Abstract
This paper describes how genetic programming can be
used to automate the synthesis of the design of both
the topology and parameter values for controllers. The
method described in this paper automatically makes
decisions concerning the total number of processing
blocks to be employed in the controller, the type of
each block, the topological interconnections between
the blocks, the values of all parameters for the
blocks, and the existence, if any, of internal feedback
between the blocks of the overall controller. This
design process can readily combine optimization of
performance (e.g., by a metric such as the integral of
the time-weighted error) with time-domain constraints
and frequency-domain constraints. Genetic programming
is applied to two illustrative problems of controller
synthesis: the design of a robust controller for a
non-minimal-phase plant and the design of a robust
controller for a three-lag plant. A previously
published PID compensator (Astrom and Hagglund 1995)
for the three-lag plant delivers credible performance.
The automatically created controller is better than 7.2
times as effective as the previous controller as
measured by the integral of the time-weighted absolute
error, has only 50% of the rise time in response to the
reference input, has only 35% of the settling time, and
is 92.7 dB better in terms of suppressing the effects
of disturbance at the plant input.
Users
Please
log in to take part in the discussion (add own reviews or comments).