With the ongoing integration of Renewable Energy Sources (RES), the complexity of power grids is increasing. Due to the fluctuating nature of RES, ensuring the reliability of power grids can be challenging. One possible approach for addressing these challenges is Demand Response (DR) which is described as matching the demand for electrical energy according to the changes and the availability of supply. However, implementing a DR system to monitor and control a broad set of electrical appliances in real-time introduces several new complications, including ensuring the reliability and financial feasibility of the system. In this work, we address these issues by designing and implementing a distributed real-time DR infrastructure for laptops, which estimates and controls the power consumption of a network of connected laptops in response to the fast, irregular changes of RES. Furthermore, since our approach is entirely software-based, we dramatically reduce the initial costs of the demand side participants. The result of our field experiments confirms that our system successfully schedules and executes rapid and effective DR events. However, the accuracy of the estimated power consumption of all participating laptops is relatively low, directly caused by our software-based approach.

翻译：随着可再生能源的不断整合,电网的复杂性正在增加。由于可再生能源的波动性质,确保电网的可靠性可能具有挑战性。应对这些挑战的一个可能办法是需求反应(DR),它被描述为根据变化和供应情况对电力的需求进行匹配。然而,实施一个DR系统,实时监测和控制一系列广泛的电器,带来了若干新的复杂因素,包括确保系统的可靠性和财务可行性。在这项工作中,我们通过设计和实施一个分布式的膝上型电脑实时DR基础设施来解决这些问题。该基础设施估算和控制一个连接的膝上型电脑网络的电耗,以应对RES的快速、不规则的变化。此外,由于我们的方法完全以软件为基础,我们大幅降低了需求方参与者的初始成本。我们的实地实验结果证实,我们的系统成功地安排和执行了快速、有效的DR事件。然而,所有参与的膝上型电脑的估计电量的准确性较低,这是我们基于软件的方法直接造成的。