论文部分内容阅读
用BP神经网络建立脉动高梯度磁选过程模型．对不同隐含层节点数的神经网络模型预测性能进行了评价，隐含层节点为13的神经网络模型选择为最佳模型．利用选择的最佳模型，对黄铜矿高梯度磁选进行模拟研究．模型研究结果表明，在相当宽操作的范围内，模型能够很好地预测磁选精矿中铜的品位和回收率．这说明建立的高梯度磁选模型合理可行．
Establishment of Pulsatile High Gradient Magnetic Separation Process Model Using BP Neural Network. The prediction performance of neural network models with different hidden layer nodes was evaluated, and the neural network model with hidden layer node of 13 was selected as the best model. Using the best choice of model, the high gradient magnetic separation of chalcopyrite was simulated. The model results show that the model can predict the grade and recovery of copper in the magnetic separation concentrates within a fairly wide range of operation. This shows that the established high gradient magnetic separation model is reasonable and feasible.