Balanced confidence intervals

Multiple types of fuel

Many plants now use several types of fuel. The conversion from fossil fuels to green biofuels can mean that the plant has different limit values ​​for each type of fuel the plant uses. If more than one type of fuel is used in the same kiln line, this means that the limit values ​​for each fuel type are weighted in relation to how much fuel is fired of each type. The confidence interval is defined as a percentage of the daily limit value, and thus also depends on the amount of fuel fired. This is called weighted confidence intervals.

ReportLoq supports weighted confidence intervals according to the same recipe as balanced limit values. If you require weighted confidence intervals or just want to know more, read on.

The 95% confidence interval is described in MEL-16 as the value that must be deducted from the mean values ​​before environmental reporting to the authorities. The confidence interval is calculated by the authorities and determined in the environmental approval of the work. A possible requirements for weighted confidence intervals will be found in the same environmental approval.

Calculation of the balanced confidence interval

The weighting of the confidence interval is done by weighing two different confidence intervals (CI) against each other compared to production (MW) for the fuel type to which the confidence interval belongs.

An example of the formula could look like a mixture of biogas and oil:

Weighted confidence interval = (CIbiogas × MWbiogas + CI oil × MW oil): (MWbiogas + MW oil)

The result is the weighted confidence interval to be used during the STA period in question.

Extra documentation

When more than one type of fuel is used, it is often seen that the authorities require an additional report as supporting documentation for reporting. The report showing the amount of fuel fired for the fuel types and the calculated confidence interval will be helpful here and is included in ReportLoq by default.

Avoid exceedances

ReportLoq uses the weighted confidence interval calculation as an input parameter to the algorithm that predicts hourly value exceedances. In the calculation, the most recent operation of the plant is used to predict the emission value in combination with a projection of the weighted limit value and the weighted confidence interval. The result is that the operator can see if there is a predicted exceedance and thus have the opportunity to correct combustion before the exceedance becomes a reality.

Share this article