In order to improve the weighing precision and efficiency of powdered feed packaging, a quantitative weighing and packaging control system is designed based on RBF neural network. Firstly, the overall structure of the system is introduced with dynamic quantitative weighing as the main research object. Aiming at the fine feeding process, the mathematical model of the controlled object is established, and the weighing controller is designed by combining the traditional PID control and RBF neural network. Through RBF neural network, PID controller parameters can be adjusted online to improve the weighing precision. Finally, the actual weighing test is carried out. The test results show that the deviation of static and dynamic weighing can be controlled within 0.5%. The measurement range is 25-50 kg and production capacity can reach 900 packages/hour. The control system has higher weighing accuracy, better reliability and stability, and can meet the packaging requirements.