Artificial Intelligence in Petroleum Industry
1. Artificial: Does it simply imply that this approach has no occurrence in a real field scenario (Is it no more a reality??)?
If it is not real, then, can’t it have the same reservoir rock and fluid properties along with in-situ ‘dynamic’ fluid drainage principles – as observed in a real field scenario?
If so, how could - the developed computer programs – which, ‘when fed into a machine’ – would be able - to make the machine to exhibit – the intelligence of ‘reservoir/production/drilling engineers’?
2. In case, if ‘actual oil/gas drainage problem of interest’ was not understood and captured “precisely”, then, what are we supposed to do with these machines - solving the so called ‘complex fluid flow problem’ @ faster rate; and thereby appealing that such an approach was able to make predictions of future production scenarios in the best possible manner with the least waste of time and effort – leading to huge savings in operating cost?
3. What does the (successful) application of AI (in part or fully) in the petroleum industry - so far - indicate us?
Whether the output/forecast – using AI - on (a) field development; (b) wellbore drilling process; (c) drilling platform selection; (d) well placement; (e) reservoir properties determination; (f) sand production prediction; (g) gas-lift performance prediction and long term gas-lift allocation optimization; (h) history matching; (i) two-phase fluid flow in pipes; (j) identification of well test interpretation models; (k) formation damage prediction; (l) completion analysis; & (m) drilling operations – remain “satisfactory”?
4. In case, if we happen to lose the knowledge gained through experience gets easily lost, when an experienced field engineer or a laboratory analyst leaves the company; then, whether, AI would efficiently make use of all the data available to map out the trends; and ultimately offer an improved understanding on petroleum reservoir characterization and engineering?
5. On a lighter note, if AI could predict the complex relationships between input and output parameters and thereby allowing to deduce the best values of reservoir rock and fluid parameters by minimizing calculation errors so smart, then, can’t AI predict, whether the resource person would remain to continue in the same company or not – by deducing the best and worst performances of that resource person, while discarding his/her emotions (similar to a human brain)?
- Badges
- Science topic