欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
将BP神经网络与遗传算法相结合，使用遗传算法优化神经网络的权值和阈值，然后用BP算法训练网络，避免了单独使用BP神经网络训练时易于陷入局部极小值的问题，建立了一种新的岩爆预测模型。采集国内外具有代表性的一些岩爆案例作为BP训练样本，将样本数据经过多次迭代之后，达到指定误差停止训练，利用训练好的模型对某铜矿部分岩爆进行预测，预测结果与实际岩爆等级一致。
英文摘要：
The authors have adopted genetic algometry （GA） and BP neural networks to build a new model for rock burst prediction. The weights and thresholds of BP neural networks were optimized using GA, then the neural networks were trained using BP algometry, so, solving the problem that traditional BP neural networks lie in absence about constringency rate slowly and easy to fall into local minimum. Then the authors selected some representative rock burst cases at home and abroad as training samples. After the multiple iteration of the sample data and when the specified error was achieve, the neural network training was stopped. The trained model was used to predict the rock bursts happened in some copper mine, and the predicted results were consistent with the actual level of the rock bursts.
同期刊论文项目
同项目期刊论文
期刊信息
位置：