Monitoring control system and control method

• US 9,621,966 B2
• Filed: 06/26/2014
• Issued: 04/11/2017
• Est. Priority Date: 06/28/2013
• Status: Active Grant

• Current Assignee: Toshiba Digital Solutions Corporation (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation), Toshiba Solutions Corporation (Toshiba Corporation)
• Inventors: Yoshimoto, Takehiro, Fukushima, Nobuyuki, Inaba, Masumi, Fujita, Shinichi
• Primary Examiner(s): WANG, TED M
• Application Number:
  US14/901,234
• Publication Number:
  US 20160227299A1
• Time in Patent Office: 1,020 Days
• CPC Class Codes: G05B 15/02 electric
• CPC Class Codes: G05B 2219/2642 Domotique, domestic, home control, automation, smart house
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H04Q 9/02 Automatically-operated arrangements

Abstract

A monitoring control system sequentially processing data collected from various data sources to find a possible phenomenon, producing an appropriate action for the found change or phenomenon, and outputting control information for causing a system to perform a notification or predetermined control. The monitoring control system includes a phenomenon detecting section configured to detect a phenomenon leading to a particular result from contiguous data received from data sources serving as data transmission element based on detecting rule definition data previously specifying a change attribute of contiguous time-series data pieces for each phenomenon leading to a particular result, and a determination section configured to extract action information associated with the particular result detected by the phenomenon detecting section to produce an event for a predetermined system based on action definition data previously specifying action information for each particular result.

Claims

1. A monitoring control system configured to detect one of phenomena each leading to a particular result based on at least one data piece received from a single or a plurality of data sources each serving as a data transmission element and to output action information to a predetermined system for the detected particular result, comprising:

a phenomenon detecting section configured to detect the phenomenon leading to the particular result from contiguous time-series data pieces received from the data source based on detecting rule definition data previously specifying a change attribute of the contiguous time-series data pieces for each of the phenomena leading to the particular result, and

a determination section configured to extract the action information associated with the particular result detected by the phenomenon detecting section to produce an event for the predetermined system based on action definition data previously specifying the action information for each of the particular results.

2. The monitoring control system according to claim 1, wherein the detecting rule definition data includes, as the change attribute, a predetermined chronological sequence of the contiguous time-series data pieces for each of the phenomena leading to the particular result, and

the phenomenon detecting section is configured to match a chronological sequence of the contiguous time-series data pieces received from the data source with the predetermined sequence in the detecting rule definition data to detect the phenomenon leading to the particular result.

3. The monitoring control system according to claim 1, further comprising a statistical analysis processing section configured to perform statistical processing of the contiguous time-series data pieces received from the data source during a fixed time period to output a statistical result,

wherein the detecting rule definition data includes, as the change attribute, a relationship between the statistical result and a threshold value for each of the phenomena leading to the particular result, and

the phenomenon detecting section is configured to compare the statistical result output from the statistical analysis processing section with the threshold value to detect the phenomenon leading to the particular result.

4. The monitoring control system according to claim 3, further comprising:

a storage section configured to store feedback information obtained from the predetermined system in response to the action information output to the predetermined system; and

a statistical analysis learning section configured to calculate a learning value of a predetermined parameter for use in calculating the statistical analysis based on the feedback information.

5. The monitoring control system according to claim 1, further comprising:

a storage section configured to store feedback information obtained from the predetermined system in response to the action information output to the predetermined system; and

a supplementary information learning section configured to change predetermined supplementary information based on the feedback information, the supplementary information being added to the action information and relating to the detected phenomenon.

7. The monitoring control system according to claim 2, further comprising a statistical analysis processing section configured to perform statistical processing of the contiguous time-series data pieces received from the data source during a fixed time period to output a statistical result,

wherein the detecting rule definition data includes, as the change attribute, a relationship between the statistical result and a threshold value for each of the phenomena leading to the particular result, and

the phenomenon detecting section is configured to compare the statistical result output from the statistical analysis processing section with the threshold value to detect the phenomenon leading to the particular result.

6. A method of controlling a monitoring control system configured to detect one of phenomena each leading to a particular result based on at least one data piece received from a single or a plurality of data sources each serving as a data transmission element and to output action information to a predetermined system for the detected particular result, comprising the steps of:

detecting the phenomenon leading to the particular result from contiguous time-series data pieces received from the data source based on detecting rule definition data previously specifying a change attribute of the contiguous time-series data pieces for each of the phenomena leading to the particular result, and

extracting the action information associated with the particular result detected by the step of the phenomenon detecting to produce an event for the predetermined system based on action definition data previously specifying the action information for each of the particular results.

8. The method of controlling the monitoring control system according to claim 6, further comprising:

storing feedback information obtained from the predetermined system in response to the action information output to the predetermined system; and

calculating a learning value of a predetermined parameter for use in calculating a statistical analysis based on the feedback information.