Joint Estimation of Behind-the-Meter Solar Generation in a Community

Kabir, Farzana; Yu, Nanpeng; Yao, Weixin; Yang, Rui; Zhang, Yingchen

doi:10.1109/tste.2020.3016896

Cited by 56 publications

(28 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We use this method as a baseline. The authors extend this work in [18] to jointly disaggregate the net load data of a group of customers into aggregated solar generation and aggregated home load.…”

Section: Related Work 21 Solar Disaggregationmentioning

confidence: 99%

A Data-Efficient Approach to Behind-the-Meter Solar Generation Disaggregation

Chen

Haji

Ardakanian

2021

Preprint

View full text Add to dashboard Cite

With the emergence of cost effective battery storage and the decline in the solar photovoltaic (PV) levelized cost of energy (LCOE), the number of behind-the-meter solar PV systems is expected to increase steadily. The ability to estimate solar generation from these latent systems is crucial for a range of applications, including distribution system planning and operation, demand response, and non-intrusive load monitoring (NILM). This paper investigates the problem of disaggregating solar generation from smart meter data when historical disaggregated data from the target home is unavailable, and deployment characteristics of the PV system are unknown. The proposed approach entails inferring the physical characteristics from smart meter data and disaggregating solar generation using an iterative algorithm. This algorithm takes advantage of solar generation data (aka proxy measurements) from a few sites that are located in the same area as the target home, and solar generation data synthesized using a physical PV model. We evaluate our methods with 4 different proxy settings on around 160 homes in the United States and Australia, and show that the solar disaggregation accuracy is improved by 32.31% and 15.66% over two state-of-the-art methods using only one real proxy along with three synthetic proxies. Furthermore, we demonstrate that using the disaggregated home load rather than the net load data could improve the overall accuracy of three popular NILM methods by at least 22%. CCS CONCEPTS• Computer systems organization → Sensors and actuators; • Computing methodologies → Modeling and simulation.

show abstract

Section: Related Work 21 Solar Disaggregationmentioning

confidence: 99%

A Data-Efficient Approach to Behind-the-Meter Solar Generation Disaggregation

Chen

Haji

Ardakanian

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…These unsupervised methods have achieved promising performance, and some of them can even compete with the stateof-the-art supervised methods. The problem of unsupervised methods is that they have several hyperparameters that are crucial to their estimation performance [18][19][20][21][22][23][24]. In previous studies, researchers apply cross-validation to find the optimal hyperparameters, which involves the use of labelled data.…”

Section: Introductionmentioning

confidence: 99%

Permutation test and bootstrap methods for unsupervised detection and estimation of behind‐the‐meter photovoltaic generation

Liu

Shi

Chen

et al. 2021

IET Renewable Power Gen

View full text Add to dashboard Cite

The penetration of unmonitored behind-the-meter photovoltaic systems has increased rapidly over the past decade. While unsupervised methods have been proposed to estimate behind-the-meter photovoltaic generation, their performance is significantly affected by hyperparameters. Existing methods for choosing hyperparameters require labelled data, which are unavailable to system operators. In this paper, a data-driven method is proposed to generate labelled data to optimize the hyperparameters for unsupervised estimation of behind-the-meter photovoltaic generation. First, a permutation-test based method is developed to detect photovoltaic installation in an unsupervised way. Second, consumers without PV are combined with limited monitored photovoltaic sites to simulate the original system through the bootstrap method. The simulated system provides labelled data for hyperparameter optimization. Finally, the near-optimal hyperparameters are searched on the simulated system and applied to the original system's estimation. Through experiments on a smart meter dataset, the effectiveness of the proposed methodology is verified. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

show abstract

“…Os métodos baseados em modelos procuram estimar a energia produzida por painéis solares a partir da estimativa do seu desempenho através de modelos detalhados, entre os quais pode-se citar o PVWatts, desenvolvido pelo Laboratório Nacional de Energias Renováveis (National Renewable Energy). Segundo vários autores estes modelos não são completamente confiáveis e acurados, devido a incerteza sobre os dados dos equipamentos efetivamente instalados e também devido à inerente degradação da performance destes equipamentos (Kabir et al, 2021). Por outro lado, métodos baseados em dados não utilizam parâmetros de equipamentos para estimar a potência e energia gerada localmente, mas se baseiam em medições realizadas online (equipamentos de medição inteligente), controle supervisório e aquisição de dados (SCADA), dados meteorológicos e irradiação solar.…”

Section: Introdu ç ãOunclassified

“…Sociedade Brasileira de Automática (SBA) XV Simpósio Brasileiro de Automação Inteligente -SBAI 2021, 17 a 20 de outubro de 2021 O ponto comum a todos os métodos é que apenas dados líquidos de consumo/geração estão disponíveis para a as empresas que operam a rede.De acordo comKabir et al (2021), há quatro métodos considerados estado da arte para a desagregação de dados de consumo e geração obtidos a partir de medições locais de consumidores que possuem geração fotovoltaica. Estes métodos são descritos emCheung et al (2018),Sossan et al (2018),Chen and Irwin (2017) eKabir et al (2019) e são aplicados com diversas finalidades, desde aplicação a estudos de fluxo de potência, controle volt-var, estimação e previsão de carga até o estabelecimento de modelos para fins de planejamento e otimização da operação.Nesse contexto, o presente trabalho apresenta um método simplificado de desagregação de energia a partir de registros líquidos de consumo e geração de prossumidores.…”

unclassified

Método Simplificado de Desagregação de Energia Aplicado a Redes contendo Prossumidores

Scheid

Haffner

Ferraz

et al. 2021

Procedings Do XV Simpósio Brasileiro De Automação Inteligente

View full text Add to dashboard Cite

The present work proposes a new, simplified method for estimating the total consumption and generation of prosumers with monthly net metering. Initially, the characteristics of electric energy measurement systems are presented, as well as the norms that regulate distributed generation in Brazil. Next, a new method is presented and detailed for energy disaggregation of monthly net metering. The method allows the estimation of a daily load and generation profiles for consumers and prosumers, using a reference generation based on climate data, a seasonality factor that describes the variation of consumption throughout the year, and an efficiency factor that describes the characteristics of each consumer unit with distributed generation. Finally, practical results obtained with the proposed method applied to prosumers located in South Brazil are presented and discussed. These results reveal not only a close agreement between the measured and estimated monthly consumed and injected energy but also that the method is enough accurate for practical purposes. Resumo: O presente trabalho propõe um novo método simplificado para estimar o consumo e a geração total de prossumidores com medição líquida mensal. Inicialmente, são apresentadas as características dos sistemas de medição de energia elétrica, assim como as normas que regulamentam a geração distribuída no Brasil. Em seguida é apresentado e detalhado o novo método para desagregação energética de medições mensais líquidas, o qual permite estimar perfis de carga e geração diários de consumidores e prossumidores utilizando uma geração de referência baseada em dados climáticos, um fator de sazonalidade que descreve a variação do consumo ao longo do ano e um fator de eficiência que descreve as características de cada unidade consumidora com geração distribuída. Por fim, são apresentados e discutidos os resultados obtidos para prossumidores localizados no sul do Brasil. Estes resultados mostram não apenas uma boa concordância entre os valores registrados e estimados para a energia consumida e injetada para cada um dos meses do ano, mas também mostram que o método é suficientemente acurado para fins práticos.

show abstract

Joint Estimation of Behind-the-Meter Solar Generation in a Community

Cited by 56 publications

References 30 publications

A Data-Efficient Approach to Behind-the-Meter Solar Generation Disaggregation

A Data-Efficient Approach to Behind-the-Meter Solar Generation Disaggregation

Permutation test and bootstrap methods for unsupervised detection and estimation of behind‐the‐meter photovoltaic generation

Método Simplificado de Desagregação de Energia Aplicado a Redes contendo Prossumidores

Contact Info

Product

Resources

About