Energy Non-intrusive Load Disaggregation Monitoring:A Search Of Practical Methods

Smart Meters can help users get an insight into energy consumption and production as well.These smart devices can send a warning to the users if the consumption is high.As the numbers of the smart meters are low,the total consumption cannot be measured.However,the user cannot change their electricity usage habits based on the measurements.The solution to this problem is providing the user with the energy consumption data for all the appliances.This also needs a separate power meter for all the appliances,which means the cost of the installation will also increase.There is another method as well to solve this issue by disaggregating the complete power signal,which is given to a home using smart meters.This research work addresses the topic of disaggregation of the energy which is consumed.This method of the consumption of energy is known as Non-Intrusive Load Monitoring(NILM).Before the availability of the smart meters,there is no algorithm to date for accurate disaggregate energy consumption.The exact disaggregation of energy is very helpful for providing insight into the appliance-wise consumption of energy.The accurate monitoring of the load can easily detect the aging appliances using a Smartphone application and suggest the user replace the appliance by the new model and save energy.There are other possible applications of the NILM that can detect the malfunctioning of the appliances or the appliances which need repair like the freezers sometimes need to the de-iced to save energy.As these sustainable power sources are available,the rate of the supply of the energy will start fluctuating&the price of the electric energy will also fluctuate.When people have insights about different appliances,they can turn them off when the energy consumption is high to ensure a low price.In short,we can say that NILM is very beneficial to regulate the market of supply and demand.Methods are characterized by the fact that they use data that is collected by smart meters since smart meters are more and more the standard in Western households.Appliances are modeled by Hidden Markov Models and households are modeled by super-state Hidden Markov Models or Factorial Hidden Markov Models.These models are trained by iterative K-means or expectation maximization,where iterative K-means turns out as the superior training method.Three disaggregation algorithms are researched:the Viterbi algorithm,the particle filter,and the newly proposed Global Transition Minimization(GTM).The disaggregation algorithms are tested on a synthetic dataset of 6 appliances and part of the Dutch Residential Energy Dataset(DRED).The results on the synthetic dataset vary between the accuracy of 81%for the GTM and 95%for the Viterbi algorithm.On the DRED the highest accuracy that is achieved is 77%for the particle filter.It turned out that the accuracy of the particle filter is not always improved by increasing the number of particles when real-world data is considered.Improvements for the disaggregation algorithms by using time information data and by using reactive power as extra input data are proposed.However,both suggestions do not lead to more accurate disaggregation results.A new modeling framework,where state transitions are also included in the model,is suggested as a future research topic.