الفهرس 
The proposed techniousOnly 14 pages are availabe for public view
 |  | from | 119 |  |  |
| | | from | 119 | | |
Abstract
Modern power system are currently operating under heavily loaded conditions due to various economic, enviromental and regulatory changes. Consequently, maintianing voltage stability has become a growing concern for electric power utilities, so, the problem of voltage satbility and voltage collapse of power systems attracts more and more attention. voltage collapse can take place in systems and subsystems and can appear quite abruptly. Continyous monitoring of the system state is therefore necessary.
There are both static and dynamic aspects involved in voltage stability. static considerations relate voltage instability to the reaching of some maximal admissible load, beyond which a load-flow solution no longer exists. In vestigations are still required into the dynamic mechanism and modelling of real systems.
The research work describes in this thesis is devoted to the study of voltage collapse phenomenon. There arew different methods used for studying such problems. these methods are reviewed and discussed in this thesis. some of these methods are the jacobain method, the voltage instability proximity index VIPI and the voltage collapse proximity indicator VCPI method. the last indicator VCPI is known for its simplicity, easy of applications and almost linear behaviour.
As the thesis is concerning with the problem of static voltage stabilty, it investigates a proposed voltage collapse proximity indicator VCPI applicable to the load points of a power system. This indicatior is generalised and applied to both simple and realistic systems. the performance of this indicator is investigated over both the stable and the unstable regions, as the load at a particular node or the system load increases.
Tests show that the indicator can provide useful information at any operating point. also, in this thesis, voltage instability is predicted using feedforward back-propagation artificial neural networks ANN`S on the basis of a voltage collapse proximity indicator. The ANN`s are trained to determine the VCPI value and the voltages at load busses. Linearity of the VCPI make this possible to cover wide range of load variation with few numbers of test cases. Different system loading strategies are also studied and evaluated. Test results on a simple and realistic power system demonstrate the merits of the proposed approach.