Sensor-Based Teaching Data Preparation for Abnormal Sensor Readings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing teaching data preparation devices fail to prepare accurate data when sensors are not normal, i.e., out of order or with deteriorated sensing accuracy.
Innovation Solution
A teaching data preparation device that includes an acquisition unit, calculation unit, and determination unit to calculate and compare sensor values with calculation values, determining sensor normality and preparing teaching data using these values even when sensors are abnormal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor values are used directly for teaching data preparation, then the process is simple, but the data accuracy deteriorates when sensors are not normal
Solution Approach 1:
The system performs preliminary determination of sensor normality before using sensor values for teaching data preparation. The determination unit checks whether each sensor is normal beforehand, and only uses sensor values from normal sensors, preventing inaccurate data from being prepared in advance.
Solution Approach 2:
The determination unit acts as an intermediary between the sensor values and the teaching data preparation unit. It verifies the normality of sensors and controls which sensor values are passed to the teaching data preparation process, ensuring data accuracy without requiring complex modifications to the overall system.
2Reliability
If all sensor values are used regardless of normality, then data quantity is maximized, but data quality deteriorates due to inclusion of abnormal sensor readings
Solution Approach 1:
The system applies different quality standards to different sensor values based on their source. Each sensor value is evaluated individually for sensor normality, and only values from normal sensors are selected for teaching data preparation. This local quality control ensures high reliability while maintaining maximum possible data quantity from functional sensors.
3Measurement precision
If sensor normality is not checked, then processing speed is fast, but data accuracy cannot be ensured when sensors fail
Solution Approach 1:
The determination of sensor normality is performed as a preliminary step before teaching data preparation. By checking sensor status in advance and caching the normality determination results, the system ensures accuracy without significant time loss during the actual data preparation process.
Data Source
AI summary
A teaching data preparation device includes: an acquisition unit configured to acquire a plurality of sensor values from a plurality of sensors; a calculation unit configured to use a sensor value included in the sensor values to calculate a calculation value corresponding to a sensor value other than the used sensor value; a determination unit configured to determine whether each of the sensors is normal; and a teaching data preparation unit configured to prepare, when the determination unit determines that at least one of the sensors is not normal, teaching data from an input-output dataset including at least one calculation value that corresponds to the sensor value of the sensor determined to be not normal.