Facility Monitoring Threshold Setting for Neural Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manufacturing facilities face challenges in predicting failures and determining the remaining useful life, leading to costly and time-consuming expert analysis, and there is a need for a technology to diagnose failures and predict anomalies using a neural network model.
Innovation Solution
A method for setting a model threshold value in a facility monitoring system that involves acquiring sensor data, extracting feature values, inputting them into a trained neural network model, and comparing the input and output data to set a threshold value for anomaly detection, using techniques such as autoencoders and machine learning to determine device state indices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If expert analysis is used to determine facility failure causes, then diagnostic accuracy is improved, but cost and time consumption increase
Solution Approach 1:
The patent replaces the mechanical expert analysis system with an automated neural network model that processes sensor data to diagnose facility failures. The system uses machine learning algorithms to substitute human expert judgment, achieving rapid automated diagnosis without requiring manual expert intervention while maintaining diagnostic accuracy through trained models.
Solution Approach 2:
The facility monitoring system performs self-diagnosis through automated neural network analysis of sensor data. The system serves itself by automatically detecting anomalies, identifying failure causes, and generating diagnostic results without external expert intervention, thereby reducing both time consumption and operational costs.
2Measurement precision
If expert analysis is used to determine facility failure causes, then diagnostic accuracy is improved, but cost increases
Solution Approach 1:
The patent replaces expensive expert analysis services with an automated neural network system that processes facility sensor data. This substitution eliminates the need to pay expert consultants while maintaining diagnostic accuracy through machine learning models trained on historical failure data.
Solution Approach 2:
The system performs self-diagnosis through automated neural network analysis, eliminating dependency on external experts and reducing operational costs. The facility monitoring system serves itself by automatically identifying failure causes through trained algorithms.
3Productivity
If neural network model is used for anomaly detection, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent segments the facility monitoring system into modular components: sensor data acquisition modules, feature extraction modules, neural network analysis modules, and anomaly detection modules. This segmentation allows the complex neural network functionality to be distributed across independent modules, making the system more manageable and easier to implement while maintaining high productivity.
Data Source
AI summary
An exemplary embodiment of the present disclosure discloses a method of setting a model threshold value for detecting an anomaly of a facility monitoring system, the method including: acquiring sensor data output from each sensor; extracting a feature value for the sensor data of each sensor; acquiring output data by inputting input data including the extracted feature value to a trained neural network model; and comparing the input data and the output data and setting a threshold value for detecting an anomaly based on a calculated comparison result value.