Test Bench Measurement Trajectories Using Safe Active Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The creation of data-based models for technical systems is hindered by the limited availability and high quality requirements of measurement data, leading to lengthy measurement campaigns and high costs, with incomplete data coverage resulting in uncertain extrapolation behavior.
Innovation Solution
An active learning method is employed to measure technical systems on a test bench, using a probabilistic regression model to optimize measurement trajectories based on a surrogate model, ensuring complete data coverage while avoiding unsafe operating points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If measurement campaigns are extended to collect more measurement data, then the quality and quantity of training data improve, but the measurement time and costs increase
Solution Approach 1:
The patent applies preliminary action by using an active learning approach that plans and optimizes measurement trajectories before actual measurement campaigns. A surrogate model predicts system behavior to pre-determine informative measurement points, allowing the system to prepare optimal measurement paths in advance without extensive trial-and-error measurement campaigns
Solution Approach 2:
The system applies self-service by automatically optimizing its own measurement trajectories using the active learning framework. The surrogate model and optimization algorithm autonomously determine the next most informative measurement points without requiring manual intervention or extensive human expertise in experimental design
2Measurement precision
If measurement trajectories are optimized to cover complete data space, then the model training quality improves, but the risk of exceeding safety limits increases
Solution Approach 1:
The patent implements feedback by continuously updating the surrogate model with measurement data obtained during the measurement process. This feedback loop allows the system to adaptively refine its understanding of the system behavior and adjust subsequent measurement trajectories to maintain both data coverage and safety constraints
Solution Approach 2:
The surrogate model acts as an intermediary between the measurement system and the actual technical system. It predicts system behavior and identifies safe measurement trajectories, serving as a mediator that enables comprehensive data collection while filtering out potentially harmful operating conditions through predictive modeling
3Reliability
If manual planning and monitoring are used to avoid safety limits, then the safety of measurement is ensured, but the preparation time and complexity increase
Solution Approach 1:
The patent replaces manual mechanical planning and monitoring processes with an automated computational system. The active learning framework with surrogate modeling substitutes human experts' manual trajectory design and safety monitoring with algorithmic optimization and predictive modeling, reducing both time and complexity
Data Source
AI summary
A computer-implemented method for providing input data points for measuring a technical system is disclosed. The technical system is measured in particular on a test bench according to predefined measurement trajectories in order to obtain measurement data, wherein the measurement data assigns one or more measured variables as labels to an input data point from one or more input variables. The method includes (i) measuring a measurement trajectory by successively controlling the technical system with input data points of the measurement trajectory and identifying the respective one or more measured variables as respective labels, (ii) training or updating a data-based surrogate model, which is designed in particular as a probabilistic regression model, with the labeled input data points, (iii) determining a further measurement trajectory to be measured by optimizing a total information measure of the input data points of the measurement trajectory in the surrogate model, and (iv) measuring the technical system with the determined further measurement trajectory to be measured. The total information measure indicates a sum of the information measures of the individual input data points of the measurement trajectory, wherein the information measure specifies the contribution of the relevant input data point to the further training of the surrogate model.