Sensor Inspection Compensation Using Big-Data Channel Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sensors and sensor ICs in electric apparatuses face issues with false positives due to connection and inspection environment deviations, leading to inconsistent inspection results.
Innovation Solution
A method is developed to compensate for these deviations by generating models based on big-data inspection results, calculating characteristic and compensation coefficients, and creating node compensation maps to adjust sensing values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensors are mechanically connected to the inspection environment for inspection, then inspection can be performed, but false positives occur due to connection deviation and inspection environment deviation
Solution Approach 1:
The patent applies preliminary action by generating compensation coefficients before the actual inspection process. The system collects inspection data from multiple sensors under various connection conditions and environmental conditions, then processes this data to create compensation coefficients that are stored and applied during subsequent inspections. This pre-processing of compensation data eliminates false positives without affecting the actual inspection workflow.
Solution Approach 2:
The patent changes parameters by introducing compensation coefficients that adjust the sensing values based on connection deviation and inspection environment deviation. Instead of maintaining fixed inspection criteria, the system dynamically modifies the sensing parameters using the compensation coefficients derived from statistical analysis of multiple measurements, thereby improving measurement precision while maintaining inspection capability.
2Adaptability or versatility
If multiple inspection environments are used to improve adaptability, then more sensors can be inspected, but inspection results become inconsistent due to environment deviations
Solution Approach 1:
The system performs preliminary actions by collecting inspection data across multiple environments and connection conditions before establishing the compensation model. This pre-collection of diverse data enables the system to adapt to various inspection environments while maintaining result consistency through the statistically derived compensation coefficients.
Solution Approach 2:
The patent achieves universality by creating a compensation coefficient system that works across multiple inspection environments and sensor types. The compensation model is designed to be environment-agnostic, using statistical analysis to derive coefficients that can be applied universally regardless of the specific inspection environment or connection method, thereby maintaining consistency across diverse conditions.
3Reliability
If compensation coefficients are generated based on big-data from multiple sensors and environments, then false positives are reduced, but processing complexity increases
Solution Approach 1:
The patent replaces complex mechanical adjustment systems with a data-driven computational approach. Instead of physically adjusting inspection parameters for each sensor and environment combination, the system uses statistical analysis and automated coefficient generation to compensate for variations, reducing mechanical complexity while improving reliability.
Solution Approach 2:
The system creates a virtual model (compensation coefficients) that replicates the effects of physical adjustments across multiple sensors and environments. By copying the statistical patterns from big-data into a compact coefficient set, the system achieves high reliability without the complexity of processing and managing raw data for each individual inspection case.
Data Source
AI summary
A method of inspecting a sensor including generating a first model based on first big-data including first inspection results for sensors of a same type connected to a current inspection environment, generating first target characteristic coefficients for channels included in the first model, generating first characteristic coefficients for channels included in a current sensor, and generating first compensation coefficients for the channels included in the current sensor based on the first target characteristic coefficients and the first characteristic coefficients.