Eman M. Hammad

A Game-Theoretic Analysis of Cyber Switching Attacks and Mitigation in Smart Grid Systems

We propose a framework for the analysis of cyber switching attacks and control-based mitigation in cyber-enabled power systems. Our model of the switching attack is simple, only requiring knowledge of the sign of the local relative rotor speed, which may be estimated. The controller is modeled to be resource constrained, choosing to act only during select intervals of time. We make use of an iterated game-theoretic formulation to describe the interactions of the parties and its effect on system stability. Analytic results indicate the potential of the constrained controller to achieve transient stabilization over time using zero-determinant strategies. Numerical results of the New England 39-bus power system demonstrate the potential for such a controller to increase system resilience during cyber-attacks.

A cyber-enabled stabilizing controller for resilient smart grid systems

A parametric controller is proposed for the frequency and phase stabilization after the occurrence of a disturbance in the power grid. The proposed controller is based on the feedback linearization control theory. To drive the frequency of the system generators to stability, the controller relies on receiving timely phasor measurement unit (PMU) readings about the power grid to employ fast-acting flywheels that are situated near the synchronous generators in order to balance a swing equation model of the synchronous generators. The advantages of the proposed controller are that it is tunable and integrates well with existing governor controls in contrast to other forms of PMU-based control. Numerical results show the effectiveness of the proposed controller when applied to the New England power system. Further, a comparison is drawn between the controller and recently-proposed nonlinear controllers for transient stability.

A Cyber-Enabled Stabilizing Control Scheme for Resilient Smart Grid Systems

A parametric controller is proposed for transient stability of synchronous generators after the occurrence of a disturbance in the power grid. The proposed controller based on feedback linearization control theory relies on receiving timely phasor measurement unit (PMU) information from selected parts of the power grid to employ fast acting flywheels that are situated near synchronous generators. The local storage devices aim to balance a swing equation model of the synchronous generator to drive the associated rotor speed to stability. The advantages of the proposed controller include that it is tunable and integrates well with existing governor controls in contrast to other forms of PMU-based control. Further, a comparison is drawn between the proposed controller and recently proposed nonlinear controllers for transient stabilization. Numerical results show the effectiveness and robustness of the proposed controller when applied to the 39-bus 10-generator New England power system.

A resilient feedback linearization control scheme for smart grids under cyber-physical disturbances

A cyber-enabled parametric control scheme is proposed for efficient transient frequency and phase stabilization in the power grid. Different implementations of the proposed control are investigated in this work. First, a centralized control scheme is proposed where the controller relies on timely phasor measurement unit (PMU) information about the grid to employ fast-acting energy storage systems for stabilization. Further, a decentralized controller implementation assumes information about the rest of the grid is not available, and hence acts based on local PMU measurements. For the case of cyber attacks targeting communication channels and resulting in large delays or absence of PMU data, we propose a robust combined control scheme where the controller operates in a centralized mode by default and switches to the decentralized scheme if PMU information is delayed or not available. Numerical results show the effectiveness and robustness of the proposed controller against physical and cyber-physical disturbances in the 39-bus 10-generator New England power system.

A Cyber-Physical Control Framework for Transient Stability in Smart Grids

Denial of service attacks and communication latency pose challenges for the operation of control systems within power systems. Specifically, excessive delay between sensors and controllers can substantially worsen the performance of distributed control schemes. In this paper, we propose a framework for delay-resilient cyber-physical control of smart grid systems for transient stability applications. The proposed control scheme adapts its structure depending on the value of the latency. As an example, we consider a parametric feedback linearization (PFL) control paradigm and make it “cyber-aware.” A delay-adaptive design that capitalizes on the features of PFL control is presented to enhance the time-delay tolerance of the power system. Depending on the information latency present in the smart grid, the parameters and the structure of the PFL controller are adapted accordingly to optimize performance. The improved resilience is demonstrated by applying the PFL controller to the New England 39-bus and WECC 9-bus test power systems following the occurrence of physical and cyber disturbances. Numerical results show that the proposed cyber-physical controller can tolerate substantial delays without noticeable performance degradation.

Practical limitations of sliding-mode switching attacks on smart grid systems

Switching attacks in smart grid systems have gained some recent attention by the research community. Based on understanding the structure of the physical system and accessing the system variables, effective sliding-mode switching attacks can be used to disrupt the normal operation of the power grid. This article investigates the practical limitations of such attacks. Sliding-mode switching attacks on a single-machine infinite-bus system are considered in this work, and the impact of some practical limitations is investigated. These limitations include sampling period, quantization level, signal-to-noise ratio, hysteresis margin and minimum time to switch of a circuit breaker, and communication latency. Results of this work detail the effectiveness of the sliding-mode switching attacks under different practical limitations.

On using distributed energy resources to reshape the dynamics of power systems during transients

This work investigates controlling the operation of distributed energy resources (DERs) in order to reshape the dynamics of the power system during instability periods. Specifically, the output of the DERs is controlled using a linear feedback optimal (LFO) controller for stabilizing the rotor speeds of the synchronous generators after the occurrence of a disturbance in the power grid. Based on optimal linear-quadratic control theory, the proposed LFO controller relies on receiving timely system information to employ fast-acting DERs that are situated near synchronous generators in order to drive the rotor speed of the generators to stability. The performance of the proposed controller is investigated on the 39-bus 10-generator New England test power system. Further, the performance of the LFO distributed controller is investigated in the presence of non-ideal practical measurement, communication, and storage constraints.

A systematic approach to delay-adaptive control design for smart grids

Communication latency poses a challenge to the operation of control systems in power systems. Specifically, excessive delay between sensors and controllers can substantially worsen the performance of distributed control schemes. In this paper we utilize the parametric feedback linearization (PFL) control to actuate energy storage systems in order to achieve transient stability. A delay-adaptive design capitalizing on the features of PFL control is presented in this work in order to enhance the time delay tolerance of the power system. Specifically, a piece-wise linear delay-adaptive property of the PFL control is investigated. The parameters of the PFL controller are adapted according to the latency value in the cyber component of the grid in order to optimize performance. The enhancements achieved by the proposed delay-adaptive scheme on the performance of the distributed controller are shown when applied to the New England 39-bus and WECC 9-bus power systems following the occurrence of a physical disturbance. Further, numerical results show that the proposed delay-adaptive control is able to tolerate substantial delay without degradation in performance.

A game-theoretic control approach to mitigate cyber switching attacks in Smart Grid systems

A parametric game-theoretic controller is proposed to stabilize power systems during and after cyber switching attacks. The attacks are based on calculated switchings of external power sources in order to destabilize the power system. The controller relies on receiving timely information from the cyber component of the smart grid to control the output of fast-acting external power sources in order to balance the swing equation of the system. In this regard, game-theoretic analysis is applied to devise reactive strategies for the controller to drive the frequency of the systems generators to stability. The proposed controller gives system operators the flexibility to meet constraints on external power while achieving an acceptable stability time. Numerical results show the usefulness of the proposed controller in stabilizing the WECC 3-generator power system during and after a switching attack.

On the Use of Energy Storage Systems and Linear Feedback Optimal Control for Transient Stability

In this paper, we study a distributed control strategy that harnesses the highly granular data available in future power systems in order to improve system resilience to disturbances. Specifically, we investigate the role of external energy storage systems (ESSs) in stabilizing the dynamics of power systems during periods of disruption. We consider an information-rich multiagent framework and focus on ESS output control via linear feedback optimal (LFO) control to achieve transient stability. The LFO control scheme relies on receiving timely state information to actuate distributed ESSs in order to drive the synchronous generators to stability. We evaluate the performance of the LFO control on the 39-bus 10-generator New England test power system in the presence of ideal and nonideal conditions including communication latency, finite sampling rate, and sensor noise. The LFO controller is found to have a simple structure, be tunable, and to have fast response to achieving transient stability while being sensitive to information latency and data rate.