Investigation of Electrochemical Processes in Solid Oxide Fuel Cells by Modified Levenberg–Marquardt Algorithm: A New Automatic Update Limit Strategy

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