Digital Twin for Well-to-Surface Optimization of Cyclic Steam Stimulation (CSS) and Sucker Rod Pump (SRP) Operations for Heavy Oil Wells of Baghewala Field.
Oil India Limited · Smart Automation · Software
Precisely described and genuinely valuable, but with no field data your twin validates against itself — take it only with petroleum engineering support, and be explicit about which physics is modelled and which is assumed.
What it actually is
Baghewala's crude is thick and the reservoir is cool, so operators inject steam to thin the oil and use rod pumps to lift it. Today the steam cycle and the pump settings are tuned separately from experience, and as the reservoir cools the oil thickens, pumps lose efficiency and rods fail. The ask is a live digital model that optimises the steam cycle and the pump together.
What to build
A well-to-surface digital twin coupling three layers: a reservoir model predicting how injected steam heats the near-wellbore region and how temperature and viscosity decay through the production cycle, a wellbore model translating that into fluid properties at pump depth, and a rod pump model predicting load, efficiency and rod float from stroke length and strokes per minute against those conditions — with an optimiser recommending steam volume, injection pressure, soak time and production cut-off for each cycle alongside continuously adjusted pump settings, a rod-float and impact-loading detector, and a dashboard tracking steam-oil ratio, energy consumption and predicted versus actual production.
Smallest thing that wins the room
Replay a steam cycle and show the twin tracking viscosity rising as the near-wellbore region cools, the optimiser progressively reducing strokes per minute to keep the rod loaded, and a flagged rod-float event that the current fixed-setting operation would have run straight into.
How crowded this one gets
A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.
Quieter than 73% of the 226 · #62 of 226 by expected field
A normal-sized field. Your idea has to be good, not miraculous.
Why: company-sponsored statements drew the smallest fields of all.
This is a guess, not a fact
Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.
The scores
The number is the shorthand. The line under it is the reason.
Acceptance potential
2/5The problem is genuine and precisely described, but with no field data the twin is calibrated against nothing and validated against itself, and Oil India judges are petroleum engineers who will probe the reservoir and rod-dynamics assumptions far more deeply than a general judge would.
Feasibility
2/5This requires coupled thermal reservoir simulation, wellbore fluid modelling and rod pump dynamics validated against real field data from one specific Rajasthan field — none of which is public, and the petroleum engineering underpinning each layer is specialist knowledge rather than something learnable alongside the build.
Innovation scope
3/5The physics of each layer is established petroleum engineering and the description names the parameters to optimise, so your room is in how you couple the three models and formulate the joint optimisation rather than in the modelling itself.
Clarity
4/5The description names the field, the crude properties, the specific operational failures — rod floating, impact loading, high steam-oil ratio — and lists seven concrete things the solution should do, so the target is well defined even though the data is not provided.
Effort
MassiveThree coupled physical models plus a joint optimiser plus a monitoring dashboard is a large systems build, and each physical layer requires domain study before any code can be written.
Demo-ability
MediumThe optimiser adjusting pump settings as modelled viscosity rises tells a clear story, but every number in it comes from a simulation you built, so the demo largely demonstrates your own assumptions.
In its favour
- Green flag: The description hands you the reservoir characteristics — API gravity, temperature range, viscosity behaviour — so the physical parameters of your model come from the statement rather than invention
- Green flag: Rod floating and impact loading are specific, well-documented failure modes with established detection signatures in dynamometer data
- Green flag: Steam-oil ratio is a standard industry metric, so your optimisation objective is unambiguous and quantifiable
- Green flag: The field is narrow and named, which keeps the scope from sprawling across the whole of oil production
Against it
- Red flag: No Baghewala field data is provided or public, so the twin is built and validated entirely against a simulator you wrote, which proves your assumptions rather than your optimisation
- Red flag: Coupled thermal reservoir simulation is specialist work and a simplified analytical model will be visibly inadequate to a reservoir engineer
- Red flag: Rod pump dynamics involve rod string elasticity and fluid inertia, and a static force model will mispredict exactly the rod-float condition the description cares most about
- Red flag: Oil India evaluators are production engineers who work on this field, so domain errors that a general judge would miss will be immediately apparent
What you will be writing
- Thermal reservoir modelling (analytical Marx-Langenheim / simplified numerical)
- Rod pump dynamics and dynamometer card simulation
- Viscosity-temperature correlation for heavy crude
- Multi-objective optimisation (OR-Tools / scipy)
- Time-series ingestion (TimescaleDB)
- React operations dashboard
- Petroleum production engineering
- Digital twin systems
- Process optimisation
Prior art to read before you start
coupled reservoir and artificial lift optimisation · thermal EOR cycle design · equipment failure prediction from operating conditions
Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.