针对长短时记忆(Long-Short-Term Memory,LSTM)网络方法预测刀具状态仅考虑过去信息而忽略未来信息的问题,无法准确预测刀具状态,提出了一种基于双向长短时记忆(Bi-directional Long Short-Term Memory,Bi-LSTM)预测刀具状态的方法,该方法首先利用主成分分析法(Principal components analysis,PCA)用以降维数据,然后,将降维处理后的数据输入到Bi-LSTM网络中,最后对刀具状态进行预测.通过实验对比,表明PCA结合Bi-LSTM网络的预测准确率达到98.3959％,优于LSTM网络和PCA结合LSTM网络,并且误差也小于其他两种模型,验证了该方法对于刀具状态预测的有效性.
Tool Condition Prediction Based on Bidirectional Long Short Memory Network
In view of the problem that the long and short term memory network method only considers the past information and ignores the future information in predicting the tool state,a principal component analysis method is proposed to reduce the dimension data,and the reduced dimension data is input into the two-way long and short term memory network for tool state prediction.The experimental comparison shows that the prediction accuracy of PCA combined with Bi-LSTM network reaches 98.3959％,which is better than LSTM network and PCA com-bined with LSTM network,and the error is also less than the other three models,which verifies the effectiveness of this method for tool state prediction.
Bidirectional long short memory networkTool statusPrincipal components analy-sisPrediction accuracy
NETL
NSTL
万方数据
