论文部分内容阅读
外加剂对充填体强度影响复杂,具有非线性特性,用数理统计的方法建立充填体强度与外加剂之间的关系模型很困难。因此,首先开展了全尾砂充填体强度与外加剂掺量的正交试验;然后采用BP神经网络进行试验样本的学习训练,建立充填体强度与外加剂种类及掺量之间的关系模型。结果表明,采用BP神经网络建立的预测模型,不仅对学习样本的预测精度高,更重要的是对测试样本的预测精度同样高,预测的最大误差仅为4.16%。实践证明,BP神经网络预测模型可以提高实验工作效率,节省人力、物力,为充填体添加外加剂的研究提供了一条有应用前景的理论设计途径。
The admixture has a complex influence on the strength of the filling body and has nonlinear characteristics. It is very difficult to establish a mathematical model for the relationship between the strength of the filling body and the admixture. Therefore, the orthogonal experiment of the strength of tailings filling agent and admixture is carried out firstly. Then the BP neural network is used to study the training of the test sample, and the relation model between the strength of filling body and the type and dosage of additive is established. The results show that the prediction model established by BP neural network not only has high prediction accuracy for learning samples, but also more importantly, the prediction accuracy of test samples is also high, and the maximum prediction error is only 4.16%. Practice has proved that BP neural network prediction model can improve experimental work efficiency, save manpower and material resources, and provide a promising theoretical design approach for the study of adding filler in filling body.