8000 GitHub - mathewjhan/soc_prediction: Electric vehicle state of charge prediction using recurrent neural networks, submitted to UToronto's 2020 ProjectX competition representing UMass Amherst
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Electric vehicle state of charge prediction using recurrent neural networks, submitted to UToronto's 2020 ProjectX competition representing UMass Amherst

Notifications You must be signed in to change notification settings

mathewjhan/soc_prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Battery State-of-Charge Estimation using RNNs

Code to train various models for state-of-charge estimation.

Data

This code is trained on the open source Panasonic 18650PF Li-ion Battery Data.

Code Setup

This code has the following dependencies:

  • PyTorch
  • Matplotlib
  • Seaborn
  • pandas
  • mat4py
  • scikit-learn

Our RNN script builds on code from https://github.com/yunjey/pytorch-tutorial/blob/master/tutorials/02-intermediate/recurrent_neural_network.

Preconfiguration

Create a data/ folder and extract the dataset mentioned in the Data section into it.

Train an MLP on the SOC Data

python mlp.py --window_size 10 --hidden_size 4 --noise_std 0.005

Train an LSTM on the SOC Data

python recurrent.py --model lstm --sequence_length 100 --hidden_size 4 --num_layers 4 --noise_std 0.005

About

Electric vehicle state of charge prediction using recurrent neural networks, submitted to UToronto's 2020 ProjectX competition representing UMass Amherst

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

0