Non-Intrusive Load Monitoring Using a CNN-LSTM-RF Model Considering Label Correlation and Class-Imbalance

Non-Intrusive Load Monitoring (NILM) is particularly important for demand response. This paper proposes an innovative method based on a deep learning model to recognize the working state of electrical appliances using low frequency load data. The approach includes a data processing step, a deep learning model and a new accuracy calculation method. The data processing step consists of a multi-feature and high-dimensional method (MFHDM) and a pre-training process. The deep learning model consists of a convolutional neural network (CNN), a long-term short-term memory network (LSTM) and a random-forest (RF) algorithm. The proposed method addresses the label correlation problem and the class-imbalance problem. To test the proposed method, the Reference Energy Disaggregation Dataset (REDD) and the Pecan Street dataset (PSD) are used. A comparative analysis with several models shows that the proposed method can effectively improve electrical appliance recognition accuracy and realize NILM.

[1]  Shirantha Welikala,et al.  Incorporating Appliance Usage Patterns for Non-Intrusive Load Monitoring and Load Forecasting , 2019, IEEE Transactions on Smart Grid.

[2]  Shufang Li,et al.  Semisupervised Multilabel Deep Learning Based Nonintrusive Load Monitoring in Smart Grids , 2020, IEEE Transactions on Industrial Informatics.

[3]  Yongheng Pang,et al.  An Event-Driven Convolutional Neural Architecture for Non-Intrusive Load Monitoring of Residential Appliance , 2020, IEEE Transactions on Consumer Electronics.

[4]  Hsueh-Hsien Chang,et al.  Particle-Swarm-Optimization-Based Nonintrusive Demand Monitoring and Load Identification in Smart Meters , 2013 .

[5]  Jing Liao,et al.  Non-intrusive appliance load monitoring using low-resolution smart meter data , 2014, 2014 IEEE International Conference on Smart Grid Communications (SmartGridComm).

[6]  Jack Kelly,et al.  Neural NILM: Deep Neural Networks Applied to Energy Disaggregation , 2015, BuildSys@SenSys.

[7]  Eduardo Gomes,et al.  PB-NILM: Pinball Guided Deep Non-Intrusive Load Monitoring , 2020, IEEE Access.

[8]  R. M. de Azevedo,et al.  Event Classification in Non-Intrusive Load Monitoring Using Convolutional Neural Network , 2019, 2019 IEEE PES Innovative Smart Grid Technologies Conference - Latin America (ISGT Latin America).

[9]  G. W. Hart,et al.  Nonintrusive appliance load monitoring , 1992, Proc. IEEE.

[10]  Bin Yang,et al.  A new approach for supervised power disaggregation by using a deep recurrent LSTM network , 2015, 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[11]  Ashutosh Saxena,et al.  Exploring Correlation between Labels to improve Multi-Label Classification , 2015, ArXiv.

[12]  Shouxiang Wang,et al.  A Load Identification Method Based on Active Deep Learning and Discrete Wavelet Transform , 2020, IEEE Access.

[13]  David J. Hill,et al.  A Hierarchical Hidden Markov Model Framework for Home Appliance Modeling , 2018, IEEE Transactions on Smart Grid.

[14]  Scott Dick,et al.  Toward Non-Intrusive Load Monitoring via Multi-Label Classification , 2017, IEEE Transactions on Smart Grid.

[15]  Yu-Hsiu Lin,et al.  Development of an Improved Time–Frequency Analysis-Based Nonintrusive Load Monitor for Load Demand Identification , 2014, IEEE Transactions on Instrumentation and Measurement.

[16]  Edoardo Fiorucci,et al.  A New Convolutional Neural Network-Based System for NILM Applications , 2021, IEEE Transactions on Instrumentation and Measurement.

[17]  Wei Wu,et al.  SGM: Sequence Generation Model for Multi-label Classification , 2018, COLING.

[18]  Chen Huang,et al.  Deep Imbalanced Learning for Face Recognition and Attribute Prediction , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Tao Wang,et al.  A non-intrusive load identification method based on convolution neural network , 2017, 2017 IEEE Conference on Energy Internet and Energy System Integration (EI2).

[20]  Chengwei Huang,et al.  Sequence-to-Sequence Load Disaggregation Using Multiscale Residual Neural Network , 2020, IEEE Transactions on Instrumentation and Measurement.

[21]  Jing Liao,et al.  Low-complexity energy disaggregation using appliance load modelling , 2016 .

[22]  Germano Lambert-Torres,et al.  Non-Intrusive Identification of Loads by Random Forest and Fireworks Optimization , 2020, IEEE Access.

[23]  Shikha Singh,et al.  Non-Intrusive Load Monitoring via Multi-Label Sparse Representation-Based Classification , 2019, IEEE Transactions on Smart Grid.

[24]  Jyoti Maggu,et al.  Simultaneous Detection of Multiple Appliances From Smart-Meter Measurements via Multi-Label Consistent Deep Dictionary Learning and Deep Transform Learning , 2019, IEEE Transactions on Smart Grid.

[25]  Yoash Levron,et al.  Modified Cross-Entropy Method for Classification of Events in NILM Systems , 2019, IEEE Transactions on Smart Grid.

[26]  Dirk Benyoucef,et al.  Frequency Invariant Transformation of Periodic Signals (FIT-PS) for Classification in NILM , 2019, IEEE Transactions on Smart Grid.

[27]  J. Zico Kolter,et al.  REDD : A Public Data Set for Energy Disaggregation Research , 2011 .

[28]  Janaka Ekanayake,et al.  Residential Appliance Identification Based on Spectral Information of Low Frequency Smart Meter Measurements , 2016, IEEE Transactions on Smart Grid.

[29]  Hyoeun Kang,et al.  Household Appliance Classification Using Lower Odd-Numbered Harmonics and the Bagging Decision Tree , 2020, IEEE Access.

[30]  Tommi S. Jaakkola,et al.  Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation , 2012, AISTATS.

[31]  Marcos J. Rider,et al.  Nonintrusive Load Monitoring Algorithm Using Mixed-Integer Linear Programming , 2018, IEEE Transactions on Consumer Electronics.

[32]  Scott Dick,et al.  Residential Household Non-Intrusive Load Monitoring via Graph-Based Multi-Label Semi-Supervised Learning , 2019, IEEE Transactions on Smart Grid.

[33]  Hsueh-Hsien Chang,et al.  A New Measurement Method for Power Signatures of Nonintrusive Demand Monitoring and Load Identification , 2011, IEEE Transactions on Industry Applications.

[34]  Howon Kim,et al.  Nonintrusive Load Monitoring Based on Advanced Deep Learning and Novel Signature , 2017, Comput. Intell. Neurosci..

[35]  Fernando Deluno Garcia,et al.  Embedded NILM as Home Energy Management System: A Heterogeneous Computing Approach , 2019, IEEE Latin America Transactions.

[36]  Nikolaos Doulamis,et al.  Bayesian-optimized Bidirectional LSTM Regression Model for Non-intrusive Load Monitoring , 2019, ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).