论文部分内容阅读
为解决煤矿材料成本预测存在的问题.将多元回归模型和RBF神经网络相结合,建立了煤矿材料成本预测的MRA-RBF耦合模型.从自然因素、技术因素、管理因素等方面选取8个变量建立煤矿成本预测指标体系.对实际煤矿材料成本进行预测分析.结果表明:MRA-RBF耦合模型预测最大误差为10.795 145 2%,平均误差为5.459 71%,最小误差仅为0.344 581 7%,预测效果较好,预测精度与单一MRA模型及RBF神经网络相比有了较大提高.验证了所提出模型的科学性、准确性,说明将线性拟合算法(MRA)和非线性拟合算法(RBF)结合起来用于煤矿材料成本预测是一种较为优越的算法,为煤矿材料成本预测及控制提供一种新的方法.
In order to solve the existing problems of coal material cost prediction, the MRA-RBF coupling model of coal material cost prediction was established by combining multiple regression model with RBF neural network. Eight variables were selected from natural factors, technical factors and management factors The forecasting index system of coal mine cost forecasting shows that the maximum forecast error of MRA-RBF coupled model is 10.795 145 2%, the average error is 5.459 71% and the minimum error is only 0.344 581 7% The prediction accuracy is greatly improved compared with the single MRA model and the RBF neural network.It is proved that the proposed model is scientific and accurate.It shows that the linear fitting algorithm (MRA) and the nonlinear fitting algorithm (RBF ) Is a superior algorithm for forecasting coal material cost, which provides a new method for coal material cost prediction and control.