Nowadays, more and more often made use of mathematical models to predict the emissions based on measured process parameters
Nowadays, more and more use of mathematical models to predict emissions based on measured process parameters. These so-called Predictive Emission Monitoring Systems (PEMS) can be used, for predicting the NOX emissions as class 1 (ETS) classified installations.
Many consultants promise mountains of gold and claim that, using neural networks, it is possible to model the emissions from all plants. Yet it proved in the past that this technique is rarely used because the line "rubbish in = rubbish out out" still applies forever. BIEM will therefore not give this guarantee. But we can boast sufficient evidence chemometric knowledge, which we were able to model the emissions of many plants. The most striking example is the many PEMS models that have been used for years in the Netherlands for the prediction of NOX emissions from gas-turbine plants.
Practical example of a Predictive Emission Monitoring System
waarin:
NOxPEMS Predicted NOx-emission (g/GJ)
QA Consumption gas A (Nm3/h)
QB Consumption gas A (Nm3/h)
QC Consumption liquid C (kg/h)
Qstoom Consumption steam (ton/h)
HiB Calorific value gas B (MJ/Nm3)
HiC calorific value liquid C (MJ/kg)
