Title :
Neural networks control by immune network algorithm based auto-weight function tuning
Author :
Kim, Dong Hwa ; Lee, Kyu Young
Author_Institution :
Dept. of Instrum. & Control Eng., Hanbat Nat. Univ., Seoul, South Korea
fDate :
6/24/1905 12:00:00 AM
Abstract :
A weight function auto-tuning scheme for neural networks has been suggested by an immune algorithm for nonlinear processes. A number of neural network structures are considered as learning methods for control systems. A general view is provided that they are special cases of either the membership functions or modifications of the network structure in the neural networks. On the other hand, since the immune network system possesses a self-organizing and distributed memory, it is thus adaptive to its external environment and allows a PDP (parallel distributed processing) network to complete patterns against the environmental situation. Also, it can provide the optimal solution. Simulation results by immune-based weight tuning reveal that an immune algorithm is an effective approach to search for optimal or near-optimal process control
Keywords :
adaptive control; content-addressable storage; evolutionary computation; learning (artificial intelligence); learning systems; neural net architecture; neurocontrollers; optimal control; process control; self-organising feature maps; tuning; adaptive system; automatic weight function tuning; control system learning methods; external environment; immune network algorithm; membership functions; neural network control; neural network structures; neurocontrol; nonlinear processes; optimal process control; parallel distributed processing; self-organizing distributed memory; simulation; Artificial neural networks; Biological neural networks; Control systems; Immune system; Intelligent networks; Learning systems; Neural networks; Neurons; Process control; Supervised learning;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007734
