Monitoring apparatus, monitoring method, computer program product, and model training apparatus

Patent metadata

Field	Value
Publication number	US-11740613-B2
Application number	US-202117185882-A
Country	US
Kind code	B2
Filing date	Feb 25, 2021
Priority date	May 12, 2020
Publication date	Aug 29, 2023
Grant date	Aug 29, 2023

Abstract

According to an embodiment, a monitoring apparatus configured to generate time-series predicted data based on time-series measured data and a prediction model that generates predicted data including one or more predicted values predicted to be output from one or more sensors; and generate, for a first sensor among the one or more sensors, a displayed image including a measured value graph representing a temporal change in a measured value included in the time-series measured data in a second period after a first period, a predicted value graph representing a temporal change in a predicted value included in time-series predicted data in the second period, past distribution information representing a distribution of a measured value in the first period, and measurement distribution information representing a distribution of the measured value included in the time-series measured data in the second period.

First claim

What is claimed is:

1. A monitoring apparatus comprising:
a hardware processor configured to:
acquire time-series measured data including one or more measured values output from one or more sensors installed in a system to be monitored;
acquire a prediction model that generates predicted data including one or more predicted values predicted to be output from the one or more sensors at a first time based on the measured data at or before first time;
acquire, as a past data, data including past information on a distribution representing a distribution of a measured value in a first period, with respect to time-series data including one or more measured values output from the one or more sensors before the time-series measured data in a time direction, the first period being a period of training data used to train the prediction model;
generate time-series predicted data based on the time-series measured data and the prediction model;
generate, for a first sensor among the one or more sensors, a displayed image including a measured value graph representing a temporal change in a measured value included in the time-series measured data in a second period that is a period after the first period, a predicted value graph representing a temporal change in a predicted value included in the time-series predicted data in the second period, the past distribution in the first period and a measurement distribution representing a distribution of the measured value included in the time-series measured data in the second period; and
cause a monitoring device to display the displayed image.

2. The apparatus according to claim 1 , wherein the hardware processor is further configured to calculate, for each of the one or more sensors, a similarity degree between a distribution of a measured value included in the past data in the first period and a distribution of a measured value included in the measured data in the second period, wherein
the hardware processor is configured to generate the displayed image including the similarity degree for each of the one or more sensors.

3. The apparatus according to claim 2 , wherein
the hardware processor is further configured to:
calculate, for each of the one or more sensors, a difference between measured value included in the measured data at the first time and a predicted value included in the predicted data at the first time and detect, as a target sensor, a sensor for which the difference is larger than a preset first threshold or smaller than a present second threshold among the one or more sensors; and
calculate a first score having a larger value as the similarity degree is higher for the target sensor, and
generate the displayed image including at least either the similarity degree or the first score for the target sensor.

4. The apparatus according to claim 3 , wherein
the system includes a plurality of sensors installed therein, and
the hardware processor is further configured to:
acquire correlation information representing an intensity of a correlation for each pair of two sensors included in the plurality of sensors installed in the system; and
identify, for the target sensor, a related sensor for which the intensity of the correlation with the target sensor is equal to or more than a prescribed value among the plurality of sensors from the correlation information, and
calculate, for the target sensor, a second score having a value larger as the similarity degree for the target sensor is higher and larger as the similarity degree for the related sensor is higher,
generate the displayed image including at least either the second score for the target sensor or the similarity degree for the related sensor with respect to the target sensor.

5. The apparatus according to claim 4 , wherein the hardware processor is configured to generate the displayed image including at least one of information indicating whether related maintenance is performed, a time at which the maintenance is performed, and contents of the maintenance, for at least one of a sensor for which the similarity degree is lower than a prescribed value, a sensor for which the similarity degree for the related sensor is lower than a prescribed value, a sensor for which the first score is smaller than a prescribed value, a sensor for which the second score is smaller than a prescribed value, and the related sensor to a sensor for which the first score or the second score is smaller than a prescribed value among the plurality of sensors.

6. The apparatus according to claim 2 , wherein
the hardware processor is further configured to:
calculate, as a shift amount, a total number of sensors for each of which the similarity degree is lower than a preset third threshold among the one or more sensors or a total of difference degrees for the one or more sensors, the difference degrees each having a smaller value as the similarity degree is higher, and
generate the displayed image including the shift amount.

7. The apparatus according to claim 6 , wherein
the hardware processor is further configured to:
detect that the shift amount is larger than a preset fourth threshold, and
generate the displayed image including information indicating that the shift amount is larger than the fourth threshold when the shift amount is larger than the fourth threshold.

8. The apparatus according to claim 4 , wherein the hardware processor is further configured to perform control to make the first threshold larger and make the second threshold smaller as any one of the similarity degree, the first score, and the second score is smaller.

9. The apparatus according to claim 4 , wherein the hardware processor is further configured to classify each of the one or more sensors into at least four states based on the similarity degree and the similarity degree for the related sensor.

10. The apparatus according to claim 9 , wherein the hardware processor is configured to:
generate a layout image representing a layout of each of the plurality of sensors in the system, and
changes at least one of information and a color representing each sensor of the plurality of sensors displayed on the layout image in accordance with a state into which the sensor is classified.

11. The apparatus according to claim 9 , wherein the hardware processor is configured to display a target sensor list including identification information identifying the target sensor among the plurality of sensors and change a color of each piece of identification information included in the target sensor list in accordance with a state into which the corresponding target sensor is classified.

12. The apparatus according to claim 1 , wherein the hardware processor is configured to:
acquire, as the past data, data including time-series data including one or more measured values output from the one or more sensors before the timeseries measured data in the time direction; and
generate the display image further including a past value graph representing a temporal change in a measured value included in the time-series past data in the first period, for the first sensor.

13. The apparatus according to claim 1 , wherein the hardware processor is configured to:
acquire, as the past data, data including a maximum value and a minimum value in time-series data including one or more measured values output from the one or more sensors before the time-series measured data in the time direction; and
generate the displayed image further including a maximum value straight line representing the maximum value or a value obtained by adding a prescribed value to the maximum value and a minimum value straight line represent

Assignees

• Toshiba Kk
• Toshiba Energy Systems & Solutions Corp

Inventors

• Taguchi Yasunori
• Nakata Kouta
• Naito Susumu
• Kato Yuichi
• Aoki Toshio
• Tominaga Shinya
• Nagura Isaku
• Miyake Ryota
• Miyamoto Chikashi

Classifications

• G05B19/4183 Primary

  characterised by data acquisition, e.g. workpiece identification · CPC title

• G06F17/18

  for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title

• G06N20/00

  Machine learning · CPC title

• G05B23/0254 Primary

  based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks · CPC title

• G05B23/0297

  Reconfiguration of monitoring system, e.g. use of virtual sensors; change monitoring method as a response to monitoring results · CPC title