Abstract
Thyristor controlled series capacitors (TCSC) damping controller is considered as a useful device to improve the stability of transmission power systems. Seeking the optimal placement of TCSC controller has become a critical issue for the power system stability. From the working point of view, the TCSC controller is used to improve the voltage magnitude, reactive power and oscillation damping of power systems. The effectiveness of the location of TCSC controller, particularly in the damping of power oscillations depends on its location and input signals in the power system during time of grid fault. In this paper, an energetic method based on controllability and observability Gramians in continuous-time is proposed to determine the optimal location of TCSC. The effectiveness of the proposed method has been verified on the 39-bus New England power system.
Introduction
Modern power grids often operate near their stability limits. When a fault occurs—such as a lightning strike or a line failure—the system experiences low-frequency oscillations (0.1 to 2.0 Hz). If these oscillations are not damped quickly, they can lead to widespread instability or even total grid collapse.
Thyristor Controlled Series Capacitors (TCSC) are powerful tools used to manage these oscillations by dynamically changing the reactance of transmission lines. However, TCSCs are expensive, and their effectiveness is highly sensitive to their physical location in the grid. Finding the "sweet spot" for placement is a complex optimization problem that previous methods have struggled to solve efficiently for both steady-state and transient stability.
Methodology: The Energetic Approach
The proposed method moves away from traditional modal analysis and instead adopts a maximum energetic approach rooted in linear control theory.
Gramian Framework
The core of the method relies on the Controllability Gramian ($W_c$). In control theory, this matrix describes how easily a system's internal state can be moved by an external input. By maximizing the trace of the Gramian, which serves as a proxy for the total "control energy" transmitted from the TCSC to the grid, the controller gains more "leverage" to influence the system's state and damp out oscillations.
The Algorithm
- Modeling: Represent the power system as a set of differential-algebraic equations.
- Candidate Placement: Place a candidate TCSC in various transmission lines.
- Disturbance Simulation: Simulate several disturbance scenarios, such as perturbations in active power.
- Gramian Calculation: Solve the Lyapunov equation to find the Controllability Gramian for each location.
- Optimization: Select the location that yields the maximum total energy across all scenarios.
Experimental Setup
The study utilizes the 39-bus New England power system, a standard benchmark used in power systems research to simulate a complex, multi-generator environment. The system consists of 10 generators and 39 buses.
The authors used PSS/E (Power System Simulation Engineering) software to perform the load flow and dynamic simulations required for the Gramian analysis.
Results and Discussion
The mathematical analysis identified Line 16-17 as the optimal location for the New England 39-bus system, yielding a total energy index of 55.200. For comparison, other candidate lines like 28-26 had an index as low as 2.535.
Time-domain simulations confirmed these findings. When a three-phase fault was simulated, the TCSC at the optimal location (Line 16-17) damped out active power oscillations significantly faster than at other locations.
As shown in the figure above, placing the TCSC in sub-optimal lines (like 28-26 or 14-4) resulted in significantly longer oscillation times or higher peak-to-peak swings, proving that the Gramian-based energy metric is a reliable predictor of real-world damping performance.
Conclusion and Future Work
This energetic method provides a robust way to optimize TCSC placement by focusing on the energy the controller can inject into the system's modes. While effective, the approach currently relies on linearized models. Future work could integrate observability Gramians more deeply to optimize sensor placement and explore the scalability of solving Lyapunov equations for massive, real-world grids with thousands of buses.
Summary updated