Traditionally, Volt/VAR optimization (VVO) is performed in distribution networks through legacy devices such as on-load tap changers (OLTCs), voltage regulators (VRs), and capacitor banks. With the amendment in IEEE 1547 standard, distributed energy resources (DERs) can now provide reactive power support to the grid. For this, renewable energy-based DERs, such as PV, are interfaced with the distribution networks through smart inverters (SIs). Due to the intermittent nature of such resources, VVO transforms into a dynamic problem that requires extensive communication between the VVO controller and devices performing the VVO scheme. This communication, however, can be potentially tampered with by an adversary rendering the VVO ineffective. In this regard, it is important to assess the impact of cyberattacks on the VVO scheme. This paper develops a real-time co-simulation setup to assess the effect of cyberattacks on VVO. The setup consists of a real-time power system simulator, a communication network emulator, and a master controller in a system-in-the-loop (SITL) setup. The DNP3 communication protocol is adopted for the underlying communication infrastructure. The results show that corrupted communication messages can lead to violation of voltage limits, increased number of setpoint updates of VRs, and economic loss.

This paper introduces an energy management system (EMS) aiming to minimize electricity operating costs using reinforcement learning (RL) with a linear function approximation. The proposed EMS uses a Q-learning with tile coding (QLTC) algorithm and is compared to a deterministic mixed-integer linear programming (MILP) with perfect forecast information. The comparison is performed using a case study on an industrial manufacturing company in the Netherlands, considering measured electricity consumption, PV generation, and wholesale electricity prices during one week of operation. The results show that the proposed EMS can adjust the prosumer's power consumption considering favorable prices. The electricity costs obtained using the QLTC algorithm are 99% close to those obtained with the MILP model. Furthermore, the results demonstrate that the QLTC model can generalize on previously learned control policies even in the case of missing data and can deploy actions 80% near to the MILP's optimal solution.

Embedded controllers, sensors, actuators, advanced metering infrastructure, etc. are cornerstone components of cyber-physical energy systems such as microgrids (MGs). Harnessing their monitoring and control functionalities, sophisticated schemes enhancing MG stability can be deployed. However, the deployment of ‘smart’ assets increases the threat surface. Power systems possess mechanisms capable of detecting abnormal operations. Furthermore, the lack of sophistication in attack strategies can render them detectable since they blindly violate power system semantics. On the other hand, the recent increase of process-aware rootkits that can attain persistence and compromise operations in undetectable ways requires special attention. In this work, we investigate the steps followed by stealthy rootkits at the process level of control systems pre- and post-compromise. We investigate the rootkits' precompromise stage involving the deployment to multiple system locations and aggregation of system-specific information to build a neural network-based virtual data-driven model (VDDM) of the system. Then, during the weaponization phase, we demonstrate how the VDDM measurement predictions are paramount, first to orchestrate crippling attacks from multiple system standpoints, maximizing the impact, and second, impede detection blinding system operator situational awareness.