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.

Reliability assessment of power semiconductor devices requires accelerated stress tests. Different test methods might produce different results. In this paper, two different test methods are used for on-state resistance degradation for Si MOSFETs: High frequency cycling and low frequency cycling. Presented results indicate that the degradation of on-state resistance occurs irrespective of which method is used. Additionally, gate-to-source capacitance degradation is shown where results show that the gate-to-source impedance degrades over time.

The theory of switched systems which focuses on stability and control synthesis of systems with multiple modes or different dynamics, has gained significant attention from control theorists, computer scientists, and practicing engineers. Switching between motor controllers due to sensor faults, faster response, or targeted dynamics, open a new dimension of shaping a motor drive's performance. There are three main types of control strategies for induction motor drives: constant volts per hertz (V/f) control, field-oriented control (FOC), and direct torque control (DTC). Multi-mode-controlled induction motor drive can be modelled as a switched system where each control mode results in different motor dynamics. In this paper, a comprehensive state-space model of an induction motor drive under different control mode is developed11This work is supported by the National Science Foundation under Award 1752297.. Detailed analysis as well as simulation results are included to demonstrate the validity of the state-space representation for induction motor under different mode.

Modeling faulty behavior of systems has benefits in diagnosis and control. In this paper a data-driven method, dynamic mode decomposition with control (DMDc), is employed for modeling an inverter-fed induction machine. Results are shown and compared for two scenarios: A step input change and an inverter fault. For both cases, the algorithm can correctly predict behavior of the system. The advantage of this model is its independence from the system parameters. The results show promise for data-drivenfault diagnostics and system modeling.

Efficient conversion of the PV energy requires better hardware and software solutions. Apart from efficient power structures, microinverters should have control algorithms like maximum power tracking and phase locked loop (PLL). In this study, an overview of push-pull based microinverter and its control structure is presented. PLL structure is given with a mathematical background and it is simulated for different cases using MATLAB. PLL algorithms are implemented in a microcontroller and they compared with experimental results. Experimental results are in agreement with the simulation results.