Search

Power electronics (PE) and high-frequency switching circuits are key to superior performance of electric vehicles. It is vital to monitor the condition of the PE components in real-time for safety and reliability. In this article, we propose two anomaly detection methods based on a combination of data preprocessing to suppress noise and outliers, multivariate regression models to predict signals of interest under nominal operation, and sequential analysis of residuals. In particular, the methods utilize median filtering to extract on-state medians in each switching cycle in nonlinear autoregressive exogenous neural network models or filtered on-state data in partial least squares-based models to represent the nominal circuit behavior. Optimal and approximate dynamic programming-based feature selection methods are developed to select the most informative signals or their transformations. Predictions from the learned models are used to generate the residuals for anomaly detection by Page's cumulative sum test. The proposed models and anomaly detection methods are validated on three accelerated aging experimental datasets, comprised of 60 power mosfet devices with low-frequency and high-frequency switching under disparate operating conditions. Due to the simplicity and efficiency of the data-driven anomaly detection schemes, the proposed methods can potentially be embedded in real-time digital platforms.

Advanced metering infrastructures (AMIs) enable two-way communications between the utility and customers opening up opportunities for an efficient and reliable smart grid. This, however, opens up the grid to a vast amount of attack vectors and intruders. Therefore, cyber security is an indispensable part of the smart grid communications’ infrastructure and must be considered in the early phases of planning. The achievement of a secure grid infrastructure requires cryptographic primitives whose strength is dependent on the utilized keys, raising the need for secure key generation/distribution mechanisms. Furthermore, the underlying security mechanisms should be scalable due to the large AMI network sizes. In this paper, we propose the usage of identity-based-cryptography as the basic block of the key distribution mechanism to generate secret keys between neighboring smart meters in a non-interactive manner. Physical unclonable functions are proposed on the hardware platform to eliminate key compromise at the hardware level. We also propose a secure mechanism for updating private keys that are utilized as a seed within the identity-based cryptosystem. A lightweight key delivery mechanism exploiting multicast network features is presented along with a new authentication algorithm for the updated keys. Hardware tests are performed on a Cortex M3-based microcontroller to mimic smart meters’ processors and assess the feasibility of the proposed key authentication. Experiments conducted on a network simulator validate the feasibility of the proposed methodology indicating key delivery latency reduction up to 80% and network traffic reduction up to 27%.