Abstract: The accuracy of machine tools is sensitive to heat, and the heat generated by the feed system causes changes in the temperature field of the machine tool structure, inducing accuracy instability. Constructing a thermal characteristic evaluation model of the feed system under complex thermal conditions is of great significance for accurately grasping the thermal characteristics of machine tools and guiding the development of thermal control and thermal stability improvement. Traditional heat generation models are established with constant parameters and fail to reflect the mapping relationship between time-varying models and process information. This paper proposes a heat generation model construction method considering multi-source process information, comprehensively considering the changes in operating parameters, process parameters, thermal state, and other multi-source information during the working process of the machine tool. The matching relationship between the input power of the machine tool and the power loss of each component of the feed system is deeply studied, and an instantaneous heat generation power calculation model for the structural components of the feed system is established. The experimental results show that compared with traditional modeling methods, the temperature field prediction accuracy based on this model is improved by 28.15%.