.png){kind=icon}
返回
Data reduction in a cloud-based AMI framework with service-replication
.png){kind=icon}
PRE
AI
DOI：10.1016/j.compeleceng.2018.02.042.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Smart grids use cloud-based Advanced Metering Infrastructure (AMI) systems to ensure high performance in collecting, storing, and processing energy consumption data originating at the smart meters. However, this data is usually huge, so its transmission to the cloud requires stable, high-bandwidth Internet connections, which may be costly or even unavailable. A solution allowing transmission on normal connections is to reduce the data by approximating it at the cloud using forecasting. The problem is that AMI data changes pattern continuously and unexpectedly, due to the constant, uncoordinated addition and removal of loads. This makes predicting the data using a single forecasting method almost impossible. We circumvent this problem by designing an adaptive, multi-method, service-replicated Framework. At any time, the Framework uses a method that is optimal for the current data. Once the data changes pattern, the Framework automatically switches to another method that suits the new pattern. Experimental results show reduction rates up to 55% on real AMI data. (C) 2018 Elsevier Ltd. All rights reserved.
Keyword:
Smart grid
Data reduction
Forecasting
Service replication
Advanced metering infrastructure
Cloud computing
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
C
IF:
4.9
论文数:
6.7K
被引数:
1.3W
机构
引用论文
分享收藏TEMPORAL TRENDS AND PREDICTORS OF VENOUS THROMBOEMBOLISM IN HOSPITALIZED CHRONIC LYMPHOCYTIC LEUKEMIA PATIENTS住院慢性淋巴细胞白血病患者静脉血栓栓塞症的时序趋势和预测因子
PREAI
分享收藏
Rule induction for forecasting method selection: Meta-learning the characteristics of univariate time series.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏 没有更多内容
没有更多内容