4 ms·
Although I'm finding it hard to find confirmation (or denial), I believe refineries have captive power plants as part of their operations, which are effectively
by mchannon 8y ago
Although I'm finding it hard to find confirmation (or denial), I believe refineries have captive power plants as part of their operations, which are effectively isolated from the grid, and kept out of aggregate electricity generation statistics (they always seem to call it "net electricity" for some reason and maybe this is part of why).
- reitzensteinm 8y agoBut those captive power plants would have to add up to be a third of the capacity of the power plants on the main grid - that's a huge amount. If the process uses a lot of heat, and you're why wouldn't you just skip the electricity generation and use the thermal energy directly? So presumably the electrical consumption to the extent it exists is for a more sophisticated use than that. I would suggest (in the absence of other evidence) the government report linked above is correct - that it's relatively efficient in terms of electricity consumption, but not in terms of overall energy consumption.
- mchannon 8y agoCheck out http://www.iaee.org/documents/denver/varela-salazar.pdf http://www.iaee.org/documents/denver/varela-salazar.pdf Texas has 2GW of refinery cogen capacity alone.
- reitzensteinm 8y agoAccording to the numbers in that document, that's equivalent to 0.46 kWh per gallon. It also notes: "The comparison of both graphics suggests that refineries in Texas have excess capacity in many cases" Compared to a "typical requirement". The slope of the line for "typical requirement" is 0.22 kWh per gallon. This either means Texan refineries are atypically inefficient, or they sell power back to the grid or shut down their power plants. Either way, we're still an order of magnitude away from the 6 kWh/gallon figure, and from sources that look fairly credible.
- mchannon 8y agoFair enough. I do know that extraction and pipelines also require prodigious amounts of power. Obviously not so much that it’s uneconomical, but seeing the electric bill for one oil well in the Permian was eye-opening, particularly at the 3c/kWh rate they somehow got. It’d be nice to know the range for different kinds of oil wells. Maybe that’s the source of the electrical use.