论文部分内容阅读
为实现对盾构机刀盘的故障预测,提出一种基于粗糙集理论与BP神经网络相结合的盾构机刀盘故障预测法。首先应用粗糙集理论的基于动态层次聚类的离散化算法对刀盘历史数据离散化处理,然后利用改进的基于差别函数的属性约简算法约简决策表,将冗余属性从决策表中删除,把经过约简的数据作为输入端输入BP神经网络,构建故障预测模型,最后以盾构机刀盘故障预测为例进行试验验证。结果表明,该模型能有效缩短训练时间,提高预测效率,具有较高的预测精度,在实际工程应用中具有良好的应用价值。
In order to realize the fault prediction of cutterhead of shield machine, a cutterhead fault prediction method based on rough set theory and BP neural network is proposed. Firstly, the discretization of the historical data of the cutterhead is discretized by using the dynamic hierarchical clustering based on the rough set theory. Then, the decision table is reduced by using the attribute reduction algorithm based on the difference function, and the redundant attributes are deleted from the decision table , The reduced data is input into the BP neural network as the input to construct the fault prediction model. Finally, the fault prediction of the cutterhead of shield machine is taken as an example to verify the model. The results show that the model can effectively shorten the training time, improve the prediction efficiency, and has a high prediction accuracy, which has a good application value in practical engineering applications.