Control Systems
A control system measures a plant, compares it to a target, and adjusts actuators to drive the difference toward zero.
The feedback loop
Control is the discipline of making a physical system behave as desired despite disturbances. A sensor measures the current state, a controller compares it to a setpoint, and an actuator applies a correction. The loop repeats fast enough to keep the system on target.
Feedback and feedforward
- Feedback: react to the measured error after it appears.
- Feedforward: anticipate a known disturbance and pre-compensate for it.
- Combined, they give both responsiveness and stability.
Stability comes first
A poorly tuned controller can amplify errors instead of correcting them, driving a system into oscillation or runaway. Control theory provides the tools to guarantee stability margins before a loop is ever connected to real hardware, which is essential when the hardware is expensive and energetic.
Determinism and timing
Control loops must be deterministic and meet strict timing: a correction that arrives late is worse than useless. This is why control logic is built as auditable, real-time software rather than left to systems whose behavior cannot be guaranteed within a fixed time budget.
Control in fusion
A fusion plasma is a fast, nonlinear system requiring active control of position, shape, and stability. For a spherical tokamak like the breeder Hyperion, magnetic control keeps the plasma in a prescribed shape — including its negative triangularity — while protecting the machine, with deterministic interlocks guarding against fault conditions.