Self-learning simulation environments

Patent metadata

Field	Value
Publication number	US-10101244-B2
Application number	US-201514636135-A
Country	US
Kind code	B2
Filing date	Mar 2, 2015
Priority date	Mar 2, 2015
Publication date	Oct 16, 2018
Grant date	Oct 16, 2018

Abstract

Systems and methods may include receiving data of a physical system. The systems and methods may include determining whether an anomalous event has occurred based on the data. The systems and methods may include, in response to determining that the anomalous event has occurred, capturing data of the physical system and of an environment. The systems and methods may include generating a virtual representation of the physical system and the environment based on the captured data. The systems and methods may include repeatedly simulating data of the virtual representation by varying parameters of the captured data. The systems and methods may include determining which parameters are a root cause of the anomalous event based on the simulated performance data. The systems and methods may include determining particular values of the parameters that are likely the root cause of the anomalous event to use as a trigger for a proactive process.

First claim

What is claimed is:

1. A method comprising:
receiving performance data of a physical system;
determining whether an anomalous event has occurred in the physical system based on the performance data;
in response to determining that the anomalous event has occurred in the physical system, capturing data of the physical system and data of an environment of the physical system;
generating a first virtual representation of the physical system and the environment based on the captured data of the physical system and the captured data of the environment, wherein the first virtual representation comprises a plurality of actual parameters and at least one simulated parameter;
wherein values of the actual parameters are determined using the data of the physical system and the data of the environment of the physical system, and values of the at least one simulated parameter is determined using simulated data;
generating at least a second virtual representation that comprises the first virtual representation with the value of the at least one simulated parameter being varied from the first virtual representation;
using the first virtual representation and the at least a second virtual representation to determine a correlation between the at least one simulated parameter and the anomalous event;
determining which particular parameter of either the actual parameters or the at least one simulated parameter is likely a root cause of the anomalous event based on the correlation; and
determining particular values of the particular parameter that are likely the root cause of the anomalous event to use as alert triggers.

2. The method of claim 1 , further comprising:
determining characteristics of the anomalous event; and
accessing predefined logic indicating a potential cause of an event with characteristics similar to the characteristics of the anomalous event,
wherein generating the at least a second virtual representation comprises selectively varying the at least one simulated parameter that is associated with the potential cause of the event with characteristics similar to the characteristics of the anomalous event.

3. The method of claim 1 , further comprising:
selectively varying the values of the actual parameters of the captured data of the environment using the captured data of the physical system in an unvaried state.

4. The method of claim 1 , further comprising:
selectively varying the values of the actual parameters of the captured data of the physical system using the captured data of the environment in an unvaried state.

5. The method of claim 1 , wherein determining which particular parameter of either the actual parameters or the at least one simulated parameter is likely the root cause of the anomalous event comprises:
determining which particular parameter of the one or more parameters have statistically significant correlation to simulated performance data indicating that the anomalous event has occurred; and
determining that the particular parameter of the one or more parameters that has the statistically significant correlation to the simulated performance data indicating that the anomalous event has occurred is likely the root cause of the anomalous event.

6. The method of claim 1 ,
wherein the captured data of the physical system comprises data, which was transmitted by a plurality of components of the physical system, that was received within a predetermined period of time including a time of occurrence of the anomalous event, and
wherein the captured data of the environment comprises data, which was obtained by a plurality of sensors monitoring the environment, that was received within the predetermined period of time.

7. The method of claim 1 ,
wherein the particular values of the particular parameter that is likely the root cause of the anomalous event are values of the parameters that are within a range of values that are highly correlated to an occurrence of the anomalous event, and
further comprising issuing an alert indicating that the anomalous event is likely to occur again in response to determining that the particular parameter that is likely the root cause of the anomalous event has reached the particular values.

8. A system comprising:
a processing system configured to:
receive performance data of a physical system;
determine whether an anomalous event has occurred in the physical system based on the performance data;
in response to determining that the anomalous event has occurred in the physical system, capture data of the physical system and data of an environment of the physical system;
generate a virtual representation of the physical system and the environment based on the captured data of the physical system and the captured data of the environment;
repeatedly simulate performance data of the virtual representation by selectively varying one or more first parameters of the captured data of at least one of the physical system and the environment, while keeping one or more second parameters of the captured data of at least one of the physical system and the environment constant;
determine which parameters of the one or more first and second parameters are likely a root cause of the anomalous event based on the repeatedly simulated performance data of the virtual representation; and
determine particular values of the parameters that are likely the root cause of the anomalous event to use as a trigger for a proactive process to address the anomalous event.

9. The system according to claim 8 , wherein the processing system is further configured to:
determine characteristics of the anomalous event; and
access predefined logic indicating a potential cause of an event with characteristics similar to the characteristics of the anomalous event,
wherein, when selectively varying the one or more first parameters, the processing system is configured to:
initially vary parameters of the one or more first parameters that are associated with the potential cause of the event with characteristics similar to the characteristics of the anomalous event.

10. The system according to claim 8 , wherein, when selectively varying the one or more first parameters, the processing system is configured to:
selectively vary the one or more first parameters of the captured data of the environment using the captured data of the physical system in an unvaried state.

11. The system according to claim 8 , wherein, when selectively varying the one or more first parameters, the processing system is configured to:
selectively vary the one or more first parameters of the captured data of the physical system using the captured data of the environment in an unvaried state.

12. The system according to claim 8 , wherein, when determining which parameters of the one or more first and second parameters are likely the root cause of the anomalous event, the processing system is configured to:
aggregate the simulated performance data of the virtual representation and associated values of the one or more first and second parameters used to simulate the simulated performance data;
determine a correlation between each parameter of the one or more first parameters and the simulated performance data;
determine which parameters of the one or more first parameters have statistically significant correlation to simulated performance data indicating that the anomalous event has occurred; and
determine that the parameters of the one or more first parameters that have statistically significant correlation to the simulated performance data indicating that the anomalous event has occurred are likely the root cause of the anomalous event.

Assignees

• Mankovskii Serge
• Greenspan Steven L
• Danielson Debra J
• Parikh Prashant
• Velez Rojas Maria C
• Reno James D
• Ca Inc

Inventors

• Mankovskii Serge
• Greenspan Steven L
• Danielson Debra J
• Parikh Prashant
• Velez-Rojas Maria C
• Reno James D

Classifications

• G06F11/079

  Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36) · CPC title

• G06F11/00

  Error detection; Error correction; Monitoring (error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00) · CPC title

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• G05B23/0275

  Fault isolation and identification, e.g. classify fault; estimate cause or root of failure · CPC title

• H04L67/12

  specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks · CPC title