T-year maintenance unit: obtaining the maintenance times of each maintenance unit in the T year according to the obtained maintenance rescue data, and calculating the historical response times in the reverse order; specifically, the historical response times of the maintenance units with the highest response times are subtracted by the historical response times of each other maintenance unit to serve as index historical response indexes of the corresponding maintenance units, and the index historical response indexes are normalized to be between 0 and 1 and multiplied by five.
Repair units and repair sites with low response times cannot represent low rescue capability, and the repair units or repair sites may be in areas where elevator faults occur less or areas where elevator numbers are small, or may be always biased to allocate faults to sites where repair tasks are frequently completed when manually dispatched according to the earlier stage. Furthermore, the index is only one of a plurality of indexes, is used as a fairness measure, is not used as a unique reference index, and can be limited by reducing the weight of the final index and the weight of other indexes when the final index and the other indexes are weighted and summed.
And thirdly, comprehensively evaluating all elevator maintenance in the first step by using a rescue capability evaluation system to obtain an optimal scheme, and rescue the faulty elevator by taking the optimal scheme as a recommended unit. The method specifically comprises the following steps:
1) Calculating Euclidean distances L between all elevator maintenance units and the fault elevators in the set area according to the position information of the fault elevators and the position information of the maintenance units or the maintenance stations; according to the calculation result, sorting from low to high according to Euclidean distance values, and selecting N maintenance units with the top ranking as candidate sets for follow-up recommendation;
2) Calculating brand assignment between N maintenance units in the candidate set and the fault elevator, sorting from high to low according to the brand assignment, and selecting M maintenance units with brand assignment ranking top as a preferred set; wherein N and M are positive integers greater than or equal to 5, and N is greater than M;
3) Calculating a weighted average value of 5 index assignment of each elevator maintenance unit or maintenance station in the optimized set, wherein the weight of each index is the same, and taking the elevator maintenance unit with the highest score as an elevator fault rescue recommendation result;
the assignment of the distance index is that the Euclidean distance L value from each elevator maintenance or maintenance station to the fault elevator in the preference set is normalized to be between 0 and 1 and multiplied by 5, namely:
wherein L is max To preferably concentrate the maximum value of the euclidean distance of all elevator maintenance units or maintenance stops to the failed elevator, L min To preferably concentrate the minimum value of the euclidean distance of all elevator maintenance units or maintenance stops to the failed elevator, L i The euclidean distance from the maintenance unit or maintenance station i to the fault elevator.
Specific examples:
in this embodiment, the location information of the failed elevator is 118.8 ° longitude and 32 ° latitude, and the failed elevator is located in south kyo city and manufactured by shanghai toshiba elevator limited, N in the candidate set is 10, and M in the candidate set is 5.
First, the distance L from all maintenance stations to the elevator in the administrative area where the fault elevator is located is calculated, and a distance list is formed. The first 10 maintenance units or maintenance stations are selected according to the sequence from the near to the far, namely the first 10 elevator maintenance units closest to the accident position are selected as candidate sets.
Taking the top 5 brand indexes according to the brand recommendation result, wherein the brand indexes are respectively as follows: jiangsu Guangli Lift installation engineering Co., ltd., jiangsu Brand of Hitachi Lift (China) Co., ltd., nanjing Brand of Jiangsu sea water chestnut mechanical and electrical equipments engineering Co., nanjing Strong mechanical and electrical engineering Co., ltd., nanjing Brand of Tongli Lift Co., ltd. The results of this 5-dimensional insurance unit are shown in table 1 using euclidean distance calculations.
TABLE 1 Euclidean distance calculation results
The normalization is then carried out and,
the specific calculation formula is as follows: (max-current)/(max-min) ×5.
The other indices were also calculated as described above, taking Jiangsu Yueli Elevator installation engineering Co., ltd.) as an example:
maintenance star level: the historical data for this repair site is [4, none, 4]. Thus, the dimension-guaranteed star level is 4.
Average rescue response time t: the maintenance time is in seconds, the total maintenance time of the maintenance site is 107174 seconds in the last few years, and the maintenance time is 123 times. The response time averages 871.33 seconds, i.e. around 14.5 minutes. The normalized result was 2.341566600800527.
Number of maintenance rescue: the number of recent responses for the repair site is 62. After normalization, the result was 4.301801801801802.
The data of the remaining 4 maintenance units were calculated in sequence in the above manner, and the final results are shown in table 2.
TABLE 2 rescue Unit recommendation results
As shown in table 2, the radar map in the python pyechorts package was used for visualization, and the results are shown in fig. 2.
And calculating a weighted average of 5 indexes of each maintenance unit, wherein the weight is 0.2 of each dimension, and calculating a comprehensive recommendation score, and the rescue unit with the highest score is Jiangsu Yueli elevator installation engineering Co. Thus, the final automatic recommendation is Jiangsu Yueli elevator installation engineering Co. The comprehensive recommended result comparison is carried out by using the 2021 elevator emergency treatment historical worksheet data, the top1/top3 accuracy is 0.336 and 0.227 respectively, and the fact that a plurality of 2022 maintenance sites are not in a list in 2021 is considered, so that the result is only used as a reference, and the recommended accuracy can be improved by adjusting the weight in the follow-up process.
By the embodiment, as shown in fig. 1-2 and table 1-2, the most reasonable rescue unit can be allocated to the failed elevator by using the method, and professional maintenance personnel can be rapidly dispatched to enable trapped masses to escape, so that the failed rescue speed of the elevator can be accelerated, the utilization efficiency of rescue resources of the elevator can be improved, secondary hazards can be reduced, unnecessary casualties can be avoided, and the method has important significance for ensuring the safe operation of the elevator.
It should be understood that the detailed description and specific examples, while indicating the invention, are intended for purposes of illustration only and are not intended to limit the invention; any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. The rescue unit automatic recommendation method for the intelligent emergency treatment process of the elevator is characterized by comprising the following steps of:
taking the position of the elevator with the fault as the center, taking the set distance R as the radius, forming a search area, and acquiring information of all elevator maintenance units in the area;
step two, constructing a rescue capability evaluation system model of an elevator maintenance unit, wherein the system model comprises a distance L between a fault elevator and the maintenance unit or a maintenance site, a matching relation between the maintenance unit and a fault elevator manufacturer, comprehensive evaluation star-level change conditions of the maintenance unit, maintenance rescue times of the maintenance unit in the recent T year and 5 indexes of average rescue response time;
judging the matching relation between the maintenance unit and the fault elevator manufacturer, comprising the following steps:
1) If a certain maintenance unit is the maintenance unit of the fault elevator manufacturer, namely the maintenance unit is matched with the fault elevator manufacturer and recorded as the original factory maintenance, the index is fully divided into 5 points;
2) Calculating the relevance coefficient between each maintenance unit and the fault elevator by adopting a word frequency-inverse text frequency index weighting method for all non-original factory maintenance units, and respectively calculating the TF value of each maintenance unit and the IDF value of the elevator brand according to formulas (2) and (3), and multiplying to obtain TF-IDF importance indexes:
for elevators of brand j, the ratio TF of the number of times maintenance is taken care of by the i-th maintenance unit in all elevator malfunctions occurring i,j ：
k is a positive integer, k is [1, D ];
in all elevator repair and rescue completed by the ith maintenance unit, the repair and rescue of the elevator with the brand of j is biased to IDF i ：
Wherein n is i,j Representing the maintenance and rescue times of maintenance units i to elevators with brands of j in the recent T years, wherein D is the maintenance unit number in the urban district where the accident elevator is located, and D j Is maintenance history data of the brand of the elevator in the last T years;
the values TF and IDF are obtained by the two formulas, importance indexes are obtained through TF multiplied by IDF, the score between 0 and 1 is obtained through min-max normalization processing, and then the score is multiplied by 5 to obtain corresponding assignment, namely:
brand assignment= (max-current)/(max-min) ×5;
and thirdly, comprehensively evaluating all elevator maintenance in the first step by using a rescue capability evaluation system to obtain an optimal scheme, and rescue the faulty elevator by taking the optimal scheme as a recommended unit.
2. The automatic rescue unit recommending method for the intelligent emergency handling process of the elevator according to claim 1, wherein in the first step, the acquired elevator maintenance unit information comprises a name of each maintenance unit, position coordinates of the maintenance unit and a lower maintenance station, elevator brands associated with the maintenance unit, all elevator maintenance rescue data information in recent T years and maintenance rescue capability comprehensive evaluation star grade data information of each year maintenance unit in recent T years.
3. The automatic rescue unit recommending method for the intelligent emergency treatment process of the elevator according to claim 2, wherein after acquiring the information of all the elevator maintenance units in the area, the elevator maintenance rescue data is required to be preprocessed, and the method specifically comprises the following steps:
1) Processing the summarized elevator maintenance rescue data json file by using a python json file package, extracting required fields and values, processing some missing values and abnormal values, and storing the data for subsequent use;
2) Extracting maintenance rescue capability comprehensive evaluation star-level data information of a maintenance unit, and recording the star-level data of the past year in an excel file;
3) And extracting the name of the maintenance unit, the elevator brand related to the name and the maintenance rescue capability of the maintenance unit by using a python openpyxl file package, comprehensively evaluating star-level data and the position coordinates of the elevator maintenance unit, and storing the star-level data and the position coordinates.
4. The automatic rescue unit recommendation method for the intelligent emergency treatment process of the elevator according to claim 3, wherein the method is characterized by processing some missing values and abnormal values and specifically comprises the following steps: the file format errors are corrected, and the individual error data and the missing data are corrected and supplemented by querying the corresponding original database.
5. The automatic rescue unit recommending method for an elevator intelligent emergency handling process according to claim 1, wherein the position coordinates of the failed elevator are set to (x 1 ，y 1 ) The position coordinates of the maintenance unit or the maintenance site arranged below the maintenance unit are (x) 2 ，y 2 ) The distance L between the failed elevator and the maintenance unit or repair station is calculated as follows:
6. the automatic rescue unit recommendation method for the intelligent emergency treatment process of the elevator according to claim 1, wherein different rescue capabilities of the maintenance unit dimension are comprehensively evaluated to correspond to different assignments, and the method is specifically as follows: one star corresponds to 1 minute, two stars corresponds to 2 minutes, three stars corresponds to 3 minutes, four stars corresponds to 4 minutes, and five stars corresponds to 5 minutes;
and calculating the comprehensive evaluation star-scale change of the maintenance rescue capability of the maintenance unit according to the following algorithm:
1) If the current star level of the maintenance unit is lower than the star level of the last year, assigning a value of 0.5 score to be subtracted on the basis of the corresponding score of the current star level;
2) If the current star grade of the maintenance unit is the same as the star grade of the last year, assigning a corresponding score of the current star grade;
3) If the current star level of the maintenance unit is higher than the star level of the last year, assigning a value of 0.5 score on the basis of the corresponding score of the current star level;
4) If the maintenance unit is currently star-level vacant, assigning the current star-level vacant as an average value of scores of the maintenance unit in the last 6 years;
5) If the elevator is newly installed in the current year, the value is the average value of all current elevator maintenance unit star-class values in the urban district where the accident elevator is located.
7. The automatic rescue unit recommendation method for the elevator intelligent emergency treatment process according to claim 1, wherein the average rescue response time T comprises an average value of the sum of time T1 from receiving rescue information to a rescue site and total rescue time T2 in all the maintenance and rescue units participating in maintenance and rescue work within nearly T years; as an indicator t of the response time, it is assigned a value normalized to between 0 and 1 and multiplied by 5.
8. The automatic rescue unit recommendation method for the intelligent emergency treatment process of the elevator according to claim 1, wherein the maintenance rescue times of the near-T-year maintenance unit are as follows: obtaining maintenance rescue times of each maintenance unit in the T year according to the obtained maintenance rescue data, and calculating historical response times in the reverse order; specifically, the historical response times of the maintenance unit with the highest response times are subtracted by the historical response times of each other maintenance unit to serve as index historical response indexes of the corresponding maintenance units, and the indexes are normalized to be between 0 and 1 and multiplied by 5, namely:
wherein k is max For the maximum value, k of maintenance and rescue times in the near-T years of all maintenance units in the area min K is the minimum value of maintenance and rescue times in the last T years of all maintenance units in the area i The maintenance rescue times are maintained for the maintenance unit i in the near T years.
9. The automatic rescue unit recommending method for the intelligent emergency treatment process of the elevator according to claim 1, wherein the third step comprises the following steps:
1) Calculating Euclidean distances L between all elevator maintenance units and the fault elevators in the set area according to the position information of the fault elevators and the position information of the maintenance units or the maintenance stations; according to the calculation result, sorting from low to high according to Euclidean distance values, and selecting N maintenance units with the top ranking as candidate sets for follow-up recommendation;
2) Calculating brand assignment between N maintenance units in the candidate set and the fault elevator, sorting from high to low according to the brand assignment, and selecting M maintenance units with brand assignment ranking top as a preferred set; wherein N and M are positive integers greater than or equal to 5, and N is greater than M;
3) Calculating a weighted average value of 5 index assignment of each elevator maintenance unit or maintenance station in the optimized set, wherein the weight of each index is the same, and taking the elevator maintenance unit with the highest score as an elevator fault rescue recommendation result;
the assignment of the distance index is that the Euclidean distance L value from each elevator maintenance or maintenance station to the fault elevator in the preference set is normalized to be between 0 and 1 and multiplied by 5, namely:
wherein L is max To preferably concentrate the maximum value of the euclidean distance of all elevator maintenance units or maintenance stops to the failed elevator, L min To preferably concentrate the minimum value of the euclidean distance of all elevator maintenance units or maintenance stops to the failed elevator, L i The euclidean distance from the maintenance unit or maintenance station i to the fault elevator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211484455.6A CN115730850B (en) | 2022-11-24 | 2022-11-24 | Rescue unit automatic recommendation method for intelligent emergency disposal process of elevator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211484455.6A CN115730850B (en) | 2022-11-24 | 2022-11-24 | Rescue unit automatic recommendation method for intelligent emergency disposal process of elevator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115730850A CN115730850A (en) | 2023-03-03 |
| CN115730850B true CN115730850B (en) | 2023-06-16 |
Family
ID=85298091
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211484455.6A Active CN115730850B (en) | 2022-11-24 | 2022-11-24 | Rescue unit automatic recommendation method for intelligent emergency disposal process of elevator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115730850B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1178915A1 (en) * | 1999-05-19 | 2002-02-13 | Giok Djien Dr.-Ing. Go | Rescue system of a train or vehicle |
| CN106892311A (en) * | 2017-04-12 | 2017-06-27 | 南京市特种设备安全监督检验研究院 | A kind of optimal site selecting method of elevator rescue website based on minimum weight time gap |
| CN109775501A (en) * | 2017-11-10 | 2019-05-21 | 广东工匠特种设备有限公司 | A kind of projection touch-control Internet of Things rescue system applied to elevator |
| WO2020093969A1 (en) * | 2018-11-05 | 2020-05-14 | 山东科技大学 | Urban elevator intelligent dispatching inspection system based on big data drive |
| CN112079223A (en) * | 2020-11-03 | 2020-12-15 | 南京市特种设备安全监督检验研究院 | Elevator maintenance-as-needed working quality evaluation method |
| CN112433786A (en) * | 2019-08-26 | 2021-03-02 | 中移(苏州)软件技术有限公司 | Information processing method, terminal and computer readable storage medium |
-
2022
- 2022-11-24 CN CN202211484455.6A patent/CN115730850B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1178915A1 (en) * | 1999-05-19 | 2002-02-13 | Giok Djien Dr.-Ing. Go | Rescue system of a train or vehicle |
| CN106892311A (en) * | 2017-04-12 | 2017-06-27 | 南京市特种设备安全监督检验研究院 | A kind of optimal site selecting method of elevator rescue website based on minimum weight time gap |
| CN109775501A (en) * | 2017-11-10 | 2019-05-21 | 广东工匠特种设备有限公司 | A kind of projection touch-control Internet of Things rescue system applied to elevator |
| WO2020093969A1 (en) * | 2018-11-05 | 2020-05-14 | 山东科技大学 | Urban elevator intelligent dispatching inspection system based on big data drive |
| CN112433786A (en) * | 2019-08-26 | 2021-03-02 | 中移(苏州)软件技术有限公司 | Information processing method, terminal and computer readable storage medium |
| CN112079223A (en) * | 2020-11-03 | 2020-12-15 | 南京市特种设备安全监督检验研究院 | Elevator maintenance-as-needed working quality evaluation method |
Non-Patent Citations (2)
| Title |
|---|
| Intelligent Guidance Method for Elevator Emergency Treatment based on Automatic Recommendation and Fault Prediction;Guangwei Qing;《2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)》;462-468 * |
| 城市电梯应急救援响应与救援对策研究;王会方;《机电工程技术》;第43卷(第4期);101-105 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115730850A (en) | 2023-03-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6671824B1 (en) | Cable network repair control system | |
| CN110782370A (en) | Comprehensive operation and maintenance management platform for power dispatching data network | |
| CN110110881A (en) | Power customer requirement forecasting analysis method and system | |
| CN106327062A (en) | Method for evaluating state of power distribution network equipment | |
| US11906112B2 (en) | Methods for safety management of compressors in smart gas pipeline network and internet of things systems thereof | |
| CN107635080A (en) | Report processing method and equipment for repairment | |
| CN112258065A (en) | Method, device and equipment for determining maintenance and dispatching sequence | |
| CN111275315A (en) | Charging pile maintenance scheduling method and system | |
| CN106803142A (en) | Repair and maintenance personnel automated decision system and method in information-based elevator operation and maintenance system | |
| CN115619382B (en) | Visual management method and system for power dispatching | |
| CN111461539A (en) | Distribution network dispatching work task difficulty grading method and system and computer equipment | |
| CN111178382A (en) | Repair service prediction analysis method based on work order data mining | |
| CN115730850B (en) | Rescue unit automatic recommendation method for intelligent emergency disposal process of elevator | |
| CN114091213A (en) | Power communication network planning scheme evaluation system | |
| CN116011827B (en) | Power failure monitoring analysis and early warning system and method for key cells | |
| CN117094487A (en) | Intelligent overall method and device for bearing capacity production and electronic equipment | |
| CN112232662A (en) | Service monitoring system and method | |
| CN110895742A (en) | Risk assessment method and system for power dispatching robot | |
| CN104281934A (en) | Operation evaluation method applicable to full life cycle management of network distribution device | |
| CN114372712A (en) | Supervision task management system based on process | |
| CN113469507A (en) | Human resource management system based on network platform | |
| CN113807653A (en) | Fault repair work order positioning method based on fuzzy matching | |
| CN113825162A (en) | Method and device for positioning fault reason of telecommunication network | |
| CN112433786A (en) | Information processing method, terminal and computer readable storage medium | |
| CN114399203B (en) | Transformer substation maintenance task processing method and system, terminal device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |