Method for Computer-Aided Analysis of an Automation System

Patent metadata

Field	Value
Publication number	US-2016110277-A1
Application number	US-201414515710-A
Country	US
Kind code	A1
Filing date	Oct 16, 2014
Priority date	Oct 16, 2014
Publication date	Apr 21, 2016
Grant date	—

Abstract

A method for computer-aided analysis of an automation system, where the automation system executes a number of jobs, each job being performed repetitively, wherein the execution durations of a respective job of the number of jobs for several repetitions of the respective job is determined to produce a plurality of execution durations, a statistical analysis on the plurality of execution durations is performed to produce at least one statistical quantity valid for the plurality of execution durations, and an action is performed for protecting the automation system and/or for generating a warning if a condition indicating an incorrect execution of the respective job is fulfilled for at least one statistical quantity.

First claim

What is claimed is:

1 . A method for computer-aided analysis of an automation system executing a number of jobs, each job being performed repetitively, the method comprising:
determining execution durations of a respective job of the number of jobs for several repetitions of the respective job being performed on the automation system to produce a plurality of execution durations of the automation system; performing a statistical analysis on the plurality of execution durations to produce at least one statistical quantity valid for the plurality of execution durations; performing an action for at least one of (i) protecting the automation system and (ii) generating a warning if a condition indicating an incorrect execution of the respective job is fulfilled for the at least one statistical quantity.

2 . The method according to claim 1 , wherein a statistical distribution of the plurality of execution durations is determined by statistical analysis, the statistical distribution being a frequency distribution of the plurality of execution durations or a probability distribution derived from the frequency distribution, at least statistical values with respect to the statistical distribution being derived as statistical quantities.

3 . The method according to claim 2 , wherein the at least one statistical value with respect to the statistical distribution comprise at least one of (i) a mean value, (ii) a standard deviation, (iii) a variance and (iv) at least one higher moment of the statistical distribution.

4 . The method according to claim 3 , wherein the condition for the at least one statistical quantity is fulfilled if at least one of the statistical values is outside a predetermined value range.

5 . The method according to claim 3 , wherein the condition for the at least one statistical quantity is fulfilled if at least one of the statistical values is greater than a predetermined threshold.

6 . The method according to claim 2 , further comprising:
determining a trend of at least one statistical value over consecutive time points of repetitive executions of the respective job; wherein the condition for the at least one statistical quantity is fulfilled if the trend of the at least one statistical value represents an increase or a decrease exceeding a predefined threshold.

7 . The method according to claim 3 , further comprising:
determining a trend of at least one statistical value over consecutive time points of repetitive executions of the respective job; wherein the condition for the at least one statistical quantity is fulfilled if the trend of the at least one statistical value represents an increase or a decrease exceeding a predefined threshold.

8 . The method according to claim 4 , further comprising:
determining a trend of at least one statistical value over consecutive time points of repetitive executions of the respective job; wherein the condition for the at least one statistical quantity is fulfilled if the trend of the at least one statistical value represents an increase or a decrease exceeding a predefined threshold.

9 . The method according to claim 1 , wherein the statistical analysis comprises fitting a function to the dependency of the plurality of execution durations from consecutive time points of the repetitive executions of the respective job; and
wherein the condition for the at least one statistical quantity is fulfilled if at least one parameter of the function fulfils a predetermined criterion.

10 . The method according to claim 9 , wherein the condition for the at least one statistical quantity is fulfilled if the trend of the plurality of execution durations over the consecutive time points according to the function represents an increase or decrease exceeding a preset threshold.

11 . The method according to claim 9 , wherein the function is a polynomial with time points of the repetitive executions of the respective job as variables and having at least one term of ascending degrees, each term comprising a coefficient; and
wherein the condition for the at least one statistical quantity is fulfilled if at least one coefficient of the polynomial lies in a predetermined value range.

12 . The method according to claim 1 , wherein the respective jobs of the number of jobs are automatically scheduled based on fulfillment of at least one logic condition.

13 . The method according to claim 1 , wherein each of the plurality of execution durations is measured by determining a time difference between generation of a message starting the respective job and generation of a message confirming that the respective job is finished, the time difference being derived from time stamps included in the messages.

14 . The method according to claim 1 , wherein the method is performed in one of (i) an automation system for manufacturing goods and (ii) an automation system for at least one of generating electric energy and producing electric energy.

15 . The method according to claim 1 , wherein the automation system is an assembly line.

16 . An apparatus for a computer-aided analysis of an automation system executing a number of jobs, each job being performed repetitively, the apparatus comprising:
means for determining the execution durations of a respective job of the number of jobs for several repetitions of the respective job to produce a plurality of execution durations; means for performing a statistical analysis on the plurality of execution durations to produce at least one statistical quantity valid for the plurality of execution durations; means for performing an action for at least one of (i) protecting the automation system and (ii) generating a warning if a condition indicating an incorrect execution of the respective job is fulfilled for the at least one statistical quantity.

17 . The apparatus according to claim 16 , wherein the apparatus is configured to:
determine a statistical distribution of the plurality of execution durations by statistical analysis, the statistical distribution being a frequency distribution of the plurality of execution durations or a probability distribution derived from the frequency distribution, at least statistical values with respect to the statistical distribution being derived as statistical quantities.

18 . An automation system, wherein the automation system is configured to execute a number of jobs during operation, each job being performed repetitively, the automation system comprising:
an apparatus for a computer-aided analysis of an automation system executing the number of jobs, each job being performed repetitively, the apparatus comprising:
means for determining the execution durations of a respective job of the number of jobs for several repetitions of the respective job to produce a plurality of execution durations of the automation system;
means for performing a statistical analysis on the plurality of execution durations to produce at least one statistical quantity valid for the plurality of execution durations;
means for performing an action for at least one of (i) protecting the automation system and (ii) generating a warning if a condition indicating an incorrect execution of the respective job is fulfilled for the at least one statistical quantity.

Assignees

• Siemens Ag

Inventors

• Edmondson Zachery
• Ludwig Hartmut
• Obradovic Dragan
• Szabo Andrei
• Wang Lingyun

Classifications

• G06F11/3452

  Performance evaluation by statistical analysis · CPC title

• G05B23/0232

  based on qualitative trend analysis, e.g. system evolution · CPC title

• G06F11/3409 Primary

  for performance assessment · CPC title

• G06F9/4881

  Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues · CPC title

• G06F2201/865

  Monitoring of software · CPC title