Automatic disaggregation of residential electrical consumption with non-intrusive methods
dc.contributor.advisor | Carrillo Caicedo, Gilberto | |
dc.contributor.advisor | Petit Suárez, Johann Farith | |
dc.contributor.advisor | Duarte Gualdrón, César Antonio | |
dc.contributor.author | Jiménez Manjarres, Yulieth | |
dc.date.accessioned | 2024-03-04T00:07:40Z | |
dc.date.available | 2018 | |
dc.date.available | 2024-03-04T00:07:40Z | |
dc.date.created | 2018 | |
dc.date.issued | 2018 | |
dc.description.abstract | La informacion detallada de los electrodomésticos individuales en el hogar, llamada desagre- ´ gacion de carga, puede motivar el ahorro energético y apoyar planes de gestión de demanda. Estaénformacion se puede estimar mediante sistemas de Monitorización No intrusiva de Carga (NILM, ´ por sus siglas en ingles), realizan procesamiento de se ´ nales y modelado matem ˜ atico a partir de ´ mediciones electricas en un solo punto. Bajo la premisa de que las se ´ nales de los electrodom ˜ esticos ´ tienen caracter´ısticas distintivas, denominadas firmas de carga, un enfoque es discriminar los electrodomesticos mediante técnicas de inteligencia artificial. Aunque la investigación en estaárea está´ en crecimiento, aun se detectan algunas brechas en la literatura cient ´ ´ıfica y esta tesis contribuye al conocimiento en varios aspectos. Primero, se presenta un marco para implementar sistemas NILM. Segundo, se propone un sistema basado en eventos que comprende las etapas de deteccion´ de eventos, extraccion más efectiva de caracter ´ ´ısticas transitorias basadas en el dominio del tiempo y de la transformada S, clasificacion a través de un enfoque no tradicional y estimación de poten- ´ cia mediante la dependencia de la tension. Tercero, se evalúa la capacidad de discriminación de ´ las firmas de carga para determinar el impacto del punto de los factores de impacto mencionados. Finalmente, se construyo una base de datos de medidas de aparatos residenciales bajo diferenteséscenarios de tension de alimentación, impedancia y operación de los aparatos. As ´ ´ı, estos sistemas NILM se vislumbran como aplicaciones de hogares inteligentes. | |
dc.description.abstractenglish | One path to enhance energy efficiency and design demand side management plans is providing detailed information about the individual appliances in houses, namely, load disaggregation. Nonintrusive Load Monitoring (NILM) Systems aim to obtain the disaggregated information from measurements in a single point through signal processing and mathematical modeling. One approach assumes that appliances could be represented by characteristics computed from the electrical signals, i.e. load signatures. Although research in this area is increasing, several gaps are detected in the scientific literature: there is not a widely accepted set of load signatures, the complexity of the traditional systems increases exponentially with the number of appliances, fully labeled datasets of electrical signals are lacking, previous work has not been focused on the development of integral algorithms, and the question about the impact of factors (voltage distortion, network impedance, etc.) on NILM algorithms remains open. This thesis contributes to knowledge in several ways. First, a framework for implementing NILM systems is presented. Second, an event-based NILM system is proposed, which comprises the following stages: event detection, feature extraction based on waveforms and the S transform, classification through a nontraditional approach and power estimation by considering the voltage dependency. Third, the discrimination capacity of the load signatures is assessed to determine the impact of point-on-wave of switching, voltage distortion and network impedance. Finally, a fully dataset of residential appliances is provided under several scenarios of voltage, impedance and operation. These NILM algorithms are envisioned as smart home applications for appliance management. | |
dc.description.degreelevel | Doctorado | |
dc.description.degreename | Doctor en Ingeniería | |
dc.format.mimetype | application/pdf | |
dc.identifier.instname | Universidad Industrial de Santander | |
dc.identifier.reponame | Universidad Industrial de Santander | |
dc.identifier.repourl | https://noesis.uis.edu.co | |
dc.identifier.uri | https://noesis.uis.edu.co/handle/20.500.14071/38872 | |
dc.language.iso | spa | |
dc.publisher | Universidad Industrial de Santander | |
dc.publisher.faculty | Facultad de Ingenierías Fisicomecánicas | |
dc.publisher.program | Doctorado en Ingeniería: Área Ingeniería Electrónica | |
dc.publisher.school | Escuela de Ingenierías Eléctrica, Electrónica y Telecomunicaciones | |
dc.rights | http://creativecommons.org/licenses/by/4.0/ | |
dc.rights.accessrights | info:eu-repo/semantics/openAccess | |
dc.rights.creativecommons | Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0) | |
dc.rights.license | Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) | |
dc.rights.uri | http://creativecommons.org/licenses/by-nc/4.0 | |
dc.subject | Consumo Electrico | |
dc.subject | Residencial | |
dc.subject | Desagregación De Carga | |
dc.subject | ´ No Intrusivo | |
dc.subject | Firma De Carga | |
dc.subject | Inteligencia Artificial | |
dc.subject | Identificacion De Elec- ´ Trodomesticos | |
dc.subject | Transformada Stockwell | |
dc.subject | Clasificación De Una Clase. | |
dc.subject.keyword | Electrical Power Consumption | |
dc.subject.keyword | Residencial | |
dc.subject.keyword | Load Disaggregation | |
dc.subject.keyword | Non-Intrussive | |
dc.subject.keyword | Load Signature | |
dc.subject.keyword | Artificial Inteligence | |
dc.subject.keyword | Appliance Identification | |
dc.subject.keyword | Stockwell Transform. | |
dc.title | Automatic disaggregation of residential electrical consumption with non-intrusive methods | |
dc.title.english | Automatic disaggregation of residential electrical consumption with non-intrusive methods | |
dc.type.coar | http://purl.org/coar/version/c_b1a7d7d4d402bcce | |
dc.type.hasversion | http://purl.org/coar/resource_type/c_db06 | |
dc.type.local | Tesis/Trabajo de grado - Monografía - Doctorado |
Files
Original bundle
1 - 3 of 3
No Thumbnail Available
- Name:
- Carta de autorización.pdf
- Size:
- 2.31 MB
- Format:
- Adobe Portable Document Format
No Thumbnail Available
- Name:
- Documento.pdf
- Size:
- 16.09 MB
- Format:
- Adobe Portable Document Format
No Thumbnail Available
- Name:
- Nota de proyecto.pdf
- Size:
- 2.43 MB
- Format:
- Adobe Portable Document Format