The immune clone evolutionary (ICE) algorithm is presented to optimize the parameters of the fuzzy neural network (FNN) controller for the complex billet heating process. First, the neural model of the process was set up via the observation data; then, based on the model, the rule optimization algorithm of the fuzzy neural network controller (FNNC) was introduced to obtain the optimal rule numbers of the FNNC. Finally, by combining the FNNC parameters obtained with the rule optimization into an individual in the initial population, the ICE algorithm was adopted to optimize the parameters of the FNNC. The proposed search strategy is a global optimization and local precision optimization algorithm. Results from simulations prove the effectiveness of the constructed system.