The Hybrid Optimization Algorithm Application Research On Thermal System Identification

System identification is an important content and foundation of control theory. Traditional system identification was all about linear system, and it is still too difficult for complex nonlinear system identification. The rising of neural network gives us a new way on system identification, and it soon becomes to be the main way on nonlinear system problems. RBF neural network is a kind of feed forward network, and it has good approximating and global optimization abilities. This paper is based on RBF neural network system identification, and a new hybrid optimization algorithm is raised to confirm the amount and location of network hidden layer nodes, and it has do a great job on network optimization to meet the requirements of speed and precision on system identification.