基于BP神经网络的土石填料压实质量控制

现行的土石填料压实质量检测方法由于自动化程度低、效率低、采样密度小等缺点,很难满足工程建设质量实时控制的需要.开发新的检测技术,实现填料高质、快速、无损和动态的检测已成为亟待解决的重要任务之一.本文阐述了BP神经网络原理,并结合PFWD时程曲线的动态参数,建立了基于BP神经网络的土石填料预测模型.实际数据对比分析表明,该方法能够满足压实质量的快速检测.

BP neural network based quality control on compaction of earth-rock fill material

Since the current methods for the detecting the compaction quality of earth-rock fill material have some defects such as lower automatic level,lower efficiency,less sampling density,etc.,they are difficult to meet the demand of the real-time control on the quality the actual construction;for which developing a new detection technology for realization of the high precision,quick,non-destructive and dynamic detection of the fill material becomes an important task to be urgently solved.The principle of BP neur...