The Univer­sity of Al­berta is part­ner­ing with some pre­mier oil sands com­pa­nies to fig­ure out what data looks like and how best to use it

The Univer­sity of Al­berta is part­ner­ing with some pre­mier oil sands com­pa­nies to fig­ure out what data looks like and how we should use it

Alberta Oil - - CONTENTS -

Data has long been a buzz word in the oil field, and for good rea­son. Oil and gas pro­duc­ers have found that they can col­lect data on just about any­thing, from mon­i­tor­ing well­head pres­sure and flow rates to mod­el­ing fu­ture pro­duc­tion zones. But with the drive to col­lect and store data on al­most every nuance of the up­stream busi­ness, con­fu­sion still abounds about what ex­actly to do with it all. Just as the per­son who doesn’t read has no ad­van­tage over the one who can’t read, to para­phrase Mark Twain, the en­ergy pro­ducer who col­lects data but doesn’t put it to use is no bet­ter off than the one who doesn’t col­lect it in the first place.

Whether in the up­stream, mid­stream or down­stream sec­tors, the best use of data is drive ef­fi­ciency and re­move costly hu­man er­rors from pro­cesses. To do that, you need more than just data-col­lect­ing sen­sors. You also need the an­a­lyt­i­cal soft­ware—or so-called “soft sen­sors”—to dis­cover trends and anom­alies, and then vi­su­al­ize the data be­fore any im­prove­ments can be made.

It’s in that soft­ware space that Biao Huang, a Univer­sity of Al­berta pro­fes­sor and se­nior oil sands process con­trol chair at the Nat­u­ral Sciences and Engi­neer­ing Re­search Coun­cil of Canada, is work­ing. The goal is a more ef­fi­cient oil

sands op­er­a­tion from top­soil re­moval to tail­ings. “You need to have soft sen­sors and al­go­rithms to op­ti­mize the op­er­a­tion of the process through con­trol sys­tems,” Huang says. “Then the con­troller can make the whole op­er­a­tion more ef­fi­cient. To have such a largescale ap­pli­ca­tion of these sen­sors in some­thing like the oil sands is not very com­mon yet.”

Be­cause of the harsh en­vi­ron­ment of the oil sands—the in­tense heat, the cold and, of course, the oily sand—phys­i­cal sen­sors tend to break down eas­ily. “For things like tem­per­a­ture, pres­sure and wa­ter lev­els, those are eas­ier to de­velop in the oil sands,” Huang says. “But other things are much more dif­fi­cult like a sen­sor to de­ter­mine how much bi­tu­men is in the frost—that’s not easy to mea­sure with any hard­ware.” The an­swer to that prob­lem, ac­cord­ing to Huang and his project part­ners at Spar­tan Con­trols, Emer­son, Syn­crude and Cen­ovus, can be found in the soft sen­sor and the power of math­e­mat­i­cal in­fer­ence. “You in­fer some­thing that is dif­fi­cult to mea­sure with a hard­ware sen­sor us­ing math and physics and chem­istry, and then you use that data to build your com­puter mod­els.”

Like Syn­crude and Cen­ovus, Spar­tan and Emer­son both earn rev­enue through the oil sands. They sup­ply many of the process con­trol prod­ucts and ser­vices such as valves and engi­neer­ing ex­per­tise to the min­ing and up­grader fa­cil­i­ties. And with the last­ing down­turn in oil prices, process ef­fi­ciency has be­come cru­cial to en­sure oil sands pro­duc­tion stays com­pet­i­tive with cheaper light oil plays. “In re­finer­ies they’ve been us­ing the soft sen­sors for a long time be­cause the re­fin­ing process is eas­ier to make in­fer­ences on,” Huang says. “Re­fin­ing has bet­ter hard­ware sen­sors too be­cause the en­vi­ron­ment is much bet­ter than in the oil sands.”

One of the big­gest hur­dles to de­ploy­ing data an­a­lyt­ics in the oil field has less to do with the an­a­lyt­ics and more to do with data visualization. Once vi­su­al­ized prop­erly and melded with other vi­su­al­ized data sources, the right course of ac­tion can be­come very clear. But that can of­ten re­quire sig­nif­i­cant man­ual in­ter­ven­tion with the data and that’s some­thing that many oil­field op­er­a­tors haven’t bud­geted for. Mon­i­tor­ing data from thou­sands of pieces of equip­ment and for­mat­ting the re­sults onto a sin­gle read­able com­puter sin­gle screen can be a full-time job. That’s one ma­jor is­sue that Huang and his as­so­ci­ates are now try­ing to tackle through the de­vel­op­ment of sim­ple user interfaces. When they do, ad­vances in cloud-based con­nec­tiv­ity will lower the cost bar­ri­ers for the in­dus­try over­all and should lessen the bur­den on oil­field op­er­a­tors enough to use these data ser­vices with­out hav­ing to hire a team of com­puter sci­en­tists.

“To have such a large-scale ap­pli­ca­tion of these sen­sors in some­thing like the oil sands is not very com­mon yet.” – Biao Huang

Biao Huang, left, and war­ren mitchell of spar­tan con­trols at the univer­sity of al­berta

Newspapers in English

Newspapers from Canada

© PressReader. All rights reserved.