In this paper, an online measurement and monitoring framework for smart-grid is presented, integrating false data injection (FDI) detection and falsified data reconstruction. The model continuously acquires operational and meteorological data, evaluates system performance through hybrid measurement indices, and updates operating setpoints in a rolling 24-hour cycle to ensure resilience against the considered FDI attacks. Initially, the microgrid is modeled in detail, incorporating equipment such as batteries as well as wind and solar resources, and connected to both critical and non-critical loads. In this modeling process, real-world nonlinear cost functions and dynamic behaviors are considered and subsequently linearized to accurately reflect the realistic behavior of the system. The microgrid interfaces with a central Energy Management System (EMS) via SCADA, utilizing two dedicated communication channels for data reception and transmission. Through the reception channel, it receives optimized operating setpoints for its equipment from the central EMS. Through the transmission channel, it sends measured microgrid data, such as load consumption, generation output, battery power, and power exchanged with the upstream grid, back to the central EMS. Both the forward and backward communication channels are susceptible to FDI attacks. The proposed model defines a set of hybrid indicators, including the Power Balance Index (PBI), Battery Setpoint Deviation (BSD), PV Power Deviation (PPD), Wind Power Deviation (WPD), and Grid Power Deviation (GPD). By combining these indicators, the model can determine the location, timing, and type of falsification in both data reception and transmission channels. It can also estimate the magnitude and direction (increase or decrease) of the falsified signal, and subsequently reconstruct the original data. This structure increases resilience against FDI attacks and physical disruptions by continuously updating operating points through a rolling-horizon framework.
