Ronald Hernandez

Adaptive Near-Optimal Compensation in Lossy Polyphase Power Systems

MS Thesis

April 2009

The role of compensation in power system efficiency optimization is to reduce the power consumption of the source (line) impedance, so that most of the source power is delivered to the load. A classical result by Fryze is used in the case when the voltage drop across the line is negligible in comparison with the load voltage.  However, when the source impedance becomes significant the traditional Fryze current is no longer the smallest (by rms) line current that supplies the same real power to the load as the original load current.

An optimal solution considering significant line impedance has been already obtained in recent works. Unfortunately, it relies on network and load parameters that are not easy to determine during operation.  This motivates our approach in searching for a sub-optimal easy-to-implement solution.

Our basic objective in this thesis is the construction of an adaptive near-optimal compensator that relies only on measurements of the load voltage and current so as to allow precise control of the compensated load as well as the real power flowing out of the compensator, while reducing line losses to within a few percent of its theoretical minimum.  Properties of the solution are illustrated for an asymmetrical three phase induction motor supplied with unbalanced non-sinusoidal voltages.

Advisor:  Hanoch Lev-Ari