A new optical microscope approach-based tracker for increasing harvested energy from hybrid photovoltaic- thermoelectric generators

• First time to track the GP of PV-TEG generating system in NUTD via suggested optical microscope algorithm. • A 9 × 9 TEG array installed in the back of PV panel at seven cases NUTD patterns has been analyzed. • A comparison to GRO, SCA, HBA, SOA, PSO, and JS is presented to evaluate the proposed OMA-MPPT. • The obtained results validate the suggested tracker’s resilience. Integrating solar photovoltaic panels and thermoelectric generators is considered one of the most promising electrical hybrid energy systems. Such hybrid systems utilize the wasted heat from the back side of solar panel to feed the thermoelectric generators (TEGs) to generate electricity. The extracted power from the TEG is mostly related to the load and the distribution of temperature. During uniform temperature distribution (UTD), there is a single maximum power point (MPP) in the current-power curve. This point can be easily tracked using any traditional MPP tracking methods such as incremental conductance or perturb and observe (P&O). However, in the case of non-uniform temperature distribution (NUTD), there is only one global maximum power point (GMPP) and some local MPPs. The GMPP under NUTD can be tracked using a variety of methods, but many of them have shortcomings such slow tracking speeds, excessive steady-state oscillations, and poor performance in situations that change quickly. Consequently, this research proposes a powerful optical microscope algorithm (OMA) to extract the global MPP in case of NUTD of hybrid photovoltaic (PV)-TEG system considering different temperature distribution scenarios. The suggested methos is evaluated on PV panel and 9 × 9 TEG array fixed at its back and operated at seven operating conditions: normal, non-homogeneous row, non-homogeneous column, diagonal, long wide, random, and internal. The optimized results are compared with gold rush optimizer (GRO), sine cosine algorithm (SCA), honey badger algorithm (HBA), particle swarm optimization (PSO), seagull optimization algorithm (SOA), and Jellyfish search (JS) approach. The suggested OMA-MPPT succeeded in achieving the least error between the fetched GP and the real one with a value of 0.0115 % during diagonal heat distribution. While the greatest error was 0.2187 % during normal operation. The results showed that OMA-MPPT is superior to other methods for obtaining the best global power (GP) from a hybrid PV-TEG system under all examined heat distribution patterns.