Facility State Change Detection Using Prior Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for monitoring facility states and detecting abnormalities require extensive data learning and parameter adjustment, leading to delayed detection and a need for specialized knowledge, resulting in inefficiencies and potential production losses.
Innovation Solution
A monitoring device that uses a data acquisition unit, state change detection unit, update unit, and information presentation unit to detect changes in facility states based on a preset prior distribution and run length probability distribution, allowing immediate operation and accurate detection without extensive learning or parameter adjustment, and enabling users to select appropriate prior distributions through a user-friendly interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a model with high accuracy is built by learning from a large amount of data and adjusting parameters, then measurement precision is improved, but loss of time increases due to several months required for system import
Solution Approach 1:
The patent applies preliminary action by pre-setting prior distribution parameters that represent initial knowledge about facility state changes. Instead of requiring extensive data learning from scratch, the system starts with pre-configured probability distributions for state changes, which are then updated as data is collected. This allows the system to begin operation immediately with reasonable detection accuracy rather than requiring months of data collection and parameter adjustment.
2Reliability
If physical quantities are monitored by sensors to detect abnormalities, then reliability is improved, but loss of time increases due to time lag from when change occurs until significant change appears in monitored physical quantity
Solution Approach 1:
The patent introduces an intermediary statistical model (prior distribution and probability calculation mechanism) between the raw sensor data and the abnormality detection decision. Instead of directly thresholding sensor readings, the system uses probability distributions to model expected state changes and compares actual observations against these models. This intermediary layer enables earlier detection of subtle changes before they become significant abnormalities, reducing the detection time lag while maintaining reliability.
3Productivity
If existing monitoring systems are implemented, then productivity is improved through preventive maintenance, but device complexity increases due to requirement of highly specialized knowledge for sensing procedures
Solution Approach 1:
The patent applies self-service by designing a system that automatically performs the complex statistical analysis and parameter updating without requiring specialized human expertise. The prior distribution parameters are automatically updated as data is collected, and the system self-adjusts to changing facility conditions. This automation eliminates the need for users to possess specialized statistical knowledge while maintaining the sophisticated analysis required for effective preventive maintenance, thereby improving productivity without increasing operational complexity.
Data Source
AI summary
A state change detection unit obtains the data generation probability on the basis of the values of observation data and the value of a parameter of a prior distribution, obtains, on the basis of the data generation probability, a run length probability distribution of the time-series observation data acquired up to the current time point as a condition, and detects a change in the state of a facility on the basis of the run length probability distribution. Furthermore, an update unit updates the value of the parameter of the prior distribution using the values of the observation data, to generate the prior distribution to be used for calculating the data generation probability at a next time point.