Self-Encoder Abnormality Detection for Correlated Sensor Time Series
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing abnormality detection systems, such as those described in Patent Literature 1, fail to accurately detect system-wide abnormalities in complex systems like manufacturing machinery and power plants due to neglecting the correlation between sensor values over time, resulting in insufficient detection accuracy.
Innovation Solution
An abnormality detection apparatus and method that utilizes a self-encoder to convert time series data from multiple sensors into multi-dimensional vector data, allowing for the identification of abnormal time periods by comparing input and output vector data, thereby enhancing detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a self-encoder is used to process sensor data as in Patent Literature 1, then noise processing is improved, but detection accuracy is insufficient because correlation between sensor values at different times is not considered
Solution Approach 1:
The patent transforms the self-encoder architecture to process multi-dimensional time series data by introducing a time dimension. The encoder receives sensor data sequences over time and outputs reconstructed sequences, enabling the model to learn temporal patterns while maintaining the noise filtering capability of the original self-encoder approach.
2Ease of operation
If individual sensor values are analyzed separately, then simple detection is achieved, but system-wide abnormalities are not detected
Solution Approach 1:
The patent merges multiple sensor data streams into a unified time series analysis framework. The self-encoder processes correlated sensor values simultaneously, combining information from multiple sensors to detect system-wide abnormalities that individual sensor analysis would miss, while maintaining operational simplicity through automated pattern recognition.
Data Source
AI summary
An abnormality detection apparatus (200) includes storage means (210) for storing a learned self-encoder (211) including predetermined number of two or more of elements as input layers, extraction means (220) for extracting a target data group of a predetermined period including data pieces from time series data measured by one or more sensors, the number of the data pieces being the predetermined number, conversion means (230) for converting the target data group into multi-dimensional vector data including the predetermined number of elements, identifying means (240) for identifying a time period in which there may be a cause of an abnormality from the predetermined period based on a difference between output vector data having the predetermined number of elements obtained by inputting the multi-dimensional vector data to the self-encoder (211) and the multi-dimensional vector data, and output means (250) for outputting abnormality detection information including the identified time period.