Abstract

In this paper an energy disaggregation problem, commonly studied as “non-intrusive load monitoring”, is presented. Non-intrusive load monitoring is a procedure for estimating the energy profiles of individual electric home appliances from the aggregated power measurement. We present this problem as an optimization-based problem in which the least-square errors are minimized to find the set of active appliances, using the instantaneous aggregated power. The convex regularization terms are also added, which exploit the information of the probability of individual appliance usage and assume the behavior of the appliances’ power profiles to be constant as piecewise signal over time. The problem is formulated using a scenario-based stochastic optimization approach to consider the underlying uncertainties in power measurements. The expected value of the problem’s objective function is computed by using the sample average approximation method, where normally distributed samples of a random variable are introduced into the problem using the Monte Carlo simulation. Moreover, the appliances’ limits on the power modes are formulated as a chance constraint in the optimization problem. The training and testing of the proposed algorithm are done by using the benchmark data: AMPds and REFIT. The simulation results show that the proposed scenario-based stochastic optimization approach successfully estimates the energy profiles of individual appliances with multiple power modes.

Highlights

  • Information of load demand at the appliance level can significantly enhance the estimation of user energy consumption

  • In Intrusive Appliance Load Monitoring (IALM), the operating condition of the appliances are predicted by attaching a sensor to each appliance, in contrast to Non-Intrusive Appliance Load Monitoring (NIALM), where only a single sensor is attached at the aggregate level

  • The simulated results of the proposed scenario-based stochastic optimization approach is compared with a benchmark approach, Factorial Hidden Markov Model (FHMM), using Non-Intrusive Load Monitoring Toolkit (NILMTK) [23]

Read more

Summary

INTRODUCTION

Information of load demand at the appliance level can significantly enhance the estimation of user energy consumption. The convex regularization terms are added to this problem, which exploit the information of the probability of the appliances’ usage and assume the behavior of the appliances to be piecewise constant These features help to improve the estimates further. The proposed algorithm is tested and trained on benchmark data of AMPds (the Almanac of Minutely Power dataset) [21] and REFIT (Personalized Retrofit Decision Support Tools For UK Homes Using Smart Home Technology) [22], which are readily available online and are used in the literature for load disaggregation purposes. The simulated results of the proposed scenario-based stochastic optimization approach is compared with a benchmark approach, Factorial Hidden Markov Model (FHMM), using Non-Intrusive Load Monitoring Toolkit (NILMTK) [23]. The results show that the proposed approach successfully estimates the energy profiles of the individual appliances with multiple power modes, and performs better than the FHMM approach.

PROBLEM DESCRIPTION
REGULARIZATION TO ACHIEVE SINGLE OPERATED
REGULARIZATION TO ACHIEVE PIECEWISE CONSTANT
CALCULATION OF THE EXPECTED VALUE OF THE OBJECTIVE FUNCTION
CHANCE CONSTRAINTS RELATED TO THE LIMITS ON APPLIANCES’ POWER MODES
PARAMETERS SELECTION
GAIN SELECTION
REDUCTION OF THE PROBLEM’S COMPUTATIONAL COMPLEXITY
APPLICATION TO REAL TIME DATA
Findings
CONCLUSION
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call