跟踪凸优化理论最新进展，基于交替方向乘子法，提出一种动态经济调度问题的分散式优化方法，以解决集中式优化调度在智能电网下面临的诸多困难。借鉴IEC 61970标准中CIM建模思路，将电网模型解耦为设备、端子和连通点模型，使传统的网络模型转换为一个分散式的电气模型。在算法上，通过每次迭代中邻近设备间少量的消息传递，实现各设备子优化问题的并行求解。所提方法无需全局协调和同步迭代，呈现即插即用和自愈的特性；所建子优化模型的目标函数和约束均为凸，保证了算法的收敛性。几个系统的计算表明，所提方法具有极高的计算效率和良好的收敛特性，是适应未来电网发展的新思路。“,”Following the advances in convex optimization, this paper proposed a decentralized method based on the alternating direction method of multipliers (ADMM) for the dynamic economic dispatch (DED) problem to overcome the difficulties arising from the traditional centralized dispatch method using in smart grid. Inspired by the modeling idea of CIM in IEC 61970 standard, which decomposes the power grid into devices, terminal and connectivity node, a decentralized model of power grid was obtained. Each device exchanges simple messages with its neighbors in the network, and solves the individual sub optimization problem by itself in parallel pattern. No global coordination and synchronizing iterations are needed in the proposed method, which possess the self-healing and plug-in features. Moreover, the convergence of the proposed method is guaranteed since the device objective function and constraints are convex. Numerical simulations on some systems confirm the extremely efficiency and good convergence of the algorithm. It offers a new idea for development of future power grid.