Trend-GRU Model based Time Series Data Prediction in Melt Transport Process

Temperature and pressure are two important performance indicators during the melt transport of polyester fiber production, which can affect the overall properties of the melt. Therefore, the accurate prediction of these two indicators is crucial for the control of melt properties. This paper proposes a data prediction model, Trend-GRU (Gated Recurrent Unit), which can extract the feature of unstable change of the melt data. On the premise that the model is not over-complicated, a new structure is designed to extract the feature of unstable change to improve the prediction accuracy. Two data sets on temperature and pressure collected from the actual production process of a spinning factory are used for comparative experiments. The results show that the accuracy of the data prediction of the proposed model is better than the original one.

[1]  Yoshua Bengio,et al.  Drawing and Recognizing Chinese Characters with Recurrent Neural Network , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Marcel van Gerven,et al.  Modeling the Dynamics of Human Brain Activity with Recurrent Neural Networks , 2016, Front. Comput. Neurosci..

[3]  Wei Wang,et al.  Adaptive Granulation-Based Prediction for Energy System of Steel Industry , 2018, IEEE Transactions on Cybernetics.

[4]  Bogdan Gabrys,et al.  Data-driven Soft Sensors in the process industry , 2009, Comput. Chem. Eng..

[5]  Witold Pedrycz,et al.  Granular Model of Long-Term Prediction for Energy System in Steel Industry , 2016, IEEE Transactions on Cybernetics.

[6]  Yoshua Bengio,et al.  Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation , 2014, EMNLP.

[7]  Lei Wang,et al.  Prediction of northern summer low-frequency circulation using a high-order vector auto-regressive model , 2016, Climate Dynamics.

[8]  Witold Pedrycz,et al.  Hybrid Neural Prediction and Optimized Adjustment for Coke Oven Gas System in Steel Industry , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[9]  Tong Wang,et al.  Using gated recurrence units neural network for prediction of melt spinning properties , 2017, 2017 11th Asian Control Conference (ASCC).

[10]  Fathi M. Salem,et al.  Gate-variants of Gated Recurrent Unit (GRU) neural networks , 2017, 2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS).

[11]  H. Tong,et al.  Threshold Autoregression, Limit Cycles and Cyclical Data , 1980 .

[12]  Yildiray Turhan,et al.  Comparison of high-volume instrument and advanced fiber information systems based on prediction performance of yarn properties using a radial basis function neural network , 2013 .

[13]  Witold Pedrycz,et al.  Effective Noise Estimation-Based Online Prediction for Byproduct Gas System in Steel Industry , 2012, IEEE Transactions on Industrial Informatics.

[14]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[15]  Kwok-wing Chau,et al.  Improving Forecasting Accuracy of Annual Runoff Time Series Using ARIMA Based on EEMD Decomposition , 2015, Water Resources Management.

[16]  Chen Jie,et al.  Wind speed forecasting using nonlinear-learning ensemble of deep learning time series prediction and extremal optimization , 2018, Energy Conversion and Management.

[17]  Jian Cao,et al.  Estimation of Optimal Blank Holder Force Trajectories in Segmented Binders Using an ARMA Model , 2003 .

[18]  Dexian Huang,et al.  Data-driven soft sensor development based on deep learning technique , 2014 .