Prediction of outlet pressure for sulfur dioxide blower based on ARX model and adaptive Kalman filter

The sulfur dioxide blower refers to a centrifugal blower that transports various gases in the sulfuric acid production process from flue gases. Accurately predicting the outlet pressure of the sulfur dioxide blower is significant for the sulfuric acid production process from flue gases. Due to the complex internal structure of the sulfur dioxide blower, it is difficult to establish a precise mechanism model. In this paper, a novel hybrid algorithm combining Autoregressive exogenous (ARX) model and Sage-Husa adaptive Kalman filter is used to establish the sulfur dioxide blower model and predict its outlet pressure. Where the Akaike Information Criterion (AIC) is used to determine the order of the ARX model, and the least square method is used to determine the ARX model parameters. Considering the high-order ARX model parameter estimation is difficult to calculate, the optimal ARX model is determined in the low-order range, and the Kalman equation of state and observation equation are constructed using this model. By combining the ARX model and the Sage-Husa adaptive Kalman filter, experiment shows that the proposed algorithm obtains better prediction effect than the traditional time series model combined with Kalman filter.