In this paper, a control-oriented model has been investigated for the evaporator in an Organic Rankine Cycle (ORC) system. Due to the nonlinearity and uncertainty existed in the evaporator, the model parameters of the evaporator have been identified by the recursive least square algorithm with forgetting factor. The improved generalized predictive control algorithm with constraints is then utilized to control the outlet temperature of the evaporator. The simulation results show the effectiveness of the proposed scheme.