The pervasive use of heterogeneous and non-proprietary information and communication technology exposes the power grid to cyber-attacks. In particular, monitoring-control attacks (MCA), which manipulate control decisions by fabricating measurements, are highly threatening, because MCA are difficult to detect and can coordinately inflict severe consequences at a large scale. To defend against MCA, a semantic analysis framework is proposed in complement to direct-setting intrusion detection. The proposed framework has the advantages of promising runtime and detection performance. The performance of the proposed framework is evaluated under different attack scenarios and compared with a direct-settings intrusion detection systems using a 6-bus test system and the New-England 39-bus test system.