Abstract

Slagging is a common phenomenon interfering with the safe and efficient operation of coal-fired boilers. Due to the lack of monitoring methods for species and temperature distribution near the water wall, which is critical for slagging formation, it is a tricky problem to detect and reduce the slagging in real time. A method for slagging distribution prediction and anti-slagging optimization of a 330 MW tangentially coal-fired boiler is developed in this paper. Proper orthogonal decomposition (POD) reduced order modeling and support vector machine (SVM) are used to predict the distribution of temperature and O2 near the water wall by taking 15 variables as input parameters (load, coal types, primary air rate, secondary air rate, excess air coefficient, and secondary air velocities). Next, fuzzy comprehensive evaluation (FCE) is employed to quantify the slagging occurrence and distribution on the water wall. Furthermore, the real time slagging prediction is realized with the combination of POD, SVM, and FCE. Finally, the conventional genetic algorithm (CGA) and simulated annealing genetic algorithm (SAGA) are adopted to optimize the air distribution schemes and coal blending schemes to reduce the slagging ratio, and their performances are compared. The results indicate that the predictions of temperature and O2 with POD show a good agreement with CFD simulations. In the optimization process, whether air distribution is optimized alone or both coal blending and air distribution are optimized simultaneously, the performance of SAGA is better than that of CGA. After optimizing the air distribution with SAGA, the slagging ratio of a typical operating condition with a load of 330 MW is reduced from 14.21% to 7.16%. The slagging ratio is further reduced to 5.36% after optimizing air distribution and coal blending schemes simultaneously. The proposed slagging prediction and optimization for anti-slagging opens a new perspective to alleviate slagging of water walls effectively.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call