Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance

Srikrishnan, Vivek; Young, George S.; Witmer, Lucas; Brownson, Jeffrey R.S.

doi:10.1016/j.solener.2015.06.004

Cited by 15 publications

(7 citation statements)

References 21 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Artificial neural networks (ANNs) are efficient tools that predict patterns too complex for humans to find and taught machines to recognize [40]. ANN-based methods have been successfully developed in the solar energy research to predict, optimize design, and evaluate different solar radiation parameters [41][42][43][44][45][46]. Alnagi et al [47] proposed a hybrid method based on ANN incorporated with particle swarm optimization for predicting the energy performance of a building integrated photovoltaic/thermal system.…”

Section: Literature Reviewmentioning

confidence: 99%

Design parameter modelling of solar power tower system using adaptive neuro-fuzzy inference system optimized with a combination of genetic algorithm and teaching learning-based optimization algorithm

Khosravi

Malekan

Pabón

et al. 2020

Journal of Cleaner Production

View full text Add to dashboard Cite

Determining the optimal sizing of a solar power tower system (SPTS) with a thermal energy storage system is subject to finding the optimum values of design parameters including the solar multiple (SM), design direct normal irradiance (DNI) and thermal storage hours. These design parameters are determined for each station separately and have remarkable effects on the thermo-economic performance of the system. This paper aims to demonstrate how artificial intelligence (AI) techniques may play an important role in addressing the above-mentioned need and help determine the optimum design parameters for different stations. For this purpose, we developed a thermo-economic model of a 100 MW SPTS with a molten salt storage system for five stations (two stations in India, and one each in Bangladesh, Pakistan, and Afghanistan). A method-based AI is utilized in this paper to ascertain the design parameters of the system. Additionally, a novel hybrid method based on adaptive neuro-fuzzy inference system optimized with a combination of genetic algorithm and teaching-learning-based optimization algorithm (ANFIS-GATLBO) is employed. The input parameters are latitude, longitude, design point DNI and SM, while the annual energy produced, levelized cost of energy and capacity factor are the target variables. The results of the study show that although the annual energy produced by SPTS rises by increasing the SM and decreasing design point DNI, optimum design parameters should be determined by the economic factors. In addition, it was found that the ANFIS-GATLBO method used in this study successfully predicted the targets with a correlation coefficient close to 1.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Design parameter modelling of solar power tower system using adaptive neuro-fuzzy inference system optimized with a combination of genetic algorithm and teaching learning-based optimization algorithm

Khosravi

Malekan

Pabón

et al. 2020

Journal of Cleaner Production

View full text Add to dashboard Cite

show abstract

“…Boyd stated that all the obtained data had a value higher than 0.98 R 2 by testing the uncertainties and sensitivities of 40 different pyranometers in 6 different models. Srikrishnan et al [14] worked on an alternative approach that can detect direct normal solar radiation (DNI) of a multiple pyranometer by means of the ANN method and examined the results on various NN topologies. Lester and Myers [15] stated that pyranometers produced an electrical signal depending on global radiation and that the ratio of the signal to irradiance was the responsivity (RS) of the device.…”

Section: Introductionmentioning

confidence: 99%

“…Experiments in their study, in which equations related to the solar density, output voltage and ambient temperature were derived by employing mathematical and thermodynamically models, were conducted in Semman, Iran. After reviewing the existing studies on solar radiation measurement systems, a new device that would help the true measurement of solar radiation on horizontal surfaces and enable the user to read solar radiation in W/m 2 , and that is easy to use and highly precise was designed and tested in a study conducted by Badran et al [14]. The device can read the solar radiation instantaneously and also show the average radiation for a certain period of time.…”

Section: Introductionmentioning

confidence: 99%

Solarmeter Design for High Solar Radiation Measurement and Experimental Validation

Ergün

Ceylan

Aydın

et al. 2019

El-Cezeri Fen Ve Mühendislik Dergisi

View full text Add to dashboard Cite

In this study, a solar meter was designed to measure high solar radiation. The solar meter uses the characteristic properties of a solar cell in order to measure solar radiation. In this way, measurements can be carried out without any limitations. The solar radiation equation was entered into the ARDUINO microcontroller in the solar meter, and the results were compared through another commercial solar meter. These results were entered into STATGRAPHIC software and three different functions with high regression values were obtained. These three functions were entered into the Arduino microcontroller and the experiments were repeated. According to the comparisons made using commercial device, R 2 values were determined as 0.944, 0.936, 0.938, and 0.986 for the first equation and for three different functions, respectively. Based on the values obtained, it has been determined that the solar meter can make highly accurate measurements and after the development of appropriate functions, it can be employed especially in high solar radiation concentrated systems.

show abstract

“…As Redes Neurais Artificiais (RNA) são um método de Aprendizado de Máquina que generaliza uma regressão logística ao alimentar variáveis em um algoritmo multicamadas dotadas de unidades de processamento que compõe um modelo de neurônio artificial (SRIKRISHNAN et al, 2015). As RNA são potentes ferramentas estatísticas compostas por unidades de processamento simples que computam funções matemáticas a partir de uma série de dados fornecidas a esta (GUARNIERI, 2006).…”

Section: Redes Neurais Artificiaisunclassified

“…Figura 2.10 -Esquema genérico de uma Rede Neural Artificial que apresenta uma camada oculta. Fonte: AdaptadoSrikrishnan (2015) …”

unclassified

Rede neural recorrente para previsão de curto prazo da usina fotovoltaica de 540 kWp da USP

Fonseca¹

View full text Add to dashboard Cite

Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance

Cited by 15 publications

References 21 publications

Design parameter modelling of solar power tower system using adaptive neuro-fuzzy inference system optimized with a combination of genetic algorithm and teaching learning-based optimization algorithm

Design parameter modelling of solar power tower system using adaptive neuro-fuzzy inference system optimized with a combination of genetic algorithm and teaching learning-based optimization algorithm

Solarmeter Design for High Solar Radiation Measurement and Experimental Validation

Rede neural recorrente para previsão de curto prazo da usina fotovoltaica de 540 kWp da USP

Contact Info

Product

Resources

About