Optimum hybrid fractional order maximum power point tracker for thermoelectric generators

Searching for new and nontraditional energy-sourcing solutions has encountered enormous focus in the last decades. One of these solutions is the thermoelectric-based generator (TEG) systems, which extract electrical DC energy using the temperature difference between the two sides of the device. According to the Seebeck theorem, TEGs can benefit and convert heat flows into usable DC power. The thermoelectric modules (TEMs) represent the basic constructing units in TEG systems. Extracted TEG power relies on temperature differences between the hot and cold sides of the unit. Moreover, load resistance and operating point condition variations affect the amount of extracted TEG power. Therefore, robust, efficient, and accurate tracking systems for TEG units’ maximum power point (MPP) tracker (MPPT) are crucial for maximizing their benefits. The task of MPPT in TEG systems is to keep their operation at MPP without dependence on other factors, guaranteeing maximum extracted power from the device. Existing methodologies suffer from slow tracking speed, high ripples at steady state points, and/or continuous oscillation around the operating point, such as incremental resistance/conductance, hill climbing, and perturbations and observation methods. Additional methods, such as fractional order-based proportional-integral (FOPI) and FOPI-derivative (FOPID) techniques, were proposed. This paper proposes a new hybrid MPPT concept of FOPID with tilt-integral-derivatives (TID) techniques from control theory, namely TFOID controller. The new proposed TFOID technique merges the best features of the FOPID and TID methods and provides a more flexible design. The proposed TFOID MPPT combines the stable tracking of MPPT with fast response and rejection of disturbances. Moreover, an advanced application of the marine predator optimization algorithm (MPA) is presented in the paper to obtain the best parameter combination for the proposed TFOID method. Comprehensive tests and comparative results are provided in the paper, which confirms the superior behavior of the proposed MPA-based TFOID MPPT algorithm in TEG applications. Furthermore, several statistical tests were performed for the MPA algorithm, with several recent and standard optimization algorithms in the literature proving its improved behavior in engineering problems. The statistical test shows that the proposed MPA-based TFOID MPPT method extracts the highest power from the TEG system as its mean value of the objective function over 30 runs was 2.4094, the highest among the compared methods. The results also show that joint MPA and TFOID can be applied to MPPT applications in other renewable energy sources.