探讨了用神经网络模型来预测组合梁中栓钉连接件抗剪强度的可行性。应用神经网络的ＢＰ模型，选择混凝土的抗压强度和弹性模量、栓钉的极限抗拉强度、栓钉的直径和长度作为关键的影响因素，提出了基于神经网络理论的栓钉连接件抗剪强度的预测方法。分析结果表明，神经网络方法比其他的经验计算方法更接近试验结果。 The feasibility of using neural network model to predict the shear strength of the stud connector in the composite beam is discussed. Using the BP model of neural network, the compressive strength and elastic modulus of concrete, the ultimate tensile strength of pegs, the diameter and length of pegs are selected as the key influencing factors, and the resistance of the peg joints based on neural network theory is proposed. Shear strength prediction method. The analysis results show that the neural network method is closer to the experimental results than other empirical calculation methods.