We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
NEURAL NETWORKS BASED CONTROL OF NONLINEAR PROCESS.
- Authors
SRINIVASAN, ANDY; N., SOWRIRAJAN
- Abstract
The artificial neural network based Predictive and Inverse model controller schemes are applied to the simulation of the time-dependent behavior of an isothermal Continuous Stirred Tank Reactor (CSTR). In the first scheme, a methodology is proposed for training and prediction of dynamic behavior of isothermal CSTR using feedforward neural network. Then a nonlinear one step predictive control strategy based on identified model is proposed for CSTR control. In the second scheme, neural network based inverse model controller is presented. Here, the controller is a rearrangement of the plant model. The performances of both neural controllers are evaluated in simulation for both servo and regulatory problems and the results are compared with PID controller.
- Subjects
SERVOMECHANISMS; FEEDFORWARD neural networks; ARTIFICIAL neural networks; PID controllers; FORECASTING; PREDICTION models
- Publication
i-Manager's Journal on Instrumentation & Control Engineering, 2019, Vol 7, Issue 4, p18
- ISSN
2321-113X
- Publication type
Article
- DOI
10.26634/jic.7.4.16816