Adversarial Attack Mitigation Strategy for Machine Learning-Based Network Attack Detection Model in Power System

Huang, Rong; Li, Yuancheng

doi:10.1109/tsg.2022.3217060

Cited by 36 publications

(8 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many countries and regions focus on wind power generation, new energy vehicles, and other aspects. [1][2][3][4][5][6] As the core component of the above system, the motor has been widely studied, among which, the permanent magnet synchronous motor stands out in a crowd of motors because of its high-power density, high efficiency, lightweight, and other advantages. [7][8][9][10][11] The control algorithm directly affects the performance of the PMSMs system.…”

Section: Introductionmentioning

confidence: 99%

Simplified robust two-step model predictive control for permanent magnet synchronous motor

Zhang,

Cen

2024

Advances in Mechanical Engineering

View full text Add to dashboard Cite

Traditional finite control set model predictive control (FCS-MPC) selects an optimal voltage vector in one control cycle. However, considering multiple control cycles, it cannot be proved that the voltage vector is optimal, and the uncertainty of PMSM parameters seriously affects the prediction accuracy. In order to balance the relationship between switching frequency, steady-state performance, and robustness, a multi-step robust FCS-MPC is proposed. Firstly, an incremental prediction model is established to eliminate the flux linkage of permanent magnets. The inductance parameters in the incremental model are identified online based on the observer and inductance extraction algorithm. Then, based on the principle of no-beat, the candidate voltage vector in two control periods is simplified to get the candidate voltage vector, and the cost function is used to determine the optimal voltage vector again. Finally, the proposed FCS-MPC method is compared with the traditional FCS-MPC method. The experimental results show the effectiveness of FCS-MPC strategy.

show abstract

Section: Introductionmentioning

confidence: 99%

Simplified robust two-step model predictive control for permanent magnet synchronous motor

Zhang,

Cen

2024

Advances in Mechanical Engineering

View full text Add to dashboard Cite

show abstract

“…The similarity between adversarial signal attacks and benign signals poses a substantial threat, as it can deceive human operators and lead to a lack of awareness regarding the presence of the attack. Previous research has explored the vulnerability of DL models to specific adversarial attacks in power systems [9], investigated the effectiveness of defense methods against untargeted attacks [10], and examined joint adversarial example and false data injection attacks in power system state estimation [11]. However, no study is currently on targeted universal adversarial perturbation of time series data (TSD).…”

Section: Introductionmentioning

confidence: 99%

AdaptEdge: Targeted Universal Adversarial Attacks on Time Series Data in Smart Grids

Khan,

Mynuddin,

Nabil

2024

IEEE Trans. Smart Grid

View full text Add to dashboard Cite

Deep learning (DL) has emerged as a key technique in smart grid operations for task classification of power quality disturbances (PQDs). Even though these models have considerably improved the efficiency of power infrastructure, their susceptibility to adversarial attacks presents potential difficulties. For the first time, we introduce a novel algorithm called Adaptive Edge (AdaptEdge), which effectively employs targeted universal adversarial attack to deceive DL models working with time series data. The unique contribution of this algorithm is its ability to maintain a delicate balance between the fooling rate and the imperceptibility of perturbations to human observers. Our results demonstrate a fooling rate of up to 90.78% in the ResNet50 model-the highest achieved thus far-while maintaining an optimal signal-to-noise ratio (SNR) of 3dB and ensuring signal integrity. We implemented our algorithm across various advanced DL models and found considerable efficacy, demonstrating its adaptability and versatility across diverse architectures. The results of our study highlight the pressing need for developing more robust DL model implementations in the context of the smart grid. Additionally, our proposed approach demonstrates its effectiveness in addressing this need.

show abstract

“…The complexities associated with trojan signals, which exhibit a notable similarity to clean signals, pose a significant threat, resulting in jeopardizing the integrity of the system. Prior research has investigated adversarial attacks on DL models and their defense in SG [9], such as joint adversarial example and false data injection attacks in power system state estimation, and the efficacy of defense methods against untargeted attacks [10]. However, there is no existing study on trojan attacks specifically targeting DL models working with time series data (TSD) in the SG.…”

Section: Introductionmentioning

confidence: 99%

Explainable Neural Dynamics Models for Motor Temperature Prediction

Liao,

Chen,

Long

et al. 2023

Preprint

View full text Add to dashboard Cite

<p> The permanent magnet synchronous motor finds extensive use in industrial applications, and the development of effective thermal management solutions is crucial to enhance its power density. Accurate temperature prediction of the permanent magnet synchronous motor serves as the fundamental basis for designing effective thermal management strategies. Model-based prediction methods exhibit superior real-time performance, but the intricate modeling process requires substantial expert knowledge guidance and lacks versatility. Conversely, data-driven prediction methods, while offering flexibility, often lack physical implications in terms of system dynamics. This paper proposed a structured linear neural dynamics model for motor temperature prediction. This model is data-driven, with prior knowledge integrated into its structure, which preserves flexibility while guaranteeing system stability through the Perron-Frobenius theorem. Additionally, this paper achieves the decoupling of control input from state transitions and the embedded deployment of this model. The method is validated with a real dataset. The lightweight feature is demonstrated by the implementation of an STM32 Microcontroller with 1.808 KB and 27 mW. The paper is accompanied with open source data and code at GitHub https://github.com/ms140429/Explainable-Neural-Dynamics-Model</p>

show abstract

Adversarial Attack Mitigation Strategy for Machine Learning-Based Network Attack Detection Model in Power System

Cited by 36 publications

References 38 publications

Simplified robust two-step model predictive control for permanent magnet synchronous motor

Simplified robust two-step model predictive control for permanent magnet synchronous motor

AdaptEdge: Targeted Universal Adversarial Attacks on Time Series Data in Smart Grids

Explainable Neural Dynamics Models for Motor Temperature Prediction

Contact Info

Product

Resources

About