Information Technology Reference
In-Depth Information
Recursive Model Predictive Control
for Fast Varying Dynamic Systems
Da Lu, Guangzhou Zhao, and Donglian Qi
College of Electrical Engineering, Zhejiang University,
310027 Hangzhou, China
Abstract.
A well known drawback of model predictive control (MPC)
is that it can only be adopted in slow dynamics, where the sample time
is measured in seconds or minutes. The main reason leads to the problem
is that the optimization problem included in MPC has to be computed
online, and its iterative computational procedure requires long computa-
tional time. To shorten computational time, a recursive approach based
on Iterative Learning Control (ILC) and Recursive Levenberg Marquardt
Algorithm (RLMA) is proposed to solve the optimization problem in
MPC. Then, recursive model predictive control (RMPC) is proposed to
realize MPC for fast varying dynamic systems. Simulation results show
the effectiveness of RMPC compared with conventional MPC.
Keywords:
Model Predictive Control, Recursive Levenberg Marquardt
Algorithm, Iterative Learning Control.
1 Introduction
The basic idea of Model Predictive Control (MPC) is to improve the future
plant behavior by computing a sequence of future manipulated variable adjust-
ments based on a dynamic model of plant. Only the first element in the optimal
sequence is applied to the system. This process is repeated at every sampling
interval to update information [1]. It is well-known that an optimization prob-
lem needs to be solved online to obtain future manipulated variable in MPC.
The heavy computational burden has limited the application of MPC to slow
dynamic systems for a long time. However, in recent years, some cheering im-
provements have been made, which lead to the possibility of applying MPC to
fast varying dynamic systems [2-4].
Generally speaking, the emerging MPC methods for fast varying dynamic sys-
tems can be divided into two groups. The first is to reduce MPC optimization
problem to the selection of system behavior from finite input sequences[2, 3].
However, since the set of possible input sequences is required to be finite, mod-
ulation methods, such as space vector modulation, are dicult to be enrolled
in the scheme. In the second approach, MPC optimization problem is solved of-
fline, and the input sequence can be obtained through simple online calculation,
Search WWH ::
Custom Search