Anomaly Detection Control Using Compressed Feature Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly predictive diagnostic devices face challenges in collecting input data in milliseconds or microseconds due to multidimensional sensor data transmission over communication networks, limiting diagnosis to slow degradation anomalies.
Innovation Solution
A control device and method that generate feature amounts for anomaly detection using machine learning, with data compression and asynchronous processing, allowing for faster anomaly detection without waiting for response, enabling monitoring of control targets in shorter cycles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multidimensional sensor data is transmitted via communication network, then anomaly predictive diagnosis can be performed, but data collection cycle cannot be shortened to milliseconds or microseconds
Solution Approach 1:
The patent segments the anomaly detection process into two parts: (1) data collection and feature amount generation performed locally by the control device at high speed, and (2) anomaly detection performed by the anomaly predictive diagnostic device. This segmentation allows the control device to collect data in milliseconds without being bottlenecked by network transmission to external diagnostic devices.
Solution Approach 2:
The control device acts as an intermediary that collects multidimensional sensor data, generates feature amounts, and performs preliminary processing before sending results to the anomaly predictive diagnostic device. This intermediary role enables high-speed local data collection while still utilizing external diagnostic capabilities.
2Productivity
If data compression is performed on feature amounts, then data transmission efficiency improves, but processing complexity increases
Solution Approach 1:
The patent changes the parameter representation by converting raw sensor data into compressed feature amounts that capture essential information in a more compact form. This parameter transformation enables efficient data transmission and storage while reducing the dimensionality of the data processed by the anomaly detection algorithm.
3Measurement precision
If anomaly detection is performed synchronously with data collection, then detection accuracy is maintained, but monitoring cycle time increases
Solution Approach 1:
The control device performs preliminary actions by collecting data, generating feature amounts, and compressing data in advance before the anomaly detection is executed. This preliminary processing allows the anomaly detection to be performed quickly without waiting for complete data preparation, thereby reducing the overall monitoring cycle time while maintaining detection accuracy.
Data Source
AI summary
A control device includes feature amount generating means for generating a feature amount suitable for detecting an anomaly that occurs in a control target from data that relates to the control target, machine learning means for carrying out machine learning using the feature amount generated by the feature amount generating means, anomaly detecting means for detecting the anomaly, based on the feature amount generated by the feature amount generating means and an anomaly detection parameter determined based on a learning result of the machine learning and used in detection of the anomaly that occurs in the control target, instructing means for instructing the anomaly detecting means to perform detection of the anomaly, and data compressing means for data-compressing the feature amount generated by the feature amount generating means, and providing the data-compressed feature amount to the machine learning means and the anomaly detecting means.