Modelling of NO formation in the combustion of coal blends

Coal blending is becoming of increasing importance in power stations firing pulverised coal as a result of increasing competition, stricter emission legislation and is an attractive way of improving plant economic and combustion performance. Presently, the two general methods used by power station operators to assess or predict the performance of an unknown coal blend to be fired in power station boilers are by the use of experimental large scale rig tests or correlation indices derived from experience of firing other coal blends in the power station environment. The first is expensive and the second is of doubtful accuracy in some cases. This paper evaluates the application of mathematical modelling of the combustion of a series of binary coal blends in the test situation of a drop tube reactor to predict the NO emissions and degree of char burnout. Its applicability to low NOx burners used in power stations is discussed and it is concluded that present mathematical coal combustion models are not developed sufficiently to enable an adequate description of the binary blends and the physical and chemical processes, which may include interactions, during combustion of the blend. This means that accurate predictions cannot be made.