15-01-2013, 01:01 PM
Fuzzy Optimization Techniques Applied to the
Design of a Digital PMSM Servo Drive
Fuzzy Optimization Techniques Applied.pdf (Size: 1.16 MB / Downloads: 55)
Abstract—
This paper presents a novel design approach by
applying gradient optimization with fuzzy step-sizing techniques
to the design of a digital permanent magnet synchronous motor
(PMSM) servo drive. The servo specifications and design variables
are specified and analyzed to formulate a controller optimization
problem. The servo responses are then fed back to evaluate the
overall system performances, which can be expressed as objective
functions with respect to the servo control parameters. According
to the objective functions and design specifications, the servo
control parameters can be properly tuned toward their optimal
values by using the proposed optimization techniques. In order
to improve the convergent rate of the optimization process, a
fuzzy-logic based step-size tuning strategy is presented. Because
of the nonlinear property of the digital servo drives, the tuned
servo control parameters may be only optimal for a particular
operating point, therefore, once the optimum design is achieved,
the proposed fuzzy optimizing controller can perform as an
intelligent tuner for on-line gain adaptation under different
loading conditions. The proposed fuzzy optimization servo tuner
has been realized under a PC-MATLAB-based environment
with an on-line controlled digital PMSM servo drive. Simulation
and experimental results indicate that the control parameters
of a digital PMSM servo drive can be optimized for its dynamic
responses under various load conditions.
I. INTRODUCTION
ALTHOUGH high performance permanent magnet synchronous
motor (PMSM) servo drives are increasingly
employed in industrial automation and home appliances, the
design of these servo drives still requires time-consuming trial
and error processes during its final tuning stage to fulfill the
application requirements. In practice, the design of a digital
PMSM servo drive involves several complex procedures which
include modeling, control scheme design, servo loop analyzes,
simulation, digital implementation, and parameter tuning.
Hence, application of intelligent optimization techniques for
simplifying the design problem remains an interesting and
important issue to be further studied.
Optimization Algorithm
After defining these objective functions, the design problem
for high-performance servo drives becomes as a multiobjective
optimization problem, which implies to minimize all these four
objective functions simultaneously. Since there are some tradeoffs
between these objective functions, the multiobjective optimization
problem is difficult to be solved. In order to solve this
problem, a weighted-sum method [17] is presented for representing
the relative importance of each objective function. By
properly combining the weighted objective functions, a convex
Relationship between objective functions and servo loop gains:
(a) position control loop (b) velocity control loop.
objective function can be formulated for determining the optimal
values of servo control parameters as