WWW.DISSERS.RU


...
    !

Pages:     | 1 |   ...   | 8 | 9 || 11 | 12 |

caoocaoa cooyoc ocoooaax e po, ooa a coceo poopeeco ocoe oppoa ey o ypeey ecy apy oeoo pa. Ceo oo oop o cyecoa aye yx paopax apa acceco ooo.

pya ocoe ooo a, o ec oaax oaoo cpyypo eoa ooy cocox aoecoa py c pyo, apep, cpeac aa ooco.

(o.) epaa xapaepca peaoax opao cocooce pooco opyae cpee. aco oeaec a oe paoe, o ec oeco ox o pepoyoo opaca ooo, ocaex ao oco. B ooo oepoa oecea xapaepca, oaaa, acoo yceo oco peae ocaey aay, ooa cooca ee o ooe c py oco.

poecc, p oo oopoo eoope oco oy opac oye o x ooca. Ooeo a ae peoe o ee yao pcocoeoc.

(o.) oa eoe H, oapyex eax.

Xpooco pcycy o cex eax opaa, xo oo eoa x ac aa ao-o opeo ee. B ooo oepoa o xpoocoo oa pae ax, coepa coe apaep. O oe pecae e apo cpo eoceoo acca.

p oppoa ceyeo ooe oy, p oopo xpooco aoee pcocoex ocoe eyeo ooe opyc ceyee ooee, e oepac ec eeecx oepaopo. apapye coxpaee oy cepxo, epexox ooe ooee, o, a pao, ycope poee oy, oa peepeeoe.

apaee aeaeco oepoa, oeee oepe eo oepoa o, a ae opocee o coy acoa e (eopeeca oo, ec aoa cyecye) pye apae pceco popapoa. Bae ce a pae eemuecue aopum, ouoe cmpameuu, ouooe npopaupoaue, ucyccmee epoe cemu, eemy ouy.

ccea, aa pa oopo opaec a p ocnpouocma, ueuocmu, copeoau u omopa.

oa epa ceo poecce.

MEHHO CCO o Theodor Bischoff Bepoc P. J. Werbos Bc R. J. Williams B Whytt Bo K. C. Wong o Camillo Golgi ya E.D. Goodman eap Rene Descartes Kenneth De Jong ao F. Jacob aa Pierre Cabanis Maao W. S. McCulloch Mao J. Monod Mc M. L Minsky Hapepa K. P. Narendra eep S. A. Papert c W. A. Pitts Pao--axa Ramon y Cajal Poea F. Rosenblatt Pyexap D. E. Rumelhart Xe D. O. Hebb Xo G. E. Hinton Xoa J. H. Holland Xoa R. B. Hollstein ee H. P. Schwefel eppo Charles Sherrington CCO TEPATP 1. Holland J. H. Adaptation in natural and artificial systems. An introductory analysis with application to biology, control, and artificial intelligence. London: Bradford book edition, 1994 211 p.

2. De Jong K. A. Genetic Algorithms: A 10 Year Perspective //In: Procs of the First Int. Conf. on Genetic Algorithms, 1985. pp.167177.

3. O eoopx eopeecx aceax ooecoo oopa coco ppoo cpe /. A. pa p. //B .: poe ooecoo oopa oepoae occe. To II. .: poeeoa, 1979. cc.729.

4. Aoaa ocooo ocpypoa (cycce ee ao poepoa) /o pe. A. . ooa. M.: Pao c, 1981. 344 c.

5. Ceec . H., Ceeo C. M. Maeaecoe oepoae ooecx poecco. .: poeeoa, 1982. 326 c.

6. Aleksander I., Morton H. An Introduction to Neural Computing. London: Chapman&Hall, 1990.

7. McCulloch W. S., Pitts W. A logical calculus of ideas imminent in nervous activity // Bulletin Mathematical Biophysics. 1943. 5. pp.115 133 (eec epeo: . Maao, . c. oecoe ccee e, oocxc epo eeoc // B .: Aoa. M.: , 1956).

8. Hebb D.O. The Organization of Behavior: A Neuropsychological Theory. NewYork: Wiley, 1949.

9. Rosenblatt F. The perceptron: A probabilistic model for information storage and organization in the brain // Psychological Review. 1958. 65. pp.386407 ( : Poea . p epoa. M.: Mp, 1966. 480 c.).

10. Minsky M. L., Papert S. A. Perceptrons. Cambridge, MA: MIT Press, 1969 (eec epeo: Mc M., eep C. epcepo. M.:

Mp, 1971. 261c.).

11. Rumelhart D. E., Hinton G. E., Williams R. J. Learning internal representation by error propagation // In: D.E.Rumelhart and J.L.McClelland (Eds.) Parallel Distributed Processing, Vol. I Foundations. Cambridge, MA: MIT Press, 1986. pp.318362.

12. Rumelhart D. E., Hinton G. E., Williams R. J. Learning representation by backpropagating errors // Nature. 1986. vol.323. pp.533536.

13. Werbos P. J. Beyond regression: New tools for prediction and analysis in the behavioral sciences. PhD Thesis, Harvard University, Cambridge, MA, 1974.

14. ape C. ., Oxo B. A. Aae ce opao opa. pacopc, 1986. 20 c. (pep / AH CCCP. C. o e. ; N 59).

15. ape C. ., e C. E., Oxo B. A. p oceoc opaa aax cee opao opa // B .: aa xecx ooecx cce. Hoocpc: Haya.

C. oe, 1989. c.655.

16. Haykin S. Neural Network. A Comprehensive Foundation. New York: Macmillan College Publishing Company, 1994. 691 p.

17. Schaffer J. D., Whitley D., Eshelman L. J. Combinations of Genetic Algorithms and Neural Networks: A Survey of the State of the Art //In: Procs.

Of the Int. Workshop on Combinations of Genetic Algorithms and Neural Networks (Eds. L. D. Whitley, J. D. Schaffer). Baltimore, Maryland, 1992. pp.137.

18. Hornik K., Stinchcomb M. and White H. Multilayer Feedforward Networks are Universal Approximators // Neural Networks. 1989. N 2. pp.359366.

19. Cybenko G. Approximation by superposition of sigmoidal functions // Math. Control, Signal Syst. 1989. N 2. pp.303314.

20. Park J., Sandberg I. W. Universal approximation using radial basis function networks // Neural Computation. 1991. vol.3. pp.246 257.

21. Heap . ., oa H. ., Caee B.. aece oe eop ypae. M.: Haya. aa pea o aeaeco epayp, 1985. 400 c.

22. Narendra K. S., Parthasarathy K. Identification and control of dynamical systems using neural networks // IEEE Trans. on Neur. Net. 1990. vol.1. N 1. pp. 427.

23. eo B. ., Maxoo . B., Cepee C. A., Bopooc. .

cyccee epoe ce: oa apaa ypae //B.:

poe aoapoaoo epopoa. Teop paa (o pe. B. . eoa p.). Xapo: Ocoa, 1995. cc.111 115.

24. Sergeev S. A., Klepikov V. B., Mahotilo K. V. AmplitudeFrequency Characteristic of a Neural Control Based DC Drive // Tagungsband Leistungselektronische Aktoren und intelligente Bewegungssteuerungen Fachtagung. Magdeburg, Deutschland, 1996. pp.4753.

25. Klepikov V. B., Sergeev S. A., Mahotilo K. V., Voronovsky G. K. Neural Technologies in Electrical Drive Control //In: II Konferencja Sterowanie w Energoelektronice i Napedzie Elektrycznym. LodzArturowek, 15listopada 1995. pp.336343.

26. Cepee C. A., Maxoo . B. eeece aop cee poapaex epox cee // Tpy XIII e. o. Hoe opaoe exoo aye, opaoa ece.

paa, p, aypy, 1524 a 1996. cc.338342.

27. Sergeev S. A., Mahotilo K. V. Evolutionary Synthesis of Dynamical Object Emulator Based on RBF Neural Network //In: Procs of the First Online Workshop on Soft Computing. Nagoya, Japan, Aug.1930, 1996. pp.3136.

28. eepo C. C. Pao o oe oo eee. Hoocpc, 1984.

29. Klepikov V. B., Sergeev S. A., Mahotilo K. V. Modification of Holland's reproductive plan for diploid populations // In: Artificial Neural Nets and Genetics Algorithms (Eds. D.Pearson et al). Springer Verlag, 1995. pp.337339.

30. Klepikov V. B. et al. Diploidybased Genetic Algorithm in Nonstationary Environment // B .: poe aoapoaoo epopoa. Teop paa. (o pe. B..eoa p.). Xapo: Ocoa, 1995. cc.108110.

31. Hollstein R. B. Artificial genetic adaptation in computer control systems (Doctoral dissertation, 1971, University of Michigan) Dissertation Abstracts International. 32.(3), 1510B.

32. NG K.P., Wong K.C. A New Diploid Scheme and Dominance Change Mechanism for NonStationary Function Optimization //Procs of the 6th Int. Conf. on Genetic Algorithms. Morgan Kaufmann, 1995. pp.159 166.

33. Voronovsky G. K., Klepikov V. B., Sergeev S. A., Mahotilo K. V.

ANN+GAbased Intelligent Control System // The 4th Int. Workshop on Optimization and Inverse Problem in Electromagnetism. Brno, Czech Republic, June 1921, 1996. p.45.

34. Voronovsky G. K., Petrashev S. N., Sergeev S. A., Mahotilo K. V.

ANNBased Intelligent System for Process Control // In: Procs of the First Int. Conf. on Evolutionary Computation and Its Applications EvCA96.

Moscow, Russia, June 2427, 1996,. pp.330334.

35. Websters school dictionary. MA, USA: MeriamWebster Inc. 1167 p.

36. Wedel E., Romanov A. RussianEnglish, EnglishRussian Dictionary. e: epy, 1993. 592 c.

37. Aopycc ec coap (o pe. .M. Tocoo). M.: o Pycc , 1978. 848 c.

38. o P. oc e: ep. c a. M.: Mp, 1993. 318 c.

POEHE 1. PETAT CHTEA AHA APAMETPECO BCTBTEHOCT HEPOOHTPOEPA Oe ypae epooe eo opoo opa (8).

apaep aecoo oea: T = 0,5, = 0,1.

e ypae oaee oeaeoc oea.

ao ypae: uk +1 = NC(xk, xk -1,urk ).

cpeoc ypae: = 0,1c.

n Aaoa y epoo cpoo co: y = th xi + w0.

w i i=oeco epoo cpo coe: 5.

Pc. 1.1 Cpyypa epoopoPc. 1.2 Cxea pao epoopoepa a ae oocooo epepa epoa NN35Taa 1.apaep epoopoepa apaep Hoep epoa cpoo co Bxoo epo 1 2 3 4 5 w0 0,1989 0,2478 0,1999 0,2038 0,0269 0,w1 0,5797 0,2454 0,7576 0,4604 0,0244 1,w2 0,9609 0,9433 0,7498 0,2493 0,5230 3,w3 0,5347 0,6090 0,1281 0,8592 0,7810 9, w4 5,w5 6,a) ) ) ) Pc. 1.3 Pea cce ypae (pa 2) c epoopoepo a xooe oece (pa 1) e apoecoo caa c eo ayo acoo a) 0; ) 0,08; ) 0,16 ) 0,32 oe apaepeco yceoc cce ee epa yoa aeca:

tmax N 1 J(x)= (x(t)- ur(t))2dt.

N tmax i= ec N=3 oeco cepeo, tmax=10 c pe eppoa epexooo poecca ccee, ur cyeaoe xooe aae ayo 1, 0 1. aee yoaa, oyeoe p oao acpoe epoopoepa, pao J=0,0235.

Ha pc. 1.4 pecae pa ee yoaa aeca p ee aoo 26 apaepo epoopoepa aaoe 100 100%. a o, eee oo 26 apaepo peeax 20 20% e po cyeceoy yxye aeca pao cce (caeca oa 5%, epepeypoae 15%, 1,22 oea epexooo poecca). Tao oyc aao pea apaepo o aaoo oaoo ae oe oe oecee p cooa caapx oypooox popo.

Pc. 1.4. acoc aeca pao cce ypae o ee apaepo epoopoepa Te e eee, pc. 1.4 o, o pae apaep epoo ce oaa paoe e a eoc ee pao. Ta, eee apaepo oep 10, 11 24 (eca epo opo ce peeo epoa cpoo co ec c c xooo epoa, cooeceo) ae aoe apye paoe epoopoepa. Bo e pe 100%oe eee 4, 17 22 apaepo (ec pee c epoo ceee oo epoo cpoo co ec epo c xooo epoa, cooeceo) paec e caaec a eoc ypae.

Ha pc. 1.5 pecaeo ceeco epexox y cce ypae p ee ae 11o apaepa epoopoepa aaoe o 100% o 100%. ee peeax 50% oaa e ocoo a caecy oy, oe e ee apya ae aece coca cce, yea oeaeoc. Ceeco, pecaeoe a pc. 1.6, oaae, o eee 17o apaepa caaec a caeco oe cce.

Pc. 1.5. epexoe y cc- Pc. 1.6. epexoe y cce ypae p pax e ypae p pax eex 11o apaepa eex 17o apaepa epoopoepa epoopoepa oo oe ee, ax appoae ae acpoex apaepo epoopoepa, oo epec pecae aa a eo paoy aap e opa ecox ce, xoe cpo epoo opa opax ce ce cce ypae.

Ha pc. 1.7 pecae pa ee yoaa aeca pao cce ypae p oceoaeo, pooo ope, xoe cpo ecox coee (ca o opeex ce cac ye). aoap oy, o opa epeaec opaaaec epoa apaeo, p opax ecox ce aaec e pea yep, a oceea epaau paoococooc cce.

Pages:     | 1 |   ...   | 8 | 9 || 11 | 12 |



2011 www.dissers.ru -

, .
, , , , 1-2 .