This paper presents a time delay neural network (TDNN) model designed for t
he prediction of nitrogen oxides (NOx) and carbon monoxide (CO) emissions f
rom a fossil fuel power plant. NOx and CO emissions of the plant are determ
ined as a function of other related time-series such as air flow rates and
oxygen levels that are measured during the system operation. Correlation an
alysis is performed on the data to determine the location and the spread of
cross-correlation between pairs of variables and this information is used
to form a variable tapped delay line at the input of the network. We also i
ntroduce a neural network based preprocessor which employs an iterative reg
ularization scheme to recover missing portions of CO data that are censored
due to saturation of the measuring device. Prediction after training with
the restored data set is observed to be significantly more accurate.