# The power disaggregation algorithms and their applications to demand dispatch

## TL;DR

This paper develops an online particle filter-based algorithm for power disaggregation to estimate individual device consumption from total household power data, enhancing demand dispatch control.

## Contribution

It introduces a scalable online disaggregation algorithm based on particle filters, improving inference efficiency for demand dispatch applications.

## Key findings

- Effective disaggregation of household power data
- Application to real-world Pecan Street dataset
- Enhanced demand dispatch control techniques

## Abstract

We were interested in solving a power disaggregation problem which comes down to estimating the power consumption of each device given the total power consumption of the whole house. We started by looking at the Factorial Hierarchical Dirichlet Process - Hidden Semi-Markov Model. However, the inference method had a complexity which scales withthe number of observations. Thus, we developed an online algorithm based on particle filters. We applied the method to data from Pecan Street https://dataport.cloud/ using Python. We applied the disaggregation algorithm to the control techniques used in Demand Dispatch.