Online Evaluation of Measurement Uncertainty in Sensor Networks: A Case Study on Voltage Transformers

Sensor networks serve as the perceptual core of industrial systems, such as the Internet of Things and smart grid, where the accuracy of individual measurements is pivotal to ensuring reliable state estimation. Nevertheless, factors, such as environmental interference and device aging, introduce measurement uncertainties to sensors, and the incompleteness of the information would propagate and accumulate across the network. Despite advances in evaluating the measurement uncertainty of individual sensors and incomplete assessments of the measurement status in sensor networks, network-level evaluation of measurement uncertainty remains an open yet practical problem. Thus, we propose a recursive framework for online evaluation of measurement uncertainties within sensor networks, demonstrated through a case study on voltage transformers in the smart grid. The proposed method incorporates a measurement model to capture interdevice dependencies and utilizes the Monte Carlo method to propagate parameter distributions considering the measurement uncertainties. Next, a recursive workflow for evaluating measurement uncertainties across the entire sensor network is proposed. Specifically, the calibrated nodes are selected based on the uncertainty propagation theory, and then the Bayesian fusion is applied to estimate the parameter distributions of extended nodes. Simulations conducted on the IEEE 30-node system, supplemented by real-world validation with power system data, indicate that the proposed method is capable of achieving high evaluation accuracy and stable uncertainty propagation.