Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
home:technologies:topology [2025/01/07 06:48] – [Anomaly detection] rahulhome:technologies:topology [2025/01/07 08:05] (current) – [Anomaly detection] rahul
Line 11: Line 11:
 https://github.com/LLNL/gridds https://github.com/LLNL/gridds
  
 +<note>MIT License
 +
 +Copyright (c) 2022 Alexander Ladd
 +</note>
 Recent research on distributed energy resources (DER) has been focused on aggregating data and feedback control to enhance the reliability and performance of energy grid. The ability to forecast load factor at transformers and substations can greatly improve demand and supply side load management. Similarly, detecting incipient failures for key devices in the system can reduce the entire grid failure, which effectively enhances the reliability of the energy grid. Machine learning models, such as fault detection and time series forecasting models, present new and innovative approaches to solving these aforementioned challenges [7, 10, 45]. Recent research on distributed energy resources (DER) has been focused on aggregating data and feedback control to enhance the reliability and performance of energy grid. The ability to forecast load factor at transformers and substations can greatly improve demand and supply side load management. Similarly, detecting incipient failures for key devices in the system can reduce the entire grid failure, which effectively enhances the reliability of the energy grid. Machine learning models, such as fault detection and time series forecasting models, present new and innovative approaches to solving these aforementioned challenges [7, 10, 45].
  
Line 23: Line 27:
 {{  :home:technologies:p542-ladd-supp4.png?nolink&  |Anomaly detection using 3 channel \mu PMU over 11 hours of train/test split}} {{  :home:technologies:p542-ladd-supp4.png?nolink&  |Anomaly detection using 3 channel \mu PMU over 11 hours of train/test split}}
  
-''https://colab.research.google.com/github/LLNL/gridds/blob/master/docs/source/autoregression.ipynb''+https://colab.research.google.com/github/LLNL/gridds/blob/master/docs/source/autoregression.ipynb 
 + 
 +https://colab.research.google.com/github/LLNL/gridds/blob/master/docs/source/autoregression_visualization.ipynb 
 + 
 +Touched up requirements.txt: 
 + 
 +<csv hdr_rows=0 file=:home:technologies:requirements.csv maxlines=10>requirements.txt</csv> 
 + 
 +requirements.txt {{ :home:technologies:requirements.csv |}} 
 + 
 +''!git clone https://github.com/LLNL/gridds''
  
-{{ :home:technologies:requirements.csv |}}+''!pip install -r requirements.txt''
  
-''!git clone https://github.com/LLNL/gridds"+At least one dependency unmet: Our proposed method uses a data-driven deep learning method called Latent Ordinary Differential Equations (LatentODE) for data imputation, followed by a Bayesian non-parametric method called **distance-dependent Chinese Restaurant Franchise** (dd-CRF) for unsupervised discovery of latent states of the energy grid.
  
-"!pip install -r requirements.txt''+<note warning>/content/gridds/experimenter.py in <module> 
 +     17 from abc import ABCMeta, abstractmethod 
 +     18 import numpy as np 
 +---> 19 from ddcrf.run_ddcrf import run_ddcrf 
 +     20 import gridds.tools.utils as utils 
 +     21 from sklearn import preprocessing
  
 +ModuleNotFoundError: No module named 'ddcrf'</note>