Fractional Hammerstein Modeling of the Heating Process With Exponential Static Nonlinearity Identified Via The Honey Badger Approach

Many thermal processes encountered in real applications display nonlinear behaviors that conventional linear models fail to represent with sufficient accuracy. This limitation highlights the need for more flexible hybrid modeling approaches, such as fractional Hammerstein structures. In this work, a thermal system is modeled using a fractional Hammerstein architecture in which the linear dynamic block is formulated as a fractional-order transfer function, while the static nonlinear block is described by an exponential function. To overcome the associated identification challenges, the recently developed Honey Badger metaheuristic optimization algorithm is employed, offering enhanced convergence properties and improved robustness compared with classical optimization approaches. Experimental data are collected from an electric oven using an STM32F407VG microcontroller, ensuring reliable real-time measurements required for accurate model identification.