A novel axle temperature forecasting method based on decomposition, reinforcement learning optimization and neural network

Abstract Axle temperature forecasting technology is important for monitoring the status of the train bogie and preventing the hot axle and other dangerous accidents. In order to achieve high-precision forecasting of axle temperature, a hybrid axle temperature time series forecasting model based on decomposition preprocessing method, parameter optimization method, and the Back Propagation (BP) neural network is proposed in this study. The modeling process consists of three phases. In stage I, the empirical wavelet transform (EWT) method is used to preprocess the original axle temperature series by decomposing them into several subseries. In stage II, the Q-learning algorithm is used to optimize the initial weights and thresholds of the BP neural network. In stage III, the Q-BPNN network is used to build the forecasting model and complete predicting all subseries. And the final forecasting results are generated by combining all prediction results of subseries. By comparing all results over three case predictions, it can be concluded that: (a) the proposed Q-learning based parameter optimization method is effective in improving the accuracy of the BP neural network and works better than the traditional population-based optimization methods; (b) the proposed hybrid axle temperature forecasting model can get accurate prediction results in all cases and provides the best accuracy among eight general models.

[1]  Jianzhou Wang,et al.  Short-term wind speed prediction using empirical wavelet transform and Gaussian process regression , 2015 .

[2]  Hongguang Li,et al.  An enhanced empirical wavelet transform for noisy and non-stationary signal processing , 2017, Digit. Signal Process..

[3]  Trevor S. Slack,et al.  A Review of Rolling Contact Fatigue , 2009 .

[4]  Ling Shao,et al.  Hyperparameter Optimization for Tracking with Continuous Deep Q-Learning , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[5]  Cheng Liu,et al.  Research and application of ensemble forecasting based on a novel multi-objective optimization algorithm for wind-speed forecasting , 2017 .

[6]  Xinhong Hei,et al.  A Prediction Method Based on Stepwise Regression Analysis for Train Axle Temperature , 2016, 2016 12th International Conference on Computational Intelligence and Security (CIS).

[7]  Nhat-Duc Hoang,et al.  Prediction of soil compression coefficient for urban housing project using novel integration machine learning approach of swarm intelligence and Multi-layer Perceptron Neural Network , 2018, Adv. Eng. Informatics.

[8]  Kai-Ying Chen,et al.  Applying back propagation network to cold chain temperature monitoring , 2011, Adv. Eng. Informatics.

[9]  Jin Huang,et al.  LSTM-Based Temperature Prediction for Hot-Axles of Locomotives , 2017 .

[10]  Andrew Ball,et al.  Modern techniques for condition monitoring of railway vehicle dynamics , 2012 .

[11]  Dietmar Stephan,et al.  Ensemble data mining modeling in corrosion of concrete sewer: A comparative study of network-based (MLPNN & RBFNN) and tree-based (RF, CHAID, & CART) models , 2020, Adv. Eng. Informatics.

[12]  Javier García,et al.  Probabilistic Policy Reuse for inter-task transfer learning , 2010, Robotics Auton. Syst..

[13]  Chun Zhang,et al.  Design of a 0.8V low power CMOS temperature sensor for RFID-based train axle temperature measurement , 2010, 2010 10th IEEE International Conference on Solid-State and Integrated Circuit Technology.

[14]  Antoine Tahan,et al.  A comparative study between empirical wavelet transforms and empirical mode decomposition methods: application to bearing defect diagnosis , 2016 .

[15]  Yitao Liu,et al.  Deep learning based ensemble approach for probabilistic wind power forecasting , 2017 .

[16]  Haiping Wu,et al.  Wind speed forecasting models based on data decomposition, feature selection and group method of data handling network , 2019, Measurement.

[17]  Zhengjia He,et al.  Wheel-bearing fault diagnosis of trains using empirical wavelet transform , 2016 .

[18]  Chao Chen,et al.  Improved pollution forecasting hybrid algorithms based on the ensemble method , 2019, Applied Mathematical Modelling.

[19]  Andrew W. Moore,et al.  Reinforcement Learning: A Survey , 1996, J. Artif. Intell. Res..

[20]  Fei Huang,et al.  Brief Introduction of Back Propagation (BP) Neural Network Algorithm and Its Improvement , 2012 .

[21]  Handan Çamdevýren,et al.  Use of principal component scores in multiple linear regression models for prediction of Chlorophyll-a in reservoirs , 2005 .

[22]  Qiyang Liu.  High-speed Train Axle Temperature Monitoring System Based on Switched Ethernet , 2017 .

[23]  Jun Fan,et al.  An EMD-SVR method for non-stationary time series prediction , 2013, 2013 International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering (QR2MSE).

[24]  Yanni Peng,et al.  Short-term traffic volume prediction using GA-BP based on wavelet denoising and phase space reconstruction , 2020 .

[25]  Qing Wang,et al.  The detection of wind turbine shaft misalignment using temperature monitoring. , 2017 .

[26]  Jianzhou Wang,et al.  A hybrid technique for short-term wind speed prediction , 2015 .

[27]  Yanxi Zhang,et al.  Online Monitoring of Welding Status Based on a DBN Model During Laser Welding , 2019, Engineering.

[28]  Crina Grosan,et al.  Experienced Gray Wolf Optimization Through Reinforcement Learning and Neural Networks , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[29]  Zhe Wang,et al.  Solar Irradiance Short-Term Prediction Model Based on BP Neural Network , 2011 .

[30]  Li Guangyao,et al.  Optimized BP neural network for Dissolved Oxygen prediction , 2018 .

[31]  Haiyan Lu,et al.  A case study on a hybrid wind speed forecasting method using BP neural network , 2011, Knowl. Based Syst..

[32]  Leemon C. Baird,et al.  Residual Algorithms: Reinforcement Learning with Function Approximation , 1995, ICML.

[33]  Michael J. Devaney,et al.  Adjustable Speed Drive Bearing Fault Detection via Wavelet Packet Decomposition , 2006, 2006 IEEE Instrumentation and Measurement Technology Conference Proceedings.

[34]  Yan Li,et al.  Short-term electricity demand forecasting with MARS, SVR and ARIMA models using aggregated demand data in Queensland, Australia , 2018, Adv. Eng. Informatics.

[35]  Zoran Miljkovic,et al.  Neural network Reinforcement Learning for visual control of robot manipulators , 2013, Expert Syst. Appl..

[36]  Chao Chen,et al.  Multi-objective data-ensemble wind speed forecasting model with stacked sparse autoencoder and adaptive decomposition-based error correction , 2019, Applied Energy.

[37]  Wenyan Wu,et al.  Streamflow prediction using LASSO-FCM-DBN approach based on hydro-meteorological condition classification , 2020 .

[38]  Yanfei Li,et al.  Wind speed forecasting method based on deep learning strategy using empirical wavelet transform, long short term memory neural network and Elman neural network , 2018 .

[39]  Jianhui Lin,et al.  Faults Diagnostics of Railway Axle Bearings Based on IMF’s Confidence Index Algorithm for Ensemble EMD , 2015, Sensors.

[40]  Jing Chen,et al.  A Resource Demand Prediction Method Based on EEMD in Cloud Computing , 2018 .

[41]  Lu-Kai Song,et al.  Dynamic neural network method-based improved PSO and BR algorithms for transient probabilistic analysis of flexible mechanism , 2017, Adv. Eng. Informatics.

[42]  Maksym Spiryagin,et al.  An overview: modern techniques for railway vehicle on-board health monitoring systems , 2017 .

[43]  Yanfei Li,et al.  Big multi-step wind speed forecasting model based on secondary decomposition, ensemble method and error correction algorithm , 2018 .

[44]  Jérôme Gilles,et al.  Empirical Wavelet Transform , 2013, IEEE Transactions on Signal Processing.

[45]  Lin Hua,et al.  Back Propagation neural network modeling for warpage prediction and optimization of plastic products during injection molding , 2011 .

[46]  Milesa Z. Sreckovic,et al.  A Stationary System of Noncontact Temperature Measurement and Hotbox Detecting , 2008, IEEE Transactions on Vehicular Technology.

[47]  Jinglong Chen,et al.  Mono-component feature extraction for mechanical fault diagnosis using modified empirical wavelet transform via data-driven adaptive Fourier spectrum segment , 2016 .

[48]  Adam P. Piotrowski,et al.  Differential Evolution algorithms applied to Neural Network training suffer from stagnation , 2014, Appl. Soft Comput..