Electrochemical Impedance Spectroscopy (EIS) is commonly used to characterize electrochemical systems such as fuel cells, supercapacitors, or batteries. Interpretation of the experimental data is mostly done using Equivalent Electrical Circuit (EEC) models whose components aim at representing the charge and mass transport phenomena occurring at the material scale. It is necessary to use a numerical solver to estimate the model parameters starting from experimental data. The simplest are based on deterministic algorithms, which are efficient in terms of convergence and repeatability with, in return, a strong dependence of their results on the initial values: finding an appropriate starting point can be difficult and time-consuming. Stochastic methods offer several advantages for global optimization: initial values are not required, or their influence is lower than for deterministic algorithms; they are also robust and thus well adapted for problems with -possibly-multiple solutions. They are however more demanding in terms of numerical resources and their results may be less reproducible than those of deterministic algorithms.In this work, we present a study on the use of three hybrid methods combining a stochastic Genetic Algorithms (GA), Particle Swarm Optimization (PSO), or Simulated Annealing (SA)- algorithm with a deterministic Nelder-Mead (NM) algorithm for the estimation of the EEC parameters of Proton Exchange Membrane Fuel Cells (PEMFC). These hybrid methods were evaluated for the interpretation of simulated and experimental PEMFC impedance spectra, modeled with EEC of different complexity including -in our case- 5 to 11 parameters. The three stochastic/deterministic methods were compared in terms of stability, efficiency, ability to explore multiple solutions, and computing resources. The results showed that all hybrid methods were able to converge toward satisfying and physically meaningful solutions in most cases and allowed an improvement compared to the use of a single NM or stochastic algorithm alone