求解计算机网络最小平均迟延和流量分配的本质是一个联合优化问题。进一步分析得出，它可近似为一个二次型优化问题。利用一种具有全局收敛性质的神经网络模型来解决这个问题，计算结果表明这种方法是可行且有效的。 Solving the problem of the minimum average delay and the distribution of traffic in a computer network is a joint optimization problem. After further analysis, it can be approximated as a quadratic optimization problem. This problem is solved by using a neural network model with global convergence property. The calculation results show that this method is feasible and effective.