给出了延迟离散Hopfield -型神经网络的收敛性定理。在广义异步运行方式下 ,证明了对称连接权阵 (只要w0 对称 )条件下的收敛性定理 ,推广了已有的延迟离散Hopfield -型神经网络的收敛性结果 ,表明网络收敛滞后于能量函数收敛最多 2n+1步。最后给出了能量函数的极大值点与延迟离散Hopfield -型神经网络的稳定态的关系。 The convergence theorem of delayed discrete Hopfield - type neural networks is given. Under the generalized asynchronous operation mode, the convergence theorem of symmetric connection weight matrix (as long as w0 symmetry) is proved, and the convergence of the existing delayed discrete Hopfield - type neural network is generalized. It shows that the network convergence lags behind the convergence of energy function Up to 2n + 1 steps. Finally, the relation between the maximum point of energy function and the steady state of discrete Hopfield - type neural networks with delay is given.