The paper presents a short overview over many different techniques for feature extraction. Feature extraction is a very important field of image processing and object recognition. Two different levels of feature extraction are also presented and the connection between them is explained

Identification of ongoing processes in solid oxide fuel cells (SOFC) enables both optimizing the operating environment and prolonging the lifetime of SOFC. The Levenberg–Marquardt algorithm (LMA) is commonly used in the characterization of unknown electrochemical processes within SOFC by extracting equivalent electrical circuit (EEC) parameter values from electrochemical impedance spectroscopy (EIS) data. LMA is an iteration optimization algorithm regularly applied to solve complex nonlinear least square (CNLS) problems. The LMA convergence can be boosted by the application of an ordinary limit strategy, which avoids the occurrence of off-limit values during the fit. However, to additionally improve LMA descent properties and to discard the problem of a poor initial parameters choice, it is necessary to modify the ordinary limit strategy. In this work, we designed a new automatic update (i.e., adaptive) limit strategy whose purpose is to reduce the impact of a poor initial parameter choice. Consequently, the adaptive limit strategy was embedded in a newly developed EIS fitting engine. To demonstrate that the new adaptive (vs. ordinary) limit strategy is superior, we used it to solve several CNLS problems. The applicability of the adaptive limit strategy was also validated by analyzing experimental EIS data collected by using industrial-scale SOFCs.

Electrochemical impedance spectroscopy (EIS) is an important electrochemical technique that is used to detect changes and ongoing processes in a given material. The main challenge of EIS is interpreting the collected measurements, which can be performed in several ways. This article focuses on the electrical equivalent circuit (EEC) approach and uses grammatical evolution to automatically construct an EEC that produces an AC response that corresponds to one obtained by the measured electrochemical process(es). For fitting purposes, synthetic measurements and data from measurements in a realistic environment were used. In order to be able to faithfully fit realistic data from measurements, a new circuit element (ZARC) had to be implemented and integrated into the SPICE simulator, which was used for evaluating EECs. Not only is the presented approach able to automatically (i.e., with almost no user input) produce a more than satisfactory EEC for each of the datasets, but it also can also generate completely new EEC configurations. These new configurations may help researchers to find some new, previously overlooked ongoing electrochemical processes.