Modeling Seebeck coefficient of Ca3−xMxCo4O9 (M=Sr, Pr, Ga, Ca, Ba, La, Ag) thermoelectric ceramics

Over the past decade, there has been considerable interest in the field of thermoelectrics, driven by the need for more efficient materials for electronic refrigeration and power generation. In the present study, a new method has been utilized to evaluate the Seebeck coefficient of different Ca3−xMxCo4O9 (M=Sr, Pr, Ga, Ca, Ba, La, Ag) thermoelectric ceramics. For this purpose, a total number of 571 data were collected from the previous experimental works, randomly divided into 462 and 109 sets, and then trained and tested, respectively by gene expression programming (GEP) process. In the set of the model, type of M (for Ca-site substitution), M content (x/3−x), process temperature, compact pressure, sintering time and sintering temperature were entered as input parameters, while the Seebeck coefficient of thermoelectric ceramics was used as output parameter. The training and testing set results of the explicit formulations obtained by the model show that artificial intelligent methods have strong potential and can be applied for the prediction of the Seebeck coefficient of different Ca3−xMxCo4O9 (M=Sr, Pr, Ga, Ca, Ba, La, Ag) thermoelectric ceramics.