Study on the Process Optimization of Synthesizing Co<sub>3</sub>O<sub>4</sub> Nanoparticles by Homogeneous Precipitation Based on Support Vector Regression

Abstract

The Co3O4 is the major raw material for fabricating lithium cobalt oxide electrode of lithium ion battery. According to the experimental dataset on grain diameter of Co3O4 nanoparticles synthesized by homogeneous precipitation under four main process parameters including the concentration of Co(NO3)2•6H2O solution, mole ratio of reactants, reaction temperature and reaction time, support vector regression (SVR) combined with particle swarm optimization (PSO) for its parameter optimization, is introduced to establish a model for estimating grain diameter of Co3O4 nanoparticles. The comparison of prediction results strongly support the prediction and generalization abilities of SVR are superior to those of multivariable gradual regression (MGR). Meanwhile, the index of grain diameter of Co3O4 nanoparticles under an independent combination of process parameters predicted by SVR model is more accurate than that by MGR model. The multi-factors analysis results based on SVR model are consistent with that of the literatures. This study suggests that SVR is a theoretical significance and potential practical value in development of smaller grain diameter of Co3O4 nanoparticles via guiding experiment.