Searches injection strategies — what to inject, where, and at what rate — against recovery and cost together.
EOR screening usually compares three or four pre-selected schemes because each evaluation costs a simulation campaign. The optimum is rarely one of the four.
The proxy simulator is placed inside a multi-objective optimiser that searches injectant, pattern, rate and timing against recovery factor and discounted cost. The Pareto front is presented to the team rather than a single recommended scheme.
Everything here inherits the proxy’s limits and the history match’s biases. Chemical EOR in particular depends on laboratory parameters — adsorption, degradation at reservoir temperature — that simulation handles crudely, and a favourable simulated result for a chemical flood should be treated as a reason to run a core flood, not as a decision.
Cada página de capacidad declara una limitación real. Si un proveedor no puede decirle dónde termina su producto, no le están diciendo la verdad sobre dónde empieza.
A learned stand-in for the full reservoir simulator that answers in seconds instead of hours, so an engineer can test fifty scenarios instead of three.
Rules out the EOR methods that cannot work in this reservoir before anyone spends a year simulating them.
Works out which injector is actually supporting which producer, and where the injected water is being wasted.
Evaluates whole development concepts — well count, phasing, facility size — as one connected problem instead of four separate studies.
Tunes the reservoir model until it reproduces what the field actually produced, automatically and with the uncertainty kept rather than discarded.