A method of dynamic matrix control based on optimal state estimation(OSE-DMC) is proposed for unmeasurable disturbances in a main steam temperature control system. Firstly, the dynamic characteristics and the mathematical model of main steam temperature with the disturbance of spray attemperator water are concluded based on the structure of the main steam temperature system. Then, a Kalman filter is designed to compensate the predictive sequence in the DMC algorithm according to the optimal state estimation theory. The influence of unmeasurable disturbance is effectively suppressed. In addition, a reduced-order algorithm for the Kalman filter is proposed to simplify calculation, which is convenient for practical application and on-line calculation. Simulation results show that the proposed method can increase the effect of disturbance suppression by 20% compared with the conventional DMC algorithm, and the capability of disturbance suppression is significantly improved.