A Fuzzy Association Rule Mining Based Approach To Identifying So2 Emission Concentration Patterns Of Coal-Fired Boilers

P. Ran,Y. Wang,Pei Liu,Zheng Li

DOI: https://doi.org/10.3303/CET1652106

2016-01-01

Abstract:Sulphur dioxide emissions control in coal-fired power plants faces significant challenges with increasingly stricter emissions standards. These challenges are further exaggerated by the fact that coal-fired power plants operate strongly irregularly in order to meet the user demand between full-load and part-load conditions, mainly due to requirements from a power grid with the increasing capacity of intermittent renewable power. Acquiring accurate SO2 emissions patterns at these variable loads, thus unstable and unpredictable, are a challenging task. In this paper, we use a fuzzy association rule mining approach, based on real-life operational data, to obtain quantitative SO2 emission concentration patterns of coal-fired boilers. Operational data from a 300 MW supercritical power plant over a one-year period are used for analysis purposes. Results show that an optimal operational strategy can be produced by applying the fuzzy association rule mining techniques. A coal-fired power plant, if operated following the optimal operation strategy, can reduce its coal consumption rate by 1.35 g/kWh and SO2 emission concentration by 145 mg/Nm(3), respectively.