# Tutorial 11: Dynamic Well Modeling

```{note}
This tutorial is available as a Python script [`examples/11_flow_dynamic_wells.py`](https://github.com/GeoBrain-Project/geobrain/blob/main/examples/11_flow_dynamic_wells.py) and an interactive Jupyter notebook [`examples/notebooks/11_flow_dynamic_wells.ipynb`](https://github.com/GeoBrain-Project/geobrain/blob/main/examples/notebooks/11_flow_dynamic_wells.ipynb).
```

Run step-by-step flow simulations with dynamic well control: add new wells,
shut in existing wells, and change injection rates during the simulation.

## What You Will Learn

- Configure wells with time-varying schedules (rates and BHP)
- Run step-by-step simulation with `FlowPropagator`
- Add and shut in wells mid-simulation
- Monitor production profiles over time
- Visualize time-lapse pressure and saturation maps

## Key Concepts

**Dynamic well control** allows modifying the well configuration between
simulation time steps. This is essential for modeling real-world scenarios
such as infill drilling, water injection optimization, and well shut-in
for maintenance.

This tutorial demonstrates three simulation modes. First, a full simulation
runs all time steps in a single call to establish a baseline. Second, a
step-by-step simulation advances the same model one time step at a time,
giving identical results while allowing inspection at arbitrary intermediate
times. Third, dynamic well operations modify the well schedule mid-simulation
-- shutting in a producer and opening a new one -- to show how the reservoir
response changes under different management strategies.

## Code

```python
from geobrain.physics.flow import ReservoirModel, FlowPropagator, Well

propagator = FlowPropagator(model)

# Step-by-step with dynamic control
for step in range(n_steps):
    if step == 50:
        model.shut_well('PROD1')
        model.add_well(Well(name='PROD2', ...))
    propagator.step(dt=1.0)
```

## Results

The full simulation runs all time steps at once, producing production curves
and state maps that serve as the reference baseline for subsequent experiments.

```{figure} ../../examples/figs/11_full_sim_production.png
:width: 100%
:name: fig-11-full-sim-production

Full simulation production curves.
```

```{figure} ../../examples/figs/11_full_sim_state_maps.png
:width: 100%
:name: fig-11-full-sim-state-maps

Full simulation state maps: pressure and saturation.
```

The step-by-step simulation reproduces the same results while allowing
intermediate inspection. The snapshot below shows the reservoir state at
t = 500 days.

```{figure} ../../examples/figs/11_step_sim_t500.png
:width: 100%
:name: fig-11-step-sim-t500

Step-by-step simulation at t=500 days.
```

When dynamic well operations are introduced -- shutting in the original
producer and opening a new well at a different location -- the saturation
front redirects toward the new producer, altering sweep efficiency.

```{figure} ../../examples/figs/11_dynamic_wells_snapshots.png
:width: 100%
:name: fig-11-dynamic-wells-snapshots

Dynamic well control: saturation snapshots.
```

```{figure} ../../examples/figs/11_dynamic_wells_final_state.png
:width: 100%
:name: fig-11-dynamic-wells-final-state

Final state after dynamic well operations.
```

## Full Example

See [`examples/11_flow_dynamic_wells.py`](https://github.com/GeoBrain-Project/geobrain/blob/main/examples/11_flow_dynamic_wells.py).