With the deepening cyber-physical integration of new power systems

cyberattacks such as false data injection attacks (FDIAs) pose a significant threat to the functional security of state estimation. Existing detection approaches often rely heavily on attack-model priors

exhibit limited adaptability to stealthy or evolving threats

and are weakly integrated with core estimation modules. This paper considers hybrid measurement-based state estimation integrating supervisory control and data acquisition (SCADA) and wide-area measurement systems(WAMS)

and proposes an attack-resilient functional design with embedded security detection based on spatiotemporal dynamic heterogeneity. A spatiotemporal disturbance mechanism is introduced to actively construct time-sampling and spatial-selection matrices

thereby injecting dynamic heterogeneity and structural uncertainty into the estimation process to invalidate the attacker’s prior assumptions. Building on statistical characterization and significance testing

this study develops a mean residual shift (MRS)-based detection index that is tightly coupled with the estimator. A robust optimization-based detection model is developed to enable coordinated adjustment of disturbance strategies and detection thresholds

which is solved via a disturbance-response adversarial optimization (DRAO) algorithm. Simulations on standard test systems demonstrate that the proposed method effectively detects a wide range of enhanced FDIA strategies while preserving the accuracy of the underlying state estimation.