Search NASA⌕ Search

SEARCH · Search NASA

Results for “Non Intrusive Load Monitoring”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records

Active Learning Framework for Non-Intrusive Load Monitoring: Preprint

Non-Intrusive Load Monitoring (NILM) is a set of techniques that estimate the electricity usage of individual appliances from power measurements taken at a limited number of locations in a building. One of the key challenges in NILM is having too much data without class labels yet being unable to label the data manually for cost or time constraints. This paper presents an active learning framework that helps existing NILM techniques to overcome this challenge. Active learning is an advanced machine learning method that interactively queries a user for the class label information. Unlike most existing NILM systems that heuristically request user inputs, the proposed method only needs minimally sufficient information from a user to build a compact and yet highly representative load signature library. Initial results indicate the proposed method can reduce the user inputs by up to 90% while still achieving similar disaggregation performance compared to a heuristic method. Thus, the proposed method can substantially reduce the burden on the user, improve the performance of a NILM system with limited user inputs, and overcome the key market barriers to the wide adoption of NILM technologies.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Multivariate exploration of non-intrusive load monitoring via spatiotemporal pattern network

Non-intrusive load monitoring (NILM) of electrical demand for the purpose of identifying load components has thus far mostly been studied using univariate data, e.g., using only whole building electricity consumption time series to identify a certain type of end-use such as lighting load. However, using additional variables in the form of multivariate time series data may provide more information in terms of extracting distinguishable features in the context of energy disaggregation. In this work, a novel probabilistic graphical modeling approach, namely the spatiotemporal pattern network (STPN) is proposed for energy disaggregation using multivariate time-series data. The STPN framework is shown to be capable of handling diverse types of multivariate time-series to improve the energy disaggregation performance. The technique outperforms the state of the art factorial hidden Markov models (FHMM) and combinatorial optimization (CO) techniques in multiple real-life test cases. Furthermore, based on two homes' aggregate electric consumption data, a similarity metric is defined for the energy disaggregation of one home using a trained model based on the other home (i.e., out-of-sample case). The proposed similarity metric allows us to enhance scalability via learning supervised models for a few homes and deploying such models to many other similar but unmodeled homes with significantly high disaggregation accuracy.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

A Siamese CNN + KNN-Based Classification Framework for Non-intrusive Load Monitoring

Through the development of smart grids, programs such as demand side response, have been presented as auxiliary services to the real-time operation of distributed networks. In order to provide consumers information on their energy consumption, so that a modulation in consumption is possible, non-intrusive load monitoring has been introduced as an solution to this pattern recognition problem. Non-intrusive load monitoring enables the modeling of electrical loads connected to the low-voltage system, considering only a single measurement point. Presented state-of-the-art solutions though, consider availability of data as well as representation of all possible classes of the environment. This is of course a most conservative hypothesis, since in real-life applications availability of such data is much difficult, as well as the dynamic behavior of models is implicitly evolving in time. Here, a framework that uses neural Siamese networks with k-nearest neighbor clustering is presented toward non-intrusive load monitoring. Online learning feature is implemented, which relaxes the hypothesis of data requirements as well addresses the evolving nature of load profile. k-nearest clustering allows nonlinear characteristic space modelling. Test results using synthetics and real-life data show that the solution, besides obtaining a good generalizability in the classification, also obtained results with an accuracy of 95.77%.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Non-Intrusive Load Monitoring of HVAC Components using Signal Unmixing

Heating, Ventilating and Air Conditioning units (HVAC) are a major electrical energy consumer in buildings. Monitoring of the operation and energy consumption of HVAC would increase the awareness of building owners and maintenance service providers of the condition and quality of performance of these units, enabling conditioned-based maintenance which would help achieving higher energy efficiency. In this paper, a novel non-intrusive load monitoring method based on group constrained non-negative matrix factorization is proposed for monitoring the different components of HVAC unit by only measuring the whole building aggregated power signal. At the first level of this hierarchical approach, power consumption of the building is decomposed to energy consumption of the HVAC unit and all the other electrical devices operating in the building such as lighting and plug loads. Then, the estimated power signal of the HVAC is used for estimating the power consumption profile of the HVAC major electrical loads such as compressors, condenser fans and indoor blower. Experiments conducted on real data collected from a building testbed maintained at the Oak Ridge National Laboratory (ORNL) demonstrate high accuracy on the disaggregation task.

Flexible Research Platform↗

Final Report On Non-Intrusive Load Monitoring Of Welding Processes

The conventional method of weld process monitoring is to monitor the process output electrical power close to the point of welding. Non-Intrusive Load Monitoring (NILM) describes the method of monitoring primary or utility electrical power into a welding process well away from the point of welding. The objective of this project was to determine if there is sufficient information within the welding process input power to understand process output power characteristics, and, if so, determine if there is sufficient resolution within the input electrical characteristics to infer some measures of weld quality. To understand the relationship between process input and output power, both input and output power were monitored for two welding processes (GMAW and GTAW) and two different power source types (inverter and SCR). Welds were made with and without intentional disturbances. The results showed that there is a strong correlation between input and output power and that the influence of process disturbances are evident within the input power. A simple method using input power only was devised and successfully demonstrated to discriminate between a weld made with no intentional disturbance (a nominal weld) from a weld made with an intentional disturbance (off-nominal weld). The primary conclusions of this work is that NILM of process input power is sensitive to process disturbances that could influence weld quality and that the approach warrants additional study.

36 MATERIALS SCIENCE↗

Draft Plan to Develop Non-Intrusive Load Monitoring Test Protocols

This document presents a Draft Plan proposed to develop a common test protocol that can be used to evaluate the performance requirements of Non-Intrusive Load Monitoring. Development on the test protocol will be focused on providing a consistent method that can be used to quantify and compare the performance characteristics of NILM products. Elements of the protocols include specifications for appliances to be used, metrics, instrumentation, and a procedure to simulate appliance behavior during tests. In addition, three priority use cases for NILM will be identified and their performance requirements will specified.

47 OTHER INSTRUMENTATION↗

Non-Intrusive Load Monitoring of EBW Processes A Second Study [Slides]

The Objective is to Determine if induced off-nominal welding disturbances: (1) Beam arc out (weld on high vapor pressure aluminum plate); (2) Pulsed beam voltage; (3) Pulsed beam current; and (4) Pulsed beam focus current can be observed in the input power to the high-voltage cabinet powering the EB gun, ie Non-Instrusive Load Monitoring (NILM).

36 MATERIALS SCIENCE↗

Lockdown impacts on residential electricity demand in India: A data-driven and non-intrusive load monitoring study using Gaussian mixture models

This study evaluates the effect of complete nationwide lockdown in 2020 on residential electricity demand across 13 Indian cities and the role of digitalisation using a public smart meter dataset. We undertake a data-driven approach to explore the energy impacts of work-from-home norms across five dwelling typologies. Our methodology includes climate correction, dimensionality reduction and machine learning-based clustering using Gaussian Mixture Models of daily load curves. Results show that during the lockdown, maximum daily peak demand increased by 150-200% as compared to 2018 and 2019 levels for one room-units (RM1), one bedroom-units (BR1) and two bedroom-units (BR2) which are typical for low- and middle-income families. While the upper-middle- and higher-income dwelling units (i.e., three (3BR) and more-than-three bedroom-units (M3BR)) saw night-time demand rise by almost 44% in 2020, as compared to 2018 and 2019 levels. Our results also showed that new peak demand emerged for the lockdown period for RM1, BR1 and BR2 dwelling typologies. We found that the lack of supporting socioeconomic and climatic data can restrict a comprehensive analysis of demand shocks using similar public datasets, which informed policy implications for India's digitalisation. We further emphasised improving the data quality and reliability for effective data-centric policymaking.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Non-Intrusive Load Monitoring Assessment: Literature Review and Laboratory Protocol

Unlike sub-metering, which requires that individual loads are equipped with their own meters, nonintrusive load monitors (NILM) use algorithms to discern the individual loads from a single metering point. Although a number of these technologies are now commercially available to utilities and homeowners, with several more to become available in the coming years, it is not yet understood how accurately these technologies can disaggregate a home’s electrical load. Pacific Northwest National Laboratory (PNNL) and its partners at the Northwest Energy Efficiency Alliance (NEEA) have identified and acquired six of these technologies and will be testing them in the PNNL Lab Homes as well as the owner-occupied Residential Test Bed operated by NEEA. As a result of this testing, PNNL and NEEA will evaluate the selected NILMs against a repeatable test protocol so that their accuracy can be compared across NILM technologies. In the field deployment in the NEEA Test Bed Homes, PNNL and NEEA will evaluate the real-world accuracy of the NILMs responding to dynamic loads. To evaluate the accuracy of NILM technologies regardless of their intended application, a standardized test protocol is required that applies representative loads and establishes consistent metrics against which the NILM can be evaluated and compared. To establish this standardized test protocol, a literature review was conducted to identify any test protocols or standardized testing approaches currently in use. The literature review indicated that no consistent conventions were currently in place for measuring the accuracy of these technologies. Consequently, PNNL developed a testing protocol and metrics to provide the basis for quantifying and analyzing the accuracy of commercially available NILM technologies. This report discusses the results of the literature review and the proposed test protocol and metrics in more detail.

NILM↗

Characteristics and Performance of Existing Load Disaggregation Technologies

Non-intrusive load monitoring (NILM) or non-intrusive appliance load monitoring (NIALM) is an analytic approach to disaggregate building loads based on a single metering point. This advanced load monitoring and disaggregation technique has the potential to provide an alternative solution to high-priced traditional sub-metering and enable innovative approaches for energy conservation, energy efficiency, and demand response. However, since the inception of the concept in the 1980’s, evaluations of these technologies have focused on reporting performance accuracy without investigating sources of inaccuracies or fully understanding and articulating the meaning of the metrics used to quantify performance. As a result, the market for, as well as, advances in these technologies have been slowly maturing. To improve the market for these NILM technologies, there has to be confidence that the deployment will lead to benefits. In reality, every end-user and application that this technology may enable does not require the highest levels of performance accuracy to produce benefits. Also, there are other important characteristics that need to be considered, which may affect the appeal of NILM products to certain market targets (i.e. residential and commercial building consumers) and the suitability for particular applications. These characteristics include the following: 1) ease of use, the level of expertise/bandwidth required to properly use the product; 2) ease of installation, the level of expertise required to install along with hardware needs that impact product cost; and 3) ability to inform decisions and actions, whether the energy outputs received by end-users (e.g. third party applications, residential users, building operators, etc.) empower decisions and actions to be taken at time frames required for certain applications. Therefore, stakeholders, researchers, and other interested parties should be kept abreast of the evolving capabilities, uses, and characteristics of NILM that make them attractive for certain building environments and different classes of end-users. The intent of this report is to raise awareness of trending NILM approaches. Additionally, three existing technologies were acquired and evaluated using the Residential Building Stock Assessment (RBSA) owner-occupied test bed operated by the Northwest Energy Efficiency Alliance (NEEA) to understand performance accuracy of current NILM products under realistic conditions. Based on this field study experience, the characteristics exhibited by the NILM products included in the assessment are also discussed in this report in terms of ease of use, ease of installation, ability to inform decisions and actions. Results of the analysis performed to investigate the accuracy of the participating NILM products in estimating energy use of individual appliances are also presented.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Improving Robustness of Spectrogram Classifiers with Neural Stochastic Differential Equations

Signal analysis and classification is fraught with high levels of noise and perturbation. Computer-vision-based deep learning models applied to spectrograms have proven useful in the field of signal classification and detection; however, these methods aren't designed to handle the low signal-to-noise ratios inherent within non-vision signal processing tasks. While they are powerful, they are currently not the method of choice in the inherently noisy and dynamic critical infrastructure domain, such as smart-grid sensing, anomaly detection, and non-intrusive load monitoring. Currently, these models can be brittle, which makes them susceptible to noisy input. This also means they have sub-optimal stability of explanation outputs. Experts and technicians using these models to make decisions in real world scenarios need assurance that a model is performing as it is supposed to. The classification or prediction outputs it generates should be sound and grounded, not likely to change in the presence of shifting noise landscapes. In this work, we explore the idea of Neural Stochastic Differential Equations (NSDE's) to improve the robustness of models trained to classify time series data and the effect of NSDE's on the explainability of outputs. We then test the effectiveness of these approaches by applying them to a non-intrusive load monitoring (NILM) dataset that consists of simulated harmonic signals injected into a real building.

Brogan, Joel↗

Classification of Unintended Radiated Emissions in a Multi-Device Environment

Here, unintended radiated emissions (URE) from electronic devices are conducted on to the power infrastructure and can be collected and analyzed for non-intrusive load monitoring (NILM) applications. Dimensionally aligned signal projections (DASP) were previously introduced as specialized signal transforms to enable device classification using machine learning on statistical features derived from the DASP transforms. In this work, we explore multi-device classification which requires more sophisticated methods owing to the complexity of the feature space such as large dynamic range, non-contiguous features, over-lapping features, and nonlinear interactions of features. In particular, we introduce an additional DASP algorithm for increased resolution and sensitivity to intermodulation products and examine the utility of convolutional neural networks (CNNs) to extract and learn features directly from DASP images for classification of single and multiple device URE captures.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Harmonic Signals Dataset

The Harmonic Signals Dataset (HSD) consists of one-dimensional time series data collected from current sensors with a goal of providing data for research and development of Non-Intrusive Load Monitoring (NILM) with high sample rate (800kHz) sensors. NILM seeks to detect and characterize electrical equipment operating within a facility by sensing changes in electrical current on the building power system associated with the equipment. To provide data with a known ground truth, but with the realism of a signal introduced within a building facility, a series of known ground truth signals were injected as voltage into the power system of a building and measurements of the signals at multiple locations across the building power system were made with electrical current sensors.

24 POWER TRANSMISSION AND DISTRIBUTION↗