论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph)

这篇具有很好参考价值的文章主要介绍了论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph)。希望对大家有所帮助。如果存在错误或未考虑完全的地方,请大家不吝赐教,您也可以点击"举报违法"按钮提交疑问。

Publisher: Energies

Publising Date: 2020

MOTIVATION OF READING: 1. zero-crossing method for data preprocessing. 2.  recurrence graph (RG).


1. Overview

Probelm statement: the performance of V–I-based approaches is still unsatisfactory as it is still not distinctive enough to recognize devices that fall into the same category.

Methodology: an appliance recognition method utilizing the recurrence graph (RG) technique and convolutional neural networks (CNNs).

2. Introduction

A V–I-trajectory is obtained by plotting one-cycle steady-state voltage and current.

we propose new feature representation for appliance classification which relies on the recurrence graph (RG), also known as the recurrence plot. The RG analyses signal dynamics in phase–space to reveal the repeating and non-linear patterns and has been used extensively for feature representation in time-series classification problems.

3. Methodology

The proposed approach consists of the following main building blocks; Feature extraction and
pre-processing, WRG generation and the CNN classifier.

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

3.1 Feature Extraction and Pre-Processing

we measure Ns = 20 cycles of v and i before and after state-transitions of appliance has been detected as shown in Figure.

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

Then piece-wise aggregate approximation (PAA) is applied to reduce the dimensional of the signal from Ts to a predefined size w with minimal information loss. It works by dividing the data into n segments of equal size, then the approximation is a vector of the median values of the data readings.(lower the resolution)

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

3.2 Weighted Recurrence Plot (WRG)

The RG feature representation uses a distance similarity matrix Dwxw to represent and visualize
structural patterns in the signal.

Consider Ts points of activation signal x = {x1, x2 . . . xTs}. The distance similarity between xk and xj uses Euclidean norm.

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

For a classification problem, the compressed distance similarity matrix that represents all
recurrences in the form of a binary matrix RGwxw = [rk,j] is usually used. The rk,j function is defined as follows:

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

However, binarizing the distance matrix Dwxw through thresholding can lead to information loss and therefore degrade classification performance. Thus in this work, we propose the generation of WRGwxw that goes beyond the traditional binary output.

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

3.3 Classifier and Training Procedure

The CNN network applied in this work consists of three-stages 2D CNN layers each with 16, 32 and 64 feature maps, 3 x 3 filter, 2 x 1 stride and padding of 1. Each CNN layer is followed by batch normalization (BN) block and Leaky relu activation functions. The final layer consists of one flatten layer and two Fully connected layers (FC) layers.

4. Experiment

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读

论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph),论文阅读文章来源地址https://www.toymoban.com/news/detail-815746.html

到了这里,关于论文阅读 Improved Appliance Classification in NILM Using Weighted RCNN (recurrence graph)的文章就介绍完了。如果您还想了解更多内容,请在右上角搜索TOY模板网以前的文章或继续浏览下面的相关文章,希望大家以后多多支持TOY模板网!

本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处: 如若内容造成侵权/违法违规/事实不符,请点击违法举报进行投诉反馈,一经查实,立即删除!

领支付宝红包 赞助服务器费用

相关文章

  • 《论文阅读》Unified Named Entity Recognition as Word-Word Relation Classification

    将NER视作是word-word间的 Relation Classification。 这个word-word 间的工作就很像是TPlinker那个工作,那篇工作是使用token间的 link。 推荐指数:★★★☆☆ 值得学习的点: (1)用关系抽取的方法做NER抽取 (2)用空洞卷积解决词间交互问题 (3)CLN(conditional LayerNorma)的使用 统一三

    2023年04月14日
    浏览(43)
  • 论文阅读:Whole slide images classification model based on self-learning sampling

    这是一篇发表在BSPC(Biomedical Signal Processing and Control)上的关于WSI分类的文章, 作者是上海科技大学的学生/老师。 论文链接为:https://www.sciencedirect.com/science/article/pii/S1746809423012594 代码:暂未开源 深度学习与计算病理学的结合的增加放大了整个WSI在现代临床诊断中的应用。然而

    2024年02月02日
    浏览(40)
  • 【论文阅读】One For All: Toward Training One Graph Model for All Classification Tasks

    会议: 2024-ICLR-UNDER_REVIEW 评分:6,6,6,10 作者:Anonymous authors 文章链接:ONE FOR ALL: TOWARDS TRAINING ONE GRAPHMODEL FOR ALL CLASSIFICATION TASKS 代码链接:ONE FOR ALL: TOWARDS TRAINING ONE GRAPHMODEL FOR ALL CLASSIFICATION TASKS  设计一个能够解决多个任务的模型是人工智能长期发展的一个目标。最近,

    2024年01月18日
    浏览(50)
  • 【论文阅读24】Better Few-Shot Text Classification with Pre-trained Language Model

    论文标题:Label prompt for multi-label text classification(基于预训练模型对少样本进行文本分类) 发表时间:2021 领域:多标签文本分类 发表期刊:ICANN(顶级会议) 相关代码:无 数据集:无 最近,预先训练过的语言模型在许多基准测试上都取得了非凡的性能。通过从一个大型的

    2024年02月14日
    浏览(43)
  • 【深度学习】语义分割:论文阅读(NeurIPS 2021)MaskFormer: per-pixel classification is not all you need

    论文:Per-Pixel Classification is Not All You Need for Semantic Segmentation / MaskFormer 代码:代码 官方-代码 笔记: 作者笔记说明 【论文笔记】MaskFormer: Per-Pixel Classification is Not All You Need for Semantic Segmentation 总结思路清晰-简洁 【MaskFormer】Per-Pixel Classification is Not All You Needfor Semantic Segmenta

    2024年02月04日
    浏览(83)
  • 论文阅读:AdaBins: Depth Estimation using Adaptive Bins

    信息的全局处理会帮助提高整体深度估计。 提出的AdaBins预测的bin中心集中在较小的深度值附近,对于深度值范围更广的图像,分布广泛。 Fu et al. 发现将深度回归任务转化为分类任务可以提升效果,将深度范围分成固定数量的bins。本文则解决了原始方法的多个限制: 计算根

    2024年04月17日
    浏览(45)
  • 论文阅读 | RePaint: Inpainting using Denoising Diffusion Probabilistic Models

    Lugmayr A, Danelljan M, Romero A, et al. Repaint: Inpainting using denoising diffusion probabilistic models[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022: 11461-11471. 左图展示了masked image逐步去噪的过程;右图展示了基于扩散模型的Inpainting方法生成图片的多样性。 现有方法的问

    2024年01月21日
    浏览(39)
  • LEARNING TO EXPLORE USING ACTIVE NEURAL SLAM 论文阅读

    题目 :LEARNING TO EXPLORE USING ACTIVE NEURAL SLAM 作者 :Devendra Singh Chaplot, Dhiraj Gandhi 项目地址 :https://devendrachaplot.github.io/projects/Neural-SLAM 代码地址 :https://github.com/devendrachaplot/Neural-SLAM 来源 :LCLR 时间 :2022 这项工作提出了一种模块化和分层的方法来学习探索 3D 环境的策略,称为

    2024年02月14日
    浏览(41)
  • 【论文阅读笔记】Detecting AI Trojans Using Meta Neural Analysis

    个人阅读笔记,如有错误欢迎指出! 会议:2021 SP        Detecting AI Trojans Using Meta Neural Analysis | IEEE Conference Publication | IEEE Xplore 问题:         当前防御方法存在一些难以实现的假设,或者要求直接访问训练模型,难以在实践中应用。 创新:         通过元分类器

    2024年01月23日
    浏览(44)
  • 论文阅读 - Detecting Social Bot on the Fly using Contrastive Learning

    目录  摘要:  引言 3 问题定义 4 CBD 4.1 框架概述 4.2 Model Learning 4.2.1 通过 GCL 进行模型预训练  4.2.2 通过一致性损失进行模型微调  4.3 在线检测 5 实验 5.1 实验设置 5.2 性能比较 5.5 少量检测研究  6 结论 https://dl.acm.org/doi/pdf/10.1145/3583780.3615468           社交机器人检测正

    2024年02月06日
    浏览(46)

觉得文章有用就打赏一下文章作者

支付宝扫一扫打赏

博客赞助

微信扫一扫打赏

请作者喝杯咖啡吧~博客赞助

支付宝扫一扫领取红包,优惠每天领

二维码1

领取红包

二维码2

领红包