Automation Anomaly Cause Analysis Using Event Interval Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly detection methods in automation systems do not automatically recognize the underlying causes of anomaly events, limiting their effectiveness in preventing damage to the system.
Innovation Solution
A method that evaluates data sets from automation systems to determine intervals between anomaly events, using Fourier transformations, autocorrelation, and cross-correlation to identify repeating process influences and their causes, allowing for timely intervention and control adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If anomaly detection methods are implemented in automation systems, then anomaly events can be detected, but the underlying causes of these anomalies cannot be automatically recognized
Solution Approach 1:
The patent segments the anomaly analysis process into distinct phases: initial anomaly detection in a first data set, interval determination between anomaly events, and subsequent searching for repeating causes in a second data set. This segmentation allows the system to methodically progress from detection to cause identification without overwhelming computational requirements.
Solution Approach 2:
The patent performs preliminary actions by first detecting anomaly events and determining their intervals before searching for underlying causes. This preliminary characterization of anomaly patterns enables more efficient and targeted cause identification in subsequent analysis steps, rather than attempting to identify causes simultaneously with detection.
2Loss of information
If multiple data sets are analyzed to identify repeating process influences, then cause recognition is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary characterization of anomaly patterns by detecting anomaly events and determining their intervals before conducting the more computationally intensive search for repeating causes in the second data set. This preliminary step reduces the search space and enables more efficient subsequent analysis.
Solution Approach 2:
The patent employs signal processing techniques such as Fourier transformations, autocorrelation, and cross-correlation to substitute complex mechanical or manual analysis methods. These mathematical transformations efficiently identify repeating patterns and causal relationships in the data without requiring exhaustive computational searches.
3Measurement precision
If Fourier transformations and correlation analyses are used to identify repeating patterns, then cause identification accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary anomaly detection and interval determination before applying computationally intensive Fourier transformations and correlation analyses. This preliminary step prepares the data and defines the search parameters, enabling more efficient execution of the advanced signal processing techniques.
Solution Approach 2:
The patent segments the data analysis into distinct phases: anomaly detection, interval determination, and cause identification using signal processing. This segmentation allows each phase to be optimized independently, with the first two phases preparing data that reduces the computational burden of the final cause identification phase.
Data Source
AI summary
The invention relates to a method for evaluating at least one data set of at least one component of an automation system, characterized by the incorporation and/or execution of the following steps: determining an interval between two anomalous events in a first data set, said first data set comprising data based on at least one first component of the automation system, and determining repeating events which are spaced according to the interval in a second data set, said second data set comprising at least one second component of the automation system.