The Investigating of Influence of Quality Criteria Coefficients on Global Complex Models
In the paper global modeling of complex systems with regard to quality of local models of simple plants are discussed. Complex systems consist of several sub-systems. As a global model multilayer feedforward neural networks were used. It is desirable to obtain an optimal global model, as well as optimal local models. A synthetic quality index as a sum of a global quality criterion and local quality indexes is defined. By optimization of the synthetic quality index the global model is obtained with regard to the quality of local models of sub-systems. The quality index of the global model contains coefficients which define the participation of the local criteria in the synthetic quality criterion. The investigation of influence of these coefficients on the quality of the global model of the complex static system is discussed. The investigation is examined by a complex system which consists of two nonlinear simple plants.
Keywordscomplex system neural networks global model
Unable to display preview. Download preview PDF.
- 2.Dahleh, M.A., Venkatesh, S.R.: System Identification of Complex Systems. IEEE Proceedings of Problem Formulation and Results 3, 2441–2446 (1997)Google Scholar
- 3.Dralus, G., Swiatek, J.: A modified backpropagation algorithm for modelling static complex systems using neural network. In: Proceedings of 5th International Conference: Neural Network and Soft Computing, Zakopane, pp. 463–468 (2000)Google Scholar
- 4.Dralus, G., Swiatek, J.: Static neural network in global modelling of complex systems. In: The 14th International Conference on Systems Engineering, Coventry, pp. 547–551 (2000)Google Scholar
- 5.Dralus, G., Swiatek, J.: Global network modeling of complex systems with respect of local models quality. In: Proceedings of Fifteenth International Conference on System Engineering ICSE 2002, August 6-8, 2002, pp. 218–226. Univ. of Nevada, Las Vegas (2002)Google Scholar
- 6.Jozefczyk, J.: Decision making problems in complex of operations systems. Wroclaw University of Technology Press, Wroclaw (2002) (in Polish)Google Scholar
- 7.Riedmiller, M., Braun, H.: RPROP – a fast adaptive learning algorithm. Technical Report, University Karlsruhe (1992)Google Scholar