AI Diagnosis System Adapts to Sensor Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI diagnostic systems are inflexible and cannot function if a sensor is replaced, as they have a fixed number of data input channels, limiting their ability to adapt to changes in sensor configuration.
Innovation Solution
The system includes a system manager with an allocator that distributes sensor data to converters, allowing virtual sensor data to be generated and input into the AI diagnosis model, ensuring continuous operation even if a sensor is disabled or removed, and enabling adaptation to changes in sensor type or number through a diagnostic model builder.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed number of data input channels are used in the AI diagnostic system, then the system structure is simple and stable, but the system cannot adapt to changes in sensor configuration and fails when a sensor is replaced
Solution Approach 1:
The converter is designed to perform multiple functions: it can process real sensor data when a sensor is present, generate virtual sensor data when a sensor is absent, and dynamically switch between these modes. This multi-functionality allows the system to adapt to different sensor configurations without requiring separate processing paths for each scenario.
Solution Approach 2:
The system implements dynamic allocation where the allocator dynamically assigns sensor data to converters based on current sensor availability. When sensors are added or removed, the allocator reconfigures the data distribution in real-time, and converters dynamically switch between processing real data and generating virtual data, making the system flexible rather than static.
Solution Approach 3:
The system generates virtual sensor data as a copy or simulation of actual sensor data when physical sensors are unavailable. This virtual data copying approach allows the AI diagnostic model to continue receiving expected input formats even when physical sensors are replaced or malfunctioning, maintaining system operation without requiring exact physical sensor presence.
2Reliability
If the AI diagnostic system requires all sensors to be present and functioning, then data completeness is ensured, but the system cannot continue operation when a sensor is replaced or fails
Solution Approach 1:
The system prepares virtual data generation capability in advance as a cushion against sensor failures. When a sensor fails or is replaced, the corresponding converter can immediately switch to generating virtual sensor data, preventing system interruption. This prior preparation of alternative data sources ensures continuous operation without requiring real-time detection and response to failures.
Solution Approach 2:
The converter acts as an intermediary between the physical sensor layer and the AI diagnostic model. When real sensor data is unavailable, the converter mediates by generating virtual sensor data that mimics the expected input format, bridging the gap between sensor failure and model requirements. This intermediary function protects the upper-level diagnostic system from lower-level sensor issues.
3Productivity
If virtual sensor data is generated when sensors are unavailable, then system continuity is maintained, but the complexity of data processing increases
Solution Approach 1:
The system extracts the virtual data generation function from the main AI diagnostic pipeline and places it in dedicated converters. This separation allows the core diagnostic model to remain unchanged while the converters handle the complexity of determining whether to process real data or generate virtual data, isolating complexity to specific modular components rather than spreading it throughout the entire system.
Data Source
AI summary
An artificial intelligence diagnosis system includes a diagnosis model responsive to data received from a plurality of sensors. Each of the sensors is a part of an input channel further including a converter operative to process the received sensor data. A system manager is provided and is operative with an allocator to selectively distribute sensor data to the converters. The system manager operates with the converters such that a converter which is allocated with the sensor data processes the allocated sensor data and inputs the processed sensor data into the artificial intelligence diagnosis model. A converter which is not allocated with the sensor data generates virtual sensor data according to an instruction of the allocator and inputs the virtual sensor data into the artificial intelligence diagnosis model.