| For the past twenty years, the theoretical advance of the matrix converters has been impeded by the complexity arising from the time-varying trigonometric functions in their transformation matrix. In addition, the switching difficulties associated with the bidirectional switches have complicated the practical implementation of this class of converters.;In this thesis, the dyadic matrix structure and the a-b-c to d-q-0 transformation have been melded together to develop the dyadic matrix converter theory which is a generalized theory for the three-phase to three-phase matrix converters.;The thesis addresses the zero-sequence interaction in the matrix converters and the role of the zero-sequence elements in the Displacement Power Factor (DPF) correction on the utility-side, based on the Static VAR Controller (SVC) principle. Also, it is proved that using all the control degrees of freedom available, the dual condition of Unity Displacement Power Factor (UDPF) on side-1 and Field Vector Control (FVC) on side-2 can be established.;In this thesis, a new matrix converter topology, based on the three-phase voltage-source converters, has been proposed in which the switching difficulties reported in the conventional nine-bidirectional-switch topology have been bypassed. The theoretical expectations have been verified by the simulation as well as experimental tests on a laboratory prototype of the new matrix converter topology composed of three units of voltage-source converters each rated at 1 kVA. |
