Including the stream quality desired predict
in the overhead product and maintain slightly less than 2% Ethane in the column bottoms.
On the other hand, if ethane, with 1.5% propane sells for $16/barrel and propane is used as fuel gas, the operating objectives of the de-ethanizer will be different from the above case: to minimize ethane in the column bottoms and keep the propane in the overhead as near 1.5% as possible, without exceeding the sales limit. To achieve either set of the above objectives, a continuous measurement of propane in the column overhead and ethane in the bottoms is necessary.
|
Tag | Range |
---|---|---|
|
TI-6 | 100-300 °F |
Overhead Temperature | TI-5 | 100-300 °F |
TI-7 | 100-300 °F | |
TI-4 | 0-200 °F | |
FI-2 | 0-20MBD | |
FI-3 | 0-10MBD | |
|
FI-6 | 0-80MBD |
|
FI-4 | 0-30MPPH |
FI-5 | 0-80MBD |
Tag | Range | |
---|---|---|
|
TI-1 | 100-400 °F |
|
TI-3 | 100-400 °F |
TI-8 | 100-300 °F | |
TI-4 | 0-200 °F | |
Re-boiler Return Temperature | TI-2 | 100-400 °F |
FI-3 | 0-10MBD | |
FI-6 | 0-80MBD | |
|
FI-4 | 0-30MPPH |
|
FI-5 | 0-80MBD |
For the naphtha 95% cut point and naphtha Reid vapor pressure stream properties, the plant data, including the stream quality desired to predict, are collected in a Microsoft Excel™ spreadsheet to facilitate data manipulation. The data are then scaled to a fraction of the transmitter range so that they are confined to a sub-interval of [0…1]. A practical region for the data is chosen to be [0.1…0.9]. In this case each input or output parameter p is normalized as pn before being applied to the neural network according to:
pn = [( 0.9 - 0.1 ) / ( pmax - pmin )] * (p - pmin ) + 0.1