Neural NILM – Deep Neural Networks Applied ToEnergy Disaggregation
Neural NILM – Deep Neural Networks Applied ToEnergy Disaggregation
0 4
BuildSys_2015_NeuralNILM
Presentation of Neural NILM for BuildSys 2015 conference in November 2015On Github JackKelly / BuildSys_2015_NeuralNILM
Neural NILM
Deep Neural Networks Applied ToEnergy Disaggregation
Jack Kelly & William Knottenbelt Imperial College London (Swipe or press right-arrow on your keyboard to change slides)
Energy Disaggregation
Aim: Itemised Energy Bills
Outline
Why use deep neural nets (DNNs) for NILM? How DNNs work Three DNN architectures for NILM Data augmentation Results SummaryOutline
Why use deep neural nets (DNNs) for NILM? How DNNs work Three DNN architectures for NILM Data augmentation Results SummaryName the Appliance?
Face Recognition
Manual Feature Extraction
Deep Neural Nets
Automatic Feature Learning
ImageNet Large Scale Visual Recognition Challenge (ILSVRC)
From: Krizhevsky, Sutskever & Hinton. ImageNet Classification with Deep Convolutional Neural Networks. NIPS (2012)
Image from devblogs.nvidia.com
Krizhevsky et al.'s DNN Results on ImageNet 2012
Krizhevsky, Sutskever & Hinton. ImageNet Classification with Deep Convolutional Neural Networks. NIPS (2012)
Outline
Why use deep neural nets (DNNs) for NILM? How DNNs work Three DNN architectures for NILM Data augmentation Results SummaryThe Artificial Neuron
Image adapted from WikiMedia Commons image by Chrislb
Feed Forward Nets
Krizhevsky et al.'s Architecture for ImageNet 2012
Krizhevsky, Sutskever & Hinton. ImageNet Classification with Deep Convolutional Neural Networks. NIPS (2012)
Training
Autoencoders
Autoencoder Examples
Hinton & Salakhutdinov. Reducing the dimensionality of data with neural networks. Science (2006)
Denoising Autoencoders
Image from Marc'Aurelio Ranzato
Vincent et al. Extracting and composing robust features with denoising autoencoders. ICML (2008)
Recurrent Neural Nets
Recurrent Neural Nets
Long Short-Term Memory (LSTM) Cells
Image from blog.otoro.net
Hochreiter & Schmidhuber. Long short-term memory. Neural Computation (1997)
Recurrent Neural Nets
Playing Volleyball :)
Outline
Why use deep neural nets (DNNs) for NILM? How DNNs work Three DNN architectures for NILM Recurrent Neural Nets (LSTM) Denoising Autoencoder 'Bounding rectangle' around the target Data augmentation Results SummaryRecurrent Neural Nets
Denoising Autoencoders
Bounding Rectangle
Outline
Why use deep neural nets (DNNs) for NILM? How DNNs work Three DNN architectures for NILM Data augmentation Results SummaryDNNs need lots of data!
Data Augmentation for Images of Plakton
Raw
Augmented
Data Augmentation for NILM
- Extract individual appliance activations from real data
- For each generated example:
- Randomly pick which appliances to include
- Randomly pick individual activations
- Randomly align activations
Outline
Why use deep neural nets (DNNs) for NILM? How DNNs work Three DNN architectures for NILM Data augmentation Results SummaryExample Output
LSTM
Autoencoder
Rectangles
Metrics
Metrics on Seen Appliances
Metrics on Unseen Appliances