Furnace Temperature Prediction Using Optimized Kernel Extreme Learning Machine

In recent years, the combustion furnace has been widely applied in different fields of industrial technology, and the accurate prediction of furnace temperature can effectively help operators adjust combustion strategies to improve combustion utilization. However, furnace combustion is an extremely complex process, and its temperature change is affected by many related factors. To effectively predict the furnace temperature, a novel furnace temperature prediction method using optimized kernel extreme learning machine (OKELM) is proposed. Firstly, the optimized kernel extreme learning machine is used to establish the relationship between the furnace temperature and its related factors. Based on this, the continuous human learning optimization (CHLO) algorithm is adopted to optimize the kernel parameter and regularization coefficient, then the best OKELM with optimal parameters is adopted to predict the furnace temperature more precisely and effectively. Finally, the experiment results show that the proposed method outperforms state-of-the-art furnace temperature prediction approaches, providing high prediction accuracy and a low false prediction error.