Sensor measurements are mission-critical for monitoring and controlling power systems because they provide real-time insight into the grid operating condition; however, confidence in these insights depends greatly on the quality of the sensor data. Uncertainty in sensor measurements is an intrinsic aspect of the measurement process. In this paper, we develop an analytical method to quantify the impact of measurement uncertainties in numerical methods that employ the Koopman operator to identify nonlinear dynamics based on recorded data. In particular, we quantify the confidence interval of each element in the push-forward matrix from which a subset of the Koopman operator's discrete spectrum is estimated. We provide a detailed numerical analysis of the developed method applied to numerical simulations and field data collected from experiments conducted in a megawatt-scale facility at the National Renewable Energy Laboratory.

翻译：传感器测量对于电力系统的监测与控制至关重要，因为它能实时提供关于电网运行状态的洞察；然而，这些洞察的可靠性在很大程度上取决于传感器数据的质量。传感器测量中的不确定性是测量过程中固有的一个方面。本文提出了一种分析方法，用于量化采用Koopman算子从记录数据中识别非线性动力学的数值方法中测量不确定性的影响。具体而言，我们量化了推进矩阵中每个元素的置信区间，Koopman算子离散谱的一个子集正是基于该矩阵进行估计的。我们对所开发的方法进行了详细的数值分析，并将其应用于数值模拟以及从美国国家可再生能源实验室兆瓦级设施进行的实验中收集的现场数据。