Computer system and method for monitoring the status of a technical system

Patent metadata

Field	Value
Publication number	US-11238371-B2
Application number	US-201916441028-A
Country	US
Kind code	B2
Filing date	Jun 14, 2019
Priority date	Dec 14, 2016
Publication date	Feb 1, 2022
Grant date	Feb 1, 2022

Abstract

A computer system can be configured to: receive, in a low-precision mode, first status data generated by one or more sensors, the first status data reflecting technical parameters of a technical system, the first status data exhibiting a first precision level; apply a low-precision machine learning model to analyze the first status data for one or more indicators of an abnormal technical status, the machine learning model having been trained with data exhibiting the first precision level; send, based on an abnormal technical status being indicated, instructions for the one or more sensors to generate second status data exhibiting a second precision level, the second precision level being associated with greater accuracy than the first precision level; receive the second status data exhibiting the second precision level based on the sent instructions; providing the second status data to a data analyzer.

First claim

We claim:

1. A computer system for monitoring a technical status of a technical system, the computer system comprising:
an interface module configured to:
receive, by the computer system while operated in a low-precision mode, status data generated by one or more sensors recorded by a monitoring system while operated in the low-precision mode, the status data reflecting technical parameters of the technical system, the low-precision mode being defined by the received status data complying with a first precision level associated with low-precision parameters including at least one of: a number of sensors associated with the low-precision mode for providing status data in the low-precision mode, a sampling frequency associated with the low-precision mode for providing status data in the low-precision mode, a transmission frequency associated with the low-precision mode for providing status data in the low-precision mode, and one of a data pre-processing or data aggregation level associated with status data in the low-precision mode;
send one or more instructions to the one or more sensors of the monitoring system, the one or more instructions being configured to cause the one or more sensors to generate further status data, the further status data complying with a second precision level, the second precision level being associated with changes applied to a selection of the low-precision parameters in response to detecting an abnormality indicator, the second precision level being associated with higher data accuracy than the first precision level; and
receive the further status data from the monitoring system complying with the second precision level and to provide the further status data to a data analyzer of the computer system;

a machine learning module configured to apply a first machine learning model to the received status data, the first model having been trained on training data complying with the first precision level and being used to analyze the received status data for one or more indicators of an abnormal technical status to detect the abnormality indicator; and
a command generator module configured to generate the one or more instructions for the one or more sensors when the abnormality indicator being detected,
wherein the data analyzer comprises a second machine learning model having been trained on training data complying with the second precision level, the first machine learning model and the second machine learning model comprising predictive machine learning models based on a predictive classification algorithm or a predictive regression algorithm.

2. The system of claim 1 , wherein the data analyzer comprises a data streamer configured to stream the second status data to a human-machine interface.

3. The system of claim 2 , wherein the data analyzer comprises a data storage component to store the received second status data.

4. A monitoring system for monitoring the technical status of a technical system, the monitoring system comprising:
one or more sensors configured to generate, in a low precision mode on the monitoring system, sensor data reflecting technical parameters of the technical system, the low precision mode being defined by status data generated from the one or more sensors complying with a first precision level associated with low precision parameters including at least one of: a number of sensors associated with the low precision mode for providing status data in the low precision mode, a sampling frequency associated with the low precision mode for providing status data in the low precision mode, a transmission frequency associated with the low precision mode for providing status data in the low precision mode, and one of a data pre-processing or data aggregation level associated with status data in the low precision mode, the one or more sensors configured to change, in response to received instructions, at least one low precision parameter to generate further status data complying with a second precision level; and
a communication module configured to:
provide the status data generated in the low precision mode to a first machine learning model of a computer system trained to identify one or more indicators of an abnormal technical status of the technical system;
receive, from the computer system, instructions for the one or more sensors to generate sensor data for the further status data complying with the second precision level if the provided status data indicates an abnormal technical status, the second precision level being associated with changes of a selection of the low precision parameters in response to a detection of an abnormal status of the technical system, the second precision level being associated with higher data accuracy than the first precision level; and
provide the further status data to a second machine learning model of the computer system trained on training data complying with the second precision level.

5. The monitoring system of claim 4 , wherein the change of the at least one low precision parameter includes any one or a combination of the following options:
the further status data includes additional status data from additional sensors, the additional status data not being included in the low precision mode;
the further status data is sent at a higher frequency to the computer system than the status data received in the low precision mode; and
the further status data is associated with a lower data pre-processing or data aggregation level than the status data sent in the low precision mode.

6. A computer-implemented method for monitoring the technical status of a technical system, the method comprising:
receiving, by a computer system while operated in a low-precision mode, status data generated by or more sensors recorded by a monitoring system while operated in the low-precision mode, the status data reflecting technical parameters of the technical system, the low-precision mode being defined by the received status data complying with a first precision level associated with low-precision parameters including at least one of: a number of sensors associated with the low-precision mode for providing status data in the low-precision mode, a sampling frequency associated with the low-precision mode for providing status data in the low-precision mode, a transmission frequency associated with the low-precision mode for providing status data in the low-precision mode, and one of a data pre-processing or data aggregation level associated with status data in the low-precision mode;
applying a first machine learning model to the received status data, the first machine learning model having been trained on training data complying with the first precision level to analyze the received status data for one or more indicators of an abnormal technical status;
if an abnormality indicator is detected, sending instructions to one or more sensors of the monitoring system to generate sensor data for further status data complying with a second precision level, the second precision level being associated with higher data accuracy than the first precision level, the second precision level being associated with changes applied to a selection of the low-precision parameters in response to the detection of the abnormality indicator;
in response to sending instructions, receiving the further status data complying with the second precision level from the monitoring system;
providing the further status data to a data analyzer having a second machine learning model trained on training data complying with the second precision level,
wherein the first machine learning model and the second machine learning model comprise predictive machine learning models based on a predictive classification algorithm or a prediction regression algorithm.

Assignees

• Abb Schweiz Ag

Inventors

• Kloepper Benjamin
• Schmidt Benedikt
• Ouertani Mohamed-Zied

Classifications

• H04W4/38

  for collecting sensor information · CPC title

• G05B23/0221 Primary

  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 title

• G06N20/00 Primary

  Machine learning · CPC title