In this paper, sensor fault detection, isolation and identification model-based approach is designed. We introduce a new state variable so that an augmented system can be constructed to treat sensor faults as actuator faults. The approach uses the model of the system and a bank of adaptive observers to generate residuals. Structured residuals are defined in such way to isolate the faulty sensor after detecting the fault occurrence. The advantage of this method is that we can treat single, multiple and simultaneous sensor faults. In this study, we consider that only abrupt faults in the system sensor can occur. The proposed strategy is validated by simulation results of a nonlinear model of a waste water treatment process.