Sees a failure before the stop. Shows which well matters most
Sentinel Lift brings the whole artificial-lift stock onto one screen: ESP, rod pump and plunger lift. The engineer gets more than an alarm: a likely cause, operating advice and the cost of delay in tonnes.
Try a dispatcher's shift
Pick a well, open "Why this?" and approve the recommendation. The tonnes at risk fall as the queue clears.
Sentinel LiftDemo · simulated dataThe whole stock in one priority list
Statuses are shown as they are. New lift types connect to the same screens and data.
ESP · rolling out
Controllers and downhole gauges: current, insulation, intake pressure and temperature.
Rod pump · prototype
Dynamometer cards from controllers, portable units and paper. A pilot on a customer's field in year one.
Progressing cavity · planned
Torque, rotation speed and wear for viscous fluid and solids.
Plunger lift · planned
Cycle and pressure control in the same priority list.
Dispatch that leads to a decision
Ordinary telemetry shows parameters. Sentinel Lift answers the shift's real question: what to do first, and why.
A queue by production at risk
Not hundreds of identical alarms, but a list of wells by tonnes at risk. The dispatcher sees where to go first.
One stock, any lift type
ESP, rod pump and plunger lift on the same screens and reports, whichever controller brand you run.
The cause, not just the alarm
Likely cause, confidence and operating advice. Every conclusion opens up to its inputs and model version.
Wells with no telemetry count too
Paper logs and photos of dynamometer cards go through FIELDPix into the same priority list.
Approval before action
The system recommends, the engineer approves. The decision goes to the journal and the well history with its author and time.
Run life and energy
Run life by manufacturer and pump size, energy per tonne, the cost of each stoppage.
Reviews in Connect
An alarm immediately opens a discussion on the well card with the right people from the shift roster.
From alarm to workover
A recommendation becomes a workover request, and the WRA result and repair report return to the well history.
Keeps working through outages
The gateway buffers data, Edge runs models on site. Deployed with us or on the customer's servers.
Conditions, signs, recommendations
| Condition | Signs in the data | Recommendation |
|---|---|---|
| Pump-off | Current and load drop, temperature rises | Lower frequency, cyclic mode |
| Gas at pump intake | Current and intake pressure fluctuate | Change mode, gas separator |
| Plugging, scale | Current rises while flow rate falls | Flushing, inhibitor |
| Insulation loss | Cable and motor insulation resistance falls | Planned repair before breakdown |
| Motor overheating | Winding temperature rises | Reduce load, check cooling |
| Pump wear | Head drops at the same frequency | Schedule replacement |
| Rod pump: valve leaks, inflow, gas, pound, parted rods | Dynamometer card shape | Pumping speed, stroke length, repair by priority |
How it works
01
Controller, gauge, test separatoror a paper log02
Industrial gatewayModbus, OPC UA, MQTT, 4G03
AEVORON Coreservers in Tashkent or at the customer04
Dispatcher and engineerscreen, alert, reportHow many tonnes your downtime costs
Enter your own values. The whole formula is visible, and you set the reduction yourself: it is an assumption to discuss, not a promised result.
Wells × failures × downtime days × rate × reduction. Based on your assumptions, not a guarantee. A pilot shows the real figure.
Start with one pad
A 1–3 month pilot: we connect the controllers, set the baseline and measure the effect in hours and tonnes. The final decision always rests with the engineer.