Article

Article title NEURAL NETWORK TECHNOLOGIES, FUZZY CLUSTERING AND GENETIC ALGORITHMS IN EXPERT SYSTEM
Authors N.A. Polkovnikova, V.M. Kureichik
Section SECTION I. EVOLUTIONARY MODELLING, GENETIC AND BIONIC ALGORITHMS
Month, Year 07, 2014 @en
Index UDC 004.891
DOI
Abstract The paper shows the development of fuzzy expert system model for fault identification of complex technical objects using data mining technology based on searching hidden patterns in database. Data mining technology allows optimizing database processing queries that retrieve the required information from the actual data in order to detect important patterns. Application of neural network technology allows to detect nonlinear dependencies of input and output data, to improve quality of object diagnostic process, which ultimately will reduce the number of accidents in operation. A method for allocating the optimal number of fuzzy clusters in the space of training examples and deducing the parameters of membership functions for input variables and output results is proposed. Considered a neuro-fuzzy clustering algorithm for multidimensional objects with incomplete and fuzzy initial information. Implementation of neural network technologies in an expert system for solving diagnosis problems will allow not only fixing the sensor readings and compare them with the reference values, but also produce a comprehensive analysis of the obtained parameters of object, predicting the possibility of failure occurrence as in separate ele- ments, and in system overall. The architecture of such expert system allows moving from normal monitoring to «information monitoring» in the specialized intelligent human-machine systems.

Download PDF

Keywords Expert system; knowledge base; fuzzy logic; linguistic variable; clustering; genetic algorithms; neural networks.
References 1. Polkovnikova N.A., Kureichik V.M. Razrabotka modeli ekspertnoy sistemy na osnove nechetkoy logiki [Development of an expert system model based on fuzzy logic], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2014, No. 1 (150), pp. 83-92.
2. Polkovnikova N.A., Kureichik V.M. Ob intellektualnom analize baz dannykh dlya ekspertnoy sistemy [Intellectual analysis of the data bases for expert systems], Informatika, vyhchislitelnaya tekhnika i inzhenernoe obrazovanie [Computer science and engineering education], 2013, No. 2 (13), pp. 39-50.
3. Yarushkina N.G. Osnovy teorii nechetkikh i gibridnykh sistem: Ucheb. posobie [Fundamentals of the theory of fuzzy and hybrid systems: textbook]. Moscow: Finansy i statistika, 2004, 320 p.
4. Silich V.A., Silich M.P., Aksenov S.V. Algoritm postroeniya nechetkoy sistemy logicheskogo vyvoda Mamdani, osnovannyy na analize plotnosti obuchayushchikh primerov [The algorithm of construction of fuzzy system logical conclusion Mamdani, based on the analysis of the density of training examples], Upravlenie, vychislitelnaya tekhnika i informatika. Doklady Tomskogo gosudarstvennogo universiteta sistem upravleniya i radioelektroniki [Management, computer engineering and Informatics. The presentations of Tomsk state University of control
systems and Radioelectronics], 2013, No. 3 (29), pp. 76-82.
5. Altukhov A.V. Formirovanie nechetkikh pravil tipa Takagi-Sugeno po rezultatam nechetkoi klasterizatsii [The formation of fuzzy rules of the type Takagi-Sugeno on the results of fuzzy clustering], Vestnik VGU: Sistemnyy analiz i informatsionnye tekhnologii [Herald of the Voronezh state University: System analysis and information technologies], 2008, No. 1, pp. 44-50.
6. Yarushkina N.G., Chekina A.V. Klasterizatsiya informatsionnykh resursov na osnove geneticheskogo algoritma [Clustering of information resources based on genetic algorithm], Avtomatizatsiya processov upravleniya [Automation of Management Processes], 2010, No. 4, pp. 66-71.
7. Demidova L.A., Konyaeva E.I. Klasterizatsiya obektov s ispolzovaniem FCM-algoritma na osnove nechetkikh mnozhestv vtorogo tipa i geneticheskogo algoritma [Clustering of objects using FCM algorithm based on fuzzy sets of the second type and genetic algorithm], Vestnik RGRTU [Herald Ryazan State Radio University], 2008, No. 4 (26), pp. 46-54.
8. Demidova L.A., Titov S.B. Podkhod k probleme nechetkoy klasterizatsii v usloviyakh neopredelennosti vybora tselevoy funktsii [Approach to the problem of fuzzy clustering in conditions of uncertainty of the choice of target functions], Vestnik RGRTU [Herald Ryazan State Radio University], 2009, No. 3 (29), pp. 54-60.
9. Gladkov L.A., Kureychik V.V., Kureychik V.M., Sorokoletov P.V. Bioinspirirovannye metody v optimizatsii [Bioinspiration methods in optimization]. Moscow: Izdatelskaya firma "Fizikomatematicheskaya literatura", 2009, 381 p.
10. Wang Y. Fuzzy clustering analysis by using genetic algorithm, ICIC Express letters, 2008. Vol. 2, No. 4, pp. 331-337.
11. Maulik U., Bandyopadhyay S. Genetic algorithm-based clustering technique, Pattern recognition, 2000, No. 33, pp. 1455-1465.
12. Mukhopadhyay A., Maulik U., Bandyopadhyay S. Multiobjective genetic algorithm-based fuzzy clustering of categorical attributes, 2009, Vol. 13, No. 5, pp. 991-1005.
13. Shi Y., Mizumoto M. An improvement of neuro-fuzzy learning algorithm for tuning fuzzy rules, Fuzzy sets and systems, 2001, Vol. 118, No. 2, pp. 339-350.
14. Zou K., Hu J., Kong X. The structure optimized fuzzy clustering neural network model and its application, International journal of innovative computing: information and control, 2008, Vol. 4, No. 7, pp. 1627-1634.

Comments are closed.