Diagnostic System Rule Update via Multi-Device Data Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional diagnostic systems face challenges in updating diagnostic rules effectively, especially when the probability of abnormal conditions in diagnostic objects is low, and detecting abnormalities during normal operation time is hindered by biased data due to usage modes or environments.
Innovation Solution
A diagnostic system where multiple diagnostic devices communicate to share diagnostic case data and update diagnostic rules, incorporating an abnormal-time estimation unit to detect time series change points or outliers, enabling the generation of new diagnostic rules based on collective data for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If diagnostic case data is collected from a single diagnostic object, then the data collection process is simple, but the diagnostic rule cannot be updated effectively when abnormal conditions occur rarely
Solution Approach 1:
The patent merges data collection from multiple diagnostic objects into a unified system. Multiple diagnostic devices connect to a server that aggregates diagnostic case data from various sources, enabling effective diagnostic rule updates even when individual objects exhibit rare abnormal conditions. This combining approach ensures sufficient training data availability while maintaining systematic organization.
Solution Approach 2:
The diagnostic system is designed with universal functionality to collect and process diagnostic data from multiple different diagnostic objects. The server can handle diverse data types and formats from various sources, making the system adaptable to different applications while maintaining a consistent diagnostic rule update mechanism.
2Measurement precision
If data is collected during normal operation time, then continuous monitoring is achieved, but the data becomes biased due to usage modes or environment
Solution Approach 1:
The system performs preliminary classification of collected data into normal and abnormal categories. By pre-organizing data and identifying abnormal cases as they occur, the system prepares representative training data without requiring extended collection periods. This preliminary sorting enables efficient diagnostic rule generation while maintaining data quality.
3Reliability
If multiple diagnostic devices are connected to share data, then diagnostic accuracy is improved through data integration, but system complexity increases
Solution Approach 1:
The patent introduces a server as an intermediary between multiple diagnostic devices. This central mediator handles data aggregation, storage, and distribution, simplifying the connections between devices while enabling comprehensive data integration. The server manages the complexity of multi-device coordination, allowing diagnostic devices to focus on their primary diagnostic functions.
Data Source
AI summary
The present invention provides a diagnostic system that diagnoses a diagnostic object by applying a diagnostic rule to data measured on the diagnostic object wherein an object of the present invention is to allow the diagnostic rule to be updated based on a variety of diagnostic case data. Each of multiple diagnostic devices 101 makes a diagnosis by applying a diagnostic rule to diagnostic object data measured on a diagnostic object 104 and sends diagnostic case data, which includes diagnostic object data and its diagnostic result, to a diagnostic rule generation device 102 via a network 103. The diagnostic rule generation device 102 generates a diagnostic rule based on the diagnostic case data received from the multiple diagnostic devices 101 and sends the generated diagnostic rule to the diagnostic devices 101 via the network 103. The diagnostic devices 101 update a diagnostic rule in their devices with the diagnostic rule received from the diagnostic rule generation device 102.