Application of the fusion of regression machines for the analog circuit state identification |
|---|
| Piotr Bilski |
- Abstract:
- The paper presents the application of the combined group of regression algorithms to identify state of the analog circuit. Implementing the fusion of regression machines is aimed at obtaining high accuracy of the diagnosed system’s state, especially compared to the single parameter identification algorithm. The large number of simple methods (such as linear regression techniques) is expected to give the high accuracy without the need of time consuming and complex optimization of the selected approach (such as Support Vector Machines – SVM). The approach consists in preparing the ensemble architecture, selecting computational methods, optimizing features extracted from the diagnosed system and testing the approach. The tests were conducted to evaluate efficiency of various fusion architectures, determine their accuracy for different sets of features and confront them against the single optimized regression algorithm. The time analysis verified the ability of the approach to use the framework in the online mode. Obtained results show the potential of the proposed framework for the accurate identification of analog system parameters, which can be used to analyse other types of systems.
- Keywords:
- regression, diagnostics of analog systems, parameter identification
- Download:
- IMEKO-TC10-2019-025.pdf
- DOI:
- -
- Event details
- IMEKO TC:
- TC10
- Event name:
- TC10 Conference 2019
- Title:
16th IMEKO TC10 Conference "Testing, Diagnostics & Inspection as a comprehensive value chain for Quality & Safety"
- Place:
- Berlin, GERMANY
- Time:
- 03 September 2019 - 04 September 2019