Using Modified Particle Swarm Optimization to Design a Controller by Time-Domain Objective Functions

A large proportion of industrial systems are represented by linear time-invariant transfer functions. The proportional-integral -derivative (PID) controller is one of the most widely used functions. The lead-lag controller is a more practical alternative. Lead-lag-like controller can be lead-lead, lag-lead or lag-lag controller. This paper focused on the design of lead-lag-like controller by optimization of the time-domain objective function. A modified particle swarm optimization (MPSO) algorithm is chosen to find the optimal solutions. MPSO tries to force all the particles to search the optimum exhaustedly. The proposed objective function includes time-domain specifications, including the delay time, rise time, first peak time, maximum peak time, maximum overshoot, maximum undershoot, setting time and steady state error. Designers can make trade-offs in a variety of specifications. As long as the plant could be modeled as a linear time-invariant transfer function, the suggested method would design the lead-lag-like controller capable of approaching the desired specifications. Computer simulations show that the performance can be fully met or very close to desired.