Possibility of Dynamical Process Linearization Using Piecewise-Linear Neural Network
This paper presents a new technique for process identification. Since once nonlinear problem is described by piecewise-linear structure, it may be solved by many well-known techniques, the result of introduced technique provides a set of linear equations, where each of these equations is valid in some region of state space and together, they approximate whole nonlinear process. In the first five paragraphs, the technique is theoretically described and the paper is finished with demonstrative example.
Keywordsartificial neural network linearization piecewise-linear neural network
Unable to display preview. Download preview PDF.
- 4.Astrom, K., Hagglund, T.: PID controllers: theory, design and tuning. International Society for Measurement and Control, Durham (1995)Google Scholar
- 5.Ljung, L.: System identification - Theory for the User. Prentice-Hall, New Jersey (1999)Google Scholar
- 6.Bobal, V., Bohm, J., Fessl, J., Machacek, J.: Digital Self-tuning Controllers. Springer, London (2005)Google Scholar