Systems and methods for optimizing refinery coker process

A method for automatic operation of a coker, the method comprising:

• obtaining a target coke rate indicating a target rate at which to accumulate coke within a coke drum of the coker;
• using a neural network model to generate a target coker feed rate predicted to result in the coke accumulating within the coke drum at the target coke rate, the target coker feed rate indicating a target rate at which to feed oil into the coke drum;
• and
• operating a drum feeder using the target coker feed rate to modulate a feed of oil into the coke drum.

A control system for automatic operation of a petroleum product refinery, the control system comprising:

• a feeder operable to modulate a feed of an input petroleum product into a production unit of the petroleum product refinery, the production unit operable to convert at least a portion of the input petroleum product into a derived petroleum product;
• a controller comprising a processing circuit configured to:
• obtain a target production rate indicating a target rate at which to produce the derived petroleum product within the production unit;
• use a predictive model to generate a target feed rate predicted to result in the derived petroleum product being produced within the production unit at the target production rate, the target feed rate indicating a target rate at which to feed the input petroleum product into the production unit;
• and
• operate the feeder using the target feed rate to modulate the feed of the input petroleum product into the production unit.