Non-Intrusive Load Monitoring Algorithm for PV Identification in the Residential Sector

Abstract

In response to the increasing penetration of distributed energy resources in the distribution network and the technical challenges this transition represents, this paper presents a novel approach for photovoltaic (PV) systems identification in the residential sector. Non-intrusive Load Monitoring (NILM) techniques have been focused mostly in identifying conventional loads on the customer side, thus more emphasis on distributed generation being integrated into the electrical grid is required to ensure system flexibility and most importantly stability of the electrical system. The proposed methodology combines basic statistics with the conventional machine learning Support Vector Machine, to identify PV load signatures from aggregated measurements in the residential sector using OpenPMU measurements. The main contributions of this paper are based on improving processing times of the conventional machine learning supervised algorithm and also providing important information for network operators based on simple techniques using electric current records from OpenPMU measurements.

Effective identification of distributed energy resources using smart meter net‐demand data

“…The time series created as windows of n samples represents a single characteristic (aggregated net load) in the time domain of the aggregated household measurements. Thus, statistical variables are extracted from each window ( W i ) to reduce noise in the net load signal, optimise computational times, and provide machine learning models with several time‐domain features [22, 34]. Namely, minimum ( W i‐ min ), maximum ( W i‐ max ), the maximum minus the minimum of W i ( W i‐ max ‐ W i‐ min ), mean (), variance (σ 2 ), standard deviation (σ) and kurtosis ( K u ) are obtained from each window.…”

“…Namely, minimum (W i-min ), maximum (W i-max ), the maximum minus the minimum of W i (W i-max -W i-min ), mean (W i ), variance (σ 2 ), standard deviation (σ) and kurtosis (K u ) are obtained from each window. The mathematical expressions to derive these features are defined as follows [22,34].…”

“…As a result, an average root‐mean‐squared error of 1.101 was achieved for 20 households. Regarding solar generation systems, the authors previously proposed a classification approach to identify the contribution of a 3.5‐kW peak (kWp) PV system [22]. SVMs and a principal component analysis were implemented as a supervised NILM method to separate PV generation from conventional loads, obtaining an F 1 score of 96%.…”

“…This study builds on preliminary research in [21] and [22]. In [21], a regression model based on RF and kNNs was proposed.…”

Load modelling and non-intrusive load monitoring to integrate distributed energy resources in low and medium voltage networks

“…Commercial Energy management systems and off the shelf solutions are commonly used, including PowerScout 24 [50], eGauge energy monitoring systems [51], EnviR energy aggregator [52], and oscilloscopes [53]. Alternatively, some researchers have developed their own experimental data acquisition systems using devices such as a Raspberry Pi 3 and a Arduino Mega 2560 [54] or the modular open-source phasor measurement unit called OpenPMU [55].…”