—We consider the improvement of dynamic security in power systems by tuning of the power plant controllers. The parameters of existing controllers are tuned like power system stabilizers in order to increase the stability reserve and oscillation damping after a dropout of power plants and power lines. The tuning is done in two steps: in the first step, the power system is stabilized after the dropout of a power plant or power line. Then, oscillation damping is increased after the dropout by an H-infinity optimization approach. Both steps use linear matrix inequality optimization methods. To perform the tuning, we introduce a modeling method for the dropout of power plants and power lines, such that the dropout of components can be considered directly as a disturbance input for the H-infinity optimization. Finally, we evaluate the approach on the IEEE 39 bus model. We show that the presented methods successfully stabilize the system and improve power oscillation damping after the dropout of different power plants and power lines.

Power system stabilizers are controllers which damp power oscillations in electrical networks. They typically reside in the automation system of the power plant. Their design and structure are typically fixed in the design of the power plant. Optimal design and tuning of these decentralized controllers such that power oscillations are avoided is a challenging task. In the first part of the paper, we outline this problem and transform it into a so called structured controller synthesis problem where the control structure is fixed and optimal controller parameters need to be found. Based on this formulation, which preserves the real controller parameters, we propose a coordinate descent method to solve the controller design and tuning problem. To this end, we consider additional steady-state constraints in the system. We show the effectiveness of the proposed approach by detailed simulations of an established power system benchmark.

The rising share of renewable generation increases uncertainty in power system operation. New methods are needed for secure operation of power systems under uncertainties. We present a robust optimal power flow algorithm for mixed ac and dc power transmission systems which considers uncertain renewable generation and/or load. The uncertain infeeds are assumed to lie inside a known norm-bounded set. The presented algorithm guarantees satisfaction of various steady-state power system constraints for all uncertainties within this set. In order to satisfy these constraints at minimal cost, the algorithm modifies both the redispatch and primary reserve. This is done through the optimization of setpoints and droop gains of generators and high-voltage dc converters. The algorithm is evaluated on a test system, the Cigre B4 DC grid, the IEEE RTS 96 grid, and on the Danish transmission system. In these examples, we show that redispatch costs can be reduced by additionally optimizing primary reserve, and demonstrate primary reserve sharing between asynchronous ac grids through dc grids. We illustrate the impact of the uncertainty set choice on the worst-case redispatch costs, perform a sensitivity analysis of the approach, and evaluate the computation time as a function of the number of uncertainties.