CN110383264B

CN110383264B - Retrieval system

Info

Publication number: CN110383264B
Application number: CN201680091113.4A
Authority: CN
Inventors: 服部智宏; 大泽奈奈穗; 坂上聪子; 松枝丰; 福永宽
Original assignee: Mitsubishi Electric Corp; Mitsubishi Electric Building Solutions Corp
Current assignee: Mitsubishi Electric Corp; Mitsubishi Electric Building Solutions Corp
Priority date: 2016-12-16
Filing date: 2016-12-16
Publication date: 2022-12-30
Anticipated expiration: 2036-12-16
Also published as: WO2018109929A1; KR20190049794A; JP6575694B2; CN110383264A; KR102152218B1; JPWO2018109929A1

Abstract

The search system has a1 st storage unit (10), a1 st calculation unit (11), a2 nd storage unit (12), and a2 nd calculation unit (13). A1 st calculation unit (11) calculates the similarity between the signals included in the 1 st signal group and the signals included in the 2 nd signal group for each of the corresponding signals. A2 nd storage unit (12) stores the importance of each signal for a plurality of signals included in the 1 st signal group. A2 nd calculation unit (13) calculates the similarity between the 1 st signal group and the 2 nd signal group based on the calculated similarity for each signal and the stored importance of each signal.

Description

Retrieval system

Technical Field

The present invention relates to a search system.

Background

For example, when a fault occurs in an elevator, a snapshot of the signal values representing the elevator status is taken. In this specification, the snapshot of the signal values is referred to as trace data.

Patent document 1 describes an example of a system for detecting an abnormality in acquired data.

Documents of the prior art

Patent document

Patent document 1: japanese patent No. 5301310

Disclosure of Invention

Problems to be solved by the invention

Conventionally, when new trace data is acquired, it is impossible to search for past trace data similar to the new trace data. Such a search is useful, for example, to estimate the cause of a fault from the trace data.

The present invention has been made to solve the above-described problems. The object of the present invention is to provide a search system capable of searching a signal group similar to a signal group including a plurality of signals with a simple configuration.

Means for solving the problems

The search system of the present invention includes: a1 st storage unit that stores a1 st signal group including a plurality of signals and a2 nd signal group including a plurality of signals corresponding to the plurality of signals included in the 1 st signal group; a1 st calculation unit that calculates a similarity between a signal included in the 1 st signal group and a signal included in the 2 nd signal group for each corresponding signal; a2 nd storage unit that stores importance of each signal with respect to the plurality of signals included in the 1 st signal group; and a2 nd calculating unit that calculates the similarity of the 1 st signal group and the 2 nd signal group based on the similarity of each signal calculated by the 1 st calculating unit and the importance of each signal stored in the 2 nd storing unit.

Effects of the invention

In the search system of the present invention, the 1 st calculation unit calculates the similarity between the signals included in the 1 st signal group and the signals included in the 2 nd signal group for each of the corresponding signals. In the 2 nd storage unit, the importance of each signal is stored for a plurality of signals included in the 1 st signal group. A2 nd calculating unit calculates the similarity of the 1 st signal group and the 2 nd signal group based on the similarity of each signal calculated by the 1 st calculating unit and the importance of each signal stored in the 2 nd storing unit. According to the retrieval system of the present invention, it is possible to retrieve a signal group similar to a signal group including a plurality of signals with a simple configuration.

Drawings

Fig. 1 is a diagram showing an example of a search system in embodiment 1 of the present invention.

Fig. 2 is a diagram showing an example of trace data.

Fig. 3 is a diagram showing an example of the search device.

Fig. 4 is a flowchart showing an example of the operation of the search system in embodiment 1 of the present invention.

Fig. 5 is a flowchart showing an example of the operation of the search system in embodiment 1 of the present invention.

Fig. 6 is a diagram for explaining the function of the 1 st calculation unit.

Fig. 7 is a diagram for explaining the function of the 1 st calculation unit.

Fig. 8 is a diagram showing a display example of the display.

Fig. 9 is a diagram showing another example of the search device.

Fig. 10 is a diagram showing another example of the search device.

Fig. 11 is a diagram showing another display example of the display.

Fig. 12 is a diagram showing an example of a search device included in a search system according to embodiment 2 of the present invention.

Fig. 13 is a flowchart showing an example of the operation of the search system in embodiment 2 of the present invention.

Fig. 14 is a diagram showing a display example of the display.

Fig. 15 is a diagram showing an example of a search device included in a search system according to embodiment 3 of the present invention.

Fig. 16 is a flowchart showing an example of the operation of the search system in embodiment 3 of the present invention.

Fig. 17 is a diagram showing an example of a search device provided in a search system according to embodiment 4 of the present invention.

Fig. 18 is a flowchart showing an example of the operation of the search system in embodiment 4 of the present invention.

Fig. 19 is a diagram showing an example of the updated importance.

Fig. 20 is a diagram showing an example of the updated importance.

Fig. 21 is a diagram showing a hardware configuration of the search device.

Detailed Description

The invention is described with reference to the accompanying drawings. Duplicate descriptions are appropriately simplified or omitted. In the drawings, the same reference numerals denote the same or equivalent parts.

Embodiment mode 1

Fig. 1 is a diagram showing an example of a search system in embodiment 1 of the present invention. The search device 1 can communicate with a plurality of remote elevator devices. Each elevator apparatus includes, for example, a car 2 and a counterweight 3. The car 2 and the counterweight 3 are suspended from the hoistway by main ropes 4. The hoisting machine of the elevator has, for example, a drive sheave 5 and a motor 6. The main rope 4 is wound around the drive sheave 5. The drive sheave 5 is driven by a motor 6. The motor 6 is controlled by a control panel 7. A communication device 8 is connected to the control panel 7. The communication device 8 communicates with an external apparatus. Each elevator device communicates with the search device 1 via the communication device 8.

When a fault occurs in the elevator apparatus, the communication device 8 acquires trace data, which is a snapshot of the signal values indicating the elevator state. The trace data is an example of a signal group including a plurality of signals. For example, the tracking data contains signals for determining the elevator installation itself. The tracking data contains a signal representing the time of day. The tracking data comprises a signal representing the current value of the control disc 7. The tracking data comprises a signal representing the voltage value of the control disc 7. The tracking data contains a signal representing the speed of the motor 6. The tracking data contains a signal representing the torque of the motor 6. The tracking data includes a signal indicating the open/close state of the door. The tracking data contains a signal representing the position of the car 2. The trace data includes a signal indicative of an operational state of the security device. The signals contained in the trace data are not limited to these examples. A portion of the illustrated signals may not be included in the trace data. The tracking data may also comprise other signals.

Fig. 2 is a diagram showing an example of trace data. Fig. 2 shows an example of 4 trace data. The labels of the signals contained in the trace data are not limited to 1. For example, a signal marked by 2, a signal marked by 16, and a signal marked by 10 may be included in one trace data. A trace data may also contain signals of various signal lengths.

When a failure occurs in the elevator apparatus, the communication device 8 acquires the tracking data at the time of the failure. The communication device 8 may acquire the trace data within a predetermined time before and after the occurrence of the failure. The timing of acquiring the trace data is not limited to the time of occurrence of a failure. For example, the communication device 8 may periodically acquire the trace data. After acquiring the trace data, the communication device 8 transmits the acquired trace data to the search device 1.

Fig. 3 is a diagram showing an example of the search device 1. The search device 1 includes, for example, a receiving unit 9, a1 st storage unit 10, a1 st calculation unit 11, a2 nd storage unit 12, a2 nd calculation unit 13, a display control unit 14, and a display 15. Next, the function and operation of the search device 1 will be described with reference to fig. 4 to 8. Fig. 4 and 5 are flowcharts showing an example of the operation of the search system in embodiment 1 of the present invention.

When a failure occurs in any of the elevator devices, the tracking data is transmitted from the communication device 8 of the elevator device to the search device 1. The tracking data transmitted from the communication device 8 is received by the receiving unit 9 in the search device 1 (S101). The trace data received by the receiving unit 9 is stored in the 1 st storage unit 10 (S102). A plurality of trace data are accumulated in the 1 st storage unit 10.

Fig. 5 shows an example of a processing flow of retrieving trace data similar to the trace data X1. The trace data X1 is one of the trace data stored in the 1 st storage unit 10. The trace data X1 may be the latest trace data stored in the 1 st storage unit 10. The plurality of trace data may be stored in the 1 st storage unit 10 after the trace data X1 is stored in the 1 st storage unit 10.

The 1 st calculation unit 11 reads the trace data X1 from the 1 st storage unit 10 (S201). The tracking data X1 includes, for example, a signal A1, a signal A2, a signal A3, …, and a signal AN. N is a natural number of 3 or more, for example. N may be a natural number of 100 or more. In the following description, the signal A1 included in the tracking data X1 is denoted as a signal a11. Similarly, the signal A2 included in the tracking data X1 is denoted as a signal a21. The signal A3 included in the tracking data X1 is denoted as a signal a31. The signal AN contained in the tracking data X1 is labeled as a signal AN1.

Next, the 1 st calculation unit 11 reads trace data other than the one trace data X1 from the trace data stored in the 1 st storage unit 10 (S202). For example, the 1 st calculation unit 11 reads the trace data X2 from the 1 st storage unit 10. The tracking data X2 includes a plurality of signals corresponding to the plurality of signals included in the tracking data X1. For example, the tracking data X2 includes a signal A1, a signal A2, a signal A3, …, and a signal AN. In the following description, the signal A1 included in the tracking data X2 is denoted as a signal a12. Similarly, the signal A2 included in the tracking data X2 is denoted as a signal a22. The signal A3 included in the tracking data X2 is labeled as a signal a32. The signal AN contained in the tracking data X2 is labeled as a signal AN2.

The signal a12 is a signal corresponding to the signal a11. For example, the signal a11 and the signal a12 are signals indicating time of day. The signal a22 is a signal corresponding to the signal a21. For example, the signal a21 and the signal a22 are signals indicating the current value of the control panel 7. The signal a32 is a signal corresponding to the signal a31. For example, the signal a31 and the signal a32 are signals indicating the torque of the motor 6. The signal AN2 is a signal corresponding to the signal AN1.

The 1 st calculating unit 11 calculates the similarity between the signal included in the tracking data X1 and the signal included in the tracking data X2 for each corresponding signal (S203). For example, the 1 st calculating part 11 calculates the similarity between the signal a11 and the signal a12. Similarly, the 1 st calculating part 11 calculates the similarity of the signal a21 and the signal a22. The 1 st calculating part 11 calculates the similarity between the signal a31 and the signal a32. The 1 st calculating unit 11 calculates the similarity between the signal AN1 and the signal AN2.

Fig. 6 and 7 are diagrams for explaining the function of the 1 st calculating unit 11. The 1 st calculation unit 11 calculates the similarity of each signal using, for example, a dynamic programming method. Fig. 6 and 7 show examples of calculating the similarity of the signal a11 and the signal a12. In the example shown in fig. 6 and 7, the signal a11 is 00000111. Signal a12 is 00001111.

In the case of calculating the similarity by the dynamic programming method, first, as shown in fig. 6 and 7, one signal is arranged in the vertical direction and the other signal is arranged in the horizontal direction. Next, values are input from the upper left grid to the lower right grid while considering the movement penalty and the inconsistency penalty. Then, a path having the smallest value is searched from the upper left grid to the lower right grid.

In the example shown in fig. 6, the penalty for movement in the right direction is +1. The penalty for movement in the down direction is +1. The penalty for movement in the oblique direction to the lower right is 0. The penalty when the values match is 0. The penalty in case of inconsistent values is +1. In the example shown in fig. 6, the similarity of the signal a11 and the signal a12 is calculated as 1.

In the example shown in fig. 7, the penalty for movement in the right direction is +1. The penalty for movement in the down direction is +1. The penalty for movement in the oblique direction to the lower right is 0. The penalty when the values match is 0. The penalty in case of inconsistent values is +3. In the example shown in fig. 7, the similarity of the signal a11 and the signal a12 is calculated as 2.

In the examples shown in fig. 6 and 7, the more similar the signals to be compared are to each other, the smaller the value of the similarity calculated by the 1 st calculating part 11 is. The calculation method may also be set as follows: when the 2 signals have the same value, the similarity calculated by the 1 st calculating unit 11 has the maximum value. The maximum value of the calculated similarity may be set to 1, 100, or the like. The method of setting the movement penalty and the method of setting the inconsistency penalty are not limited to the examples shown in fig. 6 and 7. The method of calculating the similarity by the 1 st calculating unit 11 is not limited to the dynamic planning method.

The 2 nd storage unit 12 stores the importance of each signal for a plurality of signals included in the trace data. In the example shown in this embodiment, the tracking data X1 includes a signal A1, a signal A2, a signal A3, …, and a signal AN. The tracking data X2 includes a signal A1, a signal A2, a signal A3, …, and a signal AN. Therefore, the 2 nd storage unit 12 stores the importance of the signal A1, the importance of the signal A2, the importance of the signal A3, …, and the importance of the signal AN.

The 2 nd calculating unit 13 reads the importance of each signal from the 2 nd storage unit 12 (S204). Next, the 2 nd calculating unit 13 calculates the similarity between the tracking data X1 and the tracking data X2 (S205). The 2 nd calculating unit 13 performs the above calculation based on the similarity of each signal calculated by the 1 st calculating unit 11 and the importance of each signal stored in the 2 nd storage unit 12. For example, the 2 nd calculating unit 13 calculates the similarity by the following expression.

(similarity between tracking data X1 and tracking data X2)

= (degree of similarity of signal A1) × (degree of importance of signal A1)

+ (similarity of Signal A2) × (importance of Signal A2)

+ (similarity of Signal A3) x (importance of Signal A3)

：

(similarity of Signal AN) × (importance of Signal AN)

The method of calculating the similarity by the 2 nd calculating part 13 is not limited to the above example. The maximum value of the similarity calculated by the 2 nd calculation unit 13 may be set to 1, 100, or the like.

The 1 st storage unit 10 stores a plurality of trace data. For example, the 1 st storage unit 10 stores the trace data X1, the trace data X2, the trace data X3, …, and the trace data XM. M is, for example, a natural number of 3 or more. When the similarity between the trace data X1 and a certain trace data is calculated, it is determined whether or not the similarity is calculated for all the trace data (S206). If there is trace data for which the similarity with the trace data X1 is not calculated, the processing shown in S202 to S205 is performed on the trace data.

For example, when calculating the similarity between the tracking data X1 and the tracking data X2, the 1 st calculating unit 11 reads the tracking data X3 from the 1 st storage unit 10 (S202). The tracking data X3 includes a plurality of signals corresponding to the plurality of signals included in the tracking data X1. For example, the tracking data X3 includes a signal A1, a signal A2, a signal A3, …, and a signal AN. The 1 st calculating unit 11 calculates the similarity between the signal included in the tracking data X1 and the signal included in the tracking data X3 for each corresponding signal (S203). The 2 nd calculating unit 13 calculates the similarity between the tracking data X1 and the tracking data X3 based on the similarity for each signal calculated by the 1 st calculating unit 11 and the importance for each signal stored in the 2 nd storage unit 12 (S205).

The trace data for calculating the similarity to the trace data X1 may be a part of the trace data described in the 1 st storage unit 10. For example, when the 1 st storage unit 10 stores the tracking data X1 to X1000000, the similarity with the tracking data X1 may be calculated for each of the tracking data X2 to X1000. The similarity with the tracking data X1 may be calculated for each of the tracking data X300000 to the tracking data X399999.

The display control unit 14 controls the display 15. For example, the display control unit 14 causes the display 15 to display the similarity calculated by the 2 nd calculation unit 13 (S207).

Fig. 8 is a diagram showing a display example of the display 15. Fig. 8 shows an example in which the information for specifying the trace data, the similarity calculated by the 2 nd calculation unit 13, and the correspondence content are associated with each other and displayed on the display 15. In the example shown in fig. 8, the information for specifying the trace data is displayed from the top in the order of the calculated similarity from high to low. For example, if the degree of similarity of the tracking data X1 and the tracking data X2 is the highest among the calculated degrees of similarity, information for specifying the tracking data X2, the calculated degree of similarity, and the corresponding content are displayed in the uppermost column. Fig. 8 shows an example in which information indicating the number of a building having an elevator apparatus and the date and time when a failure occurred is displayed as information for specifying tracking data. The correspondence content is information that is input by a maintenance worker who actually performed a job after the job.

In the example shown in this embodiment mode, a signal group similar to a signal group including a plurality of signals can be searched with a simple configuration. The similarity of the signal group is calculated from the similarity of each signal and the importance of each signal, and therefore, a highly accurate search can be performed.

In the present embodiment, an example in which the 2 nd storage unit 12 stores the importance of 1 group is described. This is an example. A plurality of sets of importance may be stored in the 2 nd storage unit 12. Fig. 9 is a diagram showing another example of the search device 1. As a simplest example, fig. 9 shows an example in which 2 sets of importance degrees are stored in the 2 nd storage unit 12. The 2 nd storage unit 12 may store 3 or more sets of importance degrees.

The search device 1 performs the processing shown in S201 to S203 in the same manner as in the above example. The 2 nd calculating unit 13 reads 1 group importance from the 2 groups of importance stored in the 2 nd storage unit 12 (S204). The 2 nd calculation unit 13 determines the importance to be read based on a predetermined condition. For example, the 2 nd calculation unit 13 determines the importance to be read based on the date and time when the trace data X1 is acquired.

The 2 nd calculation unit 13 may determine the importance to be read according to the presence or absence of a specific request. For example, the 2 nd calculation unit 13 reads one importance level only when there is the request. If the request does not exist, the 2 nd calculation unit 13 reads another importance level. The 2 nd calculation unit 13 may determine the importance to be read based on the value of the specific signal included in the tracking data X1. For example, if the value of the signal A9 included in the tracking data X1 is 0, the 2 nd calculating unit 13 reads one importance level. If the value of the signal A9 included in the tracking data X1 is 1, the 2 nd calculating unit 13 reads another importance level.

The 2 nd calculating unit 13 calculates the similarity between the tracking data X1 and the tracking data X2 based on the similarity for each signal calculated by the 1 st calculating unit 11 and the importance of each signal read in S204 (S205).

In the example shown in fig. 9, the importance of each signal can be changed according to the conditions.

In the present embodiment, an example in which only the importance of each signal is stored in the 2 nd storage unit 12 is described. This is an example. In the 2 nd storage unit 12, in addition to the importance of each signal, the importance of a combination of specific signals may be stored. Fig. 10 is a diagram showing another example of the search device 1. For example, in the 2 nd storage unit 12, in addition to the importance of each signal, the importance for the combination of the signal A1 and the signal A2 and the importance for the combination of the signal A6 and the signal A8 are stored. The importance of a combination of 3 or more signals may be stored in the 2 nd storage unit 12. For example, the 2 nd storage unit 12 may store the importance degree for the combination of the signal A1, the signal A2, and the signal A3.

The search device 1 performs the processing shown in S201 and S202 in the same manner as in the above example. The 1 st calculation unit 11 calculates the similarity between the signal included in the tracking data X1 and the signal included in the tracking data X2 for each corresponding signal (S203). Further, in S203, the 1 st calculating unit 11 calculates the similarity of the combination of the specific signals. In the example shown in fig. 10, the 1 st calculating unit 11 calculates the similarity between the combination of the signal A1 and the signal A2 included in the tracking data X1 and the combination of the signal A1 and the signal A2 included in the tracking data X2. The 1 st calculating unit 11 calculates the similarity between the combination of the signal A6 and the signal A8 included in the tracking data X1 and the combination of the signal A6 and the signal A8 included in the tracking data X2.

For example, the 1 st calculation unit 11 determines whether or not the signal A1 included in the tracking data X1 matches the signal A1 included in the tracking data X2. The 1 st calculation unit 11 determines whether or not the signal A2 included in the tracking data X1 matches the signal A2 included in the tracking data X2. If both of the signals coincide with each other, the 1 st calculation unit 11 calculates the similarity of the combination of the signal A1 and the signal A2 as 1. When both or only one of the signals does not match, the 1 st calculation unit 11 calculates the similarity of the combination of the signal A1 and the signal A2 as 0. The method of calculating the similarity is not limited to the above example.

The 2 nd calculating unit 13 reads the importance of each signal from the 2 nd storage unit 12 (S204). Further, in S204, the 2 nd calculation unit 13 reads the importance of the combination of the specific signals.

The 2 nd calculating unit 13 calculates the similarity between the tracking data X1 and the tracking data X2 based on the similarity for each signal calculated by the 1 st calculating unit 11 and the importance for each signal stored in the 2 nd storage unit 12 (S205). Further, in S205, the 2 nd calculating unit 13 corrects the calculated similarity based on the similarity of the combination of specific signals calculated by the 1 st calculating unit 11 and the importance of the combination of specific signals stored in the 2 nd storage unit 12. For example, the 2 nd calculation unit 13 corrects the similarity by the following expression.

(similarity after correction between tracking data X1 and tracking data X2)

= (similarity calculated from similarity of each signal and importance of each signal)

+ (similarity of combination of Signal A1 and Signal A2) × (importance of combination of Signal A1 and Signal A2)

The method of correcting the similarity by the 2 nd calculation section 13 (similarity of combination of signal A6 and signal A8) × (importance of combination of signal A6 and signal A8) is not limited to the above example.

In the example shown in fig. 10, the similarity can be calculated also in consideration of the combination of specific signals. Therefore, the accuracy of the calculated similarity can be further improved.

Fig. 11 is a diagram showing another display example of the display 15. Fig. 11 shows an example in which information indicating similar signals is displayed in association with information for specifying tracking data and the like.

The search device 1 performs the processing shown in S201 to S206 in the same manner as in the above example. The display control unit 14 causes the display 15 to display the similarity calculated by the 2 nd calculation unit 13 (S207). Further, the display control unit 14 causes the display 15 to display information indicating the similar signal. The display control unit 14 specifies the similar signal based on the similarity of each signal calculated by the 1 st calculation unit 11 and the importance of each signal stored in the 2 nd storage unit 12. For example, the display control unit 14 determines that the signal calculated by the 1 st calculation unit 11 is a similar signal when the similarity of the signal is higher than a reference value and the importance of the signal stored in the 2 nd storage unit 12 is higher than the reference value. The display control unit 14 may determine that the signal is a similar signal when the product of the similarity of the signal calculated by the 1 st calculation unit 11 and the importance of the signal stored in the 2 nd storage unit 12 is higher than a reference value. The number of similar signals may be set to 3 in advance, and 3 signals from the signal having the larger product value may be determined as similar signals.

In the example shown in fig. 11, the information of the signal contributing to the improvement of the similarity of the trace data can be displayed on the display 15 in association with the calculated similarity. By observing the information of the signal displayed on the display 15, it is possible to more accurately determine whether or not the present failure case matches the past failure case, for example.

Embodiment mode 2

Fig. 12 is a diagram showing an example of the search device 1 included in the search system according to embodiment 2 of the present invention. The overall view of the search system is the same as that of fig. 1. The search device 1 includes, for example, an input unit 16, a1 st determination unit 17, a2 nd determination unit 18, and an update unit 19 in addition to the reception unit 9, the 1 st storage unit 10, the 1 st calculation unit 11, the 2 nd storage unit 12, the 2 nd calculation unit 13, the display control unit 14, and the display 15. The functions of the reception unit 9, the 1 st storage unit 10, the 1 st calculation unit 11, the 2 nd storage unit 12, the 2 nd calculation unit 13, the display control unit 14, and the display 15 are the same as those of the arbitrary functions disclosed in embodiment 1. In the case of retrieving the trace data similar to the trace data X1, for example, the processing shown in fig. 5 is performed. In the 2 nd storage unit 12, in addition to the importance of each signal, the importance of a combination of specific signals may be stored.

Next, the function and operation of the search device 1 will be described with reference to fig. 13 and 14. Fig. 13 is a flowchart showing an example of the operation of the search system in embodiment 2 of the present invention.

When a failure occurs in any of the elevator apparatuses, the tracking data X1 is transmitted from the communication device 8 of the elevator apparatus to the search device 1. When the maintenance person arrives at the building having the elevator apparatus in which the trouble has occurred, the maintenance person operates the terminal connected to the search apparatus 1, thereby causing the display 15 to display the information of the trace data similar to the trace data X1. The maintenance person observes the contents displayed on the display 15, and estimates the cause of the failure or a reference for performing repair.

Fig. 14 is a diagram showing a display example of the display 15. Fig. 14 shows an example in which the display control unit 14 causes the display 15 to display a column of the similarity determination in S207 of fig. 5.

After the repair of the elevator apparatus is completed, the maintenance worker inputs the correspondence content for the current failure case from the input unit 16. The information input from the input unit 16 is stored in the 1 st storage unit 10 as the content corresponding to the trace data X1.

After the repair of the elevator apparatus is completed, the maintenance worker inputs the judgment of the search result from the input unit 16. For example, as in the example shown in fig. 14, a check box indicating similarity and a check box indicating dissimilarity are displayed in the column of the similarity determination. If the current fault case is considered to be similar to the past fault case with the similarity calculated as 95.8%, the maintainer checks a similar check box. If the current fault case is not similar to the past fault case with the similarity calculated as 95.8%, the maintainer checks the dissimilar check box. That is, the maintainer enters feedback on the results of the search. The display of check boxes can be made for a plurality of instances. For example, a check box may be displayed for a past failure case whose similarity is calculated to be 92.1%. Next, an example in which the trace data whose similarity is calculated to be 95.8% is the trace data X2 will be described.

The input unit 16 is a device for inputting information. In the above example, the input section 16 is a keyboard and a mouse. The input unit 16 may be a touch panel. The input unit 16 is not limited to these examples.

In the search device 1, it is determined whether or not feedback input is performed on the search result (S301). In the example shown in fig. 14, it is determined whether a similar check box or a dissimilar check box is checked with respect to the trace data X2. The 1 st determination unit 17 determines whether or not the tracking data X1 and the tracking data X2 are similar to each other based on the information input from the input unit 16 (S302). For example, when the similar check box is checked, the 1 st determination unit 17 determines that the trace data X1 is similar to the trace data X2. When the dissimilar check boxes are checked, the 1 st determination unit 17 determines that the trace data X1 and the trace data X2 are dissimilar.

The 2 nd determining unit 18 determines whether or not the signal included in the tracking data X1 is similar to the signal included in the tracking data X2 for each corresponding signal (S303). For example, the 2 nd determination unit 18 determines whether the signal a11 is similar to the signal a12. Similarly, the 2 nd determination unit 18 determines whether the signal a21 is similar to the signal a22. The 2 nd determination unit 18 determines whether the signal a31 is similar to the signal a32. The 2 nd decision unit 18 decides whether or not the signal AN1 is similar to the signal AN2.

For example, the 2 nd determination unit 18 performs the determination according to a preset condition. The 2 nd determining unit 18 may determine that the signals are similar when the similarity calculated by the 1 st calculating unit 11 exceeds a reference value. The 2 nd determination unit 18 may perform the determination based on whether or not the values of the signals match. For example, when the signal A1 is a 2-valued signal, the 2 nd determination unit 18 determines that the signal a11 is 1 and the signal a12 is 1. If the signal a11 is 0 and the signal a12 is 0, the 2 nd determination unit 18 determines that the signals are similar to each other. If the signal a11 is 1 and the signal a12 is 0, the 2 nd determination unit 18 determines that the two are not similar. If the signal A11 is 0 and the signal A12 is 1, the 2 nd determination unit 18 determines that the two are not similar. The method of determining by the 2 nd determining unit 18 is not limited to the above example.

The updating unit 19 updates the importance of each signal stored in the 2 nd storage unit 12 (S304). The updating unit 19 performs the updating based on the result determined by the 1 st determining unit 17 and the result determined by the 2 nd determining unit 18. Next, the function of the update unit 19 will be described in detail with reference to tables 1 and 2.

[ TABLE 1 ]

In the example shown in Table 1, the maintainer checks a similar check box. Therefore, the 1 st determination unit 17 determines that the tracking data X1 is similar to the tracking data X2.

In the example shown in table 1, the value of the signal A1 included in the tracking data X1 is 0. The signal A1 included in the tracking data X2 has a value of 0. The 2 nd determination unit 18 determines that the signal a11 is similar to the signal a12. The judgment by the 1 st judgment unit 17 is based on the judgment of the maintenance worker. On the other hand, the determination by the 2 nd determining unit 18 is based on the value of the signal included in the tracking data. If the determination by the 1 st determining unit 17 and the determination by the 2 nd determining unit 18 match, the updating unit 19 updates the importance level so that the importance level of the signal becomes higher. For example, the updating unit 19 stores, as the importance of the signal A1 after updating, a value in which the importance of the signal A1 stored in the 2 nd storage unit 12 is 1.5 times as large as the importance of the signal A1 stored in the 2 nd storage unit 12. The rate of change in importance is not limited to 1.5 times.

In the example shown in table 1, the value of the signal A2 included in the tracking data X1 is 0. The signal A2 included in the tracking data X2 has a value of 1. The 2 nd determination unit 18 determines that the signal a21 is not similar to the signal a22. If the determination by the 1 st determining unit 17 does not match the determination by the 2 nd determining unit 18, the updating unit 19 updates the importance level so that the importance level of the signal becomes lower. For example, the updating unit 19 stores, as the importance of the signal A2 after updating, a value in which the importance of the signal A2 stored in the 2 nd storage unit 12 is 0.5 times as large as the importance of the signal A2 stored in the 2 nd storage unit 12. The rate of change in the importance is not limited to 0.5 times.

Similarly, the updating unit 19 updates the importance of the signal A3, the importance of …, and the importance of the signal AN.

[ TABLE 2 ]

In the example shown in Table 2, the maintainer checks the dissimilar checkboxes. Therefore, the 1 st determining unit 17 determines that the tracking data X1 is not similar to the tracking data X2.

In the example shown in table 2, the value of the signal A1 included in the tracking data X1 is 0. The signal A1 included in the tracking data X2 has a value of 0. The 2 nd determination unit 18 determines that the signal a11 is similar to the signal a12. If the determination by the 1 st determining unit 17 does not match the determination by the 2 nd determining unit 18, the updating unit 19 updates the importance level so that the importance level of the signal becomes lower. For example, the updating unit 19 stores, as the importance of the signal A1 after updating, a value in which the importance of the signal A1 stored in the 2 nd storage unit 12 is 0.5 times as large as the importance of the signal A1 stored in the 2 nd storage unit 12.

In the example shown in table 2, the value of the signal A2 included in the tracking data X1 is 0. The signal A2 included in the tracking data X2 has a value of 1. The 2 nd determination unit 18 determines that the signal a21 is not similar to the signal a22. If the determination by the 1 st determining unit 17 and the determination by the 2 nd determining unit 18 match, the updating unit 19 updates the importance level so that the importance level of the signal becomes higher. For example, the updating unit 19 stores, as the importance of the signal A2 after updating, a value in which the importance of the signal A2 stored in the 2 nd storage unit 12 is 1.5 times as large as the importance of the signal A2 in the 2 nd storage unit 12.

Similarly, the update unit 19 updates the importance of the signal A3, …, and the importance of the signal AN.

[ TABLE 3 ]

In the example shown in table 3, the maintainer checks a similar check box for the tracking data X2. Therefore, the 1 st determination unit 17 determines that the tracking data X1 is similar to the tracking data X2. On the other hand, the maintainer checks dissimilar check boxes for the tracking data X3. Therefore, the 1 st determining unit 17 determines that the tracking data X1 is not similar to the tracking data X3.

In the example shown in table 3, the value of the signal A1 included in the tracking data X1 is 0. The signal A1 included in the tracking data X2 has a value of 0. The 2 nd determination unit 18 determines that the signal a11 is similar to the signal a12. The judgment by the 2 nd judging unit 18 matches the judgment by the 1 st judging unit 17 on the tracking data X1 and the tracking data X2. The signal A1 included in the tracking data X3 has a value of 1. The 2 nd determination unit 18 determines that the signal a11 is not similar to the signal a 13. The judgment by the 2 nd judging unit 18 matches the judgment by the 1 st judging unit 17 on the tracking data X1 and the tracking data X3. If the determination by the 1 st determining unit 17 and the determination by the 2 nd determining unit 18 are all matched, the updating unit 19 updates the importance level so that the importance level of the signal becomes higher. For example, the updating unit 19 stores, as the importance of the signal A1 after updating, a value in which the importance of the signal A1 stored in the 2 nd storage unit 12 is 1.5 times as large as the importance of the signal A1 stored in the 2 nd storage unit 12.

In the example shown in table 3, the value of the signal A2 included in the tracking data X1 is 0. The signal A2 included in the tracking data X2 has a value of 1. The 2 nd determination unit 18 determines that the signal a21 is not similar to the signal a22. The judgment by the 2 nd judging unit 18 does not match the judgment by the 1 st judging unit 17 on the tracking data X1 and the tracking data X2. The signal A2 included in the tracking data X3 has a value of 0. The 2 nd determination unit 18 determines that the signal a21 is similar to the signal a 23. The judgment by the 2 nd judging unit 18 does not match the judgment by the 1 st judging unit 17 on the tracking data X1 and the tracking data X3. If the determination by the 1 st determining unit 17 does not match the determination by the 2 nd determining unit 18, the updating unit 19 updates the importance level so that the importance level of the signal becomes lower. For example, the updating unit 19 stores, as the importance of the signal A2 after updating, a value in which the importance of the signal A2 stored in the 2 nd storage unit 12 is 0.5 times as large as the importance of the signal A2 stored in the 2 nd storage unit 12.

Similarly, in the example shown in table 3, the value of the signal A3 included in the tracking data X1 is 0. The signal A3 included in the tracking data X2 has a value of 0. The 2 nd determination unit 18 determines that the signal a31 is similar to the signal a32. The judgment by the 2 nd judging unit 18 matches the judgment by the 1 st judging unit 17 on the tracking data X1 and the tracking data X2. The signal A3 included in the tracking data X3 has a value of 0. The 2 nd determination unit 18 determines that the signal a31 is similar to the signal a 33. The judgment by the 2 nd judging unit 18 does not match the judgment by the 1 st judging unit 17 on the tracking data X1 and the tracking data X3. If the determination by the 1 st determining unit 17 and the determination by the 2 nd determining unit 18 do not match, the updating unit 19 updates the importance level so that the importance level of the signal becomes low. For example, the updating unit 19 stores, as the importance of the signal A3 after updating, a value in which the importance of the signal A3 stored in the 2 nd storage unit 12 is 0.5 times as large as the importance of the signal A3 stored in the 2 nd storage unit 12.

With respect to the signal A3, the judgment by the 1 st judging section 17 and the judgment by the 2 nd judging section 18 partially match. Therefore, the updating unit 19 may store the value of the importance of the signal A3 stored in the 2 nd storage unit 12, which is 0.9 times the importance of the signal A3 stored in the 2 nd storage unit 12, as the importance of the updated signal A3.

After the maintenance operator has fed back the search result, the 1 st decision unit 17 makes a decision as to whether the search results are similar or dissimilar. In the example shown in table 1 and the example shown in table 2, the importance of each signal stored in the 2 nd storage unit 12 is updated every time the 1 st determination unit 17 performs the determination. In the example shown in table 3, the importance of each signal stored in the 2 nd storage unit 12 is updated every time the 1 st determination unit 17 makes 2 determinations. The timing of updating the importance of each signal is not limited to the above example. For example, the importance may be updated every time the 1 st determining unit 17 makes 10 determinations. The importance may be updated when the set date or time is reached.

In the example shown in the present embodiment, the importance of each signal stored in the 2 nd storage unit 12 can be updated. In addition, the judgment of the maintenance worker can be reflected in the value of the importance of each signal. The importance of each signal is updated, and therefore, the accuracy of the search is improved. The importance level is updated for each signal, but the input of the maintenance person is only information as to whether the trace data is similar. Therefore, the labor and time of the maintenance worker can be reduced.

In the present embodiment, an example is described in which the maintenance person directly inputs information on whether or not the trace data X1 and the trace data X2 are similar to each other from the input unit 16. This is an example. As described above, after the repair of the elevator apparatus is completed, the maintenance worker inputs the contents of the correspondence performed by the maintenance worker from the input unit 16, for example, as a comment on the tracking data X1. The 1 st storage unit 10 stores the corresponding content performed in the past as a comment for the tracking data X2. In S302 of fig. 13, the 1 st deciding unit 17 may decide whether or not the tracking data X1 and the tracking data X2 are similar to each other based on the comment on the tracking data X1 and the comment on the tracking data X2.

For example, the 1 st determination unit 17 decomposes the comment for the tracking data X1 for each word. The 1 st judgment unit 17 decomposes the comment for the tracking data X2 for each word. The 1 st determination unit 17 determines that the tracking data X1 is similar to the tracking data X2 when an arbitrary word included in the comment for the tracking data X2 is included in the comment for the tracking data X1. The 1 st determination unit 17 determines that the tracking data X1 is not similar to the tracking data X2 when any word included in the comment for the tracking data X2 is not included in the comment for the tracking data X1. Table 4 shows an example of annotation for the trace data.

[ TABLE 4 ]

No.	Corresponding content
		Trace data X1	Replacement of YZ substrate
Trace data X2	Adjusting YZ substrates for reasons

In the example shown in table 4, both the comment on the tracking data X1 and the comment on the tracking data X2 include a word "YZ substrate". Therefore, the 1 st determination unit 17 determines that the tracking data X1 is similar to the tracking data X2.

In addition, the input of the comment is performed by various maintenance personnel, and therefore, it is possible to use a plurality of expressions for the same device. For example, sometimes a certain maintenance person inputs a certain device in japanese and a certain maintenance person inputs the device in english. In addition, some maintenance person may input a device in an industry phrase, and some maintenance person may input the device in a company interior phrase. Therefore, the search apparatus 1 may register a similar expression set in advance. Table 5 shows an example of a similar expression set.

[ TABLE 5 ]

In the examples shown in Table 5, the term "YZ substrate" and the term "ABC" are considered to be the same for the judgment part 1. Similarly, the 1 st judging section 17 is regarded as the same term for the ZZ switch and SW-ZZ.

Since a general term such as "replacement" is excluded from the target, the 1 st determining unit 17 may determine whether or not the tracking data X1 and the tracking data X2 are similar to each other only from terms listed in the similar term set.

After the repair of the elevator installation, the maintenance person must enter comments for the tracking data. Therefore, if the above example is used, the labor and time of the maintenance worker can be reduced.

Embodiment 3

Fig. 15 is a diagram showing an example of the search device 1 included in the search system according to embodiment 3 of the present invention. The overall view of the search system is the same as that of fig. 1. The search device 1 includes, for example, a3 rd determination unit 20 in addition to the reception unit 9, the 1 st storage unit 10, the 1 st calculation unit 11, the 2 nd storage unit 12, the 2 nd calculation unit 13, the display control unit 14, the display 15, the input unit 16, the 1 st determination unit 17, the 2 nd determination unit 18, and the update unit 19. The functions of the reception unit 9, the 1 st storage unit 10, the 1 st calculation unit 11, the 2 nd storage unit 12, the 2 nd calculation unit 13, the display control unit 14, and the display 15 are the same as those of the arbitrary functions disclosed in embodiment 1. In the case of retrieving the trace data similar to the trace data X1, for example, the processing shown in fig. 5 is performed. In the 2 nd storage unit 12, in addition to the importance of each signal, the importance of a combination of specific signals may be stored. Among the functions of the input unit 16, the 1 st determining unit 17, the 2 nd determining unit 18, and the updating unit 19, functions not described in detail in the present embodiment are the same as any of the functions disclosed in embodiment 2.

In embodiment 2, the following example is explained: when the 1 st decision unit 17 determines that the signals are similar and the 2 nd decision unit 18 determines that the signals are similar, the importance is set to, for example, 1.5 times, regardless of the type of the signal. In this embodiment, an example of an update method for changing the importance level according to the value of a signal will be described.

Next, the function and operation of the search device 1 will be described with reference to fig. 16. Fig. 16 is a flowchart showing an example of the operation of the search system in embodiment 3 of the present invention. The processing shown in S401 to S403 in fig. 16 is the same as the processing shown in S301 to S303 in fig. 13. The 1 st determination unit 17 determines whether or not the tracking data X1 and the tracking data X2 are similar to each other, for example, based on the information input from the input unit 16 (S402). The 2 nd determining unit 18 determines whether or not the signal included in the tracking data X1 is similar to the signal included in the tracking data X2 for each corresponding signal (S403).

The 3 rd determining unit 20 determines whether or not the value of the signal included in the tracking data X1 is a special value that satisfies a predetermined condition (S404). Next, the function of the 3 rd deciding unit 20 will be described in detail with reference to table 6.

[ TABLE 6 ]

Table 6 shows an example of the contents stored in the 1 st storage unit 10. For example, the signal A2 is less likely to have a value of 1, and is more likely to have a value of 0. If the value of the signal A2 is 1, the 3 rd decision unit 20 decides that the value is a special value.

The 3 rd determination unit 20 performs the above determination according to a preset condition. For example, if the frequency of occurrence of a value of a certain signal is equal to or less than the reference value, the 3 rd determination unit 20 determines that the value of the signal is a special value. For example, the reference value is set to 3%. The frequency of the signal A2 having a value of 0 is 99%, and the frequency of the signal A2 having a value of 1 is 1%. In this case, if the value of the signal A2 is 0, the 3 rd determining unit 20 does not determine that the value is a special value. If the value of the signal A2 is 1, the 3 rd decision unit 20 decides that the value is a special value. The reference value for determining whether or not the value of the signal is the special value is not limited to the above example. When the signal has various values, the 3 rd decision unit 20 may perform the above decision based on the degree of deviation such as the standard deviation.

The updating unit 19 updates the importance of each signal stored in the 2 nd storage unit 12 (S405). The updating unit 19 performs the updating based on the result determined by the 1 st determining unit 17, the result determined by the 2 nd determining unit 18, and the result determined by the 3 rd determining unit 20.

For example, it is assumed that the 1 st determining unit 17 determines that the tracking data X1 is similar to the tracking data X2. In the example shown in table 6, the value of the signal A2 included in the tracking data X1 is 1. The signal A2 included in the tracking data X2 has a value of 1. The 2 nd determination unit 18 determines that the signal a21 is similar to the signal a22. The updating unit 19 updates the importance of the signal determined to be similar by the 2 nd determining unit 18 so that the importance when the 3 rd determining unit 20 determines that the value of the signal is a special value is higher than the importance when the 3 rd determining unit 20 determines that the value of the signal is not a special value. That is, regarding the signal A2, the updated importance level when both the signal a21 and the signal a22 are 1 is higher than the updated importance level when both the signal a21 and the signal a22 are 0.

In the above example, when both the signal a21 and the signal a22 are 1, the updating unit 19 updates the importance level of the signal A2 by multiplying a value obtained by multiplying the importance level of the signal A2 stored in the 2 nd storage unit 12 by 1 or more constant, for example. The updating unit 19 may store, as the importance of the signal A2 after updating, a value obtained by multiplying the importance of the signal A2 stored in the 2 nd storage unit 12 by 1.5 times by the inverse of the appearance frequency in the 2 nd storage unit 12. This makes it possible to update the signal value with a high priority.

In addition, for a signal having a value of only 0 or 1, such as the signal A2 shown in table 6, the 2 nd storage unit 12 may store 2 importance levels as the importance levels of the signal. For example, the importance of the signal A2 when the 2 nd storage unit 12 stores a value of 0 and the importance of the signal A2 when the value is 1 may be set.

Embodiment 4

Fig. 17 is a diagram showing an example of the search device 1 included in the search system according to embodiment 4 of the present invention. The overall view of the search system is the same as that of fig. 1. The search device 1 further includes a 4 th determination unit 21 in addition to the reception unit 9, the 1 st storage unit 10, the 1 st calculation unit 11, the 2 nd storage unit 12, the 2 nd calculation unit 13, the display control unit 14, the display 15, the input unit 16, the 1 st determination unit 17, the 2 nd determination unit 18, and the update unit 19, for example. The functions of the receiving unit 9, the 1 st storage unit 10, the 1 st calculating unit 11, the 2 nd storage unit 12, the 2 nd calculating unit 13, the display control unit 14, and the display 15 are the same as those of the arbitrary functions disclosed in embodiment 1. In the case of retrieving the trace data similar to the trace data X1, for example, the processing shown in fig. 5 is performed. In the 2 nd storage unit 12, in addition to the importance of each signal, the importance of a combination of specific signals may be stored. Among the functions of the input unit 16, the 1 st determining unit 17, the 2 nd determining unit 18, and the updating unit 19, those not described in detail in the present embodiment are the same as those disclosed in

embodiment

2 or 3. The search device 1 may further include a3 rd determination unit 20.

In this embodiment, an example of an update method for changing the importance level according to the magnitude of the similarity of signals is shown. Next, the function and operation of the search device 1 will be described with reference to fig. 18. Fig. 18 is a flowchart showing an example of the operation of the search system in embodiment 4 of the present invention. The processing shown in S501 to S503 of fig. 18 is the same as the processing shown in S301 to S303 of fig. 13.

The 1 st determination unit 17 determines whether or not the tracking data X1 and the tracking data X2 are similar to each other, for example, based on the information input from the input unit 16 (S502). The 2 nd determination unit 18 determines whether or not the signal included in the tracking data X1 is similar to the signal included in the tracking data X2 for each corresponding signal (S503).

The 4 th decision unit 21 decides whether or not the degree of similarity between the tracking data X1 and the tracking data X2 calculated by the 2 nd calculation unit 13 is higher than a reference value (S504). The updating unit 19 updates the importance of each signal stored in the 2 nd storage unit 12 (S505). The updating unit 19 performs the updating based on the result determined by the 1 st determining unit 17, the result determined by the 2 nd determining unit 18, and the result determined by the 4 th determining unit 21.

For example, when the determination by the 1 st determining unit 17 and the determination by the 2 nd determining unit 18 match, the updating unit 19 updates the importance level of the signal so that the importance level becomes higher. Further, the updating unit 19 updates the importance of the signal so that the importance when the 4 th deciding unit 21 decides that the similarity is not higher than the reference value is higher than the importance when the 4 th deciding unit 21 decides that the similarity is higher than the reference value.

On the other hand, if the determination by the 1 st determining unit 17 does not match the determination by the 2 nd determining unit 18, the updating unit 19 updates the importance level of the signal so that the importance level becomes low. Further, the updating unit 19 updates the importance of the signal so that the importance when the 4 th deciding unit 21 decides that the similarity is higher than the reference value is lower than the importance when the 4 th deciding unit 21 decides that the similarity is not higher than the reference value.

The example in which the updating unit 19 updates the importance level based on the result determined by the 4 th determining unit 21 is not limited to the above example. For example, the updating unit 19 may update the importance level according to the following expression. The following formula is an example of calculation when 50% is used as the reference value.

Case of similarity higher than 50%

( Importance after update) = (importance before update) × (conversion rate of importance: for example, 1.5) × (the number of times of determination as dissimilar by the 1 st determining section 17/the total number of determinations by the 1 st determining section 17) × ((100-degree of similarity [% ])/50 )

The degree of similarity is 50% or less

( Importance after update) = (importance before update) × (conversion rate of importance: for example, 0.5 x (the number of times of determination as similar by the 1 st determining part 17/the total number of determinations by the 1 st determining part 17) x ((50-degree of similarity [% ])/50 )

The above equation shows an example in which the update unit 19 updates the importance of the signal also based on the number of times of determination as dissimilar or the number of times of determination as similar by the 1 st determination unit 17. When the above-described number of times does not need to be considered, the right-hand item 3 of the above expression may be set to 1 for calculation. Similarly, the above expression shows an example in which the updating unit 19 updates the importance of the signal based on the value of the similarity of the tracking data calculated by the 2 nd calculating unit 13. When the similarity value does not need to be considered, the right-hand 4 th term of the above expression may be set to 1 for calculation.

In the above calculation example, for example, in the case where the similarity is high but the similarity is apparently dissimilar to the maintainer, the importance of the signal can be significantly reduced. Also, for instances where the similarity is low but apparently similar to the maintainer, the importance of the signal can be greatly improved.

Embodiment 5

In the present embodiment, an example will be described in which the updating unit 19 further updates the importance of each signal stored in the 2 nd storage unit 12 based on the information of the input person who has performed the feedback input. The overall view of the search system is the same as that of fig. 1. The overall view of the search device 1 is the same as that of fig. 12, for example. The search device 1 may also include a3 rd determination unit 20. The search device 1 may further include a 4 th determination unit 21.

In the example shown in the present embodiment, for example, the same operation as that shown in fig. 13 is performed. In the example shown in the present embodiment, information of the input person who has performed the feedback input is input from the input unit 16. In S304, the updating unit 19 also updates the importance of each signal stored in the 2 nd storage unit 12 based on the information of the input person input from the input unit 16. For example, the maintenance person inputs his/her employee number from the input unit 16. The updating unit 19 also considers the importance of the skill level update signal of the maintenance worker who has performed the feedback input, for example. For example, the updating unit 19 updates the importance according to the following expression.

( Importance after update) = (importance before update) × (conversion rate of importance: e.g., 1.5) × (working years of inputter/constant )

As the constant, for example, an average value of the number of working years of the maintenance worker can be used. The constant may be 1. In the above equation, "years of experience of the input person" may be used instead of "years of work of the input person". In the above equation, "the number of times of failure of the input person" may be used instead of "the number of operating years of the input person". The information required by the above formula is stored in the search device 1 in association with, for example, the employee number.

The updating unit 19 may update the importance of the signal so that, for example, when the input person belongs to the specific group, the importance becomes higher than when the input person does not belong to the specific group.

When the information of the input person is applied to the example disclosed in embodiment 4, the updating unit 19 may update the importance level according to the following expression, for example.

Case of similarity higher than 50%

( Importance after update) = (importance before update) × (conversion rate of importance: for example, 1.5) × ((the number of times of determination as dissimilar by the 1 st determination unit 17 × the number of operating years of the input person)/the total determination number by the 1 st determination unit 17) × (100-degree of similarity [% ]/50 )

In the above formula, the right 4 th term may be 1. The same applies to the case where the similarity is 50% or less.

In the example shown in the present embodiment, the information of the person who has performed the feedback input can be reflected in the value of the importance of each signal. As in the above example, the labor and time of the maintenance worker can be reduced by inputting only the employee number or the like from the input unit 16.

Embodiment 6

In the present embodiment, an example will be described in which the updating unit 19 updates the importance of each signal stored in the 2 nd storage unit 12 in accordance with the number of times of determination by the 1 st determining unit 17. The overall view of the search system is the same as that of fig. 1. The overall view of the search device 1 is the same as that of fig. 12, for example. The search device 1 may also include a3 rd determination unit 20. The search device 1 may further include a 4 th determination unit 21. The updating unit 19 may also consider the information of the input person input from the input unit 16 when updating the importance.

In the example shown in the present embodiment, the updating unit 19 updates the importance level according to the following expression, for example.

(importance after update) = (importance before update) × f (x)

f (x) is a correction function. x is the number of times the 1 st deciding unit 17 decides the tracking data as the object. The example shown in embodiment 2 corresponds to a case where f (x) is 1 regardless of the value of x.

Fig. 19 and 20 are diagrams showing examples of the updated importance. Fig. 19 and 20 show an example in which the importance level is increased by updating. In the example shown in fig. 19, the number of determinations by the 1 st determining unit 17 reaches n ₁ Previously, the rate of change of the importance was a ₁ . The number of determinations in the 1 st determining part 17 is n ₁ ～n ₂ The rate of change of the importance is a ₂ . Rate of change a ₂ Greater than the rate of change a ₁ . The number of determinations by the 1 st determining unit 17 reaches n ₂ When the importance is changed, the rate of change of the importance is a ₃ . Rate of change a ₃ Less than the rate of change a ₂ . Rate of change a ₃ Can be associated with the rate of change a ₁ The same is true. The change in updated importance may be illustrated graphically. The number of determinations by the 1 st determining unit 17 reaches n ₂ After that, the importance may not be changed.

The maintainer may check the wrong checkbox when entering the feedback. By changing the importance after updating as shown in fig. 19, the importance can be updated without changing the value significantly until a certain degree of input is performed. For example, n may be ₁ Is set to 3.

In the example shown in fig. 20, the number of determinations by the 1 st determining unit 17 reaches n ₃ Previously, the rate of change of the importance was a ₄ . The number of determinations by the 1 st determining unit 17 reaches n ₃ When the importance is changed, the rate of change of the importance is a ₅ . Rate of change a ₄ Greater than the rate of change a ₅ . The change in updated importance may be illustrated graphically. The number of determinations by the 1 st determining unit 17 reaches n ₃ After that, the importance may not be changed.

In fig. 19 and 20, the horizontal axis represents the absolute value of the number of times the 1 st decision unit 17 makes a decision. In order to similarly process the trace data that is frequently feedback-inputted and the trace data that is hardly feedback-inputted, the horizontal axis may be a ratio of the number of times of determination by the 1 st determining unit 17. For example, the horizontal axis may adopt "the number of times the 1 st determining unit 17 determines the target tracking data/the total number of determinations by the 1 st determining unit 17". In the case where "the ratio of the number of times the 1 st determining unit 17 performs the determination" is adopted on the horizontal axis, when the first feedback input is performed, the ratio is 100%, and the importance may not be updated appropriately. Therefore, when the above ratio is used on the horizontal axis, the importance updating function may be stopped before a certain number of feedback inputs are performed.

Each of the parts shown by reference numerals 9 to 14 and 16 to 21 shows a function of the search device 1. Fig. 21 is a diagram showing a hardware configuration of the search device 1. The search device 1 has, as hardware resources, a processing circuit including a processor 22 and a memory 23, for example. The functions of the 1 st storage unit 10 and the 2 nd storage unit 12 are realized by the memory 23. The search device 1 implements the functions of the respective sections indicated by

reference numerals

9, 11, 13 to 14, and 16 to 21 by executing the program stored in the memory 23 by the processor 22.

The processor 22 is also called a CPU (Central Processing Unit), a Central Processing Unit, a Processing device, an arithmetic device, a microprocessor, a microcomputer, or a DSP. As the memory 23, a semiconductor memory, a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD can be used. Semiconductor memories which may be used include RAM, ROM, flash memory, EPROM, EEPROM, and the like.

Part or all of the functions of the search device 1 may be implemented by hardware. As hardware for realizing the function of the search device 1, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof may be used.

Industrial applicability

The retrieval system of the present invention can be applied to a system that retrieves a signal group similar to a specific signal group from among a plurality of signal groups.

Description of the reference symbols

1: a retrieval means; 2: a car; 3: a counterweight; 4: a main rope; 5: a drive sheave; 6: an electric motor; 7: a control panel; 8: a communication device; 9: a receiving section; 10: a1 st storage unit; 11: a1 st calculation unit; 12: a2 nd storage unit; 13: a2 nd calculation unit; 14: a display control unit; 15: a display; 16: an input section; 17: a1 st judgment unit; 18: a2 nd judgment unit; 19: an update unit; 20: a3 rd judgment unit; 21: a 4 th judgment unit; 22: a processor; 23: a memory.

Claims

1. A retrieval system, wherein the retrieval system has:

a1 st storage unit that stores a1 st signal group including a plurality of signals and a2 nd signal group including a plurality of signals corresponding to the plurality of signals included in the 1 st signal group;

a1 st calculation unit that calculates a similarity between a signal included in the 1 st signal group and a signal included in the 2 nd signal group for each corresponding signal;

a2 nd storage unit that stores the importance of each signal for the plurality of signals included in the 1 st signal group;

a2 nd calculation unit that calculates a similarity between the 1 st signal group and the 2 nd signal group based on the similarity of each signal calculated by the 1 st calculation unit and the importance of each signal stored in the 2 nd storage unit;

an input unit for inputting information;

a1 st decision unit that decides whether or not the 1 st signal group and the 2 nd signal group are similar based on information input from the input unit;

a2 nd determination unit that determines whether or not a signal included in the 1 st signal group is similar to a signal corresponding to the signal included in the 2 nd signal group; and

an updating unit that updates the importance of the signal stored in the 2 nd storage unit, based on the result determined by the 1 st determining unit and the result determined by the 2 nd determining unit.

2. The retrieval system of claim 1,

when the 1 st signal group and the 2 nd signal group are determined to be similar by the 1 st determining unit, the updating unit updates the importance degrees stored in the 2 nd storage unit so that the importance degree of the signal determined to be similar by the 2 nd determining unit becomes high and the importance degree of the signal determined to be dissimilar by the 2 nd determining unit becomes low.

3. The retrieval system of claim 2,

the search system further includes a3 rd determination unit that determines whether or not a value of a signal included in the 1 st signal group is a special value that satisfies a predetermined condition,

when the 1 st signal group and the 2 nd signal group are determined to be similar by the 1 st determining unit, the updating unit updates the importance of the signal determined to be similar by the 2 nd determining unit such that the importance when the value of the signal is determined to be a special value by the 3 rd determining unit is higher than the importance when the value of the signal is determined not to be a special value by the 3 rd determining unit.

4. The retrieval system according to claim 1 or 2,

when the 1 st decision unit decides that the 1 st signal group and the 2 nd signal group are not similar, the updating unit updates the importance levels stored in the 2 nd storage unit so that the importance level of the signal decided to be dissimilar by the 2 nd decision unit becomes high and the importance level of the signal decided to be similar by the 2 nd decision unit becomes low.

5. The retrieval system of claim 1,

the retrieval system further has a 4 th decision unit that decides whether or not the similarity of the 1 st signal group and the 2 nd signal group calculated by the 2 nd calculation unit is higher than a reference value,

the updating unit further updates the importance degree stored in the 2 nd storage unit in accordance with the result determined by the 4 th determining unit.

6. The retrieval system of claim 5,

the updating unit further updates the importance degree stored in the 2 nd storage unit, based on the value of the similarity degree calculated by the 2 nd calculation unit.

7. The retrieval system of claim 5 or 6,

the updating unit further updates the importance degree stored in the 2 nd storage unit in accordance with the number of times the 1 st signal group and the 2 nd signal group are determined to be dissimilar by the 1 st determining unit, in a case where the 4 th determining unit determines that the similarity is higher than a reference value.

8. The retrieval system of claim 5 or 6,

the updating unit may further update the importance degree stored in the 2 nd storage unit in accordance with the number of times the 1 st signal group and the 2 nd signal group are determined to be similar by the 1 st determining unit, in a case where the 4 th determining unit determines that the similarity degree is lower than a reference value.

9. The retrieval system according to any one of claims 1 to 8,

information of an inputter is input from the input unit,

the updating unit further updates the importance stored in the 2 nd storage unit in accordance with the information of the inputter input from the input unit.

10. The retrieval system according to any one of claims 1 to 9,

the updating unit further updates the importance stored in the 2 nd storage unit in accordance with the number of times of determination by the 1 st determining unit.

11. The retrieval system according to any one of claims 1 to 10,

information on whether the 1 st signal group and the 2 nd signal group are similar is input from the input unit.

12. The retrieval system according to any one of claims 1 to 11,

inputting an annotation for the 1 st signal group and an annotation for the 2 nd signal group from the input unit,

the 1 st decision unit decides whether or not the 1 st signal group and the 2 nd signal group are similar from the comment for the 1 st signal group and the comment for the 2 nd signal group input from the input unit.

13. The retrieval system according to any one of claims 1 to 12,

the 1 st calculation unit calculates a similarity between a combination of specific signals included in the 1 st signal group and a combination of signals corresponding to the combination of the specific signals included in the 2 nd signal group,

the importance of the combination of the specific signals is stored in said 2 nd memory unit,

the 2 nd calculating unit further calculates the similarity between the 1 st signal group and the 2 nd signal group based on the similarity of the combination of the specific signals calculated by the 1 st calculating unit and the importance of the combination of the specific signals stored in the 2 nd storing unit.

14. The retrieval system according to any one of claims 1 to 13,

the search system further includes a display control unit that associates the similarity between the 1 st signal group and the 2 nd signal group calculated by the 2 nd calculation unit and information indicating similar signals and displays the result on a display,

the display control unit determines the similar signal based on the similarity of each signal calculated by the 1 st calculation unit and the importance of each signal stored in the 2 nd storage unit.

15. A retrieval system, wherein the retrieval system has:

a2 nd storage unit that stores the importance of each signal for the plurality of signals included in the 1 st signal group; and

a2 nd calculation unit that calculates a similarity between the 1 st signal group and the 2 nd signal group based on the similarity of each signal calculated by the 1 st calculation unit and the importance of each signal stored in the 2 nd storage unit,

16. A retrieval system, wherein the retrieval system has:

a2 nd calculation unit that calculates a similarity of the 1 st signal group and the 2 nd signal group based on the similarity of each signal calculated by the 1 st calculation unit and the importance of each signal stored in the 2 nd storage unit; and

a display control unit for displaying on a display the similarity between the 1 st signal group and the 2 nd signal group calculated by the 2 nd calculation unit and information indicating similar signals in association with each other,