In modern power systems, sensor data is transmitted over long communication lines to the control centers. There, data is analysed and utilized by various optimization or control algorithms for power system planning, operation, and scheduling. The communication lines are not secured and the data is not encrypted, which makes these networked systems vulnerable to cyber-attacks. In this paper, the false data injection (FDI) attack is launched on the communication line transmitting power sensor data to the control centers. This data carries the measurements of load demand, and branch power flows, which is further analysed and input to the optimal power flow (OPF) algorithm. The attack vectors are designed to manipulate the line power flows. We present a vulnerability assessment of FDI attacks against OPF using case studies of IEEE 2-bus and 37-bus power systems.