To address the limitations in predictive capabilities of thermal error models built from single-source, single-structure data, this paper proposes a thermal error prediction model based on decision-level fusion of multi-source heterogeneous information to enhance prediction accuracy. First, an experimental platform for multi-source heterogeneous information acquisition was constructed to collect thermal error data from different signal sources (multi-source) and different structures (heterogeneous). Next, based on the characteristics of the multi-source and heterogeneous data, relevant features were extracted to construct the feature set. Then, using the feature information set of the multi-source and heterogeneous data, thermal error prediction sub-models were established using Nonlinear Autoregressive models with exogenous inputs (NARX) and Gated Recurrent Units (GRUs) for a vertical machining center spindle. Finally, the entropy weight method was employed to assign the weights for the linear-weighted fusion rule, achieving decision-level fusion of multi-source heterogeneous information to obtain the final prediction result. This result was then compared with experimental results and the prediction results of single-source models. The findings indicate that the proposed thermal error prediction model closely matches the actual results and outperforms the single-source and single-structure data models in terms of Root-Mean-Square Error (RMSE), Coefficient of Determination (R²), and Mean Absolute Error (MAE).

To comprehensively improve the thermal, static and dynamic characteristics and achieve lightweighting for CNC machine tools, this paper proposes a multi-objective joint optimization method based on coupled thermal–mechanical–vibration collaborative analysis. The dual-spindle component of a CNC machine tool is taken as the parameterized model. According to the theories of thermal characteristics, statics, and dynamics, the solution of thermal-mechanical coupling deformation and the solution of vibration characteristics under prestress are repeatedly conducted, that are working collaboratively with each process of parameters sensitivity computing, selection of design variables, central composite design, and multi-objective joint optimization. The response surfaces of the objective functions are established. The optimal parameter combination for improving CNC machine tool performance is effectively obtained. And the multiple objectives of improving the thermal, static and dynamic characteristics, as well as lightweighting, are achieved. The results show that the mass of the optimized component is reduced by 10.1%; the first-order natural frequency is increased by 3.9%; the coupling deformation of the end face of the left spindle seat is reduced by 5.3%; and the coupling deformation of the end face of the right left spindle seat is reduced by 9.0%, while the temperature of the component hardly increases. This indicates that this method can comprehensively improve the performance of CNC machine tool components and provide a reference for the multi-objective joint optimization design of CNC machine tools.