Radiometric Measurement Device Using AI for Self-Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In radiometric measurement devices, complex measurement problems often require simplified or unknown measurement models, leading to inaccurate calculation of process variables due to unaccounted influences and lack of explicit knowledge, resulting in suboptimal measurement quality.
Innovation Solution
The implementation of a radiometric measurement device that utilizes artificial intelligence methods like Machine Learning or Deep Learning to establish relationships between sensor data and measurement variables without an analytical measurement equation, allowing for self-learning and accurate determination of process influences, including unknown disturbances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simplified or unknown measurement models are used for complex measurement problems, then device complexity is reduced, but measurement precision deteriorates due to unaccounted influences and lack of explicit knowledge
Solution Approach 1:
The measurement device performs self-calibration and self-optimization by automatically determining calibration parameters and measurement model parameters without requiring external intervention or complex manual configuration. The system uses stored reference values and sensor data to autonomously calculate optimal parameters, reducing the need for complex external calibration procedures while maintaining high measurement precision.
Solution Approach 2:
The system stores reference values for process variables and calibration parameters in advance during a calibration phase. These pre-stored reference values are then used during actual measurements to quickly determine process variables without requiring complex real-time calculations or external reference measurements, thereby simplifying the measurement model while maintaining accuracy.
2Measurement precision
If multiple sensors and AI-based learning units are implemented, then measurement precision improves through accurate determination of process influences, but device complexity increases
Solution Approach 1:
The system combines multiple sensors (radiometric sensors, temperature sensors, acceleration sensors, etc.) and integrates them with a learning unit and storage unit into a unified measurement device. The sensors are spatially distributed but functionally integrated, sharing common processing units and storage resources, which reduces overall system complexity while enabling comprehensive multi-parameter measurement and improved precision through sensor fusion.
Solution Approach 2:
The measurement device is designed with multi-functional capabilities, where a single device can perform various types of measurements (radiometric measurements, temperature measurements, acceleration measurements) using the same core processing unit and storage system. The learning unit can adapt to different measurement tasks by loading appropriate reference values and calibration parameters, eliminating the need for separate dedicated devices for each measurement type.
3Measurement precision
If AI methods like Machine Learning or Deep Learning are used to establish relationships without analytical measurement equations, then measurement precision improves through self-learning, but loss of information increases due to reduced storage requirements
Solution Approach 1:
The system extracts and stores only the essential calibration parameters and reference values needed for accurate measurements in a compact format during a calibration phase. Instead of storing complete analytical measurement equations or large datasets, the system extracts the critical parameters that capture the relationship between sensor data and process variables, thereby reducing storage requirements while maintaining measurement precision through the use of these condensed parameter sets.
Data Source
AI summary
A radiometric measurement device includes a number n of sensors, wherein a respective sensor of the number n of sensors is configured to generate associated sensor data, such that overall a number n of sensor data is generated by means of the number n of sensors. A measurement variable calculation unit is configured to calculate a number m of measurement variable values depending on the number n of sensor data on the basis of values of a number d of parameters. A learning unit is configured to calculate the values of the number d of parameters on the basis of training data.