Analyzing the Effect of Overexcitation Limiter and Load on Short-Term Voltage Stability using Real-Time Simulations

Abstract

With the increase in interconnected systems and rapidly growing loads, power systems are being subjected to severely stressed conditions. As a result, transmission lines are being operated close to their capability, and systems are becoming much less stable as they consume more reactive power. The result of this is more incidents of voltage collapse and power systems being operated much closer to instability than previously. While some instances of system instability may not amount to much as they are confined to local areas, others can cause significant damage, impacting millions of people. As a result, stability analysis is becoming an increasingly urgent issue.

Overexcitation limiters (OEL) are typically looked at during long-term voltage stability due to their long operation time. However, with their ability to adjust the field forcing period and with dynamic summing type OELs having a period of transient operation, exploring the modeling and behaviour of overexcitation limiters during short-term voltage stability analysis is an essential issue for consideration by the industry.

In this thesis, the effect of OEL and load on short-term voltage stability was examined. This was done using the Taylor test system found in the literature. The test system is initially be subjected to a large disturbance by tripping one of the 500 kV transmission lines. It was then modeled using the commercially available real-time electromagnetic simulation platform (RTDS). RTDS was used because it can provide more accurate results.

The effect of load on short-term voltage stability was examined by modeling the load on the test system in two different ways. Firstly, they were modeled using a static load model. Secondly, they were modeled using complex electromagnetic transient models. The results of these simulations were compared to demonstrate the need to model loads as accurately as possible. This is to avoid a scenario where a static load may indicate that the system is short-term voltage stable. However, when using a dynamic model, the system is short-term voltage unstable.

Next, the effect that OELs will have on short-term voltage stability was examined. In conjunction with load models, the effect that the OEL has when subjected to a disturbance was examined. Four overexcitation limiter models found in the IEEE standards and literature were used to investigate the impact of each separate model. This showed that OEL operation can impact short-term voltage stability and should be modeled during short-term voltage analysis. It also showed that OEL's gain value affects how quickly the system becomes unstable.

The impact that different disturbances have on the OEL2C was explored. The three disturbances were: using a small reactive load to enable operation of the OEL during the static portion of the limiter; using a larger reactive load to allow operation of the OEL during the transient part of the limiter; and the addition of a reactor to trigger immediate action of the limiter. The OEL's response to these disturbances showed that it is necessary to model the OEL when performing short-term voltage stability analysis.

Finally, the thesis will show that accurate load modeling is integral in determining power systems short-term voltage stability. For different models of OELs, it was consistently shown for a static load model, the system would remain stable, and for a dynamic load model, the system would collapse. It will also demonstrate the OEL2C model can delay the voltage collapse from occurring for a short period of time as the generator can provide reactive power support during a protected time. This will be demonstrated using the real-time electromagnetic transient platform (RTDS$^{TM}$) so that it can provide more accurate results. The main contribution of this research is demonstrating that a dynamic summing type OEL will impact the short-term voltage stability as a result of its field forcing and transient operation period. It will also show that different load models can have an impact on short-term voltage stability as incorrect modeling can result in an event being missed.