Anomaly Detection System Using Statistical Deviation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Control systems often receive secondary induced anomalous value signals before detecting the signal representing the origin of the anomaly, leading to incorrect identification of the anomaly's origin.
Innovation Solution
A system with a setting unit to define normal ranges for monitoring target data, a determination unit to identify deviations, and a detection unit to determine the start time of anomalies based on the degree of deviation from mean learning data, accurately identifying when a signal begins to show anomalous behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional anomaly detection methods are used to identify anomaly-origin signals, then the system can detect anomalies, but the system may incorrectly identify secondary induced signals as the origin due to time delays
Solution Approach 1:
The system performs preliminary learning of normal signal patterns before actual anomaly detection occurs. By establishing a baseline of normal behavior through the learning unit, the system can compare subsequent signals against this pre-established norm, enabling more accurate identification of the true anomaly origin even when time delays cause secondary signals to arrive first.
Solution Approach 2:
The patent replaces conventional mechanical/time-based anomaly origin identification with a data-driven statistical approach. Instead of relying on signal arrival time or simple threshold detection, the system uses the detection unit to calculate deviation degrees based on learned normal patterns, substituting temporal precedence with statistical anomaly measurement to identify the true origin signal.
2Measurement precision
If the system waits for anomaly signals to be detected, then detection accuracy can be maintained, but the response time is delayed
Solution Approach 1:
The learning unit performs preliminary action by continuously learning and storing normal signal patterns before anomalies occur. This pre-prepared knowledge base enables the detection unit to immediately compare incoming signals against established norms, achieving both fast detection speed and high accuracy without waiting for anomaly confirmation.
Solution Approach 2:
The system performs self-service by automatically learning normal patterns from historical data and using this self-acquired knowledge for real-time anomaly detection. The learning unit continuously updates the normal pattern database, enabling the system to adapt and improve its detection capabilities without external intervention, thus maintaining both speed and accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing system includes: a setting unit to set a normal range showing a range of normal values of monitoring target data of time series signals by defining an upper limit and a lower limit; a determination unit to determine whether the monitoring target data is out of the normal range or not and to feed the determined time which is output in case of deviation and which is judged to be the time for the monitoring target data to turn to be out of the normal range; and a detection unit to determine the start time that is before the determined time entering from the determination unit and which is the time for the monitoring target data to start to show an anomaly on the basis of the degree of deviation showing a deviation of the monitoring target data from a mean of multiple learning data which consist of normal value signals from among already-acquired monitoring target data. This enables the system to more accurately determine the time for the signal to start to show an anomaly.