Virtual Sensor Data Suitability Evaluation via ML Coverage Ratio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Selecting appropriate data sets for virtual sensors is challenging due to the dependence on data quality and relevance, differences in operating conditions, and technical aspects like integrating data from various sources and processing large data sets.
Innovation Solution
A method involving providing a data set from real sensors, defining an input range, determining a coverage ratio using a machine learning model, and evaluating the data set based on this ratio to assess its suitability for determining the calculation function of a virtual sensor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data sets are collected using classical Design of Experiment methods, then the data collection process is systematic and structured, but the data may be insufficient for non-linear virtual sensors and the reliability cannot be adequately established
Solution Approach 1:
The patent replaces classical Design of Experiment methods with a machine learning-based approach. The system uses trained machine learning models to automatically evaluate data set quality and determine whether sufficient data is present, substituting systematic manual experiment design with automated intelligent assessment.
Solution Approach 2:
The patent creates virtual copies of sensor data through simulation models. Virtual sensor data is generated by simulation models that replicate real sensor behavior, allowing the system to evaluate data quality and generate additional training data without requiring extensive physical experimentation.
2Reliability
If data from various sources is integrated to improve data quality, then the relevance and completeness of data improve, but the complexity of processing and storing large data sets increases
Solution Approach 1:
The patent performs preliminary evaluation of data sets before they are used for virtual sensor training. The machine learning model assesses data quality, coverage, and suitability in advance, filtering and preparing data sets beforehand to avoid complex processing during deployment.
Solution Approach 2:
The patent introduces an intermediary evaluation layer between raw data collection and virtual sensor training. The machine learning model acts as a mediator that assesses data quality, identifies gaps, and determines whether data sets are suitable, simplifying the integration process by providing clear quality metrics.
3Measurement precision
If the input range is strictly defined to ensure data requirements are met, then the calculation function accuracy improves, but the coverage of diverse operating conditions may be limited
Solution Approach 1:
The patent dynamically adjusts the evaluation criteria based on the specific virtual sensor and application requirements. The machine learning model can adapt the input range definitions and coverage requirements dynamically, allowing the system to optimize between precision and versatility depending on the specific needs.
Solution Approach 2:
The patent applies different evaluation standards to different regions of the input range. Critical regions where accuracy is most important receive stricter evaluation, while less critical regions allow more flexibility, enabling the system to maintain high accuracy where needed while covering diverse operating conditions overall.
Data Source
AI summary
A method for evaluating a data set with regard to suitability for determining a calculation function of a virtual sensor includes providing the data set. The data set includes measurement data resulting from a measurement of measured variables by at least two real sensors. The measurement data has a particular dimension for one of the at least two real-world sensors. The method further includes providing an input range defined for the measured variables of the at least two real sensors to specify at least one requirement for determining the calculation function. The method further includes determining a coverage ratio between the data set and the provided input range using a machine learning model, and evaluating the data set based on the determined coverage ratio.