Controller Failure Prediction via Server-Side Statistical Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing failure prediction systems for controllers struggle to effectively anticipate errors in controllers with similar components or design issues, leading to potential fatal errors, especially when errors are caused by quality defects or design problems.
Innovation Solution
A failure prediction system that connects multiple controllers to a server through a network, utilizing error correction units and statistical processing based on manufacturing and environment information to identify controllers at risk of future errors, allowing for proactive countermeasures and improved design quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If error information from individual controllers is collected and statistically processed, then failure prediction accuracy is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The patent combines error information from multiple controllers into a centralized server for statistical processing. By merging data from numerous controllers and analyzing patterns across the fleet, the system achieves high prediction accuracy without requiring complex individual controller modifications. The server consolidates manufacturing information, error occurrences, and environmental data to identify correlations that would be invisible in isolated controller analysis.
Solution Approach 2:
The system segments the failure prediction function into two parts: simple error collection at the controller level and complex statistical analysis at the server level. This segmentation allows controllers to remain relatively simple while the server handles the computationally intensive statistical processing needed for accurate predictions.
2Reliability
If manufacturing information and environmental data are integrated into the statistical process, then prediction reliability is improved, but information processing requirements and time consumption increase
Solution Approach 1:
The system performs preliminary actions by collecting and storing manufacturing information for each controller in advance, before errors occur. This pre-collected data includes component lot numbers, manufacturing dates, and configuration details that are stored in the server's database ready for immediate statistical analysis when error data becomes available, reducing processing delays.
Solution Approach 2:
The statistical processing operates continuously as error information streams in from controllers, rather than batch processing periodically. The server maintains continuous analysis of error patterns, manufacturing information, and environmental data, enabling real-time prediction updates without significant time loss.
3Reliability
If the system monitors and analyzes errors across multiple controllers with similar manufacturing information, then ability to predict systematic failures is improved, but data collection and storage requirements increase
Solution Approach 1:
The system uses manufacturing information as a form of copying - storing lot numbers, component identifiers, and manufacturing batch data that represent groups of controllers. Instead of storing complete duplicate configuration data for each controller, the system copies reference manufacturing information and uses it to group controllers by manufacturing characteristics, enabling systematic failure prediction with reduced storage requirements.
Data Source
AI summary
From an error information containing a content of a correctable error that has occurred in a controller of a failure prediction system and an ID of the controller and manufacturing information of a machine to which the controller is attached, a failure of a controller belonging to a group of controllers in which such an error as indicated in the error information has not occurred yet is predicted.