PID Controller
A widely used feedback controller that combines proportional, integral, and derivative terms to correct error.
Definition
A PID controller computes a control action from the error between a setpoint and the measured output, summing three terms: proportional to the present error, integral of past error, and derivative of the error's rate of change.
Practical PID loops add refinements the textbook form omits: integral anti-windup prevents the accumulated term from saturating the actuator, and filtering tames the noise-amplifying derivative term. These details often matter more to real performance than the nominal gains.
Its enduring dominance in industry rests on delivering good performance without a detailed model, using just three intuitive gains. Real deployments add refinements the textbook omits, anti-windup to keep the integral term from saturating actuators, derivative filtering to tame noise amplification, and bumpless transfer between manual and automatic modes. These practical details often decide whether a loop performs well, and they explain why tuning PID remains a craft.
The three terms
- Proportional: responds to current error; larger gain, faster but riskier.
- Integral: eliminates steady-state offset by accumulating error.
- Derivative: anticipates change, damping overshoot.
Why it matters
PID is the most widely deployed control algorithm in industry because it is simple, robust, and effective without a detailed system model. Tuning its three gains balances responsiveness, stability, and overshoot, a practical skill across engineering.
Fusion connection
PID-style feedback regulates auxiliary systems in a fusion plant, such as coolant flow and gas fueling, where a reliable, model-light controller holds a setpoint against disturbances.