Coal pyrolysis is a complex process with multiple characterized kinetics that play a crucial role in understanding the devolatilization mechanisms, predicting the products distribution, and designing the reactors, etc. However, there is little knowledge about the pyrolysis kinetics of superfine pulverized coal. In this paper, typical thermogravimetric experiments at low heating rates are performed, which can identify individual species and provide reliable kinetic data. Then, three classical kinetic models are combined to analyze the superfine coal pyrolysis process, including Coats-Redfern (model-fitting type), Starink (model-free group) and distributed activation energy model (DAEM, continuous distribution category). Firstly, a five-reaction decomposition regime is identified through a tangent-bisecting method. Then, a novel application of piecewise Coats-Redfern model based on the tangent-bisecting method is developed, which shows that the coal particle size has a notable impact on the pyrolysis kinetics. Superfine pulverized coal can promote the primary pyrolysis process while inhibit the secondary reactions. Secondly, the Starink model is employed, and the arithmetic mean activation energy increases initially with increasing the particle size, and then declines, attributing to the contradictory effect of particle size. Finally, The distribution function f(E) is deduced based on the Miura-Maki DAEM, which spreads broadly with a single sharp peak. Additionally, the kinetic compensation effect (KCE) of k(0) and E is observed for the Miura-Maki DAEM, which can be utilized to establish certain mechanistic implications. The findings from this work give a better interpretation on the kinetic parameters, and provide us a new perspective to understand the effect of particle size on the coal pyrolysis kinetics.