中文摘要：

基于强化学习的方法，提出一种动态电源管理超时策略自适应在线优化算法．构建基于超时策略动态电源管理系统的半Markov控制过程模型，将动态电源管理问题转化为一个带约束的优化问题．利用此模型的动态结构特性，结合在线梯度估计与髓机逼近推导超时策略的在线优化算法．该算法自适应性强，计算量小，具有全局收敛性．通过无线网络通信节点动态电源管理的应用仿真验证了算法的有效性．

英文摘要：

Based on reinforcement learning, an adaptive online optimization algorithm of timeout policy is proposed for dynamic power management. A semi-Markov control processes based analytic framework is introduced for timeout policy driven power-managed systems. Then an adaptive optimization algorithm that combines gradient estimation online and stochastic approximation is derived. This algorithm doesn＇t depend on the prior knowledge of system parameters, and can achieve global optimum with less computational cost. As an illustrative example, the dynamic power management for wireless communication devices is formulated, and simulation results show the effectiveness of the proposed algorithm.