上向水平分层充填体强度预测与参数优选

Strength Prediction and Parameters Optimization of Upward Horizontal Cut and Fill Stoping Filling Body

为减少水泥用量,降低充填成本,优化充填体配比和组合。以新桥矿区为例,通过室内充填体强度实验,建立BP神经网络,以水泥:粉煤灰:全尾砂、浓度和组合为输入因子,以7天抗压强度为输出因子,以实验数据为训练和检验样本来建立神经网络模型,通过对比误差和计算步数,确定隐含层节点数为11。将参数进一步细化并输入到模型中,搜索出优选样本,其优选结果能在保证充填体整体稳定性的前提下,水泥用量减少18.67 kg/m3,降低了充填成本。

In order to reduce the amount of cement and the cost of filling, optimize the filling body proportions and combinations , taking xin qiao mine as an example, through the filling body strength test in laboratory, established BP neural network, taking cementfly ash-full tailings、concentration and combination as input data, and the 7 days compressive strength as output data. The data from laboratory test were used as samples of training and testing to build the prediction model for BP neural network. By comparing the influences of hidden layer nodes on model training process and prediction number of steps, determined the hidden layer node number is 11. By means of entering the refined ratio parameters into prediction model, optimal samples are searched, the result indicates that cement dosage reduced 18 . 67 kg/m^3 and the cost of filling also reduce under the premise of the overall stability of filling body.