A framework based on experimental-and-auxiliary data fusion and robust multi-task neural networks for the prediction of melting points in electroslag systems

The melting point of slag plays a pivotal role in determining the process stability and product quality during electroslag remelting (ESR). However, accurate prediction remains challenging due to the scarcity of experimental data, the compositional complexity of slag systems, and the limited accuracy of conventional empirical models. In this study, a robust multi-task neural network framework based on experimental-and-auxiliary data fusion is developed to predict the melting points of ESR slags. The framework combines one experimental melting-point source dataset with two auxiliary augmented datasets: (i) experimentally measured melting points, (ii) melting-point estimates derived from viscosity inflection points using the viscosity prediction model, and (iii) melting-point values generated from empirical correlations. A multi-task learning architecture is constructed, comprising one primary task (experimental data) and two auxiliary tasks (augmented datasets), enabling effective knowledge transfer and joint optimization through a custom-designed loss function and adaptive task-weighting scheme. Error analysis results demonstrate that the proposed method consistently outperforms traditional machine learning models and single-task neural networks across multiple evaluation metrics, including mean squared error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). In addition, the model exhibits excellent robustness and physical consistency, providing a promising solution for intelligent slag system design and process optimization in ESR operations.