"High-Fidelity Distributed Generation Dataset for Coordinated Stealth Attack Detection Using Quantum Machine Learning"

"This dataset includes simulated measurements for detecting coordinated stealth cyberattacks on a distributed generation (DG) unit in a microgrid. The data were generated using a dynamic simulation of a grid-connected DG system under normal operation and coordinated stealth attack scenarios. In these attacks, small and carefully designed perturbations are injected into control and measurement signals while remaining within normal operational bounds, making detection challenging.A balanced binary classification dataset of 600 samples is provided, consisting of 300 normal operating points and 300 coordinated stealth attack points. Each sample includes three physically meaningful features: reactive power output of the DG unit (Q_DG1), frequency deviation from nominal operation (f_dev = f_DG1 \u2212 50 Hz), and terminal voltage magnitude (V1). These features were selected because they capture subtle control-loop disturbances caused by stealthy cyber intrusions while preserving realistic system behavior.The dataset is intended for benchmarking classical machine learning, quantum machine learning, and hybrid quantum\u2013classical intrusion detection methods in smart grid cybersecurity applications. It is particularly suitable for studies involving low-dimensional learning, nonlinear feature mapping, and quantum feature embeddings under noisy intermediate-scale quantum (NISQ) constraints. All data are provided in comma-separated value (CSV) format to support reproducibility and easy integration with common data analysis and machine learning frameworks."