Multivariate multichannel software and measurement complex for detecting anomalous states of natural and technical objects and systems

A.V. Skatkov, A.A. Bryukhovetskiy, D.V. Moiseev

Sevastopol State University, RF, Sevastopol, Universitetskaya St., 33

E-mail: dmitriymoiseev@mail.ru

DOI: 10.33075/2220-5861-2021-2-119-130

UDC 004.56

Abstract:

   The state of the environment is an integral, key component of the generalized category of the quality of life of the population. In this regard, there is an objective need to develop methods and tools designed to implement a system of continuous monitoring of the key environmental indicators and forecasting the occurrence of abnormal states of ecosystems. In this paper, we consider an approach to multivariate classification of the states of natural and technical objects and systems, based on the development of methods for dynamic detection of anomalies in information data flows. The approach is based on an estimate of the statistical discrepancy between the probability distributions of random variables over variably variable time intervals, as well as an estimate of the probabilities of errors of the first and second kind.

   The structure of a multichannel software and measurement complex for detecting anomalous states of natural-technical objects and natural-technical systems is proposed, and the results of model calculations are presented. It is assumed that it is possible to initially form reference versions of the states of natural-technical objects and natural-technical systems based, for example, on expert assessments or a priori information. The use of a multivariate approach allows you to optimize the processing, analysis, and integration of heterogeneous data. The results of the study confirm the stability and sensitivity of the method when selecting threshold values of intervals that determine the states of objects.

   This model can be used in other subject areas where the assessment of the dynamic parameters of controlled objects is required, for example, when detecting vulnerabilities in the interfaces of unmanned vehicles in the infrastructure of a smart city.

Keywords: anomaly detection, multivariate model, statistical estimates, approximation function, errors of the first and second kind.

To quote: Skatkov, A.V., A.A. Bryukhovetskiy, and D.V. Moiseev. “Multivariate Multichannel Software and Measurement Complex for Detecting Anomalous States of Natural and Technical Objects and Systems.” Monitoring Systems of Environment no. 2 (June 24, 2021): 119–130. doi:10.33075/2220-5861-2021-2-119-130.

Full text in PDF(RUS)

REFERENCES

  1. Gajskij V.A. and Gajskij P.V. Mnogomernyj garmonicheskij analiz Fur’e pri izmereniyah polej morskoj sredy (Multidimensional harmonic analysis for measurements of marine environment) Monitoring systems of environment, 2019, No. 4 (38), pp. 33–42.
  2. Bryuhoveckij A.A., Skatkov A.V., Shishkin Yu.E. Modelirovanie processov obnaruzheniya anomalij v slozhnostrukturirovannyh dannyh monitoring (Modeling of anomaly detection processes in complex structured monitoring data). Sistemy kontrolya okruzhayushchej sredy, 2017, No. 9 (29), pp. 45–49.
  3. Bondura V.G. Aerokosmicheskij monitoring ob”ektov neftegazovogo kompleksa (Aerospace monitoring of oil and gas facilities). Moscow: Nauchnyj mir, 2012, 558 p.
  4. Skatkov A.V., Bryuhoveckij A.A., and Moiseev D.V. Intellektual’naya sistema monitoringa dlya resheniya krupnomasshtabnyh nauchnyh zadach v oblachnyh vychislitel’nyh sredah (Intelligent monitoring system for large-scale tasks in cloud environments). Informacionno-upravlyayushchie sistemy, 2017, No. 2 (87), pp. 19–25.
  5. Chandola V., Banerjee A., and Kumar V. Anomaly detection: a survey. ACM Computing Surveys, 09 2009. pp. 1–72.
  6. Chan P.K. and Mahoney M.V. Modeling multiple time series for anomaly detection. In Proceedings of the Fifth IEEE International Conference on Data Mining. IEEE Computer Society, Washington, USA, 2005, pp. 90–97.
  7. Agarwal D. Detecting anomalies in cross-classified streams: a bayesian approach. Knowledge and Information Systems, 2006. Vol. 11, No. 1. pp. 29–44.
  8. Charikar M., Chekuri C., Feder T., and Motwani R. Incremental clustering and dynamic information retrieval. SIAM Journal on Computing, 2004, Vol. 33, No. 6, pp. 1417–1440.
  9. Kul’bak S. Teoriya informacii i statistika. Moscow: Nauka, 1967, 408 p.
  10. Skatkov A.V., Bryuhoveckij A.A., and Moiseev D.V. Mera Kul’baka v zadachah dinamicheskoj klasterizacii nablyudenij sostoyaniya okruzhayushchej sredy (Measure of Kulbak in problems of dynamic clustering of observations of the environmental state). Sistemy kontrolya okruzhayushchej sredy, 2019, No. 3 (37). pp. 35–38.
  11. Orlov S.P. and Vasilchenko A.N. Intelligent measuring system for testing and failure analysis ofelectronic devices. 2016 XIX IEEE International Conference on Soft Computing and Measurements (SCM). IEEE Conference Publications, 2016, Vol. 1, pp. 401–403.
  12. Skatkov A.V., Bryuhoveckij A.A., and Moiseev D.V. Adaptivnyj metod obnaruzheniya uyazvimostej interfejsov bespilotnyh transportnyh sredstv v infrastrukture umnogo goroda (An adaptive method for detecting vulnerabilities in the interfaces of unmanned vehicles in the infrastructure of a smart city). Infokommunikacionnye tekhnologii, 2020, Vol. 18, No. 1, pp. 45–50.

 

Loading