A big data-driven hydrometallurgical full-process optimization control model was proposed. Firstly, a LSTM model based on improved attention mechanism was constructed, and the self-attention mechanism was used to enhance the model’s attention to key time steps to improve the prediction accuracy. Secondly, the dynamic adjustment and control of real-time production parameters are realized by using DDQN algorithm improved by priority experience playback. Finally, the optimal control model of economic benefit of the whole process is designed to obtain the optimal solution. The results show that the coefficient of determination R² of the model is 0.92, 0.90, 0.92 in different numerical simulations, which shows that the model is superior to the traditional method in many aspects.

LSTM  /  attention mechanism  /  DDQN  /  prioritized experience replay  /  numerical simulation