Non-Intrusive Load Monitoring of Household Devices Using a Hybrid Deep Learning Model through Convex Hull-Based Data Selection

Abstract

The availability of smart meters and IoT technology has opened new opportunities, ranging from monitoring electrical energy to extracting various types of information related to household occupancy, and with the frequency of usage of different appliances. Non-intrusive load monitoring (NILM) allows users to disaggregate the usage of each device in the house using the total aggregated power signals collected from a smart meter that is typically installed in the household. It enables the monitoring of domestic appliance use without the need to install individual sensors for each device, thus minimizing electrical system complexities and associated costs. This paper proposes an NILM framework based on low frequency power data using a convex hull data selection approach and hybrid deep learning architecture. It employs a sliding window of aggregated active and reactive powers sampled at 1 Hz. A randomized approximation convex hull data selection approach performs the selection of the most informative vertices of the real convex hull. The hybrid deep learning architecture is composed of two models: a classification model based on a convolutional neural network trained with a regression model based on a bidirectional long-term memory neural network. The results obtained on the test dataset demonstrate the effectiveness of the proposed approach, achieving F1 values ranging from 0.95 to 0.99 for the four devices considered and estimation accuracy values between 0.88 and 0.98. These results compare favorably with the performance of existing approaches.

From home energy management systems to energy communities: methods and data

“…The approach described above is applied to PV power generation, with great success. Moving from deterministic forecasting to probability forecasting, for both load demand and PV power generation 58 Test and validate different non-invasive load monitoring (NILM) algorithms, as performed in 59 , 60 . The first reference employs ApproxHull 61 , a data selection tool existing in our lab to deep learning models.…”

“…Test and validate different non-invasive load monitoring (NILM) algorithms, as performed in 59 , 60 . The first reference employs ApproxHull 61 , a data selection tool existing in our lab to deep learning models.…”

Energy Disaggregation Using Multi-Objective Genetic Algorithm Designed Neural Networks

“…The device's ground truth active power sequence was determined manually using CW and SP data. To create the ON-OFF labels of devices, an approach similar to the one proposed in [43] was used. Briefly, it was assumed that a device was switched ON when its power consumption exceeded a specified threshold value for at least a certain period of time.…”

“…The authors tested their model using the UKDALE low frequency dataset and demonstrated good generalization properties. In [43], a hybrid deep learning architecture based on a convex hull data selection approach using low frequency power data for NILM was proposed. They selected the most informative vertices of the real convex hull using a random approximation algorithm, incorporating them in the training data.…”

Advanced data-analytics for power system security enhancement

“…For example, there might be multiple Non-intrusive load monitoring (NILM) is an emerging technology that aims to infer the energy consumption of individual appliances within a household or a building by analyzing the aggregate electricity consumption data [210]. Unlike traditional approaches of load monitoring, which require the installation of dedicated meters for each appliance, NILM uses machine learning (ML) algorithms and signal processing techniques to extract the energy signature of each device from the overall power consumption data, enabling real-time energy monitoring and optimization without disrupting the existing electrical infrastructure [211].…”