This file has been truncated. show original
Energy Prediction System with a neural network (CNN-LSTM) in a Jetson Nano.
In this project we are going to build an active power meter with an Arduino Uno. The data will be sent to the Jetson-Nano with the Python script "arduino_serial.py" to establish the communication between the Jetson and the Arduino.
The second script, "neural_training.py" is to start the training for the hybrid neural network and visualize the data.
# The neural network
CNN-LSTM neural network, which combines convolutional neural network (CNN) and long short-term memory (LSTM), can extract complex features of energy consumption. The CNN layer can extract the features between several variables affecting energy consumption, and the LSTM layer is appropriate for modeling temporal information of irregular trends in time series components.
The CNN-LSTM method for predicting electric energy consumption consists of a series of connection of CNN and LSTM. CNN-LSTM can extract complex features among multiple sensor variables collected for electric energy demand forecasting and can store complex irregular trends. First, the upper layer of CNN-LSTM consists of a series of convolutional layers. The CNN layer can receive various variables that affect electric energy consumption such as voltage, intensity, and sub
metering. So, CNN consists of an input layer that accepts sensor variables as inputs, an output layer that extracts features to LSTMs, and several hidden layers.
* 1 - Jetson-Nano Developer Kit from Nvidia
* 1 - HDMI cable
* 1 - 7 inch LCD screen