Simulation and Prediction for a Satellite Temperature Sensors Based on Artificial Neural Network

Spacecrafts in the space environment are exposed to several kinds of thermal sources such as radiation, albedo and emitted IR from the earth. The thermal control subsystem in the spacecraft is used to keep all parts operating within allowable temperature ranges. A failure in one or many temperature sensors could lead to an abnormal operation. Consequently, a prediction process must be performed to replace the missing data with estimated values to prevent abnormal behavior. The goal of the proposed model is to predict the failed or missing sensor readings based on artificial neural networks (ANN). It has been applied to EgyptSat-1 satellite. A backpropagation algorithm called Levenberg-Marquardt is used to train the neural networks (NN). The proposed model has been tested by one and two hidden layers. Practical metrics such as mean square error, mean absolute error and the maximum error are used to measure the performance of the proposed network. The results showed that the proposed model predicted the values of one failed sensor with adequate accuracy. It has been employed for predicting the values of two failed sensors with an acceptable mean square and mean absolute errors; whereas the maximum error for the two failed sensors exceeded the acceptable limits.

[1]  M. K. Soni,et al.  Artificial neural network based peak load forecasting using Levenberg-Marquardt and quasi-Newton methods , 2002 .

[2]  Stef van Buuren,et al.  Flexible Imputation of Missing Data , 2012 .

[3]  Senol Gulgonul,et al.  Battery Thermal Design Conception of Turkish Satellite , 2008 .

[4]  Dario Izzo,et al.  Artificial Intelligence for Space Applications , 2007 .

[5]  Héctor Javier Vergara-Hernández,et al.  Artificial Neural Networks to Estimate the Thermal Properties of an Experimental Micro-Alloyed Steel and Their Application to the Welding Thermal Analysis , 2015 .

[6]  Huiying Liang,et al.  Machine learning applications for prediction of relapse in childhood acute lymphoblastic leukemia , 2017, Scientific Reports.

[7]  Marcin Kaminski.  Neural estimators of two-mass system optimized using the Levenberg-Marquardt training and genetic algorithm , 2016, 2016 21st International Conference on Methods and Models in Automation and Robotics (MMAR).

[8]  Nedim Sozbir,et al.  Analytical investigation of a nanosatellite panel surface temperatures for different altitudes and panel combinations , 2015 .

[9]  Jia Liu,et al.  Modeling and analysis of MEMS-based cooling system for nano-satellite active thermal control , 2008, 2008 2nd International Symposium on Systems and Control in Aerospace and Astronautics.

[10]  Qu Hongquan,et al.  Particle Filtering approach to parameter estimate and temperature prediction of satellite , 2008, 2008 7th World Congress on Intelligent Control and Automation.

[11]  David G. Gilmore,et al.  Spacecraft Thermal Control Handbook, Volume I: Fundamental Technologies , 2002 .

[12]  J. Graham,et al.  Missing data analysis: making it work in the real world. , 2009, Annual review of psychology.

[13]  Manjula Sutagundar,et al.  Artificial Neural Network modeling of MEMS cantilever resonator using Levenberg Marquardt algorithm , 2016, 2016 2nd International Conference on Applied and Theoretical Computing and Communication Technology (iCATccT).

[14]  Wenjing Li,et al.  An improved Levenberg-Marquardt algorithm with adaptive learning rate for RBF neural network , 2016, CCC 2016.

[15]  J. Schafer,et al.  Missing data: our view of the state of the art. , 2002, Psychological methods.

[16]  M. Donabedian,et al.  Spacecraft Thermal Control Handbook, Volume II: Cryogenics , 2004 .

[17]  Shu Yang.  Flexible Imputation of Missing Data, 2nd ed. , 2019, Journal of the American Statistical Association.

[18]  Yan Liu,et al.  Recurrent Neural Networks for Multivariate Time Series with Missing Values , 2016, Scientific Reports.

[19]  Volker Tresp,et al.  A Solution for Missing Data in Recurrent Neural Networks with an Application to Blood Glucose Prediction , 1997, NIPS.

[20]  David S. Fung,et al.  Methods for the estimation of missing values in time series , 2006 .

[21]  Limei Song,et al.  Temperature error compensation for open-loop fiber optical gyro using back-propagation neural networks with optimal structure , 2014, Proceedings of 2014 IEEE Chinese Guidance, Navigation and Control Conference.

[22]  Ana Maria Ambrosio,et al.  Real-Time Cubesat Thermal Simulation using Artificial Neural Networks , 2017 .

[23]  N Shivashankarappa,et al.  Data prediction in Wireless Sensor Networks using Kalman Filter , 2015, 2015 International Conference on Smart Sensors and Systems (IC-SSS).

[24]  Volker Tresp,et al.  Nonlinear Time-Series Prediction with Missing and Noisy Data , 1998, Neural Computation.

[25]  M. Napolitano,et al.  Neural-network-based scheme for sensor failure detection, identification, and accommodation , 1995 .

[26]  Vanathi Gopalakrishnan,et al.  An Overview and Evaluation of Recent Machine Learning Imputation Methods Using Cardiac Imaging Data , 2017, Data.

[27]  Gustavo E. A. P. A. Batista,et al.  An analysis of four missing data treatment methods for supervised learning , 2003, Appl. Artif. Intell..

[28]  Senol Gulgonul,et al.  Thermal Control Design of TUSAT , 2008 .