NEUROGENETIC LEARNING ALGORITHM WITH APPLICATION TO REAL CHEMICAL PLANT

Computers & Systems Dept., Faculty of Engineering, Mansoura University

: A neurocontroller system was applied to the common Tennessee Eastman plant with the help of

genetic algorithm to optimize the neural network parameters. The present algorithm differs from

standard genetic algorithms in that it evolves a population of neurons instead of complete

networks. These neurons are combined through different subpopulations to form hidden layer of

feed-forward networks that are then evaluated on a given problem. The proposed plant is a well

posed problem for analysis and control design of a nonlinear, open-loop unstable chemical

process. The plant consists of five major operations, a reactor, a product condenser, a

vapor/liquid separator, a recycle compressor and a product stripper. The nonlinear dynamics of

the plant are mainly due to the chemical reactions within the reactor. The neurocontroller

developed was found to be robust and presents high performance during set point changes and in

the presence of disturbances