Global Sensitivity Analysis Based Method for Distributed Voltage Control Performance Study

Various sources of uncertainty can jeopardize the performance of control algorithms when practically implemented in power systems. This paper proposes a novel and replicable method based on variance-based sensitivity analysis (SA) to effectively assess the performance of control algorithms. It overcomes the limitations of traditional SA, it can deal with dynamic and iterative algorithms under uncertainty, and it is computationally cheap. The method is applied to study the performance of a Docker container-based distributed voltage control algorithm affected by noisy measurements, uncertain line parameters, and asynchronism. The test results show the relevance of measurements' uncertainty, the robustness to line parameter errors of the control algorithm, and the importance of duly investigating the asynchronous implementation thereof.