To improve the performance of fuel cells, the operating temperature of molten carbonate fuel cells (MCFC) stack should be controlled within a specified range. With the RBF neural network's ability of identifying the complex non-linear system, a neural networks identification model of MCFC stack is developed based on the input-output sampled data. An online fuzzy control procedure for the temperature of MCFC stack is also developed based on the fuzzy genetic algorithm (FGA), the fuzzy controller's parameters and rules are optimized at the same time. Finally using the neural networks model as the real MCFC stack, the control simulation is carried out. The validity of the identification modeling of MCFC stack and the superior performance of the fuzzy controller are demonstrated by the simulation results.
[1]
Erkki Oja,et al.
Rival penalized competitive learning for clustering analysis, RBF net, and curve detection
,
1993,
IEEE Trans. Neural Networks.
[2]
Dingli Yu,et al.
Selecting radial basis function network centers with recursive orthogonal least squares training
,
2000,
IEEE Trans. Neural Networks Learn. Syst..
[3]
Gerrit Kateman,et al.
Optimization of calibration data with the dynamic genetic algorithm
,
1992
.
[4]
Lalit M. Patnaik,et al.
Adaptive probabilities of crossover and mutation in genetic algorithms
,
1994,
IEEE Trans. Syst. Man Cybern..