In this study, we focus on creating a precise servo feed drive model to estimate the servo gains and the mechanical dynamics. Traditionally, manufacturers of machine tools rely on empirical methodologies to facilitate the development of new products. However, the emergence of Industry 4.0 prompted a shift towards virtual machine tool technology to assist product development management. The complete servo feed drive system comprises two integral components: the servo control system and the feed drive mechanism. In this study, we propose a discrete model for the servo control system to enhance the accuracy of the virtual digital model. when subjecting the real system to frequency sweep tests, particle swarm optimization (PSO) is employed to identify the parameters of the servo controller in the model. We observe tracking errors in the servo control system to validate the discrete model, and the predicted tracking errors closely match the experimental values, with the difference being within 2%. Furthermore, we consider external loads and couplings as the feed drive mechanism. We create a dual mass-spring-damping model to simulate the complete servo feed drive system. The digital model proposed in this paper can accurately predict the load inertia with an error of only 1.5% compared with the experimental values. In summary, we present a servo feed drive model for machine tools, enabling the estimation of external loads and couplings in a virtual environment. The advantages are reduced energy consumption, development time, and costs during product development.