With grid modernization, smart inverters are increasingly used to execute advanced controls for distribution network reliability. However, this also increases the cyber-attack space. This paper focuses on the defense approaches to restore the system to normal operation circumstances in the presence of cyber-attacks. A unique deep reinforcement learning (DRL) method is developed to minimize voltage violations and reduce power losses for impacted feeders. The defense problem is reformulated as a Markov decision-making process to dynamically control DERs while minimizing load shedding. This is achieved via an improved soft actor-critic (SAC)-based DRL algorithm, which can govern DER set points and load-shedding scenarios in discrete and continuous modes via the auto-tune entropy and Gaussian policy features. Numerical comparison results on the modified IEEE 123-node system with other control approaches, such as Volt-VAR (VV), Volt-Watt (VW), and model predictive control (MPC) show that the proposed method can eliminate voltage violations and provide feasible control actions that perform complete mitigation of cyber-threats.

This paper proposes the integration of battery impedance spectroscopy (BIS) into a battery management system with a reduced number of inductor and switch components compared to existing methods. Moreover, this paper presents an internal preheating mechanism, active state of charge (SoC) equalizer, and BIS without an external power source so that there is no additional component needed. During a BIS measurement, the battery management system controls the switches in order to circulate a periodical current signal and measure the battery voltage terminals. The SoC equalizer is needed to equalize the SoC in each battery to prevent over-charging and over-discharging. The internal preheating mechanism is used during low temperatures to preheat the battery since the battery has a higher resistance at low temperatures. The prototype can measure the battery impedance from 100 – 5 kHz with the parameter errors for the internal series resistances of the battery ( R_batt ) are 1.59%, 3.72%, and 3.72% and the charge transfer resistances ( R_ct ) are 5.91%, 8.04%, and 6.89% for B₁ , B₂ , and B₃ . While the double layer capacitances (C _dl ) parameter errors are 5.62%, 7.29%, and 6.48% for B₁ , B₂ , and B₃ . A 50 kHz switching frequency is used with the equalizing duration of approximately 400s with the 90.67% efficiency. The self-heating energy consumption is 0.21 %/° C from 0 to 13.5 °C.

Internal resistance is one of the important parameters in the Li-Ion battery. This paper identifies it using two different methods: electrochemical impedance spectroscopy (EIS) and parameter estimation based on equivalent circuit model (ECM). Comparing internal resistance, the conventional parameter estimation method yields a different value than EIS. Therefore, a hysteresis-free parameter identification method based on ECM is proposed. The proposed technique separates hysteresis resistance from the effective resistance. It precisely estimated actual internal resistance, which matches the internal resistance obtained from EIS. In addition, state of charge, open circuit voltage, and different internal equivalent circuit components were identified. The least square method was used to identify the parameters based on ECM. A parameter extraction algorithm to interpret impedance spectrum obtained from the EIS. The algorithm is based on the properties of Nyquist plot, phasor algebra, and resonances. Experiments were conducted using a cellphone pouch battery and a cylindrical 18650 battery.

The direct borohydride fuel cell (DBFC) is directly fed sodium borohydride as a fuel and hydrogen peroxide as an oxidant. The output voltage of the DBFC varies with respect to current demand. Therefore, it requires a dc-dc converter for a regulated output voltage. The dc-dc converter should be designed considering both the impact of the fuel cell and load disturbances to achieve wide range voltage regulation. This paper analyzes the impact of the DBFC impedance on the dc-dc converter. Based on the converter's small signal model including the DBFC impedance, the boost converter was evaluated. Impacts of the voltage and current control loops on the boost converter were investigated. Improved response time and stability of the DBFC dc-dc converter were observed by adding a supercapacitor between the DBFC and the dc-dc converter. Finally, two converter controllers, a nonlinear feedforward controller and a state feedback controller are proposed for further improvement of the dc-dc converter response time. A prototype 20-W dc-dc converter was built and tested to show the response improvement of the proposed analysis and control scheme.

In a smart campus, the potential to reduce energy consumption in buildings is high. In this sense, campus utility operators increasingly need an energy management system capable of interacting with buildings' demand response (DR), which can be automated and cloud-based with the help of information and communication technologies (ICTs). Even so, cloud-based DR implementation still faces some challenges, including computation load, communication delay, and privacy concerns. To address these issues, we developed a cloud-based DR model, which is implemented as a cloud-edge combined computing platform. A competitive equilibrium state of load shedding among the campus buildings is designed to balance the trade-off between occupant discomfort and energy usage. In this paper, we adopt the clustering-based distributed alternating direction method of multipliers (ADMM) to decompose the campus load reduction assignment problem into a series of DR subproblems, each of which could be solved by each building with the edge layer computing devices. Two different campus study cases demonstrate the feasibility and performance of the proposed distributed DR algorithm.