Page 298 - Power Electronic Control in Electrical Systems
P. 298

//SYS21/F:/PEC/REVISES_10-11-01/075065126-CH007.3D ± 286 ± [263±289/27] 17.11.2001 10:25AM







               286 Harmonic studies of power compensating plant




















                      Fig. 7.18 Current waveform through the capacitor for different values of firing angle.

                      Figures 7.17(a) and (b) and 7.18(a) and (b), TCSC operation in the inductive region
                      produces larger currents flowing through the thyristors than through the capacitors.
                      The opposite situation occurs when the TCSC operates in the capacitive region.

                      7.6.2  TCSC harmonic domain modelling

                      Due to the circulating currents inside the TCSC module, intrinsic in the working
                      mechanism of this equipment, the TCR representation based on harmonic currents
                      injections is of limited use. Instead, the TCR is well modelled by a harmonic
                      admittance matrix, which can then be combined with the admittance of the capacitive
                      element to give the overall admittance representation of a single TCSC module.
                      From the modelling viewpoint, this is the basic block, which may be used to build
                      models for the three-phase TCSC systems. The basic building block is also useful to
                      conduct fundamental studies of how the TCSC harmonic impedances and cross-
                      coupling impedances vary as a function of the conduction angle.

                      7.6.2.1 Single-phase TCSC representation
                      In principle, equation (7.39) is valid for both the single-phase SVC and the TCSC,
                      but the latter is better represented as a two-port circuit because it is a series connected
                      element, that would be interfaced at both ends with transmission network plant, i.e.
                      high-voltage transmission lines. Using the result in equation (7.39), the nodal transfer
                      admittance matrix of the TCSC is built with ease

                                           I s     Y TCSC   Y TCSC    V s
                                               ˆ                                        (7:46)
                                           I r      Y TCSC  Y TCSC    V r
                      where Y TCSC ˆ Y SVC ˆ (H R ‡ Y C ), V s and V r are the voltages at the sending and
                      receiving ends, I s and I r are the currents at the sending and receiving ends, respect-
                      ively.

                      7.6.2.2 Impedance characteristics
                      The voltage drop expression for the single-phase TCSC may be derived using the
                      result in equation (7.39)
   293   294   295   296   297   298   299   300   301   302   303