To meet a demand for various polymer products，grade transitions are frequently required in the polymerization process.The recent studies for grade transition optimization mostly focused on adjusting the quality specifications to target values.However，the variations on quality specifications and state variables during the grade transition will greatly affect the operation stability of the system and the product qualities.In this study，the authors proposed a simple and effective approach to alleviate the fluctuations during grade transition by adding path constraints to the optimization structure of grade transition in a gas-phase fluidized bed reactor for polyethylene.The path constraints were transformed to control variable constraints by solving the differential algebraic equation，which significantly improves the traditional control vector parameterization method to solve dynamic optimization problems with path constraints.The simulation results confirm the effectiveness of the proposed approach to enhance the operation stability.