Since the complexity of sulfur capture and release during solid fuel combustion in circulating fluidized bed (CFB) boilers, especially in the oxycombustion conditions is still not sufficiently recognized, the development of a simple SO2 emission model for wide range of operating conditions is of practical significance. The paper introduces the artificial neural network (ANN) approach for the prediction of SO2 emissions from CFB boilers. The model considers a wide range of parameters influencing SO2 emissions. The [16-1-6-1] ANN model was successfully applied to predict SO2 emissions from coal combustion in several large- and small-scale CFB boilers, over a wide range of operating conditions, both in air-firing as well as oxygen-enriched and oxycombustion conditions. Since the method constitutes a quick and easy to run technique this approach makes a complementary tool in relation to the experimental procedures and the programmed computing approach. Therefore, the model can be easily applied by scientists and engineers for simulations and optimizations of CFB units. (C) 2015 Elsevier B.V. All rights reserved.