V.I. Goncharenko - Dr. Sc. (Eng.), Associate Professor, Director of Military Institute, Moscow Aviation Institute. E-mail: vladimirgonch@mail.ru D.S. Kucheryavenko - Ph.D. (Eng.), Senior Research Scientist, Military Academy of Strategic Rocket Forces. E-mail:d_kucheryavenko@mail.ru V.K. Goydenko - Engineer, Moscow Aviation Institute (National Research University). E-mail:assvard@mail.ru N.A. Skorik - Student, Moscow Aviation Institute (National Research University). E-mail:nikola.hrestofor@gmail.com

Сreation of analytical models for describing the systems behavior and simulation of complex nonlinear systems in practice is a difficult task. Since it is necessary to describe the state of complex systems, which include an unmanned spacecraft (USC), while having a small amount of information at the same time. For example, it is necessary to assess the state of the unmanned spacecraft at high uncertainty. There are several alternative solutions in this area allowing the accident type recognition in the complex and uncertain systems. The wavelet transformation for the simulation of nonlinear dynamic systems and diagnosis of their state is the most promising. The purpose of conducted research is to develop an algorithm and intellectualized software and hardware for the accident type recognition based on experimental data of the UCS flight tests. It is proposed to solve the problem by a sequence of steps. 1. Create a priori dictionary of accident type and the regular USC behavior. 2. Determine the equations of the classifier for each element of the dictionary. 3. Obtain wavelet coefficients for the telemetry data. 4. Search informative features that reflect the common properties of accident type. 5. Determine the type of accident by information features. In this paper, the wavelet transform is regarded as a tool that can get the feature space for further development of classification algorithms pulse signals. Selecting the discrete wavelet transform for solving the problems of recognition and classification is due to the versatility of the wavelet analysis, its ability to adapt to the shape of the signal, the similarity of the signals from the basis functions (wavelets). Computational simulation with noisy reference signals for each accident type was made to assess the noise immunity of a classification algorithm. This reference signal formed by averaging the training samples and additive noise with a normal dis-tribution and known variance were used. The experimental results show that the probability of detection P = 0.5 is observed in relation to the signal / noise ratio S = 2/3 dB for all types. Error-free detection was obtained for S = 16/18 dB. This indicates a good noise immunity of the algorithm for all accident types. So, the proposed method allows us to solve the problem of accident type recognition for unmanned spacecraft using a priori information about its structural and dynamic characteristics based on discrete wavelet transform. The implementation of this me-thod allows us to detect the accidents occurrence using the telemetry information and problem-oriented systems.

 

1. Betanov V.V., Lysenko JI.H., Lysenko I.V., Rjapolov S.I. EHksperimentalnaja ballistika raketno-kosmicheskikh sredstv / Pod red. L.N. Lysenko, V.V. Betanova, I.V. Lysenko. M.: VA RVSN im. Petra Velikogo. RARAN. 2000. 286 s.
2. Voskobojjnikov JU.E. Ustojjchivye metody i algoritmy parametricheskojj identifikacii. Novosibirsk: NGASU. 2006. 182 s.
3. Alkhaf M. Nader.Algoritmicheskoe obespechenie procedur ocenivanija i identifikacii kharakteristik spuskaemogo apparata na ehtape letnykh ispytanijj // Vestnik MGTU. Mashinostroenie. 2004. № 4. S. 27-41.
4. Ivanov V.EH. Algoritmy i sredstva klassifikacii monoimpulsnykh signalov na osnove diskretnogo vejjvlet-preobrazovanija v informacionno-izmeritelnykh sistemakh: Diss. ... kand. tekhn. nauk. KHabarovsk. 2007. 161 s. // Nauchnaja biblioteka dissertacijj i avtoreferatov http://www.dissercat.com/content/algoritmy-i-sredstva-klassifikatsii-monoimpulsnykh-signalov-na-osnove-diskretnogo-veivlet-pr#ixzz3uVa4zHgw.
5. Benevolskijj S.V., Goncharenko V.I., Kucherjavenko D.S. Opredelenie kinematicheskikh parametrov dvizhenija letatelnykh apparatov na osnove filtracii telemetricheskikh izmerenijj // Vestnik Kazanskogo gosudarstvennogo tekhnicheskogo universiteta im. A.N. Tupoleva. 2012. № 1. S. 10-17.
6. Tabyrca D.V., Popov S.A., Malisov N.P. Sintez algoritma raspoznavanija izobrazhenija na osnove vejjvlet-preobrazovanija // Vestnik VGTU. 2010. № 10. S.81-86.
7. Astafeva N.M.Vejjvlet-analiz: osnovy teorii i nekotorye prilozhenija // UFN. 1996. № 11. S. 1145-1170.
8. Dremin I.M., Ivanov O.V., Nechitajjlo V.A.Vejjvlety i ikh ispolzovanie // UFN. 2001. T. 171. № 5.
9. JAkovlev A. N. Vvedenie v vejjvlet-preobrazovanie: Ucheb. posobie. Novosibirsk: Izd-vo NGTU. 2003. 104 s.
10. Goncharenko V.I.Klassifikacija traektorijj letatelnykh apparatov s pomoshhju metodov iskusstvennogo intellekta // Informacionno-izmeritelnye i upravljajushhie sistemy. 2012. T. 10. № 4. S. 54-60.
11. Vorobev V.I., Gribunin V.G.Teorija i praktika vejjvlet-preobrazovanija. SPb.: VUS. 1999.
12. Baryshev P.F., Rozhnov A.V., Gubin A.N., Lobanov I.A. Obosnovanie informacionno-analiticheskojj sistemy v razvitii metodov i modelejj soglasovanija ierarkhicheskikh reshenii // Dinamika slozhnykh sistem - XXI vek. 2014. № 3. S. 43-52.
13. Rozhnov A.V., Antiokh G.M., Seliverstov D.E., Kublik E.I. Sistemnaja integracija napravlenijj nauchnojj dejatelnosti v uslovijakh formirovanija predyntellektualnojj infrastruktury // Informacionno-izmeritelnye i upravljajushhie sistemy. 2014. № 11. S. 59-63.