الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 171 |  |  |
| | | from | 171 | | |
Abstract
Coupling capacitor voltage transformers (CCVTs) are very popular for measurements of voltages of 100kv and above because of their economic advantage, and, for many years, site experience with relays has been satisfactory. Recently, however, interest in the effects of the secondary voltage errors during fault on relay operation has been growing; an interest promoted by the increasing use of faster and more sensitive relays and the need for reliable one-cycle operating times. It is widely appreciated that CCVTs cannot respond quickly to changes in primary voltage during fault. Distance relays, on the other hand, require an accurate and fast indication of the primary voltage for correct operation; a requirement which conflicts with the nature of the CCVT. Even before 1950 it was recognized that relay performance would be adversely affected, but investigations at that time concluded that performance, in general, would be satisfactory. Despite the decrease in relay operating times since that date, it is often claimed that the quality of the CCVT output remains adequate for distance relaying needs. This thesis discusses the effects of capacitor-voltage transformers transient errors during fault on high-speed distance relays, which can cause protective relay maloperation or even delay tripping and the thesis presents a program using neural network to recover the primary signal from the distorted secondary voltage. The ANN is trained to achieve the inverse transfer function of the coupling capacitor voltage transformer (CCVT), which provides a good estimate of the true primary voltage from the distorted coupling capacitor voltage transformers (CCVTs) secondary voltage. The neural network is developed using MA TLAB and trained using data from MA TLAB simulations. The accuracy of the simulation program is confirmed by comparison of its response with that of the target value obtained from the simulation data.