On-line pattern discovery in telemetry sequence of micro-satellite

Abstract In micro-satellite engineering, telemetry data is the only basis for the ground staffs to judge the condition of spacecraft in-orbit, and is also a critical reference for testing, operation, and Prognostics and Health Management (PHM). In practice, the analysis and processing of raw telemetry data is very complicated and cumbersome. To maintain the micro-satellites' performance and ensure their reliability, it is important to implement on-line interesting pattern discovery in the ground station monitoring system or Electrical Ground Support Equipment (EGSE) system. The goal of this paper is to monitor evolving telemetry sequence of micro-satellite, and to discover subsequences that are similar to the given query sequence, under the Dynamic Time Warping (DTW) distance. However, in the processing of telemetry sequence, massive amounts of data arrive continuously and it is infeasible to directly use DTW distance. Therefore, this paper improves the DTW and proposes a novel method (eDTW) that can perform pattern discovery on telemetry sequence online. First, the sub-sequence time warping matrix is constructed by taking the telemetry sequence as the X-axis and the query sequence as the Y-axis; Second, calculate the DTW distance from each unit, and the initial position of each path; Finally, determine the current local optimal subsequence and determine the next starting point of the next discovery. Experiments with telemetry sequence of Tsinghua University's NS2 satellite show that eDTW can discover interesting pattern quickly, with no buffering of stream values and without comparing pairs of streams. Our experimental case studies show that eDTW can incrementally capture correlations and discover trends, efficiently and effectively.

[1]  Piotr Indyk,et al.  Identifying Representative Trends in Massive Time Series Data Sets Using Sketches , 2000, VLDB.

[2]  Asok Ray,et al.  Anomaly Detection in Aircraft Gas Turbine Engines , 2006, J. Aerosp. Comput. Inf. Commun..

[3]  Eamonn J. Keogh,et al.  Finding surprising patterns in a time series database in linear time and space , 2002, KDD.

[4]  Fatemeh SalarKaleji,et al.  A survey on Fault Detection, Isolation and Recovery (FDIR) module in satellite onboard software , 2013, 2013 6th International Conference on Recent Advances in Space Technologies (RAST).

[5]  Yuji Sakamoto,et al.  Operation system for micro and nano satellites by low-cost ground station network , 2016 .

[6]  Felix Siegle,et al.  FDIR techniques for payload streaming applications using SpaceWire-based networks , 2014, 2014 International SpaceWire Conference (SpaceWire).

[7]  Mingliang Suo,et al.  Data-driven fault diagnosis of satellite power system using fuzzy Bayes risk and SVM , 2019, Aerospace Science and Technology.

[8]  Amir Shakouri,et al.  Fault Detection and Isolation of Satellite Gyroscopes Using Relative Positions in Formation Flying , 2018, Aerospace Science and Technology.

[9]  Hans-Peter Kriegel,et al.  LOF: identifying density-based local outliers , 2000, SIGMOD 2000.

[10]  Jianmin Wang,et al.  An anomaly detection approach for multiple monitoring data series based on latent correlation probabilistic model , 2015, Applied Intelligence.

[11]  Robert X. Gao,et al.  Deep learning and its applications to machine health monitoring , 2019, Mechanical Systems and Signal Processing.

[12]  Jimeng Sun,et al.  Streaming Pattern Discovery in Multiple Time-Series , 2005, VLDB.

[13]  Christos Faloutsos,et al.  Stream Monitoring under the Time Warping Distance , 2007, 2007 IEEE 23rd International Conference on Data Engineering.

[14]  Wei Xie,et al.  Fragment Anomaly Detection With Prediction and Statistical Analysis for Satellite Telemetry , 2017, IEEE Access.

[15]  Feng Han,et al.  Attitude tracking control for a space moving target with high dynamic performance using hybrid actuator , 2018, Aerospace Science and Technology.

[16]  Xibin Cao,et al.  Flexible platform based micro-satellite design method , 2016 .

[17]  Søren Vind,et al.  Motif trie: An efficient text index for pattern discovery with don't cares , 2018, Theor. Comput. Sci..

[18]  Marco Vieira,et al.  Applying Data Mining for Detecting Anomalies in Satellites , 2012, 2012 Ninth European Dependable Computing Conference.

[19]  Mehmet Yüceer.  Emerging technologies for communication satellite payloads , 2012 .

[20]  Y. Kawahara,et al.  Telemetry-mining: a machine learning approach to anomaly detection and fault diagnosis for space systems , 2006, 2nd IEEE International Conference on Space Mission Challenges for Information Technology (SMC-IT'06).

[21]  Silvio Simani,et al.  Robust FDI applied to thruster faults of a satellite system , 2010 .

[22]  Amirreza Kosari,et al.  “Wasted performance” minimization of the multi-purpose mini-satellite platform for an EO mission using a dynamic simulation-based model , 2018 .

[23]  Asok Ray,et al.  Anomaly detection in flight recorder data: A dynamic data-driven approach , 2013, 2013 American Control Conference.

[24]  Yong-Ping Zhao,et al.  Retargeting extreme learning machines for classification and their applications to fault diagnosis of aircraft engine , 2017 .

[25]  Yuefeng Li,et al.  Effective Pattern Discovery for Text Mining , 2012, IEEE Transactions on Knowledge and Data Engineering.

[26]  José M. Martínez.  New Telemetry Monitoring Paradigm with Novelty Detection , 2012 .

[27]  Bassem R. Mohammad,et al.  A novel approach of health monitoring and anomaly detection applied to spacecraft telemetry based on PLSDA multivariate latent technique , 2014, 15th International Workshop on Research and Education in Mechatronics (REM).

[28]  Shijie Xu,et al.  Active vibration suppression in flexible spacecraft with optical measurement , 2016 .

[29]  Paul Zetocha,et al.  Comparison of AI technologies for satellite anomaly FDIR , 1999 .