AI based Power Consumption Analysis for Electrical Appliances


AI based Power Consumption Analysis

Abstract: AI-based Power Consumption Analysis – Electricity consumption has been extensively studied in the computer architecture field for many years. While the acquisition of energy as a measure in machine learning is emerging, most of the experimentation is still primarily focused on obtaining hoisted levels of accuracy without any computational restrictions. We believe that one of the grounds for this shortage of interest is due to their absence of ease with access to evaluate energy consumption. The main objective of this study is come to evaluate useful regulations to the machine learning community that permits them the fundamental realization to use and build energy estimation methods for machine learning algorithms.

Problem Statement

A non-intrusive monitoring system estimates the behavior of individual electric appliances from the measurement of the total household load demand curve. The total load demand curve is measured at the entrance of the power line into the house. The power consumption of individual appliances can be estimated using several machine learning techniques by analyzing the characteristic frequency contents from the load curve of the household.

Proposed Work

Using the individual household power consumption datasets, we evaluated the proposed method using an individual household dataset that is publicly available from the UCI machine learning repository, which contains data on electric power consumption between 2006 ND 2010 The Dataset has 1048575 rows and 9 columns. We train each supervised regressor model on the train set using all features and then evaluate them on the entire test set. To measure performance over time. We use the scikit-learn implementation of logistic regression (LR) and random forest (RF) and LSTM. 


  • Separate measuring equipment for an individual appliance is not needed
  • It is in-expensive 
  • It is a non-intrusive monitoring system
  • Estimates the on/off states of appliances include inverter


AI-based Power Consumption Analysis


AI-based Power Consumption Analysis


       1 GB RAM

       80 GB Hard Disk

       Intel Processor


Software : 

     Windows OS

     Python GUI or Anaconda Navigator

System Requirement:

Operating System: Windows 7 Ultimate 32 bit / Windows XP




Data used in this paper is a set of meter data having several appliances. 

  • Dataset is a multivariate time series dataset that describes the electric consumption for a household 
  • Global_active_power, global_reactive power,voltage,global_intensity, ST, ET, etc are the attributes used
  • The dataset is split randomly into two parts: training set, test set.
  • Mean Absolute Errors (MAE), Root Mean Squared Errors (RMSE) is calculated for the datasets.
  • The algorithm which gives the lowest errors is considered to  be accurate

Customer Reviews

There are no reviews yet.

Be the first to review “AI based Power Consumption Analysis for Electrical Appliances”

This site uses Akismet to reduce spam. Learn how your comment data is processed.