Transient stability assessment of hybrid distributed generation using computational intelligence approaches

Due to increasing integration of new technologies into the grid such as hybrid electric vehicles, distributed generations, power electronic interface circuits, advanced controllers etc., the present power system network is now more complex than in the past. Consequently, the recent rate of blackouts recorded in some parts of the world indicates that the power system is stressed. The real time/online monitoring and prediction of stability limit is needed to prevent future blackouts. In the last decade, Distributed Generators (DGs) among other technologies have received increasing attention. This is because DGs have the capability to meet peak demand, reduce losses, due to proximity to consumers and produce clean energy and thus reduce the production of CO₂. More benefits can be obtained when two or more DGs are combined together to form what is known as Hybrid Distributed Generation (HDG). The challenge with hybrid distributed generation (HDG) powered by intermittent renewable energy sources such as solar PV, wind turbine and small hydro power is that the system is more vulnerable to instabilities compared to single renewable energy source DG. This is because of the intermittent nature of the renewable energy sources and the complex interaction between the DGs and the distribution network. Due to the complexity and the stress level of the present power system network, real time/online monitoring and prediction of stability limits is becoming an essential and important part of present day control centres. Up to now, research on the impact of HDG on the transient stability is very limited. Generally, to perform transient stability assessment, an analytical approach is often used. The analytical approach requires a large volume of data, detailed mathematical equations and the understanding of the dynamics of the system. Due to the unavailability of accurate mathematical equations for most dynamic systems, and given the large volume of data required, the analytical method is inadequate and time consuming. Moreover, it requires long simulation time to assess the stability limits of the system. Therefore, the analytical approach is inadequate to handle real time operation of power system. In order to carry out real time transient stability assessment under an increasing nonlinear and time varying dynamics, fast scalable and dynamic algorithms are required. Transient Stability Assessment Of Hybrid Distributed Generation Using Computational Intelligence Approaches These algorithms must be able to perform advanced monitoring, decision making, forecasting, control and optimization. Computational Intelligence (CI) based algorithm such as neural networks coupled with Wide Area Monitoring System (WAMS) such as Phasor Measurement Unit (PMUs) have been shown to successfully model non-linear dynamics and predict stability limits in real time. To cope with the shortcoming of the analytical approach, a computational intelligence method based on Artificial Neural Networks (ANNs) was developed in this thesis to assess transient stability in real time. Appropriate data related to the hybrid generation (i.e., Solar PV, wind generator, small hydropower) were generated using the analytical approach for the training and testing of the ANN models. In addition, PMUs integrated in Real Time Digital Simulator (RTDS) were used to gather data for the real time training of the ANNs and the prediction of the Critical Clearing Time (CCT).

Grantor dc:publisher.institution

Department of Electrical Engineering

Year dc:date.issued

2014

Author dc:creator

• Olulope, Paul Kehinde

Advisor dc:contributor.advisor

• Folly, Komla A

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/9288

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/9288