JP2017211682A - Data search eye point generation method, facility management device, and facility management system using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem that a conventional online analysis process requires that a search eye point adjusted to a purpose by a user be generated in advance, or a search pattern or the like be acquired by machine learning and a search point be generated in advance, but because an item to be searched and information relating thereto change, a search eye point cannot be fixed for the extraction of facility abnormality information in facility management, causing application production costs to increase, e.g., a search eye point is produced every time, and making the adoption of an online analysis process difficult.SOLUTION: The present invention, designed to solve a problem such as the one stated above, automatically generates data search eye point information needed for an online analysis process making most of facility information and significantly reduces an application production volume for adoption to a facility management system.SELECTED DRAWING: Figure 2

Description

The present invention relates to an analysis apparatus that performs online analysis processing for equipment management, particularly a data search viewpoint generation method that generates a search viewpoint when performing online analysis processing, a data search viewpoint generation apparatus, and an equipment management system that uses this apparatus About.

  Due to an increase in equipment management costs due to an increase in aging equipment and a decrease in skilled technicians, in addition to regular equipment inspections for conventional equipment management, the introduction of equipment management utilizing sensors is progressing. Such sensor information is time-series data that changes over time, and a technique for efficiently analyzing large-scale data in time series is required by accumulating such sensor information. For example, Patent Literature 1 discloses analysis of large-scale data related to hardware failure due to various factors. Patent Document 2 discloses an analysis method for information that changes one after another.

JP 2003-233686 A JP 2001-134575 A

The analysis of hardware failure in Patent Document 1 is a typical analysis method of OLAP (Online Analytical Processing), and the cause of failure is composed of a database of three viewpoints: customer viewpoint, product viewpoint, and parts viewpoint. The failure rate is analyzed using these multidimensional databases. However, the information accumulated in these databases is information from a so-called static search viewpoint that was created in advance by the user according to the purpose at the beginning of analysis. Therefore, an efficient search cannot be performed from such a static search viewpoint.
Patent Document 2 realizes high-speed data search from large-scale information by machine learning of search patterns and the like to generate search viewpoints. The search viewpoint cannot be fixed because the items to be searched and the information related thereto are changed, and the application production cost is large and difficult to apply such as creating the search viewpoint every time the online analysis process is applied.

  This invention is an equipment management analyzer for solving the above-mentioned problems, and in addition to the conventional equipment abnormality analysis from a static viewpoint, the equipment is searched for abnormal equipment from the sensor measurement data, and the equipment attributes Create a dynamic perspective of data by separately retrieving information and influencing equipment information. As a result, the number of searches or the development amount of search applications can be reduced, and the omission of search for data necessary for business can be reduced.

The facility management apparatus according to the present invention includes a sensor installed in the facility, an input unit for inputting information for managing the facility such as a history of inspection and repair, and a memory for storing information from the sensor and the input unit as a plurality of databases. A data analysis unit that generates a hierarchical search viewpoint information related to a search viewpoint item that is determined to be abnormal by comparing updated information with a threshold value among stored information, A multidimensional data generation unit is provided that extracts multidimensional data from a plurality of databases based on search viewpoint information generated by the data analysis unit and generates a new database.
Further, the data retrieval viewpoint generation method according to the present invention includes a sensor installed in equipment, an input device for inputting information for managing equipment such as inspection and repair history, and information from the sensors and input devices in a plurality of databases. Search viewpoint information in which search viewpoint items related to the abnormal value are hierarchized when the updated sensor information or the inspection history information is determined to be an abnormal value. Generate a new database by extracting multidimensional data from a plurality of databases based on the search viewpoint information.
An equipment management system according to the present invention includes the equipment management apparatus described above and an analysis apparatus that analyzes multidimensional data generated by the equipment management apparatus, and performs failure diagnosis of equipment in the equipment. And

According to the facility management apparatus or the data retrieval viewpoint generation method of the present invention, a new database is generated by extracting multidimensional data from a plurality of databases based on hierarchical retrieval viewpoint information related to a retrieval viewpoint item. Can be performed at high speed.
Further, according to the facility management system of the present invention, failure diagnosis can be performed in a short time by the facility management device that enables high-speed search.

BRIEF DESCRIPTION OF THE DRAWINGS It is a structure schematic diagram of the equipment management system of Embodiment 1 of this invention. It is a functional block diagram of the equipment management apparatus of Embodiment 1 of this invention. It is a flowchart until it produces | generates the multidimensional data of Embodiment 1 of this invention. It is an image figure of sensor data DB of Embodiment 1 of this invention. It is an image figure of inspection history information DB of Embodiment 1 of this invention. It is an image figure of equipment ledger information DB of Embodiment 1 of this invention. It is an image figure of the repair log | history information DB of Embodiment 1 of this invention. It is an image figure of the threshold value information data of Embodiment 1 of this invention. It is an image figure of the threshold determination result of Embodiment 1 of this invention. It is an image figure of the data search viewpoint information of Embodiment 1 of this invention. It is an image figure of the multidimensional data of Embodiment 1 of this invention. It is a flowchart until it produces | generates the multidimensional data of Embodiment 2 of this invention. It is an image figure of the item extraction result of Embodiment 2 of this invention. It is a functional block diagram of the equipment management apparatus of Embodiment 3 of this invention. It is a flowchart until it produces | generates the multidimensional data of Embodiment 3 of this invention. It is a functional block diagram of the equipment management apparatus of Embodiment 4 of this invention. It is a flowchart until it produces | generates the multidimensional data of Embodiment 4 of this invention, and outputs it externally. It is an image figure of the output device of Embodiment 4 of this invention. It is a functional block diagram which shows Embodiment 5 of this invention. It is a flowchart until it produces | generates the multidimensional data of Embodiment 5 of this invention. It is a whole structure figure of the equipment management system of Embodiment 6 of this invention.

Embodiment 1
Embodiment 1 of the present invention will be described below.
FIG. 1 is a schematic configuration diagram showing an equipment management system according to Embodiment 1 of the present invention. Reference numerals 1A and 2B denote facilities in the facility management system. AA1, AA2, BB1, and BB2 indicate devices in the facility 1A or 2B. 3 is a current sensor, 4 is a voltage sensor, 5 is a temperature sensor, and 6 is a vibration sensor. Although typical sensors are listed, various sensors for performing facility management are conceivable. An input device 7 is used to input data other than the sensor output. Here, the inspection history, repair history, and the like are input. Reference numeral 8 denotes an output device that outputs the result of data analysis. Reference numeral 10 denotes an equipment management device, which can automatically generate data search viewpoint information from data of various sensors 3 to 6 and data input by the input device 7 as will be described later. The hardware constituting the equipment management apparatus 10 includes an input / output control apparatus 11 that controls input / output of data, analyzes input data, automatically generates data search viewpoint information, and generates multidimensional data. And a memory 13 in which various input data are stored, and a CPU 12 that performs analysis described in detail below by this program. The communication between the input / output control device 11 and the various sensors 3 to 6 and the input device, or the communication between the input / output control device 11 and the output device 8 may be wired or wireless.

  FIG. 2 is a functional block diagram of the facility management apparatus 10 described above. In FIG. 2, the facility management apparatus 10 is acquired from a data search viewpoint automatic generation apparatus 100 that automatically generates a data search viewpoint, and sensors 3, 4, 5, and 6 that are data referred to by the data search viewpoint automatic generation apparatus. A sensor data database (hereinafter referred to as DB) 110 that records sensor data, an inspection history information DB 111 that records periodic inspection results, an equipment ledger information DB 112 that records ledger data that is attribute information of equipment, The repair history information DB 113 that records the repair history of the equipment, the threshold information DB 121 that stores threshold information for judging abnormality of the equipment sensors and inspection results, the threshold information and the data in the DBs 110, 111, 112, and 113 Compare and extract unique data items that exceed threshold information, and a tree-like hierarchy centered on these unique data items From the data search viewpoints of the data search viewpoint information DB 122 and the data search viewpoint information DB 122 that record the results analyzed by the data analysis apparatus 120, the data analysis viewpoint 120 that generates data search viewpoint information having a structure, the DBs 110, 111, 112, and 113 The multi-dimensional data generation device 130 that processes the data of the above and generates multi-dimensional data, and the multi-dimensional data DB 131 generated by the multi-dimensional data generation device 130.

Next, the operation will be described with reference to FIGS.
FIG. 3 is a flowchart for automatically generating data retrieval viewpoint information from acquired facility information and generating multidimensional data.
The data retrieval viewpoint automatic generation apparatus 100 includes a sensor data DB 110 (data image is shown in FIG. 4), an inspection history information DB 111 (data image is shown in FIG. 5), an equipment ledger information DB 112 (data image is shown in FIG. 6), repair. History information DB 113 (data image shown in FIG. 7) is acquired and transmitted to data analysis apparatus 120 (ST1001).

  The data analysis device 120 determines whether or not the threshold value of the threshold information DB 121 (data image is shown in FIG. 8) is exceeded from the data items of the sensor data DB 110 and the inspection history information DB 111, and extracts the exceeded items (ST1002). . An image of the threshold determination result is shown in FIG.

  The data analysis apparatus 120 includes the sensor data DB 110, the inspection history information DB 111, the equipment ledger information DB 112, the repair history related to the items extracted in step ST1002 described above (in the image of FIG. 9, the equipment current value and equipment vibration value are targets). Tree-like data search viewpoint information for referring to the information DB 113 is generated and recorded in the data search viewpoint information DB 122 (ST1003). The generated tree-shaped data search viewpoint information is shown in FIG.

  The multidimensional data generating apparatus 130 combines multidimensional data obtained by combining information from the sensor data DB 110, the inspection history information DB 111, the equipment ledger information DB 112, and the repair history information DB 113 along the data search viewpoint information generated in step ST1003. It is generated and recorded in the multidimensional data DB 131 (ST1004). An image of the combined multidimensional data is shown in FIG.

  As described above, in the first embodiment, data retrieval viewpoint information related to a data item corresponding to an abnormal value is automatically generated from large-scale data such as sensors and inspection history information. Therefore, it is possible to search as a multidimensional data DB extracted and combined from each DB, so that a high-speed search is possible.

Embodiment 2
In the second embodiment, an online analysis pattern is added to the data analysis apparatus 120 of the first embodiment described above by adding an analysis function that compares the characteristics of digitized information such as sensor information and inspection history information. To increase.

Hereinafter, the operation of the second embodiment will be described with reference to FIGS.
FIG. 12 is a flowchart for automatically generating data search viewpoint information from acquired facility information and generating multidimensional data.
A process (ST2001) in which the data search viewpoint automatic generation apparatus 100 acquires equipment data, a process in which the data analysis apparatus 120 generates data search viewpoint information using a threshold (ST2002), and a multidimensional data generation apparatus 130 The process of generating dimension data (ST2005) is the same as step ST1001, step ST1002, and step ST1004 described in the first embodiment.

The data analysis device 120 creates a numerical graph from numerical data items in the sensor data DB 110 and the inspection history information DB 111, and extracts items different from past graph shape features. An example of the image of the item extraction result is shown in FIG. 13 (ST2003). In FIG. 13, the vertical axis represents the power value, and the horizontal axis represents the time of the day. For example, series 1 indicates past data on January 1, and series 2 indicates current data on January 3. Since such past pattern data can obtain a logical average value compared to the threshold value of sensor data for normal sensing having a large experiential element, the reliability of numerical values increases. Furthermore, it can be used in combination with a threshold value of sensor data, and it becomes possible to extract specific data with higher accuracy.
Furthermore, the data analysis device 120 is a tree-shaped data search viewpoint information for referring to the sensor data DB 110, the inspection history information DB 111, the equipment ledger information DB 112, and the repair history information DB 113 related to the items extracted in Step ST2002 and Step ST2003. Is recorded in the data search viewpoint information DB 122 (ST2004).

  Thus, in Embodiment 2, the accuracy of extracting abnormal information can be increased by increasing the number of data analysis patterns.

Embodiment 3
As shown in FIG. 14, the third embodiment is different from the first embodiment in the data update device 200 that manages the update information for the facility inspection history, ledger information, and repair information, and the data difference information that records the update information. DB201 is added.

Hereinafter, the operation of the third embodiment will be described with reference to FIG. FIG. 15 is a flowchart from when the facility information is updated to when multidimensional data is generated.
The processing (ST3005) in which the multidimensional data generation device 130 generates multidimensional data from the data search viewpoint information generation processing (ST3004) by the data analysis device 120 is the same as steps ST1003 and ST1004 in the first embodiment.

The data update device 200 extracts update information input from the outside and difference information between the sensor data DB 110, the inspection history information DB 111, the equipment ledger information DB 112, and the repair history information DB 113, and records them in the data difference information DB 201 (ST3001). .
The data search viewpoint automatic generation device 100 acquires difference information from the data difference information DB 201 and transmits the difference information to the data analysis device 120 (ST3002).
Data analysis apparatus 120 extracts items that exceed the threshold value in threshold information DB 121 from data items that are difference information in sensor data DB 110 and inspection history information DB 111 (ST3003).

  As described above, in the third embodiment, by acquiring the difference information, the amount of information necessary for the automatic generation of the data search viewpoint can be minimized, and the automatic generation of the data search viewpoint can be accelerated.

Embodiment 4
In the fourth embodiment, as shown in FIG. 16, a device for outputting information on equipment extracted as an abnormality to the outside is added to the first embodiment described above.
Hereinafter, the operation of the fourth embodiment will be described with reference to FIG. FIG. 17 is a flowchart for automatically generating data search viewpoint information from acquired facility information, generating multidimensional data, and outputting the information externally.
From the process (ST4001) in which the data retrieval viewpoint automatic generation apparatus 100 acquires equipment data, the process in which the multidimensional data generation apparatus 130 generates multidimensional data (ST4005) is the same as step ST1001 to step ST1004 in the first embodiment. is there.
The output device 8 outputs information relating to abnormal data items such as the items extracted in step ST4002 and step ST4003 and their values, or the analysis information described in the second embodiment, in an external file format that can be viewed on a PC. (ST4004). For example, as shown in FIG. 18, the image output image is highlighted on the numerical value so that an abnormal numerical value or location can be easily visually recognized.

As described above, in the fourth embodiment, there is an effect that the equipment manager can find an early equipment abnormal state by outputting the equipment information list having the abnormal data items to the outside.

Embodiment 5
In the fifth embodiment, as shown in FIG. 19, a schedule device 220 that schedules the operations of the data retrieval viewpoint automatic generation device 100 and the multidimensional data generation device 130 is added to the above-described first embodiment. is there.

Hereinafter, the operation of the fifth embodiment will be described with reference to FIG. FIG. 20 is a flowchart for automatically generating data search viewpoint information from facility information at the acquired scheduled time and generating multidimensional data.
From the process (ST5002) in which the data search viewpoint automatic generation device 100 acquires equipment data, the process (ST5005) in which the multidimensional data generation apparatus 130 generates multidimensional data is the same as the steps ST1001 to ST1004 in the first embodiment. It is.
The schedule device 220 operates the data search viewpoint automatic generation device 100 or the multidimensional data generation device 130 at the set time (ST5001).

  As described above, in the fifth embodiment, the schedule management of the operations of the data retrieval viewpoint automatic generation device 100 and the multidimensional data generation device 130 with a large amount of processing can be performed at an arbitrary timing. This is effective for efficient resource utilization of the terminal.

Embodiment 6
The sixth embodiment implements an equipment management system that includes an analysis device that analyzes multidimensional data generated by the equipment management device 10 and that performs failure diagnosis of the devices AA1, AA2, BB1, and BB2 in the equipment. An analysis device that analyzes multidimensional data may store an analysis program in the memory 13 and cause the CPU 12 to analyze the analysis program. Alternatively, as shown in FIG. May be installed, multi-dimensional data may be transmitted, and failure diagnosis using larger-scale data may be performed together with multi-dimensional data transmitted from other equipment management apparatuses 10 that are also installed.

  It should be noted that within the scope of the present invention, the embodiments can be freely combined, or the embodiments can be appropriately modified or omitted.

  8 Output device, 100 Data retrieval viewpoint automatic generation device, 110 Sensor data DB, 111 Inspection history DB, 112 Equipment ledger information DB, 113 Repair history information DB, 120 Data analysis device, 121 Threshold information DB, 122 Data retrieval viewpoint information DB , 130 Multidimensional data generator, 131 Multidimensional data DB

Claims (7)

  In an equipment management apparatus that performs online analysis processing of equipment, an input unit that inputs information for managing the equipment, such as a sensor installed in the equipment, a history of inspection and repair, and the information from the sensor and the input section A plurality of databases as storage units, the updated information among the information is compared with a threshold value to determine whether there is an abnormality, and hierarchical search viewpoint information related to the search viewpoint item determined to be abnormal An equipment management apparatus comprising a data analysis unit for generating, and a multidimensional data generation unit for extracting multidimensional data from the plurality of databases based on the search viewpoint information generated by the data analysis unit and generating a new database .   The equipment management apparatus according to claim 1, wherein instead of the threshold value or together with the threshold value, sensor information or past information of the history is used as a reference value for determining whether there is an abnormality.   The equipment management apparatus according to claim 1, wherein difference information before and after the update of the information is compared with a threshold value.   The equipment management apparatus according to any one of claims 1 to 3, further comprising an output device that displays an output of the data analysis unit as an image.   The facility management apparatus according to any one of claims 1 to 4, further comprising a schedule device that schedules the operation timing of input of information in the input unit or the operation timing of the multidimensional data generation unit. .   A sensor installed in the facility, an input device for inputting information for managing the facility such as inspection and repair history, a storage device for storing the information from the sensor and the input device as a plurality of databases, Of the information, when the updated sensor information or inspection history information is determined to be an abnormal value, search viewpoint information in which search viewpoint items related to the abnormal value are hierarchized is generated, and based on the search viewpoint information A data retrieval viewpoint generation method for generating a new database by extracting multidimensional data from the plurality of databases.   An equipment management system comprising: the equipment management apparatus according to claim 1; and an analysis apparatus that analyzes multidimensional data generated by the equipment management apparatus, and performs failure diagnosis of equipment in the equipment.