Adaptive Parameter Estimation for a Polymer Electrolyte Membrane Fuel Cell with Nonlinear Parameterization

Parameter estimation of Polymer Electrolyte Membrane Fuel Cell (PEMFC) is important for its modeling and controller design. However, the nonlinear parameterization property of PEMFC makes it difficult to obtain the precise value of the parameter embedded in the nonlinear function. In this paper, we propose an adaptive nonlinear parameter estimation algorithm to solve the nonlinear problem of PEMFC. To separate the parameter embedded in the nonlinear function, a linear parameterization method based on direct differential and filter design is first applied to the PEMFC model. In this way, the nonlinear function is transformed into a linearly parameterized function with an exponentially decaying remainder that can guarantee the unknown parameters’ convergence. Because the direct differential operation introduces the derivation signal of the system input in the linearly parameterized function, the prescribed-time robust differentiator is used to achieve its fast convergence. Later, the adaptive parameter estimation algorithm driven by the parameter estimation error is applied. Because there is no need to construct any observer/predictor, the calculation efficiency is enhanced. Finally, the comparative experiment of the differentiator and the model validation experiment are conducted. The experiment results show that the prescribed-time robust differentiator has a fast convergence rate and the parameter estimation algorithm achieves the nonlinear parameter estimation with a high accuracy.