Accurate prediction of mill load parameters is crucial to improving grinding efficiency and saving energy. Traditional prediction models have challenges such as poor interpretability, low prediction efficiency and differences in data distribution. This study innovatively proposed a multi-task prediction model that integrates physical information and domain adaptation. By constructing a physical-data-driven hybrid model, the physical relationship between mill load parameters is embedded into the model as prior knowledge to improve the prediction accuracy of the model. At the same time, multi-task learning is used to predict the material-to-ball volume ratio and the pulp density at the same time, which improves efficiency and reduces repetitive work. The domain adaptation method is introduced to ensure that the model maintains stable prediction performance when the data distribution changes. Laboratory ball mill data verification shows that the proposed model not only improves the prediction accuracy, but also adapts well to variable working conditions, showing significant superiority.

In the grinding industry, accurate prediction of the mill load is the key to increasing mill income and reducing mill failure. It is difficult to improve the prediction accuracy of the model due to insufficient information on single-source domain data and distribution differences among different data. A multi-source domain unsupervised domain adaptation method based on common and special features is proposed. Multi-source domain data has both common and special characteristics. If only common features are emphasized, some useful information will be discarded. If only special features are used, the model generalization is not good. To solve this problem, a common feature extraction block is used to extract the common domain invariant representation of multiple source domains and target domains, and special features are obtained through the special feature extraction block. After the features are fused and input into the common regressor, the multi-source domain predicted values are obtained. Finally, the predicted values of multiple source domains are added and averaged to get the final prediction result. The effectiveness of this method is proved by cross-experiments on the ball mill data set collected in the laboratory.