Optimal Control of Electric Motors
Project Summary
The objective of this proposal is to develop a set of energy
efficient optimal controllers for the three major classes of
electric motors: separately excited dc-motors, ac-synchronous
motors, and ac-induction motors. The motivating application for
these new controllers is electric vehicle propulsion systems.
Improving the efficiency of the drive system is essential for
the development of new electric and hybrid electric vehicles to
meet emerging stringent environmental regulations. Although
extremely important, electric vehicle drive systems are only one
possible application of the results of this research project.
Industrial and manufacturing applications as well as consumer
products will also be able to benefit from this research. Most
electric motor controllers in use today are not designed to
operate cost optimal. The reason for this is clear: the design
of such optimal controllers is computationally intensive. However,
with recent technological advances in adaptive and parallel
computation, neural networks, and genetic algorithms, the design
of optimal motor controllers is now possible.
To achieve these optimal electric motor controllers, we propose
a hybrid genetic algorithm/neural network design approach. In this
case, the global optimization properties of the genetic
algorithm are combined with the learning and generalization
abilities of neural networks to produce a smooth controller
which globally minimizes some specified cost or criterion
function. By implementing the resulting controller in dedicated
parallel hardware, an extremely fast controller may be obtained.
Such a parallel realization is not necessary, however, as the
neural net controller may be simulated with current microprocessor
technology in the form of a digital control.
Due to Wayne State University's participation in two recent
hybrid-electric vehicle competitions, we have the potential to
implement the results of this research project in an actual
hybrid-electric vehicle environment. Additionally, with Wayne
State's strategic location in the automobile capital of the world:
Detroit, Michigan, there is potential for significant
cooperative projects with the major U.S. automobile companies.
Preliminary results indicate that such optimal controllers can
significantly improve motor efficiency. In particular, for the
4000 lb hybrid-electric vehicle constructed at Wayne State
University, the optimal controller produced by our hybrid
genetic algorithm-neural network approach can improve the
efficiency of the motor by as much as 28.7% over conventional
controllers. Such significant increases in motor efficiency are
encouraging and worth pursuing; in fact, an optimal controller
which could improve motor efficiency by a only few percent would
be worth pursuing, especially in the automobile industry, where
each percent improvement in efficiency is multiplied by a factor
of a million due to the large number of products sold each year.
Clearly, both industrial and environmental interests would benefit
significantly from our proposed research. Based on these
encouraging results, we propose the following tasks as part of
this research project: