Saunder January 22, 2014 10:43 K15053˙Book
176 Disturbance Observer-Based Control: Methods and Applications
optimization problem possesses the following form
min
u(t)···u(t+M−1)
J =
P
$
j=1
[e
T
(t + j)Qe(t + j)] +
M−1
$
j=0
[u
T
(t + j)Ru(t + j)],
e(t + j) =
ˆ
y(t + j) −r (t + j), (11.3)
subject to the following constraints:
u
L
≤ u(t + j) ≤ u
H
: input saturation limits,
y
L
≤ y(t + j) ≤ y
H
: output specification limits, (11.4)
u(t + j) = 0 for j > M,
where P and M represent the prediction horizon and the control horizon, respec-
tively, e(t + j)isthe prediction error, r (t + j)isthe desired reference trajectory, Q
and R represent the error weighting matrix and input weighting matrix, respectively.
MPC scheme is implemented in a receding horizon framework. At any sampling ...