JPWO2018122921A1 - Recovery system - Google Patents

Abstract

The classifier (11) classifies the input time-series data and outputs a plurality of estimated failure contents. The classifier (13) classifies the input attribute data and outputs a plurality of estimated failure contents. The failure identification unit (14) identifies the content of a failure that has occurred in the elevator apparatus based on the estimated failure content output by the classifier (11) and the estimated failure content output by the classifier (13). The work specifying unit (15) specifies the work content for the failure content specified by the failure specifying unit (14).

Description

  The present invention relates to a recovery system.

  Patent Document 1 describes a system for remotely restoring an elevator apparatus after an earthquake occurs. In the system described in Patent Document 1, when the operation of the elevator apparatus is stopped due to an earthquake, a signal indicating that the operation is stopped and a signal indicating the state of the car are transmitted to the management center. In the management center, the received signal is displayed on the display. The manager of the management center sees the contents displayed on the display and transmits a signal for resetting the earthquake detector.

Japanese Unexamined Patent Publication No. 2007-254039

  Patent Document 1 discloses a system for restoring an elevator apparatus whose operation has been stopped due to an earthquake. When the operation is stopped by an earthquake, the elevator apparatus often has no failure. For this reason, an elevator apparatus can be easily restored by resetting an earthquake detector.

  On the other hand, a failure occurs in the elevator apparatus. When a failure occurs in the elevator apparatus, it is necessary to perform work corresponding to the failure that has occurred. Conventionally, when a failure occurs in the elevator apparatus, it is not possible to grasp appropriate work contents, and the elevator apparatus cannot be restored remotely.

  The present invention has been made to solve the above-described problems. An object of the present invention is to provide a recovery system that can specify an appropriate work content for a failure that has occurred when a failure occurs in an elevator apparatus.

  The recovery system according to the present invention includes a receiving means for receiving time series data from the elevator device, time series data received by the receiving means is input, the input time series data is classified, and a plurality of estimated failure contents are output. A first classifier, an acquisition means for acquiring attribute data related to the elevator apparatus, and an attribute data acquired by the acquisition means is input, the input attribute data is classified, and a plurality of estimated failure contents are output. , A failure specifying means for specifying the failure content generated in the elevator apparatus based on the estimated failure content output by the first classifier and the estimated failure content output by the second classifier, and the failure specified by the failure specifying means A work specifying means for specifying the work content for the content.

  The recovery system according to the present invention includes a storage unit storing a plurality of failure contents, a plurality of work contents, and a probability that each of the work contents is selected for each of the failure contents, and time series data from the elevator apparatus. A receiving means for receiving; a first classifier for receiving the time series data received by the receiving means; classifying the inputted time series data; and outputting an estimated occurrence probability for each of the failure contents stored in the storage means; The acquisition means for acquiring the attribute data related to the elevator apparatus, the attribute data acquired by the acquisition means is input, the input attribute data is classified, and the estimated occurrence probability for each of the failure contents stored in the storage means is output The second classifier, the estimated occurrence probability output by the first classifier, the estimated occurrence probability output by the second classifier, and the storage means. Based on the probability, and a specifying means for specifying one of the work stored in the storage means.

  In the restoration system according to the present invention, for example, the content of a failure that has occurred in the elevator apparatus is identified based on the estimated failure content output by the first classifier and the estimated failure content output by the second classifier. Further, the work content for the specified failure content is specified. With the recovery system according to the present invention, when a failure occurs in the elevator apparatus, it is possible to specify appropriate work content for the failure that has occurred.

It is a figure which shows the example of the recovery system in Embodiment 1 of this invention. It is a figure which shows the example of a management center. It is a flowchart which shows the operation example of the recovery system in Embodiment 1 of this invention. It is a flowchart which shows the other operation example of the recovery system in Embodiment 1 of this invention. It is a figure which shows the other example of a management center. It is a flowchart which shows the other operation example of the recovery system in Embodiment 1 of this invention. It is a figure which shows the example of a decision tree. It is a figure which shows the example of a management center. It is a flowchart which shows the operation example of the recovery system in Embodiment 2 of this invention. It is a figure which shows the hardware constitutions of a management center.

  The present invention will be described with reference to the accompanying drawings. The overlapping description will be simplified or omitted as appropriate. In each figure, the same reference numerals indicate the same or corresponding parts.

Embodiment 1 FIG.
FIG. 1 is a diagram showing an example of a recovery system according to Embodiment 1 of the present invention. The management center 1 can communicate with a number of remote elevator apparatuses. Each elevator device includes, for example, a car 2 and a counterweight 3. The car 2 and the counterweight 3 are suspended from the hoistway by the main rope 4. The hoisting machine includes, for example, a driving sheave 5 and an electric motor 6. The main rope 4 is wound around the driving sheave 5. The drive sheave 5 is driven by an electric motor 6. The electric motor 6 is controlled by the control device 7. A communication device 8 is connected to the control device 7. The communication device 8 communicates with an external device. Each elevator device communicates with the management center 1 through the communication device 8.

  When a failure occurs in the elevator apparatus, a snapshot of a signal value indicating the state of the elevator, that is, trace data is acquired by the communication apparatus 8. Trace data is an example of a signal group including a plurality of signals. For example, the trace data includes a signal for specifying the elevator apparatus itself. The trace data includes a signal indicating the time. The trace data includes a signal indicating the current value and voltage value of the control device 7. The trace data includes a signal indicating the speed and torque of the electric motor 6. A signal indicating the position of the car 2 is included in the trace data. The signals included in the trace data are not limited to these examples. Some of the exemplified signals may not be included in the trace data. Other signals may be included in the trace data.

  When a failure occurs in the elevator device, the communication device 8 acquires trace data for a certain time before and after the failure occurs. For example, when a failure occurs in the elevator apparatus, the communication device 8 acquires trace data every 5 ms for a period from 50 ms before the failure occurs to 50 ms after the failure occurs. When acquiring the trace data, the communication device 8 transmits the acquired trace data to the management center 1.

  Further, when a failure occurs in the elevator device, the communication device 8 transmits a diagnostic code together with the acquired trace data to the management center 1. The diagnosis code includes data indicating the cause of the failure.

  FIG. 2 is a diagram illustrating an example of the management center 1. The management center 1 includes, for example, a storage unit 9, a reception unit 10, a classifier 11, an acquisition unit 12, a classifier 13, a failure identification unit 14, a work identification unit 15, a determination unit 16, a transmission unit 17, and a notification control unit 18. . Hereinafter, the function and operation of the recovery system will be described with reference to FIG. FIG. 3 is a flowchart showing an operation example of the recovery system according to the first embodiment of the present invention.

  When a failure occurs in any elevator device, trace data and a diagnostic code are transmitted from the communication device 8 of the elevator device to the management center 1. Data transmitted from the communication device 8 is received by the receiving unit 10 in the management center 1 (S101). When the data from the communication device 8 is received by the receiving unit 10, the management center 1 determines whether or not the received data is time-series data (S102).

  The classifier 11 outputs an estimated failure content from the input data. Time series data is input to the classifier 11. The trace data received by the receiving unit 10 includes a large number of time-series data. For example, the transition of the signal indicating the current value of the control device 7 is time series data. The transition of the signal indicating the speed of the electric motor 6 is time series data. That is, the receiving unit 10 is an example of a unit that receives time-series data from the elevator apparatus. Of the data received by the receiving unit 10 in S101, the data determined to be time-series data in S102 is input to the classifier 11.

  The classifier 11 classifies the input time series data (S103), and estimates the failure content (S104). The classifier 11 outputs a plurality of estimated failure contents based on the classification result. Additional information such as an estimated failure occurrence probability may be attached to the estimated failure content output from the classifier 11.

  The classifier 13 outputs an estimated failure content from the input data. Attribute data is input to the classifier 13. The attribute data is data that is not time series data among the data related to the elevator apparatus. For example, the diagnostic code is attribute data. Data indicating the model of the elevator apparatus is attribute data. The data indicating the rated speed is attribute data. Data indicating the floor of a building equipped with an elevator device is attribute data.

  The acquisition unit 12 acquires attribute data. When the receiving unit 10 receives the trace data and the diagnostic code, the acquiring unit 12 acquires attribute data related to the elevator apparatus that has transmitted the data. For example, the acquisition unit 12 acquires attribute data from the data received by the reception unit 10. The acquisition unit 12 may acquire attribute data from data stored in the storage unit 9. Data determined to be not time-series data in S102 is input to the classifier 13.

  The classifier 13 classifies the input attribute data (S105), and estimates the failure content (S106). The classifier 13 outputs a plurality of estimated failure contents based on the classification result. Additional information such as the estimated failure probability may be attached to the estimated failure content output from the classifier 13.

  The failure identification unit 14 identifies the content of the failure that has occurred in the elevator apparatus (S107). The failure specifying unit 14 performs the above specification based on the estimated failure content output from the classifier 11 and the estimated failure content output from the classifier 13, for example. For example, if the estimated failure content output from the classifier 11 has the highest occurrence probability and the estimated failure content output from the classifier 13 has the highest occurrence probability, the failure specifying unit 14 Identifies the content of the estimated failure as the content of the failure that has occurred in the elevator apparatus. The failure identification unit 14 may identify one failure content based on a value obtained by multiplying the occurrence probability of the estimated failure content.

  The work specifying unit 15 specifies the work content for the failure content specified by the failure specifying unit 14 (S108). For example, the operation content to be selected is set in advance for each failure content that can be specified by the failure specifying unit 14. The setting contents are stored in the storage unit 9, for example. The work specifying unit 15 may refer to the setting content stored in the storage unit 9 and specify the work content for the failure content.

  The determining unit 16 determines whether or not the work content specified by the work specifying unit 15 can be performed by remote operation (S109). For example, when the work content such as “reset elevator device” is specified in S108, it is determined Yes in S109. When the determination unit 16 determines that the operation can be performed by remote operation, the transmission unit 17 transmits an instruction for causing the work specifying unit 15 to execute the work content specified in S108 to the elevator apparatus in which the failure has occurred. (S110).

  On the other hand, if the work content requiring dispatch of maintenance personnel such as “replacement of board” is identified in S108, it is determined No in S109. If the determination unit 16 does not determine that the operation can be performed by remote operation, for example, the notification control unit 18 causes the notification unit 22 to notify the work content specified by the work specifying unit 15 in S108 (S111). The notification control unit 18 may cause the notification unit 22 to notify the result determined by the determination unit 16 in S109 together with the work content. When it is determined by the determination unit 16 that the notification control unit 18 can be implemented by remote operation, the notification of the work content specified by the work specifying unit 15 in S108 and the result determined by the determination unit 16 in S109 from the notification device 22. It may be notified. For example, the notification device 22 is provided in the management center 1.

  If it is an example shown to this Embodiment, when a failure generate | occur | produces in an elevator apparatus, the suitable work content with respect to the generate | occur | produced failure can be specified. In particular, in the example shown in the present embodiment, the work content is specified based on the estimated failure content obtained by classifying the time series data and the estimated failure content obtained by classifying the attribute data. Since the classification by a plurality of classifiers using independent data for the same event is performed, the reliability of the specific result can be improved.

  In the example shown in FIG. 3, in S110, a command for performing the work content is automatically transmitted. This is an example. When it is determined Yes in S109, the notification of the work content specified by the work specifying unit 15 in S108 and the result determined by the determination unit 16 in S109 may be notified from the notification device 22 without performing automatic command transmission. . In such a case, a command for performing the work content is transmitted at the discretion of the administrator.

  In the example shown in FIG. 3, in S109, it is determined whether or not the work content specified by the work specifying unit 15 can be performed by remote operation. This is an example. The work determination unit 15 may notify the work content specified in S108 from the notification device 22 without performing the above determination after the process of S108. In such a case, the administrator determines whether the work content can be performed by remote operation.

  FIG. 4 is a flowchart showing another operation example of the recovery system according to Embodiment 1 of the present invention. FIG. 4 shows a specific example of the process shown in S109 of FIG. The determination unit 16 may determine whether the work content specified by the work specification unit 15 can be performed by remote operation based on the time-series data received by the reception unit 10 in S101.

  For example, the determination unit 16 determines whether or not a person is on the elevator car 2 in which a failure has occurred from a signal indicating a scale value included in the trace data (S201). If a person is on the car 2, the determination unit 16 does not determine that the work content specified by the work specification unit 15 can be performed by remote operation. The determination unit 16 may determine whether or not the main rope 4 is normal from the signal included in the trace data (S202). If an abnormality has occurred in the main rope 4, the determination unit 16 does not determine that the work content specified by the work specification unit 15 can be performed by remote operation.

  The determination unit 16 may determine whether or not the brake for stopping the drive sheave 5 is normal from the signal included in the trace data (S203). If an abnormality has occurred in the brake, the determination unit 16 does not determine that the work content specified by the work specification unit 15 can be performed by remote operation. The determination items shown in S201 to S203 are an example. Only one or more of the determination items shown in S201 to S203 may be adopted. You may employ | adopt another determination item as needed. For example, the determination unit 16 can perform the work content specified by the work specification unit 15 by remote operation based on the attribute data acquired by the acquisition unit 12 as well as the time series data received by the reception unit 10. It may be determined whether or not. If the determination unit 16 determines Yes for all the determination items that have been adopted, the transmission unit 17 transmits a command for causing the work specification unit 15 to perform the work content specified in S108 to the elevator apparatus in which the failure has occurred. (S110).

  FIG. 5 is a diagram illustrating another example of the management center 1. In the example illustrated in FIG. 5, the management center 1 further includes an update unit 19 and an update unit 20 in addition to the elements illustrated in FIG. 2. The update unit 19 updates the classifier 11. The update unit 20 updates the classifier 13. Hereinafter, functions and operations of the update unit 19 and the update unit 20 will be described with reference to FIGS. FIG. 6 is a flowchart showing another example of operation of the recovery system according to Embodiment 1 of the present invention.

  When a failure occurs in any elevator device, trace data and a diagnostic code are transmitted from the communication device 8 of the elevator device to the management center 1. Data transmitted from the communication device 8 is received by the receiving unit 10 in the management center 1 (S301). Data received by the receiving unit 10 is stored in the storage unit 9.

  The storage unit 9 stores data received from a number of elevator devices. The storage unit 9 also stores other data related to the elevator apparatus. For example, the storage unit 9 stores details of work performed specifically in the elevator apparatus in which the failure occurred. This work content is, for example, data input after work by a maintenance worker who actually performed the work. The storage unit 9 may store specific failure contents determined by the maintenance personnel. This failure content is data input after maintenance by the maintenance personnel who actually performed the operation. As another example, the storage unit 9 stores data such as the model, rated speed, and rated speed of each elevator apparatus. The storage unit 9 stores data such as floors of buildings equipped with the elevator devices. The update unit 19 and the update unit 20 perform update processing based on these data stored in the storage unit 9.

  For example, the updating unit 19 performs learning based on time-series data received by the receiving unit 10 from each elevator apparatus (S303), and updates the classifier 11 (S304). The process of S302 is the same as the process of S102 shown in FIG. Data that is determined to be time-series data in S302 is input to the update unit 19.

The update unit 19 may update the classifier 11 by any method. For example, regarding time series data, [x ₁ , x ₂ ,..., X _M ] ^{T in} which time series data exists for each of M signals is input to the updating unit 19. Here, [•] ^T indicates transposition. The update unit 19 performs supervised learning based on data input after maintenance work by a neural network (NN), for example.

  For example, the update unit 20 performs learning based on attribute data regarding each elevator apparatus (S305), and updates the classifier 13 (S306). Data that is determined not to be time-series data in S302 is input to the update unit 20.

  The update unit 20 may update the classifier 13 by any method. For example, Table 1 below shows an example of attribute data stored in the storage unit 9. For example, the updating unit 20 performs learning by generating a decision tree as shown in FIG. FIG. 7 is a diagram illustrating an example of a decision tree. Note that the learning method by the updating unit 20 is not limited to a decision tree. For example, a method such as pattern mining may be employed as learning by the update unit 20.

  In the example shown in FIGS. 5 and 6, since learning using independent data is performed on each classifier for the same event, the reliability of the specific result can be improved.

Embodiment 2. FIG.
An example of the recovery system in the present embodiment is the same as the example shown in FIG. FIG. 8 is a diagram illustrating an example of the management center 1. In the example shown in the present embodiment, the management center 1 includes, for example, the storage unit 9, the reception unit 10, the classifier 11, the acquisition unit 12, the classifier 13, the identification unit 21, the determination unit 16, the transmission unit 17, and the notification control unit. 18 is provided.

  For example, a failure list, a work list, and a probability list are stored in the storage unit 9. Table 2 shows an example of the failure list. The failure list includes a plurality of failure contents that may occur in the elevator apparatus.

  Table 3 shows an example of a work list. The work list includes a plurality of work contents that can be performed by the elevator apparatus in which the failure has occurred.

  Table 4 shows an example of the probability list. The probability list includes a probability that each of the work contents included in the work list is selected for each of the failure contents included in the failure list.

  Hereinafter, functions and operations of the recovery system will be described with reference to FIG. FIG. 9 is a flowchart showing an operation example of the recovery system according to the second embodiment of the present invention.

  When a failure occurs in any elevator device, trace data and a diagnostic code are transmitted from the communication device 8 of the elevator device to the management center 1. Data transmitted from the communication device 8 is received by the receiving unit 10 in the management center 1 (S401). When the data from the communication device 8 is received by the receiving unit 10, the management center 1 determines whether the received data or the like is time-series data (S402).

In the present embodiment, the classifier 11 outputs an estimated occurrence probability for each of the fault contents included in the fault list from the input data. For example, the time series data received by the receiving unit 10 is input to the classifier 11. The classifier 11 classifies the input time series data (S403), and estimates the occurrence probability (S404). The classifier 11 outputs the estimated occurrence probability P _u (x _i ) for each failure content based on the classification result. Here, x _i is the i-th fault condition described in the fault list.

Further, the classifier 13 outputs an estimated occurrence probability for each of the fault contents included in the fault list from the input data. For example, the attribute data acquired by the acquisition unit 12 is input to the classifier 13. The classifier 13 classifies the input attribute data (S405), and estimates the occurrence probability (S406). The classifier 13 outputs the estimated occurrence probability P _v (x _i ) for each failure content based on the classification result.

The identification unit 21 identifies one of the work contents included in the work list (S408). The identifying unit 21 performs the above identification based on the estimated occurrence probability output from the classifier 11, the estimated occurrence probability output from the classifier 13, and the probability included in the probability list. For example, a result obtained by multiplying the estimated occurrence probability P _u (x _i ) output from the classifier 11 and the probability included in the probability list is P _u (x _i ) P (x _i | z _j ). A result obtained by multiplying the estimated occurrence probability P _v (x _i ) output from the classifier 11 and the probability included in the probability list is defined as P _v (x _i ) P (x _i | z _j ). Here, z _j is the i-th work content described in the work list. For example, the specifying unit 21 selects the work content having the highest average value of P _u (x _i ) P (x _i | z _j ) and P _v (x _i ) P (x _i | z _j ).

  The determination unit 16 determines whether or not the work content specified by the specifying unit 21 can be performed by remote operation (S409). When the determination unit 16 determines that the operation can be performed by remote operation, the transmission unit 17 transmits a command for causing the specification unit 21 to execute the work content specified in S408 to the elevator apparatus in which the failure has occurred ( S410).

  On the other hand, when the determination unit 16 does not determine that the operation can be performed by remote operation, for example, the notification control unit 18 causes the notification unit 22 to notify the work content specified by the specifying unit 21 in S408 (S411). The notification control unit 18 may cause the notification unit 22 to notify the result determined by the determination unit 16 in S409 together with the above-described work content. When the determination unit 16 determines that the notification control unit 18 can be implemented by remote operation, the notification control unit 18 notifies the notification unit 22 of the work content specified by the specifying unit 21 in S408 and the result determined by the determination unit 16 in S409. You may let them.

  If it is an example shown to this Embodiment, when a failure generate | occur | produces in an elevator apparatus, the suitable work content with respect to the generated failure can be specified. In particular, in the example shown in the present embodiment, the work content is specified based on the estimated occurrence probability obtained by classifying time-series data and the estimated occurrence probability obtained by classifying attribute data. Since the classification by a plurality of classifiers using independent data for the same event is performed, the reliability of the specific result can be improved.

  In the example shown in FIG. 9, in S410, a command for performing the work content is automatically transmitted. This is an example. When it is determined Yes in S409, automatic transmission of the command may not be performed, and the work content specified by the specifying unit 21 in S408 and the result determined by the determining unit 16 in S409 may be notified from the notification device 22. In such a case, a command for performing the work content is transmitted at the discretion of the administrator.

  In the example shown in FIG. 9, in S409, it is determined whether or not the work content specified by the specifying unit 21 can be performed by remote operation. This is an example. After the process of S408 is performed, the determination may not be performed, and the work content specified by the specifying unit 21 in S408 may be notified from the notification device 22. In such a case, the administrator determines whether the work content can be performed by remote operation.

  Also in the example shown in the present embodiment, the determination unit 16 determines whether the work content specified by the specification unit 21 can be performed by remote operation based on the time-series data received by the reception unit 10 in S401. You may do it. That is, the operation example shown in FIG. 4 may be adopted. Further, the determination unit 16 determines whether the work content specified by the specifying unit 21 can be performed by remote operation based on not only the time-series data received by the receiving unit 10 but also the attribute data acquired by the acquiring unit 12. It may be determined.

  The management center 1 may further include an updating unit 19 and an updating unit 20 in addition to the elements shown in FIG. The updating unit 19 performs learning based on time-series data received by the receiving unit 10 from each elevator apparatus, for example, and updates the classifier 11. The update unit 19 may update the classifier 11 by any method. The update unit 19 performs supervised learning based on data input after maintenance work by a neural network (NN), for example. The update unit 20 performs learning based on attribute data regarding each elevator device, for example, and updates the classifier 13. The update unit 20 may update the classifier 13 by any method. For example, the update unit 20 performs learning by generating a decision tree.

  Each element indicated by reference numerals 9 to 21 represents a function of the management center 1. FIG. 10 is a diagram illustrating a hardware configuration of the management center 1. The management center 1 includes a processing circuit including, for example, a processor 23 and a memory 24 as hardware resources. The functions of the storage unit 9 are realized by the memory 24. The management center 1 implements the functions of the units indicated by reference numerals 10 to 21 by executing the program stored in the memory 24 by the processor 23.

  The processor 23 is also called a CPU (Central Processing Unit), a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, or a DSP. As the memory 24, a semiconductor memory, a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD may be employed. Semiconductor memories that can be used include RAM, ROM, flash memory, EPROM, EEPROM, and the like.

  Some or all of the functions of the management center 1 may be realized by hardware. As hardware for realizing the function of the management center 1, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof may be employed.

  The recovery system according to the present invention can be applied to a system capable of communicating with an elevator apparatus.

  DESCRIPTION OF SYMBOLS 1 Management center, 2 cage | basket, 3 counterweight, 4 main rope, 5 drive sheave, 6 motor, 7 control apparatus, 8 communication apparatus, 9 memory | storage part, 10 receiving part, 11 classifier, 12 acquisition part, 13 classifier , 14 failure identification unit, 15 work identification unit, 16 determination unit, 17 transmission unit, 18 notification control unit, 19 update unit, 20 update unit, 21 identification unit, 22 alarm unit, 23 processor, 24 memory

The storage unit 9 stores data received from a number of elevator devices. The storage unit 9 also stores other data related to the elevator apparatus. For example, the storage unit 9 stores details of work performed specifically in the elevator apparatus in which the failure occurred. This work content is, for example, data input after work by a maintenance worker who actually performed the work. The storage unit 9 may store specific failure contents determined by the maintenance personnel. This failure content is data input after maintenance by the maintenance personnel who actually performed the operation. As another example, the storage unit 9 stores data such as the model, rated speed, and rated capacity of each elevator apparatus. The storage unit 9 stores data such as floors of buildings equipped with the elevator devices. The update unit 19 and the update unit 20 perform update processing based on these data stored in the storage unit 9.

The identification unit 21 identifies one of the work contents included in the work list (S408). The identification unit 21 performs the identification based on the estimated occurrence probability output from the classifier 11, the estimated occurrence probability output from the classifier 13, and the probability included in the probability list. For example, a result obtained by multiplying the estimated occurrence probability P _u (x _i ) output from the classifier 11 and the probability included in the probability list is P _u (x _i ) P (x _i | z _j ). A result obtained by multiplying the estimated occurrence probability P _v (x _i ) output from the classifier 13 and the probability included in the probability list is defined as P _v (x _i ) P (x _i | z _j ). Here, z _j is the i-th work content described in the work list. For example, the specifying unit 21 selects the work content having the highest average value of P _u (x _i ) P (x _i | z _j ) and P _v (x _i ) P (x _i | z _j ).

Claims (10)

Receiving means for receiving time series data from the elevator device;
A first classifier that receives the time-series data received by the receiving means, classifies the input time-series data, and outputs a plurality of estimated fault contents;
Obtaining means for obtaining attribute data relating to the elevator device;
A second classifier that receives the attribute data acquired by the acquisition unit, classifies the input attribute data, and outputs a plurality of estimated failure contents;
A failure identifying means for identifying a failure content generated in the elevator apparatus based on the estimated failure content output by the first classifier and the estimated failure content output by the second classifier;
Work specifying means for specifying work contents for the failure contents specified by the failure specifying means;
Recovery system with   The recovery system according to claim 1, further comprising determination means for determining whether or not the work content specified by the work specification means can be implemented by remote operation.   The restoration according to claim 2, further comprising: a transmission unit that transmits a command for executing the work content specified by the work specifying unit to the elevator device when the determination unit determines that the operation can be performed by remote operation. system. Storage means storing a plurality of failure contents, a plurality of work contents, and a probability that each of the work contents is selected for each of the failure contents;
Receiving means for receiving time-series data from the elevator device;
A first classifier that receives the time series data received by the receiving unit, classifies the input time series data, and outputs an estimated occurrence probability for each of the failure contents stored in the storage unit;
Obtaining means for obtaining attribute data relating to the elevator device;
A second classifier that receives the attribute data acquired by the acquisition unit, classifies the input attribute data, and outputs an estimated probability of occurrence of each failure content stored in the storage unit;
Based on the estimated occurrence probability output by the first classifier, the estimated occurrence probability output by the second classifier, and the probability stored in the storage unit, one of the work contents stored in the storage unit is selected. A specific means of identifying one,
Recovery system with   The recovery system according to claim 4, further comprising determination means for determining whether or not the work content specified by the specifying means can be implemented by remote operation.   6. The recovery system according to claim 5, further comprising a transmission unit that transmits a command for performing the work content specified by the specifying unit to the elevator device when the determination unit determines that the operation can be performed by remote operation. .   6. The recovery system according to claim 2, wherein the determination unit determines whether or not the work content can be performed by remote operation based on the time series data received by the reception unit.   6. The determination unit according to claim 2 or 5, wherein the determination unit determines whether the work content can be implemented by remote operation based on the time series data received by the reception unit and the attribute data acquired by the acquisition unit. The recovery system described.   The recovery system according to claim 2 or 5, further comprising notification control means for notifying a result determined by the determination means from an alarm device. Learning based on the time-series data received by the receiving means, and a first updating means for updating the first classifier;
Learning based on attribute data relating to the elevator device, second updating means for updating the second classifier;
The recovery system according to any one of claims 1 to 9, further comprising:

Images