Aiming at the quantitative weighing system developed by Shanghai Cohere Electronic Technology Co., Ltd. (consisting of a high-precision powder shaking device and a 1/10000 balance), which has nonlinearity and time delay problems due to factors such as powder density, powder fluidity, particle size, humidity, and balance delay during operation, the Smith predictive fuzzy PID controller is used to optimize the control method. Firstly, based on theoretical analysis, the transfer function of the system is obtained, and then a Smith predictive fuzzy PID controller is constructed to adapt to the nonlinearity, time delay, and other characteristics of the system. Finally, this algorithm is substituted into MATLAB to perform simulation, and a 1 g quantitative shaking experiment is carried out in the system. The standard deviation of the Smith predictive fuzzy PID control algorithm and the traditional PID algorithm is 0.0020 and 0.0042, respectively. In conclusion, the Smith predictive fuzzy PID control algorithm has better stability in practical environments and can effectively reduce the weighing error of the system.