Online learning for demand response
Online learning for demand response
复制标题
需求响应在线学习
DOI：
10.1109/allerton.2015.7447007
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
R. Rajagopal
中科院分区：
文献类型：
--
作者：
D. Kalathil;R. Rajagopal
需求响应是现有和未来网格系统的关键组成部分，面临着更多的可变性和高峰需求。扩大需求响应需要有效地预测大量消费者的个人响应，同时选择正确的响应来发出信号。本文提出了一种新的在线学习问题，该问题捕捉到了消费者多样性、消息传递疲劳和响应预测。我们使用多武装匪徒模型的框架来解决这一问题。这产生了简单且易于实现的基于索引的学习算法，并且具有可证明的性能保证。
Demand response is a key component of existing and future grid systems facing increased variability and peak demands. Scaling demand response requires efficiently predicting individual responses for large numbers of consumers while selecting the right ones to signal. This paper proposes a new online learning problem that captures consumer diversity, messaging fatigue and response prediction. We use the framework of multi-armed bandits model to address this problem. This yields simple and easy to implement index based learning algorithms with provable performance guarantees.