An elevator and an abnormality detection method, device and storage medium thereof

Abstract

This application discloses an elevator and its anomaly detection method, device, and storage medium. The method includes: acquiring operating data corresponding to different operating stages within the elevator door opening and closing cycle; calculating feature values ​​corresponding to the operating data; inputting the calculated feature values ​​into an anomaly diagnostic device corresponding to the feature values; the anomaly diagnostic device calculating upper and lower boundary values ​​based on the acquired feature values; detecting whether the feature values ​​are abnormal based on the upper and lower boundary values; determining whether an operating stage is abnormal based on the judgment result of whether the feature values ​​in each operating stage are abnormal; and determining that the elevator has malfunctioned if any one or more operating stages are detected to be abnormal, and judging the fault type based on the magnitude of a specified feature value. This embodiment enables timely detection of elevator anomalies and determination of fault types, ensuring people's safety.

Description

Technical Field

[0001] This application relates to elevator design technology, and more particularly to an elevator and its anomaly detection method, device and storage medium. Background Technology

[0002] Elevator door operators are the core components of elevators. More than 80% of elevator malfunctions and more than 70% of elevator accidents are related to the door system. Real-time detection of abnormalities and fault prediction of door operator operation are crucial to ensuring the normal operation of the door operator and thus protecting people's lives. Summary of the Invention

[0003] This application provides an elevator and its abnormality detection method, device and storage medium, which can detect elevator abnormalities in a timely manner and determine the type of fault, thus ensuring people's life safety.

[0004] This application provides a method for detecting elevator anomalies, the method may include:

[0005] Acquire operational data corresponding to different operating stages within the elevator door opening and closing cycle;

[0006] Calculate the feature values ​​corresponding to the running data;

[0007] The calculated feature values ​​are input into the corresponding anomaly diagnostic tool, which determines whether the feature values ​​are abnormal. The anomaly diagnostic tool can calculate the upper and lower boundary values ​​of the feature values ​​based on the obtained feature values, and check whether all feature values ​​are abnormal based on the upper and lower boundary values.

[0008] Whether an operation phase is abnormal is determined based on the judgment result of whether the characteristic value in each operation phase is abnormal;

[0009] If any one or more of the aforementioned operating stages are found to be abnormal, it can be determined that the elevator is malfunctioning.

[0010] In one embodiment, the running phase may include:

[0011] The door opening process includes the following operational phases: the door is in a closed state, the first slow opening phase, the accelerated opening phase, the constant speed opening phase, the deceleration opening phase, and the second slow opening phase; and,

[0012] The closing process consists of the following stages: the door is in the open state, the first slow closing stage, the accelerated closing stage, the uniform closing stage, the deceleration closing stage, and the second slow closing stage.

[0013] The operating data may include any one or more of the following: torque, speed, and displacement;

[0014] The calculation of the feature values ​​of the running data may include:

[0015] Calculate any one or more of the following values ​​for the running data: mean, standard deviation, and kurtosis.

[0016] In one embodiment, inputting the calculated feature value into a corresponding anomaly diagnostic tool, and using the anomaly diagnostic tool to determine whether the feature value is abnormal, may include:

[0017] The anomaly diagnostic tool continuously acquires the corresponding feature values ​​for each feature value corresponding to each running data in each running phase.

[0018] When the number of feature values ​​collected by each of the anomaly diagnostic devices has reached a preset threshold, each of the anomaly diagnostic devices can calculate the upper and lower boundary values ​​for feature value anomaly detection based on all the collected feature values ​​and a preset upper and lower boundary determination strategy; the upper boundary value refers to the upper boundary value of the feature value, and the lower boundary value refers to the lower boundary value of the feature value.

[0019] Based on the upper and lower boundary values, determine whether each of the collected feature values ​​is abnormal.

[0020] In one embodiment, determining whether each of the acquired feature values ​​is abnormal based on the upper boundary value and the lower boundary value may include:

[0021] If the feature value is outside the feature value range formed by the upper boundary value and the lower boundary value, the feature value can be determined to be abnormal.

[0022] If the feature value is within the range of the feature value, the feature value can be determined to be normal.

[0023] In one embodiment, the method may further include:

[0024] Each of the aforementioned anomaly diagnostics can push an anomaly result and remove the feature value or add the feature value to a pre-set anomaly cluster when it determines that the feature value is abnormal.

[0025] Each of the aforementioned anomaly diagnostic devices, upon determining that the feature value is normal, can add the feature value to a pre-set cluster of normal values.

[0026] In one embodiment, the step of inputting the feature value into a corresponding anomaly diagnostic device and determining whether the feature value is abnormal through the anomaly diagnostic device may further include:

[0027] After the anomaly diagnostic tool for each feature value in each running phase obtains a new feature value, it can calculate the sum of the number of the new feature values ​​and the number of all feature values ​​in the normal value cluster.

[0028] If the sum of the number of all new feature values ​​acquired by each anomaly diagnostic tool and the number of all feature values ​​in the corresponding normal value cluster is greater than or equal to the number threshold, the upper boundary value and the lower boundary value can be recalculated based on the acquired all new feature values, the all feature values ​​in the normal value cluster corresponding to the anomaly diagnostic tool, and the upper and lower boundary determination strategy.

[0029] Based on the recalculated upper and lower boundary values, it can be determined whether there are any abnormal feature values ​​among all the new feature values ​​collected by the anomaly diagnostic tool and among all the feature values ​​in the normal value cluster corresponding to the anomaly diagnostic tool.

[0030] In one embodiment, the upper and lower boundary determination strategy may include:

[0031] Arrange all the aforementioned feature values ​​in ascending order, and divide the arranged feature values ​​into m equal parts. The feature values ​​located at the m-1 division points of the m equal parts are taken as m quantiles; where m is a positive integer.

[0032] Get the first m-quantile Q1 and the (m-1)th m-quantile Q m-1 Q1 m-1 ;

[0033] Calculate the upper and lower quantile differences (IQ) R =Q m-1 -Q1;

[0034] Calculate the lower boundary value LowLim = Q1 - K1·IQ R Calculate the upper boundary value UpLim = Q m-1 +K2·IQ R .

[0035] In one embodiment, m = 4, K1 = K2 = 1.5.

[0036] In one embodiment, determining whether an operational phase is abnormal based on the judgment result of whether the feature value is abnormal in each operational phase may include:

[0037] After inputting the feature values ​​corresponding to a set of running data for the running phase into the corresponding anomaly diagnostics, the diagnostic results output by each anomaly diagnostics are obtained.

[0038] ​Count the first number of the first results that are judged to have abnormal feature values ​​and the second number of the second results that are judged to have normal feature values ​​among all the diagnostic results corresponding to this operation phase.

[0039] If the first quantity is greater than the second quantity, the operation phase can be determined to be abnormal; if the first quantity is less than or equal to the second quantity, the operation phase can be determined to be normal.

[0040] In one embodiment, the method may further include:

[0041] Based on the fault type to be determined, obtain the characteristic value of the specified type corresponding to the specified operating data within the specified operating phase required to determine the fault type;

[0042] Compare the feature value of the specified type with the corresponding feature threshold;

[0043] If the comparison result between the specified feature value and the feature threshold meets the preset requirements, it can be determined that the elevator has experienced this type of fault.

[0044] In one embodiment, the elevator malfunction type may include any one or more of the following: timing belt loosening, door knife jamming, and increased overall resistance;

[0045] The specified operating phase corresponding to the door knife jamming may include: the first slow-opening door phase;

[0046] The specified operating phase corresponding to the increase in overall resistance may include: the accelerated door opening phase and the accelerated door closing phase;

[0047] The specified operating phase corresponding to the timing belt detachment may include: the accelerated door opening phase and the second slow door opening phase;

[0048] The characteristic values ​​of the specified type corresponding to the gate knife jamming may include: mean torque, mean square of torque, and torque kurtosis;

[0049] The characteristic values ​​of a specified type corresponding to the increase in overall resistance may include: mean torque and mean square error of torque;

[0050] The characteristic values ​​of the specified type corresponding to the timing belt detachment may include: torque mean square error, torque kurtosis, and velocity kurtosis.

[0051] This application embodiment also provides an elevator anomaly detection device, which may include a processor and a memory. The memory stores instructions, and when the instructions are executed by the processor, the elevator anomaly detection method can be implemented.

[0052] This application also provides an elevator that may include the aforementioned elevator anomaly detection device.

[0053] This application also provides a non-transient computer-readable storage medium storing a computer program that, when executed by a processor, can implement the elevator anomaly detection method.

[0054] This application embodiment may include: acquiring operational data corresponding to different operating stages within an elevator door opening and closing cycle; calculating feature values ​​corresponding to the operational data; inputting the calculated feature values ​​into a corresponding anomaly diagnostic device, and using the anomaly diagnostic device to determine whether the feature values ​​are abnormal; wherein, the anomaly diagnostic device calculates upper and lower boundary values ​​of the feature values ​​based on the acquired feature values, and checks whether all feature values ​​are abnormal one by one based on the upper and lower boundary values; determining whether an operational stage is abnormal based on the judgment result of whether the feature values ​​in each operational stage are abnormal; if any one or more operational stages are detected to be abnormal, it can be determined that the elevator has malfunctioned, and the fault type can be determined based on the magnitude of a specified feature value. Through this embodiment, timely detection of elevator anomalies and determination of fault types are achieved, ensuring people's safety.

[0055] Other features and advantages of the embodiments of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the application. Other advantages of this application can be realized and obtained through the solutions described in the description and drawings. Attached Figure Description

[0056] The accompanying drawings are used to provide an understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0057] Figure 1 This is a flowchart of the elevator anomaly detection method according to an embodiment of this application;

[0058] Figure 2 This is a schematic diagram illustrating different operating stages within the elevator door opening and closing cycle according to an embodiment of this application.

[0059] Figure 3(a) is a schematic diagram of determining the door opening start point based on historical door opening signals according to an embodiment of this application;

[0060] Figure 3(b) is a schematic diagram of determining the door opening endpoint based on historical door opening position signals according to an embodiment of this application;

[0061] Figure 3(c) is a schematic diagram of determining the door closing start point based on historical door closing signals according to an embodiment of this application;

[0062] Figure 3(d) is a schematic diagram of determining the door closing endpoint based on historical door closing signals according to an embodiment of this application;

[0063] Figure 4 The flowchart of the method for determining whether the feature value is abnormal by inputting the calculated feature value into the anomaly diagnostic tool corresponding to the feature value in an embodiment of this application;

[0064] Figure 5 This is a schematic diagram illustrating the upper and lower boundary determination strategy of an embodiment of this application;

[0065] Figure 6 This is a block diagram of the elevator anomaly detection and fault type determination device according to an embodiment of this application;

[0066] Figure 7 This is a block diagram of the elevator components according to an embodiment of this application. Detailed Implementation

[0067] This application describes several embodiments, but these descriptions are exemplary and not restrictive, and it will be apparent to those skilled in the art that many more embodiments and implementations are possible within the scope of the embodiments described herein. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment may be used in combination with, or may replace, any feature or element of any other embodiment.

[0068] This application includes and contemplates combinations of features and elements known to those skilled in the art. The embodiments, features, and elements disclosed in this application may also be combined with any conventional features or elements to form a unique inventive scheme as defined by the claims. Any feature or element of any embodiment may also be combined with features or elements from other inventive schemes to form another unique inventive scheme as defined by the claims. Therefore, it should be understood that any feature shown and / or discussed in this application may be implemented individually or in any suitable combination. Therefore, the embodiments are not limited except by the limitations imposed by the appended claims and their equivalents. Furthermore, various modifications and changes may be made within the scope of the appended claims.

[0069] Furthermore, in describing representative embodiments, the specification may have presented methods and / or processes as a specific sequence of steps. However, the method or process should not be limited to the specific order of steps described herein, to the extent that it does not depend on such a specific order. As will be understood by those skilled in the art, other sequences of steps are also possible. Therefore, the specific order of steps set forth in the specification should not be construed as a limitation of the claims. Moreover, the claims concerning the method and / or process should not be limited to the steps performed in the written order, and those skilled in the art will readily understand that these orders can be varied and still remain within the spirit and scope of the embodiments of this application.

[0070] This application provides an elevator anomaly detection method, such as... Figure 1 As shown, the method may include steps S101-S105:

[0071] S101. Obtain the operating data corresponding to different operating stages within the elevator door opening and closing cycle;

[0072] S102. Calculate the feature values ​​corresponding to the running data;

[0073] S103. Input the calculated feature value into the corresponding anomaly diagnostic tool, and use the anomaly diagnostic tool to determine whether the feature value is abnormal; wherein, the anomaly diagnostic tool can calculate the upper boundary value and lower boundary value of the feature value based on the obtained feature value, and check whether all feature values ​​are abnormal one by one based on the upper boundary value and lower boundary value.

[0074] S104. Determine whether the operating stage is abnormal based on the judgment result of whether the characteristic value in each operating stage is abnormal;

[0075] S105. If any one or more of the aforementioned operating stages are found to be abnormal, it can be determined that the elevator has malfunctioned.

[0076] In one embodiment, before acquiring the operating data corresponding to different operating stages within the elevator door opening and closing cycle, the elevator door opening and closing cycle can be pre-divided into different operating stages, such as... Figure 2 As shown, it can be divided into 12 segments from s0 to s11.

[0077] In one embodiment, the running phase may include:

[0078] Based on the sequence of the door opening process, the operational phases of the door opening process may include, but are not limited to: the door being in a closed state (s0), the first slow opening phase (s1), the accelerated opening phase (s2), the uniform speed opening phase (s3), the deceleration opening phase (s4), and the second slow opening phase (s5); and,

[0079] According to the sequence of the closing process, the operation phases of the closing process may include, but are not limited to: the door is in the open state phase s6, the first slow closing phase s7, the accelerated closing phase s8, the uniform closing phase s9, the deceleration closing phase s10, and the second slow closing phase s11.

[0080] In one embodiment, after the second slow-closing stage s11 ends, the door will be completely closed, thus re-entering the closed-door stage s0.

[0081] In one embodiment, the method for dividing different operational phases is described in detail below.

[0082] In one embodiment, determining the different operating phases within the elevator's door opening and closing cycle may include:

[0083] Displacement curves of the elevator door operator during the opening and closing processes are obtained based on historical operating data;

[0084] The starting and ending positions of the door opening displacement curve in the displacement curve are determined by the rising edge timestamp of the historical door opening signal and the rising edge timestamp of the historical door opening position signal; the starting and ending positions of the door closing displacement curve in the displacement curve are determined by the rising edge timestamp of the historical door closing signal and the rising edge timestamp of the historical door closing position signal.

[0085] According to the preset division criteria, the door opening displacement curve and the door closing displacement curve are divided into multiple different operating stages, and the time period corresponding to each operating stage is determined; wherein, the division criteria include: the displacement range corresponding to the different operating stages as determined in advance.

[0086] In one embodiment, CAN signals can be acquired through a CAN (Controller Area Network) communication module. These CAN signals may include: the displacement curve of the elevator during the door opening and closing cycle, historical door opening signals, historical door opening-to-position signals, historical door closing signals, and historical door closing-to-position signals; and the operating data collected during the door opening and closing cycle can be acquired.

[0087] In one embodiment, the aforementioned CAN signal is acquired synchronously with the operating data to ensure that the timestamps of the CAN signal and the vibration data are consistent.

[0088] In one embodiment, the starting and ending positions of the door opening displacement curve can be located using the rising edge timestamps of historical door opening signals and historical door opening position signals; similarly, the starting and ending positions of the door closing displacement curve can be located using the rising edge timestamps of historical door closing signals and historical door closing position signals. Figure 3(a) , 3(b)As shown in 3(c) and 3(d).

[0089] In one embodiment, the elevator door opening is determined by historical door opening signals, and the elevator door closing is determined by historical door closing signals.

[0090] In one embodiment, segmenting the door opening displacement curve may include: obtaining the maximum displacement MaxDisp of the door opening displacement curve; defining the period from 0 to 0.08*MaxDisp as the "first slow door opening stage s1" and recording a timestamp; defining the period from 0.081*MaxDisp to 0.4*MaxDisp as the "accelerated door opening stage s2" and recording a timestamp; defining the period from 0.4*MaxDisp to 0.67*MaxDisp as the "uniform speed door opening stage s3" and recording a timestamp; defining the period from 0.67*MaxDisp to 0.98*MaxDisp as the "decelerated door opening stage s4" and recording a timestamp; and defining the period from 0.98*MaxDisp to MaxDisp as the "second slow door opening stage s5" and recording a timestamp.

[0091] In one embodiment, segmenting the door closing displacement curve may include: obtaining the maximum displacement MaxDisp of the door closing displacement curve; defining the range of MaxDisp to 0.98*MaxDisp as "first slow closing stage s7" and recording a timestamp; defining the range of 0.98*MaxDisp to 0.67*MaxDisp as "accelerated closing stage s8" and recording a timestamp; defining the range of 0.67*MaxDisp to 0.4*MaxDisp as "uniform closing stage s9" and recording a timestamp; defining the range of 0.4*MaxDisp to 0.081*MaxDisp as "decelerated closing stage s10" and recording a timestamp; and defining the range of 0.08*MaxDisp to 0 as "second slow closing stage s11" and recording a timestamp.

[0092] In one embodiment, the displacement range corresponding to the different operating stages described above is only one example, and the displacement range corresponding to each operating stage can be adjusted according to different application scenarios.

[0093] In one embodiment, the operating data collected during the door opening and closing cycle can be divided according to different operating stages within the elevator's door opening and closing cycle to obtain the operating data corresponding to different operating stages.

[0094] In one embodiment, dividing the operating data collected within the door opening and closing cycle according to different operating stages within the elevator's door opening and closing cycle, and obtaining the operating data corresponding to different operating stages may include: obtaining the timestamps corresponding to the operating data collected within the door opening and closing cycle; determining the time periods corresponding to different operating stages within the elevator's door opening and closing cycle; and for each operating stage, using the operating data within the time period corresponding to that operating stage as the operating data corresponding to that operating stage.

[0095] In one embodiment, since the timestamps of the CAN signal and the running data are consistent, the timestamps in the door opening displacement curve and the door closing displacement curve are also consistent with the timestamps of the running data. The time periods corresponding to different running stages can be compared with each time period generated by the running data, and the running data collected within the time period corresponding to the running stage can be used as the running data corresponding to that running stage, thereby obtaining the running data corresponding to different running stages.

[0096] In one embodiment, the operating data may include any one or more of the following: torque, speed, and displacement;

[0097] The calculation of the feature values ​​of the running data may include:

[0098] Calculate any one or more of the following values ​​for the running data: mean, standard deviation, and kurtosis.

[0099] In one embodiment, for example, std (mean deviation), mean (mean), and kur (kurtosis) of Iq (torque), V_Real (velocity), and Disp (displacement) can be calculated.

[0100] In one embodiment, the formula for calculating the mean squared error (std) may include:

[0101]

[0102] Where σ represents the mean square error, n represents the total number of data points, and x i This represents the i-th running data, where i is a positive integer, and i and n are positive integers, i ≤ n. This represents the mean of the running data.

[0103] In one embodiment, the formula for calculating kurtosis kur may include:

[0104]

[0105] In one embodiment, the following parameters can be calculated for each operating stage in the door opening displacement curve and the door closing displacement curve: Iq_std (torque mean square deviation), Iq_mean (torque mean), Iq_kur (torque kurtosis), V_Real_std (velocity mean square deviation), V_Real_mean (velocity mean), V_Real_kur (velocity kurtosis), Disp_std (displacement mean square deviation), Disp_mean (displacement mean), and Disp_kur (displacement kurtosis).

[0106] In one embodiment, a historical data manager can be set up, which may contain multiple branch managers. Each characteristic value corresponding to each type of running data in each running stage of the segmented door opening displacement curve and door closing displacement curve can correspond to a branch manager, and the branch manager can be used as an anomaly diagnostic tool for that characteristic value.

[0107] In one embodiment, the door opening displacement curve and door closing displacement curve, excluding the closed state stage s0 and the open state stage s6 (the elevator door is not in operation during these two stages, and there is no corresponding operation data), can contain a total of 10 operation stages. If three types of operation data (e.g., torque, speed, and displacement) and their corresponding three feature values ​​(e.g., root mean square deviation, mean, and sheath length) are obtained in each operation stage, then a total of 9 feature values ​​are obtained in each operation stage: Iq_std (torque root mean square deviation), Iq_mean (torque mean), Iq_kur (torque kurtosis), V_Real_std (speed root mean square deviation), V_Real_mean (speed mean), V_Real_kur (speed kurtosis), Disp_std (displacement root mean square deviation), Disp_mean (displacement mean), and Disp_kur (displacement kurtosis), then there are a total of 90 feature values ​​in the 10 operation stages. If each feature value corresponding to each type of operation data in each operation stage corresponds to an anomaly diagnostic tool, then the embodiment of this application can contain a total of 90 anomaly diagnostic tools.

[0108] In one embodiment, the anomaly diagnostic tool corresponding to each feature value of each running data in each running phase is only used to detect whether the feature value corresponding to the running data collected in that running phase is abnormal.

[0109] In one embodiment, such as Figure 4 As shown, the step of inputting the calculated feature value into the corresponding anomaly diagnostic tool, and using the anomaly diagnostic tool to determine whether the feature value is abnormal, may include steps S201-S203:

[0110] S201. The corresponding feature value is continuously obtained by the anomaly diagnostic tool corresponding to each feature value of each running data in each running stage.

[0111] S202. When the number of feature values ​​collected by each of the anomaly diagnostic devices has reached a preset threshold, each of the anomaly diagnostic devices can calculate the upper and lower boundary values ​​for feature value anomaly detection based on all the collected feature values ​​and a preset upper and lower boundary determination strategy; the upper boundary value refers to the upper boundary value of the feature value, and the lower boundary value refers to the lower boundary value of the feature value.

[0112] S203. Determine whether each of the collected feature values ​​is abnormal based on the upper boundary value and the lower boundary value.

[0113] In one embodiment, when using the anomaly diagnostic tool for the first time, a certain number of feature values ​​can be collected first, that is, the number of feature values ​​collected by the anomaly diagnostic tool can reach at least the above-mentioned number threshold (this number threshold can be defined according to different needs, and the detailed value is not limited here, for example, 100). Only then will the anomaly diagnostic tool calculate the upper and lower boundary values ​​of the feature values ​​based on all the collected feature values ​​using a preset upper and lower boundary determination strategy. At this time, the upper and lower boundary values ​​are the initial upper and lower boundary values.

[0114] In one embodiment, the upper and lower boundary determination strategy may include:

[0115] Arrange all the aforementioned feature values ​​in ascending order, and divide the arranged feature values ​​into m equal parts. The feature values ​​located at the m-1 division points of the m equal parts are taken as m quantiles; where m is a positive integer.

[0116] Get the first m-quantile Q1 and the (m-1)th m-quantile Q m-1 Q1 m-1 ;

[0117] Calculate the upper and lower quantile differences (IQ) R =Q m-1 -Q1;

[0118] Calculate the lower boundary value LowLim = Q1 - K1·IQ R Calculate the upper boundary value UpLim = Q m-1 +K2·IQ R .

[0119] In one embodiment, m = 4, K1 = K2 = 1.5 can be selected.

[0120] In one embodiment, such as​ Figure 5 As shown, when m=4, all the eigenvalues ​​are arranged in ascending order, and the arranged eigenvalues ​​are divided into 4 equal parts. The eigenvalues ​​at the 3 dividing points of these 4 parts are taken as quartiles; the first quartile Q1 and the third quartile Q3 are obtained; the difference between the upper and lower quartiles IQ is calculated. R =Q3-Q1; Calculate the lower boundary value LowLim = Q1-1.5IQ R Calculate the upper boundary value UpLim = Q3 + 1.5IQ R .

[0121] In one embodiment, determining whether each of the acquired feature values ​​is abnormal based on the upper boundary value and the lower boundary value may include:

[0122] Each of the aforementioned anomaly diagnostic tools can determine that the feature value is abnormal if the feature value is outside the feature value range formed by the upper boundary value and the lower boundary value.

[0123] Each of the aforementioned abnormality diagnostic devices can determine that the characteristic value is normal if the characteristic value is within the range of the characteristic value.

[0124] In one embodiment, the method may further include:

[0125] If the feature value is determined to be abnormal, the abnormal result can be pushed and the feature value can be removed, or the feature value can be added to a pre-set abnormal value cluster.

[0126] If the feature value is determined to be normal, it can be added to a pre-set cluster of normal values.

[0127] In one embodiment, based on the feature value range determined by the initial upper and lower boundary values, it is possible to detect whether the initially collected 100 feature values ​​are abnormal. If one or more feature values ​​are detected as abnormal, these feature values ​​can be removed.

[0128] In one embodiment, as the number of feature values ​​collected by the anomaly diagnostic tool gradually increases, if the total number of feature values ​​is greater than or equal to a set threshold (e.g., 100), the previously determined upper and lower boundary values ​​can be automatically updated each time according to the above-described upper and lower boundary determination strategy. The detailed automatic update strategy can be described as follows.

[0129] In one embodiment, the step of inputting the feature value into a corresponding anomaly diagnostic device and determining whether the feature value is abnormal through the anomaly diagnostic device may further include:

[0130] After the anomaly diagnostic tool for each feature value in each running phase obtains a new feature value, it can calculate the sum of the number of the new feature values ​​and the number of all feature values ​​in the normal value cluster.

[0131] If the sum of the calculated quantities is less than a preset quantity threshold, new feature values ​​can continue to be acquired until the sum of the total number of all new feature values ​​and the total number of all feature values ​​in the normal value cluster is greater than or equal to the quantity threshold. Each of the anomaly diagnostics can then recalculate the upper boundary value and the lower boundary value based on the acquired new feature values, the total number of feature values ​​in the normal value cluster corresponding to the anomaly diagnostics, and the upper and lower boundary determination strategy.

[0132] Based on the recalculated upper and lower boundary values, it can be determined whether there are any abnormal feature values ​​among all the new feature values ​​collected by the anomaly diagnostic tool and among all the feature values ​​in the normal value cluster corresponding to the anomaly diagnostic tool.

[0133] In one embodiment, after the anomaly diagnostic tool determines whether the initially collected 100 feature values ​​are abnormal based on the initial upper and lower boundary values, if some abnormal feature values ​​exist and are removed, then the number of normal feature values ​​is less than 100 (e.g., 97 normal feature values ​​remain). In subsequent anomaly diagnosis, each time the anomaly diagnostic tool receives a new feature value, it can accumulate the number of feature values ​​based on the existing number of normal feature values ​​(97 normal feature values). When the number of accumulated feature values ​​reaches 100 again, the upper and lower boundary values ​​can be recalculated based on the 100 feature values ​​and the aforementioned upper and lower boundary determination strategy to achieve automatic updating of the upper and lower boundary values.

[0134] In one embodiment, the anomaly diagnostic tool can detect whether there are anomalies in 100 feature values ​​based on the updated upper and lower boundary values. If anomalies are found, the abnormal feature values ​​can be removed. If no anomalies are found, the 100 feature values ​​are stored in the normal value cluster, and new feature values ​​are collected. At this time, the 100 feature values ​​in the normal value cluster plus the newly collected feature values ​​make up 101 feature values, which is greater than 100. At this time, the upper and lower boundary values ​​can be recalculated based on the 101 feature values ​​and the upper and lower boundary determination strategy to realize the automatic updating of the upper and lower boundary values. The tool can then detect whether there are anomalies in the 101 feature values ​​based on the updated upper and lower boundary values. If one or more feature values ​​are detected to be abnormal, it can be determined whether the abnormality is due to the newly collected feature values ​​or the feature values ​​stored in the original normal value cluster. If the abnormality is due to the newly collected feature values, the abnormality result can be reported. If the abnormality is due to the old feature values ​​(i.e., the feature values ​​originally stored in the normal value cluster), the abnormal value can simply be removed. This embodiment of the scheme enables continuous optimization of normal feature values ​​by automatically updating the upper and lower boundary values ​​and removing outliers from the old feature values.

[0135] In one embodiment, the above scheme can be repeated to implement the anomaly detection of feature values ​​in the embodiments of this application. During the execution of the above scheme, as the accumulated normal feature values ​​continue to increase, the number of normal feature values ​​will remain consistently above 100. Therefore, in the subsequent scheme, it means that after each new feature value is collected, the existing normal feature values ​​will be added to the new feature value to perform an upper and lower boundary determination strategy, that is, the upper boundary value and the lower boundary value will be automatically updated once, and then anomaly detection will be performed on the feature values ​​in the normal value cluster and the new feature value based on the updated upper boundary value and the lower boundary value.

[0136] In one embodiment, all 90 anomaly diagnostics described above will execute the same anomaly detection strategy, determining whether a feature value is abnormal upon receiving it.

[0137] In one embodiment, determining whether an operational phase is abnormal based on the judgment result of whether the feature value is abnormal in each operational phase may include:

[0138] After inputting the feature values ​​corresponding to a set of running data for the running phase into the corresponding anomaly diagnostics, the diagnostic results output by each anomaly diagnostics are obtained.

[0139] Count the first number of the first results that are judged to have abnormal feature values ​​and the second number of the second results that are judged to have normal feature values ​​among all the diagnostic results corresponding to this operation phase.

[0140] If the first quantity is greater than the second quantity, the operation phase can be determined to be abnormal; if the first quantity is less than or equal to the second quantity, the operation phase can be determined to be normal.

[0141] In one embodiment, each running stage corresponds to 9 anomaly diagnostics, which are anomaly diagnostics corresponding to the 9 feature values: Iq_std (torque mean square deviation), Iq_mean (torque mean), Iq_kur (torque kurtosis), V_Real_std (velocity mean square deviation), V_Real_mean (velocity mean), V_Real_kur (velocity kurtosis), Disp_std (displacement mean square deviation), Disp_mean (displacement mean), and Disp_kur (displacement kurtosis). When judging whether each operational stage is abnormal, the diagnostic results of the nine anomaly diagnostic tools can be statistically analyzed. For example, if the statistics show that only the characteristic values ​​Iq_std (torque mean square deviation) and Iq_mean (torque mean) are abnormal, while the other characteristic values ​​are normal, then there are 2 abnormal values ​​and 7 normal values. In this case, 2 is less than 7, so the operational stage can be judged as normal. As another example, if the statistics show that the characteristic values ​​Iq_mean (torque mean), Iq_kur (torque kurtosis), V_Real_std (speed mean square deviation), V_Real_mean (speed mean), and V_Real_kur (speed kurtosis) are abnormal, while the other characteristic values ​​are normal, then there are 5 abnormal values ​​and 4 normal values. In this case, 5 is greater than 4, so the operational stage can be judged as abnormal.

[0142] In one embodiment, for the operating stages in the door opening displacement curve and the door closing displacement curve: the first slow door opening stage s1, the accelerated door opening stage s2, the uniform door opening stage s3, the deceleration door opening stage s4, and the second slow door opening stage s5; and the first slow door closing stage s7, the accelerated door closing stage s8, the uniform door closing stage s9, the deceleration door closing stage s10, and the second slow door closing stage s11, the above-mentioned scheme can be used individually to detect whether the operating stage is abnormal.

[0143] In one embodiment, if any one or more abnormalities are detected in any operating phase, the current elevator door operator can be determined to be abnormal. Only when normality is detected in each operating phase can the elevator door operator be determined to be normal.

[0144] In one embodiment, when the current elevator door operator is determined to be abnormal, a fault location process can be initiated to determine the fault type.

[0145] In one embodiment, the fault type can be determined based on the magnitude of a specified type of characteristic value.

[0146] In one embodiment, the method may further include:

[0147] Based on the fault type to be determined, obtain the characteristic value of the specified type corresponding to the specified operating data within the specified operating stage required to determine the fault type;

[0148] Compare the numerical value of the specified feature type with the corresponding feature threshold;

[0149] If the comparison result between the specified feature value and the feature threshold meets the preset requirements, it can be determined that the elevator has experienced this type of fault.

[0150] In one embodiment, different fault types can be detected using different feature values ​​corresponding to different operating data at different operating stages. To determine whether a certain fault type exists in the current door operator, the feature value of the specified type corresponding to the specified operating data at the specified operating stage corresponding to the fault type can be obtained, and it can be detected whether the feature value of the specified type meets the preset requirements. If the preset requirements are met, it can be determined that the current elevator has experienced a fault of that type. If the preset requirements are not met, it can be determined that the current elevator has not experienced a fault of that type.

[0151] In one embodiment, the elevator malfunction type may include, but is not limited to, any one or more of the following: timing belt loosening, door knife jamming, and increased overall resistance;

[0152] The specified operating phase corresponding to the door knife jamming may include: the first slow-opening door phase;

[0153] The specified operating phase corresponding to the increase in overall resistance may include: the accelerated door opening phase and the accelerated door closing phase;

[0154] The specified operating phase corresponding to the timing belt detachment may include: the accelerated door opening phase and the second slow door opening phase;

[0155] The characteristic values ​​of the specified type corresponding to the gate knife jamming may include: mean torque, mean square of torque, and torque kurtosis;

[0156] The characteristic values ​​of a specified type corresponding to the increase in overall resistance may include: mean torque and mean square error of torque;

[0157] The characteristic values ​​of the specified type corresponding to the timing belt detachment may include: torque mean square error, torque kurtosis, and velocity kurtosis.

[0158] In one embodiment, the fault determination strategy for door knife jamming may include:

[0159] Calculate Iq_std (torque mean square deviation), Iq_kur (torque kurtosis), and Iq_mean (torque mean) for the "first soft opening stage S1" segment respectively.

[0160] When Iq_std>0.078, Iq_kur<7.31 and Iq_mean>0.034 are satisfied simultaneously, it can be determined that the elevator has a door knife jamming fault.

[0161] In one embodiment, the strategy for determining a fault due to increased overall resistance may include:

[0162] Calculate Iq_std1 (torque mean square deviation) for the "acceleration opening stage s2", Iq_std2 (torque mean square deviation) for the "acceleration closing stage s8", and Iq_mean (torque mean) for the "acceleration closing stage s8".

[0163] When Iq_std1<0.122, Iq_std2>0.032 and Iq_mean<-0.055 are satisfied simultaneously, it can be determined that the elevator has a fault of increased overall resistance.

[0164] In one embodiment, the strategy for determining a synchronization belt detachment fault may include:

[0165] Calculate Iq_std1 (torque mean square error) for the "accelerated door opening stage s2", Iq_kur (torque kurtosis) for the "accelerated door opening stage s2", and RealSpeed_kur (speed kurtosis) for the "second slow door opening stage s5" respectively.

[0166] When Iq_std>0.134, Iq_kur<3.004, and RealSpeed_kur>3.082 are all satisfied simultaneously, it can be determined that the elevator has a fault of loose timing belt.

[0167] This application also provides an elevator anomaly detection and fault type determination device 1, such as... Figure 6 As shown, it may include a processor 11 and a memory 12. The memory 12 stores instructions. When the instructions are executed by the processor 11, the elevator anomaly detection method described above can be implemented.

[0168] In one embodiment, any of the embodiments in the foregoing method embodiments are applicable to this device embodiment, and will not be described in detail here.

[0169] This application also provides an elevator 2, such as... Figure 7 As shown, it may include the elevator anomaly detection and fault type determination device 1.

[0170] In one embodiment, any of the embodiments in the foregoing method embodiments are applicable to this elevator embodiment, and will not be described in detail here.

[0171] This application also provides a non-transient computer-readable storage medium storing a computer program that, when executed by a processor, can implement the elevator anomaly detection method.

[0172] In one embodiment, any of the embodiments in the foregoing method embodiments are applicable to the non-transient computer-readable storage medium embodiment, and will not be described in detail here.

[0173] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

Claims

1. An elevator abnormality detection method characterized by comprising: The method comprises: obtaining operation data of different operation stages in a door opening and closing cycle of an elevator; wherein the operation stages include a plurality of door opening operation stages and a plurality of door closing operation stages, and each operation stage includes one or more operation data; calculating a feature value corresponding to each operation data in each operation stage; inputting the feature value corresponding to each operation data in each operation stage into the corresponding abnormality diagnostic device, and determining whether the feature value corresponding to each operation data in each operation stage is abnormal through the corresponding abnormality diagnostic device; wherein the abnormality diagnostic device calculates an upper boundary value and a lower boundary value of the feature value according to the obtained feature value, and detects whether the corresponding feature value is abnormal according to the upper boundary value and the lower boundary value; determining whether each operation stage is abnormal according to the determination result of whether the feature value corresponding to each operation data in each operation stage is abnormal; in the case that one or more operation data corresponding to one or more operation stages in all the operation stages are detected to be abnormal, determining that the elevator is faulty; determining whether each operation stage is abnormal according to the determination result of whether the feature value corresponding to each operation data in each operation stage is abnormal, comprising: obtaining the diagnosis result of each abnormality diagnostic device after inputting the feature value corresponding to a group of operation data in the operation stage into the corresponding abnormality diagnostic device; respectively counting a first number of abnormal feature values and a second number of normal feature values in all the diagnosis results corresponding to the operation stage; in the case that the first number is greater than the second number, determining that the operation stage is abnormal; in the case that the first number is less than or equal to the second number, determining that the operation stage is normal.

2. The elevator abnormality detection method according to claim 1, wherein the plurality of door opening operation stages include a door closed state stage, a first slow door opening stage, an accelerating door opening stage, a constant speed door opening stage, a decelerating door opening stage, and a second slow door opening stage; and the plurality of door closing operation stages include a door open state stage, a first slow door closing stage, an accelerating door closing stage, a constant speed door closing stage, a decelerating door closing stage, and a second slow door closing stage; the operation data include any one or more of the following: torque, speed, and displacement; the calculation of the feature value of the operation data includes: calculating any one or more of the following numerical values of the operation data: mean square error, mean value, and kurtosis.

3. The elevator abnormality detection method according to claim 2, characterized by, inputting the feature value into the corresponding abnormality diagnostic device, and determining whether the feature value is abnormal through the abnormality diagnostic device, comprising: continuously acquiring the corresponding feature value through the abnormality diagnostic device corresponding to each feature value in each operation stage; in the case that the number of feature values collected by each abnormality diagnostic device reaches a preset number threshold, each abnormality diagnostic device calculates an upper boundary value and a lower boundary value for abnormality detection according to all the collected feature values and a preset upper and lower boundary determination strategy; determining whether each collected feature value is abnormal according to the upper boundary value and the lower boundary value.

4. The elevator abnormality detection method according to claim 3, characterized by The method further comprises: In a case where the feature value is located outside a feature value range formed by the upper boundary value and the lower boundary value, it is determined that the feature value is abnormal; In a case where the feature value is located within the feature value range, it is determined that the feature value is normal.

5. The elevator abnormality detection method according to claim 3, characterized by, The method further comprises: In a case where it is determined that the feature value is abnormal, each of the abnormality diagnosers pushes an abnormality result, and eliminates the feature value or adds the feature value to a preset abnormal value cluster; In a case where it is determined that the feature value is normal, each of the abnormality diagnosers adds the feature value to a preset normal value cluster.

6. The elevator abnormality detection method according to claim 5, characterized by The method further comprises: In a case where a sum of a number of all new feature values obtained by each of the abnormality diagnosers and a number of all feature values in a corresponding normal value cluster is greater than or equal to the number threshold, the abnormality diagnosers recompute the upper boundary value and the lower boundary value corresponding to the abnormality diagnosers according to the all new feature values, the all feature values in the corresponding normal value cluster, and the upper and lower boundary determination strategy; and determine whether there is an abnormal feature value in the all new feature values obtained by the abnormality diagnosers and the all feature values in the normal value cluster corresponding to the abnormality diagnosers according to the recomputed upper boundary value and lower boundary value.

7. The elevator abnormality detection method according to claim 3 or 6, characterized by, The upper and lower boundary determination strategy comprises: arranging all the feature values in ascending order and dividing them into m equal parts, and taking feature values at m-1 segmentation points as m quantiles; m is a positive integer; obtaining a first m-quantile Q1 and an m-1th m-quantile Q m-1 ; Q1 < Q m-1 ; calculating an upper and lower quantile difference IQ R = Q m-1 - Q1; calculating the lower boundary value LowLim = Q1 - K1 IQ R , calculating the upper boundary value UpLim = Q m-1 + K2 IQ R .

8. The elevator abnormality detection method according to claim 7, characterized by, m=4, K1=K2=1.

5.

9. The elevator abnormality detection method according to claim 2, characterized by, The method further comprises: According to a fault type to be determined, obtaining feature values of a specified type corresponding to specified running data in a specified running phase required for determining the fault type; Comparing the feature values of the specified type with corresponding feature thresholds; In a case where a comparison result meets a preset requirement, it is determined that the elevator has a fault of the fault type.

10. The elevator abnormality detection method according to claim 9, characterized by, The fault types of the elevator include any one or more of the following: door blade blocking, overall resistance increase, and synchronous belt loosening; The specified running phase corresponding to the door blade blocking includes the first slow door opening phase; The specified running phase corresponding to the overall resistance increase includes the accelerating door opening phase and the accelerating door closing phase; The specified running phase corresponding to the synchronous belt loosening includes the accelerating door opening phase and the second slow door opening phase; The specified type of feature values corresponding to the door blade blocking includes torque mean value, torque mean square error, and torque kurtosis; The specified type of feature values corresponding to the overall resistance increase includes torque mean value and torque mean square error; The specified type of feature values corresponding to the synchronous belt loosening includes torque mean square error, torque kurtosis, and speed kurtosis.

11. An elevator abnormality detecting device characterized by comprising: An elevator abnormality detection device comprises a processor and a memory, and the memory stores instructions which, when executed by the processor, implement the elevator abnormality detection method according to any one of claims 1 to 10.

12. An elevator characterized by An elevator abnormality detection device according to claim 11.

13. A non-transitory computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a processor of a device, cause the device to perform operations comprising: The computer program, when executed by a processor, is capable of implementing the elevator anomaly detection method according to any one of claims 1 to 10.

Images