Információ feldolgozás intelligens irányítási rendszerekben  részletek

súgó  nyomtatás 
vissza »

 

Projekt adatai

 
azonosító
42741
típus K
Vezető kutató Vajk István
magyar cím Információ feldolgozás intelligens irányítási rendszerekben
Angol cím Information processing in intelligent conrol systems
zsűri Informatikai–Villamosmérnöki
Kutatóhely Automatizálási és Alkalmazott Informatika Tanszék (Budapesti Műszaki és Gazdaságtudományi Egyetem)
résztvevők Bars Ruth
Barta András
Hetthéssy Jeno
Tuschák Róbert
projekt kezdete 2003-01-01
projekt vége 2007-12-31
aktuális összeg (MFt) 7.389
FTE (kutatóév egyenérték) 0.00
állapot lezárult projekt





 

Zárójelentés

 
kutatási eredmények (magyarul)
A kutatás eredményei négy területre koncentrálódnak. Ezek: • Robusztus rendszeridentifikáció, modellezés sajátérték-sajátvektor bázisú megközelítésben, • Predikciós irányítási algoritmusok újszerű megközelítésben, • Objektum modellezés intelligens irányítási rendszerekben, valamint • Adatbányászati algoritmusok mint speciális tanulási algoritmusok kutatása. Néhány az elért eredményekből: Új identifikációs módszercsaládot dolgoztunk ki lineáris dinamikus rendszerek identifikációjára. A módszerek a szinguláris érték dekompozíció (SVD), illetve az általánosított SVD algoritmusokra épülnek. Megadtuk a módszerek direkt és indirekt megoldását. Megmutattuk, hogy a predikciós irányítási algoritmusok számítási igényét csökkenhetjük nem egyenlő közű predikciós pontok megválasztásával. Kimutattuk, hogy a predikciós számítási pontok exponenciális elrendezése rendkívül kedvező eredményt ad. A kutatást végeztünk egy egyszerűen kezelhető és tanítható gépi látórendszer kialakítása érdekében. Előírásokat adtunk meg egy általános képfeldolgozási rendszer kialakítására. Bayes-hálón alapuló algoritmust fejlesztettünk ki magas szintű alakfelismerési problémákra. Olyan adatbányászati algoritmusokat dolgoztunk ki, amelyek segítésével hatékonyan megtalálhatók a statisztikai értelemben gyakori elemhalmazok, továbbá gyakori szekvenciák, valamint az adatok között fellelhető kapcsolatok, szabályok.
kutatási eredmények (angolul)
Research results focus to the following four areas: • Robust system identification using eigenvalue-eigenvector based approach, • Predictive control algorithms in a novel approach, • Object modeling in intelligent control systems, as well as • Developing data mining algorithms as special learning algorithms. Some of the results are listed below: A novel class of identification algorithms has been elaborated to model linear dynamic system The new algorithms are based on singular value decomposition and (SVD), as well as on generalized singular value decomposition (GSVD) procedures. In terms of the predictive control algorithms it has been shown that the computing demand of the control algorithms can be reduced by applying non-equidistant point distribution along the prediction horizon. Specifically, it has been verified that arranging the prediction points along the prediction horizon according to an exponential law leads to a highly advantageous performance. Further research activity has been devoted to develop machine vision systems with easy use and training. To handle high-level pattern recognition problems a Bayesian-network based algorithm has been developed. Finally, data mining algorithms have been developed to efficiently discover frequent data patterns in statistical sense, as well as frequent sequential data sets and relations/rules existing among the data entries in the database.
a zárójelentés teljes szövege http://real.mtak.hu/775/
döntés eredménye
igen





 

Közleményjegyzék

 
Juhász S, Iváncsy R, Vajk I: Performance Modelling of the Apriori Association Rule Mining Algorithm, MicroCAD 2004 International Scientific Conference, University of Miskolc, March 18-19, Miskolc, 2004. pp. 205-210., 2004
Iváncsy R, Vajk I: Size Restricted Association Rule Mining, MicroCAD 2004 International Scientific Conference, University of Miskolc , March 18-19, Miskolc , 2004. pp. 191-196., 2004
Kovács F, Iváncsy R, Vajk I: Evaluation of the Serial Association Rule Mining Algorithms, 22ND IASTED International Conference on Databases and Applications, February 17-19, Innsbruck, Austria, 2004., 2004
Iváncsy R, Kovacs F,Vajk I: Dynamic Itemset Counting in PC Cluster Based Association Rule Mining, Proc. of Fourth International ICSC Symposium on Engineering of Intelligent Systems (EIS 2004), Island of Madeira, Portugal, Feb. 29-Mar. 2, 2004, 2004
Iváncsy R, Kovacs F, Vajk I: An Analysis of Association Rule Mining Algorithms, Proc. of Fourth International ICSC Symposium on Engineering of Intelligent Systems (EIS 2004), Island of Madeira, Portugal, Feb. 29-Mar. 2, 2004 ., 2004
Iváncsy R, Vajk I: Fast Discovery of Frequent Patterns in Market Basket Data, Proc. of 4th International Conference on Intelligent Systems Design and Applications (ISDA’04), Budapest , Hungary , Augustus 26-28. 2004. pp. 575-580., 2004
Haber R, Schmitz U, Bars R: Nonlinear Process Identification and Predictive Control by Multi-parameter Models, NOLCOS, 6th IFAC Symp. on Nonlinear Control Systems, 2004. Stuttgart, FRG. pp. 931-936., 2004
Preitl Z, Bars R, Haber R: Internal Model Control Structures Using Delta Domain Representation, Scientific Bulletin of „Politehnica” Univ. of Timisoara, Romania, Transactions on Automatic Control and Computer Science, Vol. 49(63) 2004, No.1, ISSN1224-600X, pp.29-34., 2004
Preitl Z, Bars R, Vajk I,Haber R: Internal Model Representation for Generalized Predictive Control with Constraint Handling, 4th International Conference on Systems Design and Application ISDA 2004, Budapest, 2004 aug. 26-28. pp. 681-685., 2004
Preitl Z, Bars R, Haber R: Case Study for Using Advanced Control Strategies for Level Control in a Two Tank System, IFAC Conference Advanced Control Strategies for Social and Economic Systems ACS 2004, Bécs, Ausztria, 2004 szept. 2-4. Paper No. ACS-002, 2004
Barta A, Vajk I: Structural Object Recognition by Probabilistic Feedback, International Conference on Computervision, Varsaw, Poland, 2004., 2004
Haber R, Schmitz U, Bars R: Optimal Choice of Horizons for Predictive Control by Using Genetic Algorithm, Annual Reviews in Control, Vol. 28, Issue 1, 2004, pp. 53-58., 2004
Ivancsy R, Vajk I: Finding the Size-Restricted Frequent Itemsets in Market Basket Data, Transactions on Automatic Control and Computer Science, Vol.49 (63) 2004 No. 3, ISSN 1224-600X, Editura Politechnica, Timisoara, Romania, 2004, pp. 39-44., 2004
Moore CF, Hetthéssy J,. Bars R: Control Modes – Closed-Loop Response, In: Béla G. Lipták. ed. Instrument Engineers’ Handbook, 4th edition, Process Control and Optimization, Vol. II., CRC Taylor & Francis, 2006, Chapter 2.5, pp.135-147, 2006
Preitl Zs, Bars R, Haber R:: An applied GPC cascade control solution for hydro-turbines Proceedings, 13th Workshop on Control Applications of Optimisation, CAO’2006, Paris-Cachan, France, April 26-28 2006: 171-176., 2006
Hetthéssy J, Bars R: Sampled Data Control Systems, In: Béla G. Lipták. ed. Instrument Engineers’ Handbook, 4th edition, Process Control and Optimization, Vol. II., CRC Taylor & Francis, Chapter 2.27, pp. 326-335, 2006
Arousi F, Bars R, Haber R:: Optimization of a PI controller for a predicted cost function assuming a first-order process with dead time, Proceedings of the 13th Workshop on Control Applications of Optimisation, CAO’2006, Paris-Cachan, France, April 26-28 2006: 412-417., 2006
Shinskey FG, Lipták BG, Bars R, Hetthéssy J: Control Systems – Cascade Loops, In: Béla G. Lipták. ed. Instrument Engineers’ Handbook, 4th edition, Process Control and Optimization, Vol. II., CRC Taylor & Francis, 2006, Chapter 2.6, pp.148-156, 2006
Vajk I, Hetthéssy J: Generalization of the Koopmans-Levin algorithm, Proceedings of the Workshop on System Identification and Control Systems. Budapest, 2006: 181-199., 2006
Vajk I., Hetthéssy J: Covariance matching criteria in EIV identification, 14th IFAC Symposium on System Identification, Newcastle, Australia, 2006: 184-189., 2006
Barta A., Vajk I.: Integrating low and high level object recognition steps, International Conference of Knowledge Mining, Prága, 2006, Febr. 28 - March 2. Computer Science Vol. 11: 126-132., 2006
Tóthfalussy B, Vajk I:: Framework for automatic composition of semantic web services, Proceedings of the 7th International Conference of Technical Informatics, CONTI’2006, Timisoara, Romania, 8-9 June, 2006: 195-199., 2006
Haber R, Schmitz U, Bars R:: Optimization strategies for predictive on-off control, Proceedings of the 13th Workshop on Control Applications of Optimisation, CAO’2006, Paris-Cachan, France, April 26-28 2006: 57-62., 2006
F. Arousi, R. Bars, R. Haber: Predictive PI controller based on first-order dead time models, ICCC’05, International Carpatian Control Conference, Lillafüred, May 24-27, 2005. Proc. ISBN 963-661-644-2, Vol.1, pp. 113-118., 2005
Zs. Preitl, R. Bars: Influence of tuning parameter variations upon GPC control loop, 1st IFAC Workshop on Advanced Control Circuits and Systems, ACCS’05, Cairo, Egypt, March 6-10, 2005., 2005
A. Barta, R. Bars, I. Vajk, Zs. Preitl: Practical controller design considering input saturation, ICCC’05, International Carpatian Control Conference, Lillafüred, May 24-27, 2005. Proc. ISBN 963-661-644-2, Vol.1, pp. 51-56., 2005
Zs. Preitl, R. Bars, I. Vajk, R. Haber: Cascade GPC with minimax optimal inner control loop, ICCC’05, International Carpatian Control Conference, Lillafüred, May 24-27, 2005. Proc. ISBN 963-661-644-2, Vol.1, pp. 323-328., 2005
R. Bars, P. Colaneri, C.E. de Souza, F. Allgöwer, A. Kleimenov, C. Scherer: Theory, Algorithms and Technology in the Design of Control Systems, Milestone Report of IFAC CC2 Coordinating Committee on Design Methods. 16th IFAC World Congress, Prague, July 3-8, 2005. Volume of Selected Plenaries, Milestones and Surv, 2005
Barta A., Vajk I.,: Document Processing with Bayesian Network and Agent-based Programming, Proceedings of the 5th WSEAS Int. Conf. on Signal, Speach and Image Processing, Corfu, Greece, pp. 265-269, 2005, 2005
Barta A., Vajk I.,: Document Image Analysis by Probabilistic Network and Circuit Diagram Extraction, Informatica - International Journal of Computing and Informatics, Issue 29, pp. 291-301, 2005, 2005
Barta A., Vajk I: Processing Circuit Diagrams with Belief Network and Intelligent Agents, WSEAS Transactions on Information Science and Applications, Issue 9, Vol. 2, September, pp. 1321-1329, 2005, 2005
Vajk I.: State space control, In Instrument engineers’handbook (IEH, Volume 2, 4th edition), Process control, Session 2.33. Edited B.G.Liptak, CRC Press, pp. 393-404. (könyvrészlet), 2006
Vajk I: Identification methods in a unified framework, Automatica 41(8) pp. 1385-1393., 2005
Iváncsy R., Vajk I.: Efficient sequential pattern mining algorithms, WSEAS Transactions on Computers, Issue 2, Vol.4, February 2005, ISSN: 1109-2750 pp. 96-101., 2005
Vajk I., Hetthéssy J.: Load forecasting using nonlinear modelling, Control Engineering Practice. Vol 13/7 2005. pp 895-902., 2005
Iváncsy R., Vajk I.: PD-tree: a new appoarch to subtree discovery, WSEAS Transactions on Information Sciences and Applications, 2(11), ISNN 1790-0832, pp. 1772-1779., 2005
Iváncsy R., Vajk I.: Efficient Sequential Pattern Mining Algorithms, In Proc. of the 4th WSEAS International Conference on Artificial Intelligence, Knowledge Engineering and Databases (AIKED 2005), Salzburg, Austria , Feb. 13 -15, 2005. IS, 2005
Iváncsy R., Juhász S., Vajk I.: Fast Discovery of Sequential Patterns by Level-Wise Approach, MicroCAD 2005 International Scientific Conference, University of Miskolc, March 10-11, Miskolc, 2005 pp.161-166, 2005
Ivancsy R., Vajk I.: Seqtreeminer: a new approach to sequence mining, International Carpathian Control Conference, ICCC’2005, Miskolc, pp.507-512., 2005
Iváncsy R., Vajk I.: Fast discovery of tree patterns, In Proc. of the 5th International Conference for PhD Students, Miskolc pp.67-72., 2005
Vajk I., Hetthéssy J.: Subspace identification methods: review and re-interpretation, The 5th International Conference on Control and Automation ICCA2005, Budapest (organized by IEEE) 2005. pp.113-118., 2005
Vajk I., Hetthéssy J.: On the generalization of the Koopmans-Levin estimation method, Proceedings of the 44th IEEE Conference on Decision and Control, and European Control Conference 2005, Seville, Spain, pp.4134-4139., 2005
Iváncsy R., Vajk I.: Efficient subtree inclusion testing in subtree discovering application, Proc. of the 5th WSEAS Int. Conf. on Signal, Speech and Image Processing, Corfu, Greece, pp. 265-269., 2005
Iváncsy R., Juhasz S, Vajk I.: Sequential pattern mining using finite state machine-based approach, In Proc. of 10th Panhellenic Conference of Informatics, Volos, Greece. pp.13-23., 2005
Iváncsy R., Vajk I.: Automata theory approach solving frequent pattern discovery problems, Transactions on Enformatika, Systems Sciences and Engineering. Vol.8 pp.203-208., 2005
Bars R, Colaneri P, de Souza CE, Dugard L, Kleimenov A, Scherer C: Theory, algorithms and technology in the design of control systems, Annual Reviews in Control, 2006 30: 619-30, 2006
Bars R, Preitl Zs.: Predikciós irányítási algoritmusok IMC reprezentációja, Irányítástudomány és Infokommunikáció konferencia, Győr, Széchenyi István Egyetem, 2005
Barta A., Vajk I.: Integrating Low and High Level Object Recognition Steps by Probabilistic Networks, International Journal of Information Technology, 2006, Volume 3 Number 1 ISSN 1305-2403 : 64-74., 2006
Iváncsy R., Vajk I.: Frequent pattern mining in web log data., Acta Polytechnica Hungarica, 2006, Vol. 3. Issue 1, ISSN 1785-8860 : 77-90., 2006
Iváncsy R, Vajk I.: A time- and memory-efficient frequent itemset discovering algorithm for association rule mining, Intern. Journal of Computer Applications in Technology, Special Issue on Data Mining Applications” by Inderscience Enterprises Ltd. Vol. 27. No. 4., 2006, pp. 270-2, 2006
Preitl Zs, Bars R: A Youla-parametrization approach for controller design based on ESO and 2E-SO methods for low order benchmarks, Studies in Informatics and Control. 2006 Sept.,Vol. 15, No.3. http://www.ici.ro/ici/revista/sic2006_3/index.html, 2006
Moore CF, Lipták BG, Bars R, Hetthéssy J: Control Modes – PID Controllers,, In: Béla G. Lipták. ed. Instrument Engineers’ Handbook, 4th edition, Process Control and Optimization, Vol. II., CRC Taylor & Francis, 2006, Chapter 2.2, pp.114-123, 2006
Nagy I, Bars R, Gertler J.: IFAC’50 Anniversary, Historic Control Textbooks, IFAC 1956-2006, Hungary, pp. 111-124, 2006
Barta A., Vajk I.: Adaptive multi-scale vision system for robotic applications,, IEEE International Conference on Mechatronics (ICM), Budapest, 2006, július 3-5.: 468-472., 2006
Arousi F, Bars R, Haber R:: Predictive PI(D) controllers based on first- and second-order models with dead time, In Proc. of the Workshop on System Identification and Control Systems in honor of László Keviczky on his 60th birthday, ed. Bokor J and Hangos KM, 2006: 165-180., 2006
Dávid Z, Vajk I.:: Applying reinforcement learning techniques in multiagent systems, MicroCAD 2006 International Scientific Conference, University of Miskolc, 2006: 55-60., 2006
Csorba K, Vajk I:: Supervised term clustering and confidence measurement in document clustering, Proceedings of the 7th International Conference of Technical Informatics, CONTI’2006, Timisoara, Romania, 2006: 183-188., 2006
Csorba K, Vajk I:: Greedy term selection for document classification with given minimal precision, Proceedings of the 7th International Symposium of Hungarian Researchers on Computational Intelligence, 2006: 247-254., 2006
Tuschák R, Vajk I, Bars R, Hetthéssy J, Barta A., Charaf H: Teaching basic control theory: a novel approach using web – matlab based teaching materials, In Innovations 2003: World innovations in engineering education and research. Edited by: W.Aung, M.H.W.Hoffmann, N.W.Jern, R.W.King, L.M.S.Ruiz, published by iNEER, Begel, 2003
István Vajk, Jenő Hetthéssy: Identification of a nonlinear errors-in-variables model, IIFAC Workshop on Control Application of Optimisation (CAO), 2003. Visegrád, Hungary, pp.21-26., 2003
István Vajk, Jenő Hetthéssy: Identification of nonlinear errors-in-variables models, Automatica, Vol 39., Issue 12, 2003. pp. 2099-2107, 2003
Barta A., Vajk I.: Controller design by direct numerical optimization, Prepr. of the 12th IFAC Workshop on Control Applications of Optimisation (CAO), Visegrád, Hungary, 30 June – 2 July, 2003. pp. 62-67, 2003
István Farkas, István Vajk: Experiments with robust internal model-based controller for Acurex field, The improving human potential programme. Research results at Plataforma Solar de Almeria within the year 2002 access campaign. In Serie Ponencias. Madrid, 2003. pp. 1-8., 2003
István Vajk: Identification methods in a unified framework, 13th IFAC Symposium on System Identification, Rotterdam, The Netherlands, 2003, pp.2010-2015., 2003
Barta A., Vajk I.: Structural, data driven edge detection, IASTED International Conference on Visualization, Imaging and Image Processing (VIIP 2003), pp.365-371, Spain, 2003., 2003
Preitl Zs, Bars R.: Control system design aspects using the delta transformation., Prepr. of the 12th IFAC Workshop on Control Applications of Optimisation (CAO), Visegrád, Hungary, 30 June – 2 July, 2003. pp. 249-254, 2003
Haber R, Schmitz U, Bars R.: Long-range optimal multi-step-ahead prediction identification for predictive control., 13th IFAC Symposium on System Identification (SYSID), Rotterdam, The Netherlands, 27-29 August, 2003. pp. 501- 506, 2003
Preitl Zs, Bars R, Vajk I, Haber R.: Hybrid IMC dead-beat controller design in delta domain, 2nd IFAC Conference on Control Systems Design, Bratislava, Slovak Republik, 7-10 September, 2003., 2003




vissza »