Accurate temperature-sensitive circuit modeling of thermoelectric generators for efficient energy harvesting

Only 30-40% of the energy in global industrial systems is utilized, while the rest is mostly dissipated as waste heat. More than 50% of this waste heat is low-grade, yet less than 10% is effectively recovered. Thermoelectric generators (TEGs) offer a promising solution for harvesting low-grade thermal energy, but their output is highly sensitive to temperature gradients. Here, we propose a temperature-sensitive circuit modeling method for TEGs based on nonequilibrium thermodynamics. The Seebeck coefficient and internal resistance are parameterized as explicit functions of temperature to construct a temperature-sensitive TEG circuit model (TsTEGCM). Compared with conventional models, TsTEGCM reduces voltage and power prediction errors from 7.77% and 6.46% to 3.58% and 0.91%, respectively. System-level validation under varied thermal conditions shows simulation-experiment deviations within 5%, reducing the need for repeated experiments and parameter adjustments. Furthermore, we demonstrate a synergistic perovskite solar cell (PSC)-TEG hybrid energy harvesting system, in which the TEG recovers waste heat from the PSC and, during a 43 min experimental test, helps to moderate its operating temperature. Experiments show a 34.65% increase in output power and a 2.2 °C reduction in steady-state PSC temperature, improving overall efficiency and potentially extending device lifetime.