MULTI-AGENT APPROACH IN PREDICTION OF RELIABILITY PARAMETERS FOR ELECTRONIC MODULES

I. B. Bondarenko, A. I. Ivanov


Read the full article 
Article in Russian


Abstract

The paper deals with design principles of the multi-agent system for prediction and calculation of reliability parameters of heterogeneous, diversified electronic modules. One-parameter models of failures distribution are less accurate and do not take into account the specifics of failures occurring in the base of modules. That’s why the issue has for an object the possibility to use two-parametric models of the timing intensity distribution of failures and determination of parameters for models on the basis of equipment test results on actuating factors. Prediction of reliability parameters for modules is carried out by means of numerous agents, united in a system. It includes heterogeneous agents carrying out information exchange with the user. Developed system structure is presented and the work of separate subsystems is described. The facility to add new agents-prediction techniques is available and they form a library. Prediction results depend on the chosen calculation method of reliability parameters on a specific enterprise, therefore, the developed prediction system use both standard distribution of reliability parameters and received by the method of group assessment of arguments. Selection of models for prediction is done on the basis of the calculated values of standard deviation. The output of work results is performed in a graphical form. Testing carried out on the data received during the electronic module checkout makes it possible to confirm the correctness and effectiveness of the developed approach. The developed system will give the possibility to reduce time for checkout and processing of testing results for complex electronic modules, especially manufactured at the enterprises of electronic industry for the first time.


Keywords: multi-agent system, LN-, DM- and DN-distributions, reliability, electronic modules, group assessment of arguments, tests

References
 
1.     Bondarenko I.B., Kalyaeva E.A., Koksharov D.N. Adaptаtsiya parametrov geneticheskogo algoritma dlya optimizatsii slozhnykh funktsii [Adaptation of parameters of genetic algorithm for complex function optimization]. Izv. vuzov. Priborostroenie, 2011, vol. 54, no. 9, pp. 5–9.
2.     Bondarenko I.B., Gatchin Yu.A., Trofimova E.Yu. Prognozirovanie nadezhnosti uzlov pechatnykh plat pri uskorennykh ispytaniyakh [Reliability prediction of PCB assemblies with accelerated tests]. Trudy kongressa po intellektual'nym sistemam i informatsionnym tekhnologiyam (IS-IT’12)[Proc. of the Congress on Intelligent Systems and Information Technology (IS-IT'12)]. Moscow, FIZMATLIT Publ., 2012, vol. 2, pp. 92–97. (In Russian)
3.     Romanov V. Kolichestvennaya otsenka nadezhnosti integral’nykh mikroskhem s uchetom matematicheskoi modeli otkazov [Quantitative assessment of the reliability of integrated circuits based on mathematical models of failures]. Electronnye komponenty i sistemy, 2005, no. 4, pp. 4–7.
4.     Wooldridge M. An introduction to multiagent systems. 2nd ed. Wiley, NY, 2009, 484 p.
5.     Russel S., Norving P. Artificial intelligence. A modern approach. 2nd ed. Prentice Hall Publ., 2010, 1152 p.
6.     Gaston M.E., Desjardins M. The effect of network structure on dynamic team formation in multi-agent systems. Computational Intelligence, 2008, vol. 24, no. 2, pp. 122–157. doi: 10.1111/j.1467-8640.2008.00325.x
7.     Barton L. & Allan V.H. Methods for coalition formation in adaptation-based social networks. Lecture Notes in Computer Science, 2007, vol. 4676, pp. 285–297. doi: 10.1007/978-3-540-75119-9_20
8.     Shoham Y., Leyton-Brown K. Multiagent systems: Algorithmic, game-theoretic, and logical foundations. Cambridge: Cambridge University Press, 2008, 532 p.
9.      Dignum F., Bradshaw J., Silverman B., van Doesburg W. Agents for games and simulations: Trends in techniques, concepts and design. Springer, 2010, 273 p.
10.  deWeerdt M.M., Zhang Y., Klos T. Multiagent task allocation in social networks. Autonomous Agents and Multi-Agent Systems, 2012, vol. 25, no. 1, pp. 46–86. doi: 10.1007/s10458-011-9168-3
11.  Gatchin Yu.A., Bondarenko I.B., Solov’ev D.V. Primenenie mnogoagentnogo podkhoda pri prinyatii optimal’nykh reshenii [Application of multi-agent approach for optimal decisions creation]. Trudy kongressa po intellektual'nym sistemam i informatsionnym tekhnologiyam (IS-IT’13). [Proc. of the Congress on Intelligent Systems and Information Technology (IS-IT'13)]. Moscow, FIZMATLIT Publ., 2013, vol. 1, pp. 19–22. (In Russian)
12.  Воndarenko I.B., Korobeynikov A.G., Prokhozhev N.N. Mikhailichenko O.V. Prinyatie tekhnicheskikh reshenii s pomoshch’yu mnogoagentnykh sistem [Adoption of technical solutions using multi-agent systems]. NB: Kibernetikaiprogrammirovanie, 2013, no 1, pp. 16–20.
13.  RD 50-690-89. Metodicheskie ukazaniya. Nadezhnost’ v tekhnike. Metody otsenki pokazatelei nadezhnosti po eksperimental’nym dannym[Guidance document 50-690-89. Methodical instructions. Reliability in engineering. Estimation methods for reliability by experimental data]. Moscow, Izdatel'stvostandartovPubl., 1990. 133 с.
14.  GOST 27.410-87. Nadezhnost’ v tekhnike. Metody kontrolya pokazatelei nadezhnosti i plany kontrol’nykh ispytanii na nadezhnost’[State Standard 27.410-87. Reliability in engineering. Control methods of reliability and monitoring plans for reliability tests]. Moscow, Izdatel'stvo standartov Publ., 1987, 79 p.
15.  Rumbell T., Barnden J., Denham S., Wennekers T. Emotions in autonomous agents: comparative analysis of mechanisms and functions. Autonomous Agents and Multi-Agent Systems, 2012, vol. 25, no. 1, pp. 1–45. doi: 10.1007/s10458-011-9166-5
16.  Strogonov A. Otsenka dolgovechnosti BIS po resultatam uskorennykh ispytanii [Estimation of durability of microcircuits based on the results of accelerated tests]. Tekhnologii v elektronnoi promyshlennosti, 2007, no. 3, pp. 90–96.
17.  Zelenkov A.A., Golik A.P. Otsenka nadezhnosti bortovoi avioniki na osnove DN-raspredeleniya [Evaluation of avionic’s reliability using DN-based distribution]. Elektronika ta sistemy upravleniya, 2009, no. 2 (20), pp. 12–17.
18.  Strel’nikov V.P., Antipenko K.A. O metodicheskikh pogreshnostyakh prognozirovaniya resursa vysokonadezhnykh izdelii elektronnoi tekhniki [On methodological errors of resource forecasting at highly reliable electronic products]. Matematicheskie mashiny i sistemy, 2004, no. 3, pp. 164–167.
19.  RD 26.260.004-91.Metodicheskie ukazaniya. Prognozirovanie ostatochnogo resursa oborudovaniya po izmeneniyu parametrov ego tekhnicheskogo sostoyaniya pri ekspluatatsii [Guidance document 26.260.004-91. Methodical instructions. Prediction of residual life of equipment to change the parameters of its technical condition during operating]. Moscow, Izdatel'stvostandartovPubl., 1991,38 с.
20.  Воndarenko I.B., Gatchin Yu.A., Geranichev V.N. Sintez optimal’nykh iskusstvennykh neironnykh setei s pomoshch’yu modifitsirovannogo geneticheskogo algoritma [Synthesis of optimal artificial neural networks by modified henetic algorithm]. Scientific and Technical Journal of Information Technologies, Mechanics and Optics,2012, no 2 (78), pp. 51–55.
21.  GOST 25359-82. Izdeliya elektronnoi tekhniki. Obshchie trebovaniya po nadezhnosti i metody ispytanii[State Standard 25359-82. Electronic products. General requirements for reliability and test methods]. Moscow, Izdatel'stvo standartov Publ., 1982, 7 p.
22.  Ivanov A.I., Bondarenko I.B. Modelirovanie nadezhnosti elektronnoi tekhniki pri uslovii edinichnykh otkazov [Modeling reliability of electronic equipment subject to single failures]. Materialy VIII mexhdunarodnoi nauchno-practicheskoi konferentsii «Obrazovanie i nauka v ХХI veke–2012»[Proc. VIII international scientific and practical conference “Education and science at XIX century - 2012”]. Sofia: 2012, pp. 49–54.
Copyright 2001-2017 ©
Scientific and Technical Journal
of Information Technologies, Mechanics and Optics.
All rights reserved.

Яндекс.Метрика