基于MA-BP分析机制砂形貌对砂浆流动性的影响

Effect of Manufactured Sand Morphology on Mortar Fluidity Based on MA-BP Analysis Mechanism

摘要:

基于多算法（MA）交叉与BP神经网络模型，研究机制砂形貌对砂浆流动性的影响。结果表明：机制砂形貌特征参数对流动性的影响顺序为球形度>磨圆度>棱角度>比表面积>伸长度>纵横比>分形维数>扁平度；优选参数为球形度、磨圆度和棱角度；流动性与球形度、磨圆度呈正相关，与棱角度呈负相关；基于优选参数的BP神经网络模型能够准确预测砂浆流动性。这对于建立或完善机制砂砂浆性能的预测模型，实现机制砂的规模化应用具有指导性作用。

Abstract:

The effect of manufactured sand morphology on mortar fluidity based on a multi-algorithm（MA） cross and BP neural network model was studied. The results indicate that the order of influence of morphology feature parameters on fluidity is sphericity>roundness>edge angle>specific surface area>elongation>aspect ratio>fractal dimension>flatness. The preferred parameters are sphericity， roundness and edge angle. Fluidity is positively correlated with sphericity and roundness， while negatively correlated with edge angle. The BP neural network model based on the preferred parameters can accurately predict the mortar fluidity. It is instructive to establish or improve the prediction model of the performance of manufactured sand mortar and realize the large-scale application of manufactured sand.