Abstract: The intrinsic reaction kinetic model of coal char is a key submodel required for computational fluid mechanics (CFD) simulation of coal ignition combustion rates in boilers, typically obtainable from thermal analysis. Although the study of thermal analysis reaction kinetics has been developed for decades, three main issues persist: ① Inconsistency between nonisothermal and isothermal kinetic parameters; ② Inconsistency between fixed kinetic parameters and mean variable kinetic parameters; ③ Lack of a universal kinetic model without uncertain parameters, such as the commonly used nth-order kinetic model, random pore model, and autocatalytic model, each containing uncertain parameters like reaction order n, structural parameter ψ, and reaction index a and c. To address these issues, a generalized surface activation function model (GSAFM) was proposed. Its reaction mechanism function f(X) = 1 − X (where X is the conversion ratio), with activation energy E_X and pre-exponential factor A_X varying with conversion ratio. Four isoconversional methods (variable kinetic parameter models) were employed, including isothermal (ISO) GSAFM, nonisothermal (NON) GSAFM, Flynn-Wall-Ozawa (FWO), and Kissinger-Akahira-Sunose (KAS), alongside a fixed kinetic parameter model (ISAFM) to predict the intrinsic reaction rates of nonisothermal and isothermal combustion of Jiangjunmiao (JJM) and Hongshaquan (HSQ) coal chars. Results indicated that ISO GSAFM had the best predictive performance; FWO and KAS had poor predictive performance, primarily due to the approximation error introduced in E_X determination by temperature integration in these models, which exponentially amplified when determining A_X. The mean E_X of coal char for isothermal combustion obtained by ISO GSAFM was 141 kJ/mol, close to the 146 kJ/mol obtained by ISAFM, thus resolving issue ②; its kinetic parameters could effectively predict the intrinsic reaction rates of coal char for nonisothermal combustion, indicating that ISO GSAFM parameters could be shared between nonisothermal and isothermal experiments, resolving issue ①; its f(X) = 1 − X demonstrated generality without uncertain parameters, addressing issue ③. The E_X of coal char reached its maximum value in the ignition zone, explaining the fact that coal ignition in the boiler is the most difficult segment in the combustion process; subsequent E_X gradually decreased as coal char combustion entered a stable combustion stage but showed a rapid increase in the late reaction stage, consistent with the deactivation phenomenon in the later stages of coal char combustion. Therefore, ISO GSAFM simultaneously solves issues ①‒③ while explaining experimental phenomena, potentially providing new intrinsic reaction kinetics submodels for CFD simulation based on fixed kinetic models. Although JJM and HSQ coal chars of the same rank exhibit differences in physicochemical structure, the trend of E_X variation with conversion ratio is similar, with differences of values less than 5%. Accurate prediction of their intrinsic reaction rates can be achieved by arithmetically averaging their respective variable kinetic parameters, indicating that the intrinsic reaction rates of different coal chars of the same rank can be predicted using mean kinetic parameters based on GSAFM, thus potentially realizing the use of the same variable kinetic parameters (mean of multiple samples) for predicting the reaction rates of coal chars of the same rank. Furthermore, GSAFM indicates that the E_X of coal char is only related to its chemical structure; the use of characteristic values of variable activation energy can provide insight into the underlying mechanism of the influence of coal char chemical structure on its intrinsic reactivity, such as ignition difficulty and deactivation mechanism. Once E_X of coal char is predicted based on its chemical structure, its intrinsic reaction rates can be directly predicted using GSAFM. This correlation method provides a new perspective for studying the relationship between coal char structure and reactivity.