System Identification of an Industrial Steam Generator of Power Plant Using Wavelet Neural Networks

In heat generation process at a power plant, performance improvement is a critical factor, to achieve this objective it is essential to have an accurate model for the steam generator of power plant. The model of a system can be represented by a mathematical model based on the physical principles that govern the problem, or by using the experimental data measured in the system, which is called system identification. The problem of identification consists of choosing anidentification model and adjusting the parameters such that the response of the model approximates the response of the real system to the same input. Nowadays, soft computing techniques such as neural networks (NNs) and wavelet neural networks (WNNs) have become very effective tools for identification of nonlinear plants. In this paper, we present a WNN model to identify an industrial nonlinear multivariable steam generator. Simulation results show that the model provides an accurate representation of the real system and demonstrate the effectiveness of the proposed method.