基于BP神经网络的堆肥物料抗剪强度预测模型  

Prediction Model of Compost Materials Shear Strength Based on BP Neural Network

作　　者：李成杰 王涛 李海漪 李加文 邢家乐 杨非 LI Cheng-jie;WANG Tao;LI Hai-yi;LI Jia-wen;XING Jia-le;YANG Fei(China Academy of Machinery Science and Technology Group Co.Ltd.,Beijing 100044,China;Machinery Technology Development Co.Ltd.,Beijing 100044,China)

机构地区：[1]机械科学研究总院集团有限公司,北京100044 [2]机科发展科技股份有限公司,北京100044

出　　处：《中国给水排水》2020年第7期108-113,共6页China Water & Wastewater

摘　　要：针对目前传统方式采集堆肥物料抗剪强度数据过程中环境恶劣、数据采集困难、试验误差大等问题,提出一种基于BP神经网络的抗剪强度预测模型。通过现场试验得到堆肥物料抗剪强度和堆体高度、温度、含水率、密度等参数共39组有效数据,以其中35组作为训练样本,其余4组用于评价模型的预测性能。结果表明,该模型预测值与实测值的平均误差为11. 35%,基于BP神经网络的抗剪强度预测模型具有较高的预测精度,为抗剪强度的预测提供了一种新方法。A prediction model of compost materials shear strength based on BP neural network was proposed to solve the problems of the harsh environment,difficulty in data collection and a large test error in the process of collecting shear strength data by using traditional methods.A total of 39 sets of effective data were obtained through field test of shear strength and other four parameters(pile height,temperature,moisture content and density of the compost materials),among which 35 groups were used as training samples and the remaining 4 groups were used to evaluate the prediction performance of the model.The results showed that the average prediction error between the predicted value and the measured value was 11.35%.The proposed shear strength prediction model based on BP neural network has high prediction accuracy,which provides a new method for shear strength prediction.

关 键 词：污泥堆肥 抗剪强度 BP神经网络 预测模型 

分 类 号：TU992[建筑科学—市政工程]