Abstract
This paper discusses how a cyber attack could take advantage of torsional resonances in the shaft of turbo-generators to inflict severe physical damage to a power system. If attackers were able to take over the control of a battery energy storage device, they could modulate the injection of this device at a frequency that matches one of the sub-synchronous resonance frequencies of a generator. Small changes in injection might be sufficient to excite one of these mechanical resonances, resulting in metal fatigue and ultimately a catastrophic failure in the shaft of the generator. Using a state-space model of the electromechanical system, the paper develops transfer functions linking the magnitude of the malicious injections to the magnitude of oscillations in the speed and angle of the various masses connected to the shaft. Numerical results from a two-area power system demonstrate the existence of vulnerable frequencies and show that damaging mechanical oscillations can be triggered without causing easily detectable signals at the generator terminals.
Introduction
Traditional cyber-attacks typically target data integrity or software availability, while physical attacks on infrastructure usually require an adversary's physical presence. This research explores a "cyber-physical" middle ground: leveraging remote access to control systems to inflict permanent physical damage. The primary mechanism explored is Subsynchronous Resonance (SSR).
In a power plant, a massive shaft connects various turbine stages to the electrical generator. This shaft possesses specific natural mechanical frequencies. If the electrical grid injects power at a frequency matching one of these mechanical "sweet spots," the shaft can begin to vibrate violently. Over time, these oscillations lead to metal fatigue or immediate "catastrophic failure," potentially snapping the shaft of a multi-million dollar generator. Crucially, these oscillations can be "stealthy"; if the injection magnitude is small, terminal signals monitored by grid operators may appear normal while internal mechanical components suffer severe stress.
Methodology
The authors developed a mathematical framework to prove the feasibility of this attack using state-space modeling and transfer function analysis.
Torsional Modeling
The study combines generator "swing equations" with a five-mass torsional model. This model represents the High Pressure (HP), Intermediate Pressure (IP), and two Low Pressure (LP_A, LP_B) turbine sections, along with the generator rotor (Gen), all connected by elastic shafts.
Analytical Framework
Transfer functions were derived to link malicious power injections from a Battery Energy Storage System (BESS) to two sets of outputs:
- Terminal Variables ($y_1$): Observable metrics such as speed and angle at the generator terminal.
- Torsional Variables ($y_2$): Internal stress metrics, specifically speed and angle differences between adjacent masses on the shaft.
To quantify stealth, the authors introduced the $R_M$ Ratio, defined as the ratio between internal torsional oscillation magnitude and external terminal oscillation magnitude. A high $R_M$ indicates an attack that is high-damage but low-visibility.
Numerical Simulation
The theory was validated using a standardized Two-Area Power System model. The simulation involved a 1 MW square-wave injection from a BESS—a relatively small amount of power (approx. 0.1% of total load) compared to the generator's capacity. Mechanical parameters for the torsional system were sourced from established literature for turbo-generators.
Key Results and Insights
The research identified specific "vulnerable ranges" of frequencies where the attack is most effective and hardest to detect.
Quantified Stealth and Frequency Specificity
The calculated $R_M$ values were significantly greater than 1 at certain subsynchronous frequencies. For instance, in the 25.42–25.47 Hz range, the internal speed deviation was shown to be several times higher than the terminal deviation, demonstrating a high degree of stealth.
Physical Feasibility and Detection Difficulty
Time-domain simulations confirmed that even a small 1 MW injection could trigger sustained, high-amplitude mechanical oscillations. Because terminal measurements remained close to steady-state values, these oscillations would likely fall below the detection thresholds of conventional SCADA alarms and protective relays.
BESS as an Attack Vector
Battery Energy Storage Systems are ideal for such attacks due to their power electronics, which can be modulated at high frequencies. As BESS integration into the grid grows, so does the potential attack surface for exploiting these mechanical resonances.
Conclusion and Future Work
While this study assumes a "white-box" scenario where the attacker has complete knowledge of the system's mechanical parameters, it highlights a critical vulnerability in modern power systems. Future work should focus on identifying the most susceptible generators in real-world grids and developing high-resolution monitoring techniques or specialized protective algorithms to detect and mitigate these high-frequency malicious injections.
Summary updated