Data utilization support system, data utilization support method, and program

Abstract

To provide a data utilization support system for supporting the utilization of stored data.SOLUTION: A data utilization support system according to the present embodiment includes a plurality of types of sensors that detect a situation of a manufacturing site, a log acquisition unit that acquires log data that is a detection result of the sensor, and a preprocessing unit that performs preprocessing according to a range of values that can be taken by the log data acquired by the log acquisition unit or the number of values that can be taken by the log data acquired by the log acquisition unit. The pre-processing unit includes a value conversion unit that converts a value of the binary data acquired by the log acquisition unit, and the visualization unit visualizes the log data acquired by the log acquisition unit using the value converted by the value conversion unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a data utilization support system, a data utilization support method, and a program. [Background technology]

[0002] For example, Patent Document 1 discloses a bridge operation status monitoring system in which a monitoring device that sequentially collects sensor output signals from multiple sensors that detect the operation status of the bridge being monitored and a system management center that analyzes the multiple sensor output signals to create and store operation status information for the bridge are connected via a data collection network, the system management center is connected via a user network to a user terminal for receiving or viewing the operation status information, and the system management center overlays the created operation status information on a single graph with a common time axis and displays it on the user terminal.

[0003] Patent Document 2 discloses a greenhouse management system for managing a greenhouse in which plants are grown, comprising: an image display means; a storage means for storing at least two of the following: event information relating to events that affect the growth of plants grown in the greenhouse; environmental information including physical quantities related to the greenhouse environment; and growth information relating to the growth of plants grown in the greenhouse; a greenhouse operation analysis diagram creation means for creating, based on the information stored in the storage means, a greenhouse operation analysis diagram that simultaneously displays, on a common time axis, at least two of the following: an event information graph consisting of an axis indicating the occurrence or non-occurrence of the event and a time axis; an environmental information graph consisting of an axis indicating physical quantities related to the greenhouse environment and a time axis; and a growth information graph consisting of an axis indicating the growth amount of the plants and a time axis; and a greenhouse operation comparison assistance means for simultaneously displaying a plurality of different greenhouse operation analysis diagrams on the image display means. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-25246 [Patent Document 2] Patent Publication No. 2021-40491 Summary of the Invention [Problem to be solved by the invention]

[0005] The present invention aims to support the utilization of accumulated data. [Means for solving the problem]

[0006] The data utilization support system of the present invention includes multiple types of sensors that detect the situation at a manufacturing site, a log acquisition unit that acquires log data that is the detection result of the sensors, and a pre-processing unit that performs pre-processing according to the range of values ​​that the log data acquired by the log acquisition unit can take or the number of values ​​that it can take.

[0007] Preferably, the system further includes a visualization unit that integrates and visualizes the log data of the multiple sensors that has been preprocessed by the preprocessing unit, the preprocessing unit including a value conversion unit that converts the values ​​of the binary data acquired by the log acquisition unit, and the visualization unit visualizes the log data acquired by the log acquisition unit using the values ​​converted by the value conversion unit.

[0008] Preferably, the log acquisition unit acquires log data at different times for each type of sensor, the preprocessing unit includes a time series synchronization unit that aligns the time axes of the multiple types of log data acquired by the log acquisition unit, and the visualization unit integrates the multiple types of log data using the time axes aligned by the time series synchronization unit.

[0009] Preferably, the visualization unit creates a graph according to the values ​​of the multiple log data to be integrated, and the value conversion unit converts the binary data into a value that increases the difference between one value and the other value in the graph created by the visualization unit.

[0010] Preferably, the system has a significant point detection unit that detects a pattern different from a normal pattern from the log data acquired by the log acquisition unit, and a point movement unit that instantly moves the display position where the log data is displayed to the data position showing the different pattern detected by the significant point detection unit.

[0011] Preferably, the system further comprises a display unit for displaying a specific range of data including log data showing different patterns detected by the significant point detection unit.

[0012] Preferably, the system further includes a proposing unit that proposes relationships between the sensors, the proposing unit extracting log data from sensors of similar types among the sensors, or from other sensors that have detected a different pattern at a timing close to the time of the log data detected by the significant point detection unit, and the visualization unit provisionally displays the log data extracted by the proposing unit.

[0013] Preferably, the visualization unit displays log data of a sensor specified by a user, and if the specified sensor is a single sensor, the visualization unit displays status information of the sensor in association with the log data.

[0014] The data utilization support method according to the present invention includes a log acquisition step in which a computer acquires log data, which is the detection results of multiple types of sensors that detect the situation at a manufacturing site, and a preprocessing step in which the computer performs preprocessing according to the range of values ​​that the log data acquired by the log acquisition step can take or the number of values ​​that the log data can take.

[0015] The program of the present invention causes a computer to execute a log acquisition step of acquiring log data, which is the detection results of multiple types of sensors that detect the situation at the manufacturing site, and a preprocessing step of performing preprocessing according to the range of values ​​that the log data acquired by the log acquisition step can take or the number of values ​​that the log data can take. [Effects of the Invention]

[0016] According to the present invention, it is possible to support utilization of accumulated data. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a diagram illustrating an example of the overall configuration of a data utilization support system 1. FIG. [Figure 2] 2 is a diagram illustrating an example of a hardware configuration of a log management server 2. FIG. [Figure 3] 2 is a diagram illustrating an example of a functional configuration of a log management server 2. FIG. [Figure 4] 10 is a graph illustrating log data obtained by converting binary data values. [Figure 5] FIG. 1 illustrates significant data points. [Figure 6] 10 is a flowchart illustrating a process (S10) for visualizing log data by the data utilization support system 1. [Figure 7] 10 is a flowchart illustrating a significant point display process (S20) performed by the data utilization support system 1. [Figure 8] 10 is a diagram illustrating an example of a display of log data extracted by a proposing unit 260. FIG. [Figure 9] FIG. 10 is a diagram illustrating an example of the display of log data of the sensor 3 alone. [Figure 10] 10 is a graph illustrating log data in which values ​​of binary data are identified by regions. DETAILED DESCRIPTION OF THE INVENTION

[0018] First, the background to the invention will be explained. Various types of sensors are installed in manufacturing sites, and huge amounts of log data are collected and stored, but without specialized skills to analyze it, it is difficult to effectively utilize the log data. Specifically, because there are many types of sensors, the amount of data is large, and the data format varies depending on the sensor, it is difficult to analyze the correlation between sensors. For example, because the timing of log data transmission differs depending on the sensor, the time axis differs depending on the acquired log data, making it difficult to integrate the log data. Also, because the log data from the sensor is either analog data or binary data, when these data are displayed on the same graph, the binary data is buried in the analog data, making it difficult to understand the correlation between sensors. Here, binary data is data that has only two values, for example, a data format that has a value of either "0" or "1." On the other hand, analog data is data that has a continuous range of values, for example, temperature or measured values ​​of physical quantities. In view of the above, the present invention has made it possible to facilitate the analysis of correlations between sensors by preprocessing log data, thereby supporting the utilization of log data.

[0019] Hereinafter, embodiments of the present invention will be described with reference to the drawings, but the scope of the present invention is not limited to the illustrated examples. FIG. 1 is a diagram illustrating an example of the overall configuration of a data utilization support system 1. As illustrated in FIG. 1, the data utilization support system 1 includes a log management server 2, a plurality of sensors 3, and a client terminal 4. The sensor 3 is a device that is installed at the manufacturing site, detects the surrounding environment, equipment, and device status, and transmits the detection results to the log management server 2 at predetermined intervals. Hereinafter, the detection results transmitted by the sensor 3 to the log management server 2 will be referred to as log data. The predetermined intervals at which the sensor 3 transmits the log data to the log management server 2 differ for each sensor 3. The sensor 3 is, for example, a power meter, a temperature sensor, a humidity sensor, a barometric pressure sensor, a weight sensor, an optical sensor, etc. The log management server 2 is a computer terminal that performs preprocessing on log data acquired from multiple sensors 3 and displays the multiple pieces of preprocessed log data on the client terminal 4. The log management server 2 may be a cloud server. The client terminal 4 is a computer terminal used by a user, and requests the log management server 2 to view the log data of a specified sensor 3 .

[0020] FIG. 2 is a diagram illustrating an example of the hardware configuration of the log management server 2. As illustrated in FIG. 2, the log management server 2 includes a CPU 500, a memory 502, an HDD 504, a network interface 506 (network IF 506), a display device 508, and an input device 510, which are connected to each other via a bus 512. The CPU 500 is, for example, a central processing unit. The memory 502 is, for example, a volatile memory, and functions as a main storage device. The HDD 504 is, for example, a hard disk drive device, and serves as a non-volatile storage device for storing computer programs and other data files. The network IF 506 is an interface for wired or wireless communication, and realizes communication with the sensor 3, for example. The display device 508 is, for example, a liquid crystal display. The input device 510 is, for example, a keyboard and a mouse.

[0021] FIG. 3 is a diagram illustrating an example of the functional configuration of the log management server 2. As shown in FIG. 3, the log management server 2 of this example has a log management program 20 installed therein and also has a log data database 600 (log data DB 600) configured therein. The log management program 20 has a log acquisition unit 200, a preprocessing unit 210, a visualization unit 220, a significant point detection unit 230, a point movement unit 240, a time frame display unit 250, and a proposal unit 260, and the preprocessing unit 210 includes a value conversion unit 212 and a time series synchronization unit 214. Note that part or all of the log management program 20 may be realized by hardware such as an ASIC. The log management program 20 is stored on a recording medium such as a CD-ROM, and is installed into the log management server 2 via this recording medium.

[0022] In the log management program 20, the log acquisition unit 200 acquires log data that is the detection results from a plurality of different sensors 3 installed at the manufacturing site. For example, the log acquisition unit 200 acquires information on the amount of power, temperature and humidity, atmospheric pressure, and angle as log data. Because the timing at which the log data is transmitted differs for each sensor 3, the timestamps of the log data acquired by the log acquisition unit 200 do not match. Furthermore, the acquired log data has different data formats depending on the detection target of the sensor 3, and for example, analog data and binary data are mixed. The log acquisition unit 200 stores the acquired log data in the log data DB 600.

[0023] The preprocessing unit 210 performs preprocessing according to the range of values ​​that the log data acquired by the log acquisition unit 200 can take or the number of values ​​that can take. For example, the preprocessing unit 210 makes it easier to compare log data acquired in different data formats. Specifically, the value conversion unit 212 converts the value of the binary data acquired by the log acquisition unit 200. As illustrated in FIG. 4(a), if binary data values ​​are integrated into a graph of analog data without converting them, the binary data values ​​will be buried in the analog data, making it difficult to analyze the correlation between the binary data and the analog data. Therefore, the value converter 212 converts the binary data values ​​into values ​​that increase the difference between one value and the other value in the graph created by the visualization unit 220. For example, as illustrated in FIG. 4(b), the value converter 212 converts one value of the binary data into the minimum value that the graph created by the visualization unit 220 can take, and converts the other value into the maximum value that the graph created by the visualization unit 220 can take. In this way, the value converter 212 can prevent the binary data from being buried in the analog data by correlating the binary data values ​​with the values ​​of the graph based on the analog data. Furthermore, the value converter 212 may convert the value of the binary data by multiplying it by a predetermined coefficient. For example, if the binary data has a value of "0" or "1", the value converter 212 may multiply it by a predetermined coefficient of "10" and plot the value of the binary data as "0" or "10".

[0024] The time series synchronization unit 214 aligns multiple types of log data acquired at different times by the log acquisition unit 200 on a common time axis. Specifically, the log data has a value corresponding to time, and the time series synchronization unit 214 aligns the multiple types of log data to values ​​corresponding to the same timestamp. The visualization unit 220 displays the log data of the sensor 3 specified by the user on the client terminal 4. Specifically, the visualization unit 220 integrates and visualizes multiple pieces of log data that have been preprocessed for the log data of the specified sensor 3. More specifically, the visualization unit 220 displays the data acquired by the log acquisition unit 200 on the client terminal 4 using the values ​​converted by the value conversion unit 212. Furthermore, the visualization unit 220 integrates and displays multiple types of log data on the client terminal 4 using the time axis aligned by the time series synchronization unit 214. Furthermore, the visualization unit 220 displays the log data extracted by the proposal unit 260 on the client terminal 4 using a display method different from that used for the log data of the sensor 3 specified by the user.

[0025] The significant point detection unit 230 detects a pattern that differs from a normal pattern from the log data acquired by the log acquisition unit 200. A pattern that differs from a normal pattern indicates a change in trend, for example, as illustrated in Fig. 5, when binary data DEVICE_1 has a value of "0" or "1," and a value that was "1" changes to "0." Also, a different pattern refers to a case where the fluctuation in the value of sensor 3 that has analog data is larger than normal. The point moving unit 240 immediately moves the display position of the log data displayed on the client terminal 4 to a data position showing a different pattern detected by the significant point detection unit 230. The log data position where a different pattern is detected is called a significant point. Furthermore, the point moving unit 240 moves the display position of the log data from a specific significant point to the next or previous significant point in response to a user instruction.

[0026] The time frame display unit 250 displays a specific range of data including log data showing different patterns detected by the significant point detection unit 230. Specifically, the log data for a period range specified by the user (for example, several hours before and after a significant point, several days before and after a significant point, etc.) is displayed or output in CSV format. The time frame display unit 250 is an example of a display unit according to the present invention. The proposing unit 260 proposes associations between the sensors 3. Specifically, the proposing unit 260 extracts log data of sensors of similar types, or other sensors 3 that have detected significant points with timestamps that are the same as or close to the timestamps of significant points detected by the significant point detecting unit 230.

[0027] FIG. 6 is a flowchart illustrating the log data visualization process (S10) performed by the data utilization support system 1. 6, in step 100 (S100), the log acquisition unit 200 of the log management server 2 acquires log data from a plurality of installed sensors 3 at different timings for each sensor 3. The log acquisition unit 200 stores the acquired log data in the log data DB 600. In step 105 (S105), the user specifies, via the client terminal 4, the sensor 3 having the log data that the user wishes to view, and the log management server 2 extracts from the log data DB 600 the log data specified by the user. In step 110 (S110), the time series synchronization unit 214 aligns the log data having different timestamp values ​​for each sensor 3 on a common time axis.

[0028] In step 115 (S115), if the log data of the sensor 3 specified by the user contains a mixture of binary data and analog data, the visualization process of the log data (S10) proceeds to S120 (S115: Yes), and if the log data does not contain a mixture of binary data and analog data, it proceeds to S125 (S115: No). In step 120 (S120), the value converter 212 converts the value of the binary data according to the value of the analog data. Specifically, one value of the binary data is converted to the maximum value that the graph can take when the analog data is graphed, and the other value of the binary data is converted to the minimum value that the graph can take. In step 125 (S125), the visualization unit 220 graphs the log data of the sensor 3 specified by the user on the time axis matched by the time series synchronization unit 214, and displays the graph on the client terminal 4. If the log data to be graphed includes binary data, the visualization unit 220 plots the values ​​of the binary data converted by the value conversion unit 212.

[0029] FIG. 7 is a flowchart illustrating the significant point display process (S20) performed by the data utilization support system 1. As shown in FIG. 7, in step 200 (S200), the log acquisition unit 200 of the log management server 2 acquires log data and stores it in the log data DB 600. In step 205 (S205), the log management server 2 extracts from the log data DB 600 the log data specified by the user. In step 210 (S210), the significant point detection unit 230 searches for a pattern different from a normal pattern from the log data acquired by the log acquisition unit 200. If the significant point detection unit 230 detects a significant point (S210: Yes), the significant point display process (S20) proceeds to S215, and if no significant point is detected (S210: No), the process ends.

[0030] In step 215 (S215), the point moving section 240 immediately moves the display position to the log data position showing the different pattern detected by the significant point detecting section 230 in the log data displayed in chronological order. In step 220 (S220), the proposing unit 260 identifies the timestamp of the significant point detected by the significant point detecting unit 230, and searches the log data DB 600 for other log data having a significant point at the identified timestamp. In step 225 (S225), if there is a significant point at the timestamp identified from other log data (S225: Yes), the significant point display process (S20) proceeds to S230, and if there is no significant point (S225: No), the process ends. In step 230 (S230), the visualization unit 220 integrates the log data of other sensors 3 having significant points searched for by the suggestion unit 260 as related log data with the log data specified by the user and provisionally displays it on the client terminal 4, as illustrated in FIG.

[0031] As described above, the data utilization support system 1 according to this embodiment displays multiple log data on the same graph along a common time axis, making it easier to analyze correlations between log data. Furthermore, when analog log data and binary log data are displayed on the same graph, the values ​​of the binary data are converted according to the values ​​of the analog data and then displayed on the graph. This allows the two to be compared without the values ​​of the binary data being obscured by the analog data. Furthermore, the data utilization support system 1 displays potentially related log data other than the log data of the sensor 3 specified by the user, thereby enabling new correlations between sensors to be discovered. Therefore, the data utilization support system 1 performs preprocessing and suggestions for analyzing acquired log data, making it easier to analyze the log data, enabling efficient analysis and leading to the utilization of log data.

[0032] In the above embodiment, the case where log data from multiple sensors 3 is integrated and displayed has been described. However, this is not limiting. The visualization unit 220 may switch the display method of the log data depending on whether a single sensor 3 or multiple sensors 3 are specified by the user. Specifically, when a single sensor 3 is specified by the user, the visualization unit 220 displays status information of the specified sensor 3 in association with the log data. More specifically, the visualization unit 220 switches from the log data display method shown in FIG. 5 , which illustrates the display of log data from multiple sensors 3, to a display method shown in FIG. 9 , which displays additional information in addition to the measurement values ​​of the sensor 3. For example, the visualization unit 220 associates the identification information, measurement date and time, version information, status, measurement value type, measurement value, and network information (radio signal strength, IP address, SSID) of the sensor 3 and displays them on the client terminal 4. Since information about the sensor 3 other than the measurement date and time and measurement value is displayed, when an error is detected, it is possible to first infer the cause from this information before checking on-site.

[0033] In the above embodiment, the value converter 212 converts the value of the binary data so that the binary data is not buried in the analog data, but the present invention is not limited to this. For example, as shown in Fig. 10, it is also possible to distinguish areas of the graph according to the value of the binary data. This clarifies the significant points of the log data, which is binary data. [Explanation of symbols]

[0034] 1. Data utilization support system 2...Log management server 3...Sensor 4...Client terminal 20...Log management program 200...Log acquisition section 210...Pre-processing section 220...Visualization section 230...Significant point detection unit 240...Point movement section 250...Time frame display 260…Proposal Department

Claims

1. Multiple types of sensors that detect the situation at the manufacturing site, a log acquisition unit that acquires log data that is a detection result of the sensor; a pre-processing unit that performs pre-processing according to the range of values ​​that the log data acquired by the log acquisition unit can take or the number of values ​​that the log data can take; A data utilization support system with

2. a visualization unit that integrates and visualizes the log data of the plurality of sensors that have been preprocessed by the preprocessing unit; and The pre-treatment unit a value conversion unit that converts the value of the binary data acquired by the log acquisition unit; Including, The visualization unit visualizes the log data acquired by the log acquisition unit using the values ​​converted by the value conversion unit. The data utilization support system according to claim 1.

3. the log acquisition unit acquires log data at different times for each type of sensor, The pre-treatment unit a time series synchronization unit that synchronizes the time axes of the plurality of types of log data acquired by the log acquisition unit; Including, The visualization unit integrates multiple types of log data using the time axes aligned by the time series synchronization unit. The data utilization support system according to claim 2.

4. the visualization unit creates a graph according to values ​​of the plurality of log data to be integrated; The value conversion unit converts the binary data into a value that increases the difference between one value and the other value in the graph created by the visualization unit. The data utilization support system according to claim 2.

5. a significant point detection unit that detects patterns that are different from normal patterns from the log data acquired by the log acquisition unit; a point moving unit that immediately moves a display position where the log data is displayed to a data position showing a different pattern detected by the significant point detecting unit; The data utilization support system according to claim 2, comprising:

6. a display unit for displaying a specific range of data including log data showing different patterns detected by the significant point detection unit; The data utilization support system according to claim 5, further comprising:

7. a proposal unit for proposing associations between the sensors; and the suggestion unit extracts log data of a sensor of a similar type among the sensors or of another sensor in which a different pattern has been detected at a timing close to the time of the log data detected by the significant point detection unit, The visualization unit provisionally displays the log data extracted by the suggestion unit. The data utilization support system according to claim 5.

8. the visualization unit displays log data of a sensor designated by a user; When the designated sensor is a single sensor, the visualization unit displays the status information of the sensor in association with the log data. The data utilization support system according to claim 2.

9. a log acquisition step in which a computer acquires log data that is detection results from multiple types of sensors that detect conditions at the manufacturing site; a pre-processing step in which the computer performs pre-processing according to the range of values ​​that the log data acquired by the log acquisition step can take or the number of values ​​that the log data can take; A data utilization support method having the above.

10. a log acquisition step of acquiring log data that is detection results from a plurality of types of sensors that detect the situation at the manufacturing site; a pre-processing step of performing pre-processing according to the range of values ​​that the log data acquired by the log acquisition step can take or the number of values ​​that the log data can take; A program that causes a computer to execute the following.

Images