论文部分内容阅读
本文提出了用人工神经网络求解具有约束条件的非线性优化问题的具体方法，分析了神经网络能量函数的构成形式，并在常规的Ｈｏｐｆｉｅｌｄ网络模型的基础上构造了一个非全局连接的神经网络动力学模型。这种修改的Ｈｏｐｆｉｅｌｄ网络克服了常规的Ｈｏｐｆｉｅｌｄ网络在求解非线性优化问题时权值不好映射的困难，具有结构清晰．易于软件模拟和硬件实现的优点。
In this paper, a method to solve the nonlinear optimization problem with constraints is proposed by artificial neural network. The formation form of energy function of neural network is analyzed. Based on the conventional Hopfield network model, a non-global connected neural network power is constructed Learn model The modified Hopfield network overcomes the difficulty of the traditional Hopfield network in solving the non-linear optimization problems when the weights are not well mapped, and has a clear structure. Easy software simulation and hardware implementation of the advantages.