JPWO2018122928A1 - Recovery support system - Google Patents

Abstract

The recovery support system includes a storage unit (10), a learning unit (12), a reception unit (11), and a determination unit (13). The storage unit (10) stores failure data and work data. The learning unit (12) performs machine learning on the failure data and work data stored in the storage unit (10). When the receiving unit (11) receives the failure data, the determination unit (13) dispatches maintenance personnel to correct the failure of the device that has transmitted the failure data based on the learning result by the learning unit (12). Determine whether it is necessary.

Description

  The present invention relates to a recovery support 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 monitoring center. In the monitoring center, the received signal is displayed on a display. The monitor at the monitoring center looks at the contents displayed on the display and transmits a signal for resetting the seismic 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. This work includes extremely simple work such as restarting the elevator apparatus. Conventionally, when a failure occurs in the elevator apparatus, it has not been possible to accurately determine whether or not it is necessary to dispatch a maintenance staff to fix the failure. For this reason, when a failure occurs in the elevator apparatus, maintenance personnel rush to the site and perform appropriate work.

  Such a problem can occur in other devices that require dispatch of maintenance personnel when a failure occurs.

  The present invention has been made to solve the above-described problems. An object of the present invention is to provide a recovery support system capable of accurately determining whether or not it is necessary to dispatch a maintenance staff in order to correct a failure of the device when a failure occurs in the device.

  The recovery support system according to the present invention includes: a storage unit that stores failure data that indicates a failure state of a device in which a failure has occurred; work data that indicates details of work performed to correct the failure; Learning means for machine learning of stored failure data and work data, receiving means for receiving failure data, and when the receiving means receives the failure data, the failure data has been transmitted based on the learning result by the learning means. Determination means for determining whether or not a maintenance person needs to be dispatched to correct the failure of the apparatus.

  The recovery support system according to the present invention includes a monitoring center and a plurality of devices capable of communicating with the monitoring center. The monitoring center includes failure data indicating failure status of the device in which the failure has occurred, storage means storing work data indicating work contents performed to correct the failure, and failure data stored in the storage means. And learning means for machine learning of work data. Each of the plurality of devices has an acquisition means for acquiring failure data, and maintenance is performed to correct the failure when the failure data is acquired based on a learning result by the learning means when the acquisition means acquires the failure data. Determining means for determining whether or not dispatch of an employee is necessary.

  The recovery support system according to the present invention includes, for example, a learning unit and a determination unit. The learning means performs machine learning on the failure data and work data stored in the storage means. The determination means determines whether or not a maintenance person needs to be dispatched to correct the failure of the device that has transmitted the failure data, based on the learning result of the learning means. With the recovery support system according to the present invention, when a failure occurs in an apparatus, it can be accurately determined whether or not it is necessary to dispatch a maintenance person to correct the failure of the apparatus.

It is a figure which shows the example of the recovery assistance system in Embodiment 1 of this invention. It is a flowchart which shows the operation example of the recovery assistance system in Embodiment 1 of this invention. It is a figure which shows the example of a learning result. It is a flowchart which shows the other operation example of the recovery assistance system in Embodiment 1 of this invention. It is a figure for demonstrating the function of a determination part. It is a figure for demonstrating the other function of a learning part and a determination part. It is a flowchart which shows the other operation example of the recovery assistance system in Embodiment 1 of this invention. It is a flowchart which shows the other operation example of the recovery assistance system in Embodiment 1 of this invention. It is a figure for demonstrating the other function of a determination part and an alerting | reporting control part. It is a figure which shows the other example of a recovery assistance system. It is a figure which shows the example of the recovery assistance system in Embodiment 2 of this invention. It is a flowchart which shows the operation example of the recovery assistance system in Embodiment 2 of this invention. It is a flowchart which shows the other operation example of the recovery assistance system in Embodiment 2 of this invention. It is a figure which shows the hardware constitutions of a monitoring 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 support system according to Embodiment 1 of the present invention. The monitoring 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 monitoring center 1 by the communication device 8.

  When a failure occurs in the elevator apparatus, a snapshot of the signal value indicating the state of the elevator apparatus at the time of failure, that is, trace data is acquired by the communication device 8. The trace data is an example of the failure data described in the claims. 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.

  The trace data may include a signal represented by a bit string of 0 or 1, a signal represented by a hexadecimal numeric string, and a signal represented by a decimal numeric string. Signals of various signal lengths may be mixed in the trace data. Digital values and analog values may be mixed 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 monitoring center 1.

  When a failure occurs in the elevator apparatus, maintenance personnel may be dispatched to correct the failure. The maintenance staff performs appropriate work according to the failure that has occurred and restores the elevator apparatus. When the restoration work is completed, the maintenance staff registers work data from the maintenance terminal 9, for example. The work data is data indicating the content of work performed to correct a failure that has occurred in the elevator apparatus. For example, the work data includes data indicating when and who performed what work. The work data may include data indicating details of the failure that has occurred. The work data may include data indicating the replaced part. The registered work data is transmitted from the maintenance terminal 9 to the monitoring center 1.

  The monitoring center 1 includes, for example, a storage unit 10, a reception unit 11, a learning unit 12, a determination unit 13, a transmission unit 14, and a notification control unit 15. Hereinafter, functions and operations of the recovery support system will be described in detail with reference to FIGS. FIG. 2 is a flowchart showing an operation example of the recovery support system according to Embodiment 1 of the present invention. FIG. 2 shows an example of the learning function of the recovery support system.

  The monitoring center 1 determines whether or not trace data has been received (S101). When a failure occurs in any elevator device, trace data is transmitted from the communication device 8 of the elevator device to the monitoring center 1. The trace data transmitted from the communication device 8 is received by the receiving unit 11 in the monitoring center 1 (Yes in S101). The trace data received by the receiving unit 11 is stored in the storage unit 10 (S102).

  In the monitoring center 1, it is determined whether or not work data has been received (S103). The maintenance staff transmits work data from the maintenance terminal 9 when the repair of the elevator apparatus is completed. The work data transmitted from the maintenance terminal 9 is received by the receiving unit 11 in the monitoring center 1 (Yes in S103). The work data received by the receiving unit 11 is associated with the corresponding trace data and stored in the storage unit 10 (S104). That is, the storage unit 10 stores failure data indicating the state of the elevator apparatus in which the failure has occurred and work data indicating the work content performed to correct the failure.

  The receiving unit 11 receives trace data from a number of elevator apparatuses. The receiving unit 11 receives work data from many maintenance terminals 9. Trace data and work data are accumulated in the storage unit 10.

  The learning unit 12 performs machine learning on the failure data and work data stored in the storage unit 10. In the monitoring center 1, it is determined whether or not it is a learning timing (S105). The learning timing is set in advance. If it is determined in S105 that it is the learning timing, machine learning is performed by the learning unit 12, and a learning result is output (S106). As an example, the learning unit 12 outputs a determination criterion as a learning result.

  For example, the trace data stored in the storage unit 10 is classified in advance into a first group and a second group. The first group includes trace data of cases in which maintenance personnel can recover without going to the site. For example, trace data of a case where the elevator apparatus is restored by restarting the elevator apparatus from the monitoring center 1 is classified into the first group. In addition, the first group includes trace data of a case where the maintenance staff went to the site but could be recovered without the maintenance staff going to the site. For example, trace data of a case where the elevator apparatus is restored simply by restarting the elevator apparatus at the site is classified into the first group.

  On the other hand, the second group includes trace data of cases that could not be recovered unless the maintenance staff went to the site. For example, trace data of a case where the elevator apparatus is restored by replacement of parts on site by the maintenance staff is classified into the second group. The classification of the trace data is performed based on, for example, work data. Classification flags such as “recovery only by restart” and “parts replacement” may be prepared in advance in the work data transmitted from the maintenance terminal 9. In such a case, the trace data can be classified based on the classification flag. If the work data includes contents freely described by the maintenance staff, for example, a technique such as text mining may be used to perform the trace data classification process based on the description contents.

  The learning unit 12 determines a determination criterion for the trace data classified into two groups by using a technique such as supervised learning or clustering. Examples of the techniques include support vector machines, deep learning, hierarchical clustering, and the like. FIG. 3 is a diagram illustrating an example of a learning result. The horizontal axis in FIG. 3 is the value of a certain signal included in the trace data. The vertical axis in FIG. 3 is the value of another signal included in the trace data. Black circles shown in FIG. 3 indicate the trace data classified into the first group. White circles shown in FIG. 3 indicate the trace data classified into the second group. The learning unit 12 may output a boundary line expression such as a straight line A as a learning result. A straight line A shows an example of a learning result obtained using a support vector machine. The learning unit 12 may output a standard deviation for specifying the region B1 and the region B2 as a learning result. The learning unit 12 may output the center point C1 and the center point C2 as learning results.

  FIG. 3 shows an example in which the determination criterion is determined based on the values of two signals included in the trace data as the simplest example. Since the trace data includes a large number of signals, the same learning may be performed by treating the trace data as a multidimensional vector.

  FIG. 4 is a flowchart showing another example of operation of the recovery support system according to Embodiment 1 of the present invention. FIG. 4 shows an example of the determination function of the recovery support system.

  The monitoring center 1 determines whether or not the trace data has been received (S201). When a failure occurs in any elevator device, trace data is transmitted from the communication device 8 of the elevator device to the monitoring center 1. The trace data transmitted from the communication device 8 is received by the receiving unit 11 in the monitoring center 1 (Yes in S201).

  When the trace data is received by the reception unit 11, the determination unit 13 determines whether or not it is necessary to dispatch maintenance personnel to correct the failure of the elevator apparatus that has transmitted the trace data (S202). The determination unit 13 performs the above determination based on the learning result from the learning unit 12. When the learning unit 12 outputs a determination criterion as a learning result, the determination unit 13 performs the above determination based on the determination criterion determined by the learning unit 12, for example.

  FIG. 5 is a diagram for explaining the function of the determination unit 13. For example, consider the case where the learning unit 12 outputs the straight line A as a learning result. If the coordinate indicating the trace data to be determined is above the straight line A, the determination unit 13 determines that the trace data is closer to the trace data classified into the first group than the trace data classified into the second group. To do. For example, if the coordinate D indicating the trace data received by the receiving unit 11 in S201 is above the straight line A, the determining unit 13 determines that it is not necessary to dispatch maintenance personnel. On the other hand, if the coordinate indicating the trace data to be determined is below the straight line A, the determination unit 13 converts the trace data into the trace data classified into the second group from the trace data classified into the first group. Judge as close. For example, if the coordinate D indicating the trace data received by the receiving unit 11 in S201 is below the straight line A, the determining unit 13 determines that a maintenance person needs to be dispatched.

  The determination method of the determination unit 13 may be an appropriate method according to the determination criterion determined by the learning unit 12. For example, the determination unit 13 may determine whether the region B1 includes the coordinates D indicating the trace data received by the reception unit 11 in S201. If the coordinate D is included in the area B1, the determination unit 13 determines that it is not necessary to dispatch a maintenance staff. If the coordinate D is not included in the area B1, the determination unit 13 determines that a maintenance person needs to be dispatched. The determination unit 13 may output that the determination cannot be made unless the coordinates D are included in both the region B1 and the region B2.

  As another example, the determination unit 13 may compare the distance between the coordinate D indicating the trace data received by the reception unit 11 in S201 and the center point C1, and the distance between the coordinate D and the center point C2. If the distance between the coordinate D and the center point C1 is shorter than the distance between the coordinate D and the center point C2, the determination unit 13 determines that it is not necessary to dispatch a maintenance staff. If the distance between the coordinates D and the center point C1 is longer than the distance between the coordinates D and the center point C2, the determination unit 13 determines that a maintenance person needs to be dispatched.

  The determination unit 13 may output a continuous value as a determination result instead of the binary value of dispatching maintenance personnel or not dispatching maintenance personnel. For example, the determination unit 13 may output the probability of dispatching maintenance personnel. The determination unit 13 may calculate the probability based on the distance between the coordinate D and the center point C1 and the distance between the coordinate D and the center point C2. The determination unit 13 may calculate the probability in a stepwise manner. The determination unit 13 may normalize the distance using the standard deviation.

  When the determination unit 13 determines that it is not necessary to dispatch maintenance personnel, the notification control unit 15 causes the notification device 21 to notify the result determined by the determination unit 13 in S202 (S203). For example, the alarm device 21 is provided in the monitoring center 1. If the determination unit 13 determines that it is not necessary to dispatch maintenance personnel, the transmission unit 14 sends an instruction for performing an operation necessary to correct the failure to the elevator apparatus that has transmitted the trace data. Transmit (S204). In the elevator apparatus that has received the command, an operation necessary to correct the failure is performed. For example, in the elevator apparatus that has received the command, restart is performed.

  When the determination unit 13 determines that a maintenance person needs to be dispatched, the notification control unit 15 causes the notification device 21 to notify the result determined by the determination unit 13 in S202 (S205). When the determination unit 13 determines that a maintenance person needs to be dispatched, the transmission unit 14 transmits a maintenance staff dispatch command to the maintenance staff base or the like (S206).

  In the example shown in the present embodiment, when a failure occurs in the elevator apparatus, it can be accurately determined whether or not it is necessary to dispatch a maintenance person to correct the failure of the apparatus. In particular, in the example shown in the present embodiment, machine learning using both failure data and work data stored in the storage unit 10 is performed. Since the determination based on many past cases can be performed, the determination accuracy can be improved.

  In the example shown in FIG. 4, in S204, a command for performing an operation necessary for correcting the failure is automatically transmitted. This is an example. When it is determined No in S202, only the notification of the determination result may be performed. In such a case, the command is transmitted based on the judgment of the supervisor.

  In the example shown in FIG. 4, a maintenance staff dispatch request is automatically made in S206. This is an example. When it determines with Yes in S202, you may perform only alerting | reporting of a determination result. In such a case, a dispatch request for maintenance personnel is made at the discretion of the supervisor.

  In the example illustrated in FIG. 4, if it is determined No in S202, both the processing in S203 and the processing in S204 are performed. This is an example. When the automatic command transmission is performed in S204, the process of S203 may not be performed. Similarly, in the example shown in FIG. 4, if it is determined Yes in S202, both the processing in S205 and the processing in S206 are performed. This is an example. If a maintenance staff dispatch request is automatically made in S206, the process of S205 may not be performed.

  FIG. 6 is a diagram for explaining other functions of the learning unit 12 and the determination unit 13. FIG. 6 shows an example in which the learning unit 12 determines a plurality of determination criteria as learning results. In the example shown in FIG. 6, the learning unit 12 determines a straight line A1, a straight line A2, and a straight line A3 as determination criteria. FIG. 6 shows an example. The number of determination criteria determined by the learning unit 12 may be two or four or more. The learning unit 12 performs machine learning on the data stored in the storage unit 10 and determines a plurality of determination criteria.

  In FIG. 6, symbols filled in with black indicate the trace data classified into the first group. White circles shown in FIG. 6 indicate the trace data classified into the second group. A straight line A1 shown in FIG. 6 is the same as the straight line A shown in FIG. The straight line A1 is a boundary line between the trace data classified into the first group and the trace data classified into the second group.

  The straight line A2 and the straight line A3 are boundary lines for further finely dividing the trace data classified into the first group. For example, the black symbols in FIG. 6 indicate the trace data of the case that is restored by only restarting. Among these, black circles indicate trace data of a case where a failure occurred again within one week after restart. The black triangle shows trace data of a case where a failure did not occur for one week after the restart but failed again within one month after the restart. Black squares indicate trace data of a case where no failure occurred during one month after restart. The straight line A2 is a boundary line between a black circle and a black triangle. The straight line A3 is a boundary line between the black triangle and the black square.

  In order for the learning unit 12 to determine the straight line A2 and the straight line A3, data on an interval at which a failure has occurred in each elevator apparatus is required. For example, data on the elapsed time since the failure occurred last time is stored in the storage unit 10.

  FIG. 7 is a flowchart showing another example of operation of the recovery support system according to Embodiment 1 of the present invention. FIG. 7 shows an example of the determination function of the recovery support system. The processes shown in S301 and S302 of FIG. 7 are the same as the processes shown in S201 and S202 of FIG. When the trace data is received by the reception unit 11, the determination unit 13 determines whether or not it is necessary to dispatch maintenance personnel to correct the failure of the elevator apparatus that has transmitted the trace data (S302). The determination unit 13 performs the above determination based on the learning result from the learning unit 12. In the example illustrated in FIG. 6, the determination unit 13 performs the above determination based on the straight line A <b> 1 determined by the learning unit 12.

  When it is not necessary to dispatch maintenance personnel, the determination unit 13 determines when the failure occurs again in the elevator apparatus (S307). The determination unit 13 performs the above determination based on the determination criterion determined by the learning unit 12. In the example shown in FIG. 6, the determination unit 13 performs the determination in S307 based on the straight line A2 and the straight line A3. For example, when the trace data indicated by the coordinate D is received in S301, the determination unit 13 determines that it is not necessary to dispatch maintenance personnel based on the straight line A1. Further, the determination unit 13 determines that the failure estimation time is “one week later and within one month” because the coordinates D are arranged between the straight line A2 and the straight line A3.

  When the failure estimation time is determined by the determination unit 13, the notification control unit 15 causes the notification unit 21 to notify the result determined by the determination unit 13 in S302 and S307 (S303). When the failure estimation time is determined by the determination unit 13, the transmission unit 14 transmits a command for performing an operation necessary for correcting the failure to the elevator apparatus that has transmitted the trace data (S304). . In the elevator apparatus that has received the command, an operation necessary to correct the failure is performed. For example, in the elevator apparatus that has received the command, restart is performed.

  On the other hand, when the determination unit 13 determines that a maintenance person needs to be dispatched, the notification control unit 15 causes the notification unit 21 to notify the result determined by the determination unit 13 in S302 (S305). When the determination unit 13 determines that a maintenance person needs to be dispatched, the transmission unit 14 transmits a maintenance staff dispatch command to the maintenance staff base or the like (S306).

  FIG. 6 shows an example in which the learning unit 12 determines the straight line A1, the straight line A2, and the straight line A3 as determination criteria. This is an example. The learning unit 12 may output a standard deviation or a center point as a learning result. Further, the determination data may be determined by grouping trace data for each cause of failure. The learning unit 12 may output both a determination criterion based on the failure reoccurrence time and a determination criterion based on the failure occurrence factor.

  FIG. 6 shows an example in which the determination unit 13 outputs a binary value indicating that a maintenance worker is dispatched or a maintenance worker is not dispatched as a determination result. This is an example. The determination unit 13 may output a continuous value as a determination result. For example, the determination unit 13 may determine “probability P1% that a failure will occur again within one week” and “probability P2% that a failure will occur again within one month”. The determination unit 13 determines “the probability P3% that a failure occurs again within one week due to a specific factor” and “the probability P4% that a failure occurs again within one week due to a factor other than the specific factor”. Also good.

  In the example shown in FIGS. 6 and 7, for example, when maintenance personnel are not dispatched, it is possible to determine when the failure occurs again. For this reason, other measures such as preparing replacement parts in advance are possible.

  FIG. 8 is a flowchart showing another operation example of the recovery support system according to Embodiment 1 of the present invention. FIG. 8 shows an example of the determination function of the recovery support system. FIG. 9 is a diagram for explaining other functions of the determination unit 13 and the notification control unit 15. The processes shown in S401 and S402 of FIG. 8 are the same as the processes shown in S201 and S202 of FIG. When the trace data is received by the reception unit 11, the determination unit 13 determines whether or not it is necessary to dispatch maintenance personnel to correct the failure of the elevator apparatus that has transmitted the trace data (S402). The determination unit 13 performs the above determination based on the learning result from the learning unit 12. In the example illustrated in FIG. 9, the determination unit 13 performs the determination in S <b> 402 based on the straight line A determined by the learning unit 12.

  When it is not necessary to dispatch maintenance personnel, the determination unit 13 determines the transition of a failure that has occurred in the elevator apparatus (S408). For example, the determination unit 13 specifies how the trace data received by the reception unit 11 in S401 has changed with respect to the determination criterion from the trace data received by the reception unit 11 in the past from the same elevator apparatus. In the example shown in FIG. 9, it is specified that the coordinates indicating the trace data are approaching the straight line A. That is, it is specified that the elevator apparatus has a tendency to deteriorate.

  In order for the determination unit 13 to identify the change, past determination data is required for each elevator apparatus. For example, when the determination by the determination unit 13 is performed, the determination result is stored in the storage unit 10.

  When the determination by the determination unit 13 is performed, the notification control unit 15 causes the notification device 21 to notify the result determined by the determination unit 13 in S402 and the result specified in S408 (S403). In S403, the notification control unit 15 may display a graph as shown in FIG. 9 on the display. Moreover, if the determination by the determination part 13 is performed, the transmission part 14 will transmit the instruction | command for performing operation required in order to correct a failure to the elevator apparatus which has transmitted trace data (S404). In the elevator apparatus that has received the command, an operation necessary to correct the failure is performed. For example, in the elevator apparatus that has received the command, restart is performed.

  On the other hand, when the determination unit 13 determines that it is necessary to dispatch a maintenance staff, the notification control unit 15 causes the notification unit 21 to notify the result determined by the determination unit 13 in S402 (S405). If the determination unit 13 determines that a maintenance staff needs to be dispatched, the transmission section 14 transmits a maintenance staff dispatch command to the maintenance staff base or the like (S406).

  FIG. 10 is a diagram illustrating another example of the recovery support system. FIG. 10 shows an example of the monitoring center 1. In the example illustrated in FIG. 10, unique data is stored in the storage unit 10 in addition to failure data and work data. The unique data includes data whose value does not change among the data related to the elevator apparatus. The inherent data includes data such as the installation years whose values do not change instantaneously. In addition to the installation year, for example, the installation environment of the elevator device, the number of floors of the building, the maximum number of passengers, the model, and the like are stored in the storage unit 10 as the unique data. The unique data is not limited to these examples. The storage unit 10 stores unique data for each elevator device.

  In such a case, the learning unit 12 performs machine learning using the unique data stored in the storage unit 10. The learning unit 12 may group the trace data for each model, for each installation year, or for each installation environment, and determine a determination criterion.

  When the reception unit 11 receives the trace data, the determination unit 13 determines whether or not it is necessary to dispatch a maintenance person to correct the failure of the elevator apparatus that has transmitted the trace data. In the example illustrated in FIG. 10, the determination unit 13 performs the above determination based on, for example, the unique data of the elevator device stored in the storage unit 10 and the learning result by the learning unit 12. For example, for a certain model, it is assumed that a learning result is obtained that the probability that a failure will occur again within one week after restarting increases by 5% for each year of installation. In such a case, the determination unit 13 can calculate the determination criterion for the device whose installation year is 20 years from the determination criterion for the device whose installation year is 15 years for the model. The determination unit 13 may calculate, for the model, a determination criterion for a device whose installation year is 0 years from a determination criterion for a device whose installation year is 10 years.

As another example, it is assumed that a learning result as shown in the following equation is obtained with respect to a probability that a failure occurs again within one week after the restart.
(Occurrence probability on model α) = (Occurrence probability on model β) × F (x)
Here, x is vectorized trace data. F (x) is a conversion formula calculated by the learning unit 12. In such a case, the determination unit 13 can calculate the occurrence probability of the model α from the occurrence probability of the model β.

  In the example shown in FIG. 10, for example, even when the reception unit 11 receives trace data from a newly installed elevator apparatus, it is possible to accurately determine whether or not it is necessary to dispatch maintenance personnel. it can.

Embodiment 2. FIG.
FIG. 11 is a diagram showing an example of a recovery support system according to Embodiment 2 of the present invention. In the present embodiment, an example in which the learning function disclosed in the first embodiment is performed in the monitoring center 1 and the determination function is performed in each elevator apparatus will be described.

  In the example shown in the present embodiment, the monitoring center 1 includes, for example, a storage unit 10, a reception unit 11, a learning unit 12, and a transmission unit 14. Each elevator apparatus includes, for example, an acquisition unit 16, a transmission unit 17, a reception unit 18, a storage unit 20, a determination unit 13, a notification control unit 15, and an operation control unit 19. The acquisition unit 16, the transmission unit 17, the reception unit 18, the storage unit 20, the determination unit 13, and the notification control unit 15 are provided in the communication device 8, for example. The operation control unit 19 is provided in the control device 7, for example. Hereinafter, functions and operations of the recovery support system will be described in detail.

  FIG. 12 is a flowchart showing an operation example of the recovery support system according to Embodiment 2 of the present invention. FIG. 12 shows an example of the learning function of the recovery support system. FIG. 12 shows an operation example of each elevator apparatus. In the elevator apparatus, it is determined whether or not trace data has been acquired (S501). For example, when a failure occurs in the elevator apparatus, trace data is acquired by the acquisition unit 16 (Yes in S501). The transmission unit 17 transmits the trace data acquired by the acquisition unit 16 to the monitoring center 1 (S502).

  In the monitoring center 1, the same operation as that shown in FIG. 2 is performed. That is, the monitoring center 1 determines whether or not trace data has been received (S101). The trace data transmitted by the transmission unit 17 is received by the reception unit 11 in the monitoring center 1. The trace data received by the receiving unit 11 is stored in the storage unit 10 (S102).

  In the monitoring center 1, it is determined whether or not work data has been received (S103). The work data transmitted from the maintenance terminal 9 is received by the receiving unit 11 in the monitoring center 1. The work data received by the receiving unit 11 is associated with the corresponding trace data and stored in the storage unit 10 (S104). Trace data and work data are accumulated in the storage unit 10.

  In the monitoring center 1, it is determined whether or not it is a learning timing (S105). If it is determined in S105 that it is the learning timing, machine learning is performed by the learning unit 12, and a learning result is output (S106). The learning unit 12 performs machine learning on the failure data and work data stored in the storage unit 10. As an example, the learning unit 12 outputs a determination criterion as a learning result. The transmission part 14 transmits the learning result by the learning part 12 to each elevator apparatus.

  In the elevator apparatus, it is determined whether or not the learning result by the learning unit 12 has been received from the monitoring center 1 (S503). The learning result transmitted by the transmission unit 14 is received by the reception unit 18 in the elevator apparatus (Yes in S503). The learning result received by the receiving unit 18 is stored in the storage unit 20 (S504).

  FIG. 13 is a flowchart showing another example of operation of the recovery support system according to Embodiment 2 of the present invention. FIG. 13 shows an example of the determination function of the recovery support system. FIG. 13 shows an operation example of each elevator apparatus.

  In the elevator apparatus, it is determined whether or not trace data has been acquired (S601). For example, when a failure occurs in the elevator apparatus, trace data is acquired by the acquisition unit 16 (Yes in S601). When the acquisition unit 16 acquires the trace data, the determination unit 13 determines whether or not a maintenance person needs to be dispatched in order to correct the failure when the trace data is acquired (S602). The determination unit 13 performs the above determination based on the learning result stored in the storage unit 20 in S504.

  When the determination unit 13 determines that dispatch of maintenance personnel is not necessary, the notification control unit 15 causes the notification unit 21 to notify the result determined by the determination unit 13 in S602 (S603). The alarm device 21 is provided, for example, in an elevator apparatus. When the determination unit 13 determines that dispatch of maintenance personnel is not necessary, the operation control unit 19 causes each device to perform an operation necessary to correct the failure (S604). For example, the operation control unit 19 performs restart.

  On the other hand, when the determination unit 13 determines that the maintenance staff needs to be dispatched, the notification control unit 15 causes the notification unit 21 to notify the result determined by the determination unit 13 in S602 (S605). If the determination unit 13 determines that the maintenance staff needs to be dispatched, the transmission section 17 transmits a maintenance staff dispatch command to the maintenance staff base or the like (S606).

  In the example shown in the present embodiment, when a failure occurs in the elevator apparatus, it can be accurately determined whether or not it is necessary to dispatch a maintenance person to correct the failure of the apparatus. Moreover, if it is an example shown to this Embodiment, the load of the monitoring center 1 can be reduced. If it is an example shown in this Embodiment, even if an elevator apparatus and the monitoring center 1 become disconnected by a disaster or a power failure, the necessity for dispatch of a maintenance worker can be determined in an elevator apparatus. In S604, the elevator apparatus can be automatically restored.

  In the example shown in FIG. 13, in S604, an operation necessary for correcting the failure is automatically performed. This is an example. When it is determined No in S602, only the determination result may be notified. In such a case, the operation is performed based on the judgment of a building administrator, for example.

  In the example shown in FIG. 13, a maintenance staff dispatch request is automatically made in S606. This is an example. When it is determined Yes in S602, only the determination result may be notified. In such a case, the dispatch request for the maintenance staff is made, for example, based on the judgment of the building manager.

  In the example illustrated in FIG. 13, if it is determined No in S602, both the processing in S603 and the processing in S604 are performed. This is an example. If the operation necessary for correcting the failure is automatically performed in S604, the process of S603 may not be performed. Similarly, in the example illustrated in FIG. 13, when it is determined Yes in S602, both the processing in S605 and the processing in S606 are performed. This is an example. If a maintenance staff dispatch request is automatically made in S606, the process in S605 may not be performed.

  For functions and operations not specifically disclosed in this embodiment, any of the functions and operations disclosed in Embodiment 1 may be employed.

  For example, the learning unit 12 may determine a plurality of determination criteria as learning results. In such a case, when the trace data is acquired by the acquisition unit 16, the determination unit 13 determines whether it is necessary to dispatch a maintenance staff based on the determination criterion determined by the learning unit 12. Moreover, the determination part 13 determines the time when a failure occurs again based on the other determination criteria determined by the learning part 12, if it is not necessary to dispatch maintenance personnel.

  As another example, when it is not necessary to dispatch maintenance personnel, the determination unit 13 determines which trace data acquired by the acquisition unit 16 in S601 corresponds to the determination criterion from the trace data acquired by the acquisition unit 16 in the past. You may specify how it changed. In such a case, when the determination by the determination unit 13 is performed, the determination result is stored in the storage unit 20. The notification control unit 15 may cause the notification device 21 to notify the result specified by the determination unit 13.

  As another example, in addition to the failure data and work data, the storage unit 10 may store unique data of each elevator apparatus. In such a case, the learning unit 12 performs machine learning using the unique data stored in the storage unit 10. For example, the learning unit 12 groups the trace data for each model, for each installation year or for each installation environment, and determines a determination criterion.

  In the first and second embodiments, the example in which the elevator apparatus is connected to the monitoring center 1 has been described. The device connected to the monitoring center 1 is not limited to the elevator device. Other devices maintained by maintenance personnel may be connected to the monitoring center 1.

  In the first and second embodiments, the example in which the trace data is acquired by the communication device 8 when a failure occurs in the elevator apparatus has been described. The communication device 8 may acquire trace data constantly or periodically in addition to when a failure occurs. In other words, the communication device 8 may acquire normal trace data. Also in such a case, when the communication device 8 acquires the trace data, the communication device 8 transmits the acquired trace data to the monitoring center 1. Moreover, the data which shows the content of the driving | operation performed with the elevator apparatus may be contained in trace data. For example, a signal indicating a command from the control device 7 to the electric motor 6 may be included in the trace data.

  In Embodiment 1, each part shown to the codes | symbols 10-15 shows the function which the monitoring center 1 has. FIG. 14 is a diagram illustrating a hardware configuration of the monitoring center 1. The monitoring center 1 includes a processing circuit including, for example, a processor 22 and a memory 23 as hardware resources. The functions of the storage unit 10 are realized by the memory 23. The monitoring center 1 implements the functions of the units indicated by reference numerals 11 to 15 by executing the program stored in the memory 23 by the processor 22.

  In Embodiment 2, each part shown to the codes | symbols 10-12 and 14 shows the function which the monitoring center 1 has. The hardware configuration of the monitoring center 1 in the second embodiment is the same as the example shown in FIG. The monitoring center 1 includes a processing circuit including, for example, a processor 22 and a memory 23 as hardware resources. The functions of the storage unit 10 are realized by the memory 23. The monitoring center 1 implements the functions of the units indicated by reference numerals 11, 12, and 14 by executing the program stored in the memory 23 by the processor 22.

  In Embodiment 2, each part shown to the codes | symbols 13 and 15-20 shows the function which an elevator apparatus has. The hardware configuration of the elevator apparatus in the second embodiment is the same as the example shown in FIG. Each elevator apparatus includes a processing circuit including, for example, a processor and a memory as hardware resources. The functions of the storage unit 20 are realized by a memory. The elevator apparatus implements the functions of the units indicated by reference numerals 13 and 15 to 19 by executing a program stored in the memory by a processor.

  The processor 22 provided in the monitoring center 1 and the processor provided in the elevator apparatus are also referred to as 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 23 provided in the monitoring center 1 and the memory provided in the elevator apparatus, 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 monitoring center 1 may be realized by hardware. As hardware for realizing the function of the monitoring 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. Similarly, some or all of the functions of the elevator apparatus may be realized by hardware. As hardware for realizing the function of the elevator apparatus, a single circuit, a composite circuit, a programmed processor, a processor programmed in parallel, an ASIC, an FPGA, or a combination thereof may be employed.

  The recovery support system according to the present invention can be used to determine whether or not it is necessary to dispatch a maintenance person to fix a failure of the apparatus.

DESCRIPTION OF SYMBOLS 1 Monitoring center 2 Car 3 Balance weight 4 Main rope 5 Drive sheave 6 Electric motor 7 Control apparatus 8 Communication apparatus 9 Maintenance terminal 10 Storage part 11 Receiving part 12 Learning part 13 Determination part 14 Transmission part 15 Notification control part 16 Acquisition part 17 Transmission Unit 18 receiving unit 19 operation control unit 20 storage unit 21 alarm device 22 processor 23 memory

Claims (11)

Storage means for storing failure data indicating a failure state of the device in which the failure has occurred and work data indicating the work content performed to correct the failure;
Learning means for machine learning of failure data and work data stored in the storage means;
Receiving means for receiving failure data;
When the receiving unit receives the failure data, the determination unit determines whether or not it is necessary to dispatch a maintenance person to correct the failure of the device that has transmitted the failure data based on the learning result by the learning unit. When,
Recovery support system with The learning means determines a first determination criterion and a second determination criterion as learning results,
The determination means includes
When the receiving means receives the failure data, it is determined based on the first determination criterion whether or not it is necessary to dispatch maintenance personnel to correct the failure of the device that has transmitted the failure data,
2. The recovery support system according to claim 1, wherein when it is not necessary to dispatch a maintenance staff, the time when the failure occurs again in the device that has transmitted the failure data is determined based on the second determination criterion.   If the determination means determines that it is not necessary to dispatch maintenance personnel to correct the failure of the device that has transmitted the failure data, a command for performing an operation necessary to correct the failure is issued. The recovery support system according to claim 1, further comprising transmission means for transmitting to a device that has transmitted the message. A notification control means;
The learning means determines a determination criterion as a learning result,
When the receiving means receives the failure data, the notification control means determines how the failure data corresponds to the determination criterion from the failure data received by the receiving means in the past from the device that has transmitted the failure data. The restoration support system according to claim 1, wherein a change is notified from an alarm. The receiving means receives failure data from a plurality of devices,
The restoration support system according to any one of claims 1 to 4, wherein the failure data received by the receiving unit is stored in the storage unit. A monitoring center,
A plurality of devices capable of communicating with the monitoring center;
With
The monitoring center is
Storage means for storing failure data indicating a failure state of the device in which the failure has occurred and work data indicating the work content performed to correct the failure;
Learning means for machine learning of failure data and work data stored in the storage means;
With
Each of the plurality of devices is
Acquisition means for acquiring failure data;
When the acquisition unit acquires failure data, the determination unit determines whether or not a maintenance staff needs to be dispatched to correct the failure when the failure data is acquired based on the learning result of the learning unit. When,
Recovery support system with The learning means determines a first determination criterion and a second determination criterion as learning results,
The determination means includes
When the acquisition means acquires the failure data, it is determined based on the first determination criterion whether or not a dispatch of maintenance personnel is necessary to correct the failure when the failure data is acquired,
The recovery support system according to claim 6, wherein when a maintenance person is not required to be dispatched, a time when the failure occurs again is determined based on the second determination criterion. Each of the plurality of devices further comprises operation control means,
7. The operation control means, when it is determined by the determination means that a dispatch of maintenance personnel is not required to correct a failure when failure data is acquired, performs an operation necessary for correcting the failure. Or the recovery assistance system of Claim 7. Each of the plurality of devices further includes notification control means,
The learning means determines a determination criterion as a learning result,
When the acquisition unit acquires the failure data, the notification control unit notifies the notification unit of how the failure data has changed from the failure data acquired by the acquisition unit in the past with respect to the determination criterion. The recovery support system according to claim 6.   10. The recovery support system according to claim 6, wherein each of the plurality of apparatuses further includes a transmission unit that transmits failure data acquired by the acquisition unit to the monitoring center. 11. In the storage means, unique data of the device is stored in association with the failure data of the device,
The recovery support system according to any one of claims 1 to 10, wherein the learning unit performs machine learning using unique data stored in the storage unit.

Images