China
Africa
Explore chapters and articles related to this topic
Energy efficiency and predictive maintenance applications using smart energy measuring devices
Published in Stein Haugen, Anne Barros, Coen van Gulijk, Trond Kongsvik, Jan Erik Vinnem, Safety and Reliability – Safe Societies in a Changing World, 2018
S. Kotsilitis, E.C. Marcoulaki, E. Kalligeros, Y. Mousmoulas
Monitoring of energy consumption at appliance level is essential for predicting energy needs and monitoring appliance operation in a household, a building or an industrial system. Energy disaggregation refers to using data analytics and signal processing, to identify specific patterns and to break down electricity consumption to individual appliances. This is usually done in a non-intrusive manner by monitoring the utility connection meter, and has been a field of significant research work for over twenty years. Non-Intrusive Load Monitoring (NILM) is a process where the aggregated electricity consumption is metered at the Grid-consumer connection point, and by analyzing the changes in voltage and current wavelengths tries to identify which appliances are being used at a certain time. Still, NILM technology’s main goal is to provide insights into energy consumption at appliance level, mainly to support energy efficiency actions with economic and environmental impact. There are novel techniques using various approaches of NILM for a great number of applications, like safety on industrial environments, device health monitoring and predictive maintenance and demand response applications.
EEMS2015 organizing committee
Published in Yeping Wang, Jianhua Zhao, Advances in Energy, Environment and Materials Science, 2018
In recent years, automatic electricity load moni- toring and decomposition method based on meas- uring sensor technology has obvious advantages than manual investigation, which is widely con- cerned. There are two implementations: 1). Intru- sive residential Load Monitoring (ILM): Each appliance is equipped with sensors, to collect and send consumption information. This way has a high cost and is difficult to promote. 2). Non-Intrusive Load Monitoring and Decomposition (NILMD): By installing a set of current and voltage sensors at the power service entry in a residence and analyz- ing the measured electrical signals, the electricity energy consumption of individual appliances and working status can be identified. Using this way, the financial costs are less.
A scalable and practical method for disaggregating heating and cooling electrical usage using smart thermostat and smart metre data
Published in Journal of Building Performance Simulation, 2022
Sang woo Ham, Panagiota Karava, Ilias Bilionis, James Braun
Energy disaggregation, called nonintrusive load monitoring (NILM), provides appliance-level energy consumption from net energy consumption, and it can be used for eco-feedback design (Batra, Singh, and Whitehouse 2015; Carrie Armel et al. 2013; Gopinath et al. 2020; Kimura et al. 2018). The main advantage of this technique is that it does not require appliance- or circuit-level power sensors and expensive data collection infrastructure. Various NILMs (e.g. event-based/event-less or supervised/unsupervised) have been proposed (Gopinath et al. 2020; Kelly and Knottenbelt 2016; Pereira and Nunes 2020). While NILM research has shown success based on datasets obtained in laboratories, there are several challenges for large-scale field deployment. For example, a supervised method requires many steps such as data labelling, event detection, feature extraction, and disaggregation (Giri and Bergés 2015; Gopinath et al. 2020). An unsupervised method needs to be validated with a few weeks of operation in coordination with residents (Gopinath et al. 2020). In other words, both methods require labelled observations during the training stage. In addition, most NILMs are developed based on high-resolution power, voltage, current, and phase information (Carrie Armel et al. 2013; Gopinath et al. 2020).
Real-time remote energy consumption location for power management application
Published in Advances in Building Energy Research, 2021
Sam Moayedi, Hamed Nabizadeh Rafsanjani, Subhaditya Shom, Mahmoud Alahmad, Changbum R. Ahn
Non-Intrusive Load Monitoring (NILM) is the process of monitoring the total consumed energy using one sensor or meter typically at the main electrical panel. The authors in (Alahmad et al., 2011a, 2011b) investigated Time Domain Reflectometry (TDR) as a NILM approach to determine the physical location of each load in the non-energized electrical system and correlates this data with power consumption to provide an alternative and effective approach to ILM. Collected data should go through a disaggregation process to discover the energy profile of individual loads (Ridi et al., 2014). However, disaggregation is a challenging method and requires significant studies, which usually can be tackled using machine learning techniques (Devlin & Hayes, 2019; Zhang, Chen, Ng, Lai, & Lai, 2019). Moreover, NILM is an important topic in Demand-Side Management studies (Chen & Yi, 2015). Since NILM method relies on the load’s energy signature for disaggregation, it cannot provide individual load’s data when similar loads are running at the same time. To address this issue, authors in (Rafsanjani, Ahn, & Chen, 2018) investigated Non-Intrusive Occupant Load Monitoring (NIOLM). NIOLM aims to monitor occupant-specific energy-use in a particular location (e.g. a workstation) right after user’s entrance to a building and right before user’s departure from the building. This method correlates occupancy data with building-wide energy-consuming information to disaggregate energy information down to a user level (Rafsanjani, Ahn, & Chen, 2018).
Low frequency residential non-intrusive load monitoring based on a hybrid feature extraction tree-learning approach
Published in Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 2022
Christos Timplalexis, Georgios-Fotios Angelis, Stelios Krinidis, Dimosthenis Ioannidis, Dimitrios Tzovaras
Electricity grids are moving toward sustainability based on three fundamental principles: a) Decarbonization, as an increasing number of Renewable Energy Sources (RES) are entering the grid resulting into a “greener” power generation strategy, b) Decentralization, which is a concept that marks the transition from a centralized to a distributed power generation schema with multiple users being consumers and producers at the same time and c) Digitalization of the grid since smart devices and IoT equipment are used for the monitoring and control of the grid, generating huge amounts of data, that are available to be exploited by AI algorithms. Non-intrusive load monitoring (NILM) can leverage from the grid digitalization, offering a solution that improves energy efficiency, providing multiple benefits to customers and utilities. In terms of signal processing, NILM is a single source separation problem with every sample being a mixture of signals that are not mutually exclusive (Nalmpantis and Vrakas 2020). NILM can be described as the extraction of power consumption of individual appliances out of aggregated power data (Figure 1). The aggregated power consumption is measured from a single point and none of the appliances are equipped with metering devices. The method is characterized as non-intrusive because the complex installation of smart meters at every single home appliance is avoided, so the occupants maintain their privacy. Moreover, installing multiple smart meters is not viable economically in most cases. On the contrary, disaggregating energy from a single-point meter using AI methods, is a low-cost solution that can be deployed with minimum external interference into the user’s residence.