• Journal of Biosystems Engineering
• /
• v.43 no.4
• /
• pp.352-361
• /
• 2018

Purpose: The objective of this study was to develop a multilayer perceptron (MLP) model to estimate solar radiation using a solar module. Methods: Data for the short-circuit current of a solar module and other environmental parameters were collected for a year. For MLP learning, 14,400 combinations of input variables, learning rates, activation functions, numbers of layers, and numbers of neurons were trained. The best MLP model employed the batch backpropagation algorithm with all input variables and two hidden layers. Results: The root-mean-squared error (RMSE) of each learning cycle and its average over three repetitions were calculated. The average RMSE of the best artificial neural network model was $48.13W{\cdot}m^{-2}$. This result was better than that obtained for the regression model, for which the RMSE was $66.67W{\cdot}m^{-2}$. Conclusions: It is possible to utilize a solar module as a power source and a sensor to measure solar radiation for an agricultural sensor node.

• Journal of Biosystems Engineering
• /
• v.37 no.1
• /
• pp.11-18
• /
• 2012

Purpose: Short-circuit current of a solar module that is widely used as a power source for wireless environmental sensors is proportional to solar radiation although there are a lot of factors affecting the short-circuit current. The objective of this study is to develop a model for estimating solar radiation for using the solar module as a power source and an irradiance sensor. Methods: An experiment system collected data on the short-circuit current and environmental factors (ambient temperature, cloud cover and solar radiation) during 65 days. Based on these data, two linear regression models and a non-linear regression model were developed and evaluated. Results: The best model was a linear regression model with short-circuit current, angle of incidence and cloud cover and its overall RMSE(Root Means Square Error) was 66.671 $W/m^2$. The other linear model (RMSE 69.038 $W/m^2$) was also acceptable when the cloud cover data is not available.

• Ocean and Polar Research
• /
• v.25 no.1
• /
• pp.9-20
• /
• 2003

A solar radiation model was used to investigate the UV radiation at the surface offing Sejong Station in West Antarctica. The results calculated by this model were compared with the values measured by UV-Biometer and UV-A meter during 1999-2000. In this study, the parameterization of solar radiative transfer process was based on Chou and Lee(1996). The total ozone amounts measured by Breve. Ozone Spectrophotometer and the aerosol amounts by Nakajima et al.(1996) was used as the input data of the solar radiative transfer model. And the surface albedo is assumed to be 0.20 in summer and 0.85 in winter. The sensitivity test of solar radiative transfer model was done with the variation of total ozone, aerosol amount, and surface albedo. When the cosine of solar zenith angle is 0.3, Erythemal UV-B radiation decreased 73% with the 200% increase of total ozone from 100 DU to 300 DU, but the decrease of UV-A radiation is about 1%. Also, for the same solar zenith angle, UV-A radiation was decreased 31.0% with the variation of aerosol optical thickness from 0.0 to 0.3 and Erythemal UV-B radiation was decreased only 6.1%. The increase of Erythemal W-B radiation with the variation of surface albedo was twice that of UV-A increase. The surface Erythemal UV-B and UV-A radiation calculated by solar raditive transfer model were compared with the measured values fer the relatively clear day at King Sejong Station in West Antarctica. The model calculated Erythemal UV-B radiation at the surface coincide well with the measured values except for cloudy days. But the difference between the model calculated UV-A radiation and the measured value at the surface was large because of cloud scattering effect. So, the cloud property data is needed to calculate the UV radiation more exactly at King Sejong Station in West Antarctica.

• KIEAE Journal
• /
• v.14 no.4
• /
• pp.35-45
• /
• 2014

The main objective of this study is to provide an effective algorithm of the transmitted solar radiation calculation through window glazing on a clear day. This algorithm would be used in developing a computer program for fenestration system analysis and shading device design. Various simulation methods have been evaluated to figure out the most accurate and effective procedure in estimation of transmitted solar radiation on a tilted surface on a clear day. Characteristics of simulated results of each step have been scrutinized by comparing them with measured results of the site as well as results from other simulation programs. Generally, the Duffie & Beckman's solar calculation method introducing the HDKR anisotropic model provided the most reliable simulation results. The DOE-2 program usually provided over-estimated simulation results. The estimation of extraterrestrial solar radiation and beam normal radiation were conducted pretty accurately. However, the solar radiation either on horizontal surface or on tilted surface involves complicated factors in estimation. Even though the estimation results were close to the real measured data during summer when solar intensity is getting higher, the estimation provided more error when solar intensities were getting weaker. The convex polygon clipping algorithm with homogeneous coordinates was fastest model in calculation of sunlight to shaded area ratio. It could not be applied because of its shape limitation.

• Journal of the Korean Solar Energy Society
• /
• v.30 no.2
• /
• pp.16-21
• /
• 2010

Computer simulation of buildings and solar energy systems are being used increasingly in energy assessments and design. This paper discusses the possibility of using sunshine duration data instead of global hourly solar irradiation (GHSI) data for localities with abundant data on sunshine duration. For six locations in South Korea where global radiation is currently measured, the global radiation was calculated using Sunshine Duration Radiation Model (SDRM), compared and analyzed. Results of SDRM has been compared with the measured data on the coefficients of determination (R2), root-mean-square error (RMSE) and mean bias error (MBE). This study recommends the use of sunshine duration based irradiation models if measured solar radiation data is not available.

• Ocean and Polar Research
• /
• v.26 no.2
• /
• pp.133-143
• /
• 2004

For the purpose of understanding the cloud scattering effect of UV radiation at King Sejong station In West Antarctica, we analyzed the data measured by UV-Biometer at surface and compared its result with solar radiation model. The parameterization of UV radiation by cloud ice crystal was applied to solar radiation model and the sensitivity of this model for the variation of ice crystal was tested. The cloud optical thickness was calculated by using this solar radiation model. It was compared the result from calculation with CERES satellite data. In solar radiation model, the UV radiation was less scattered with increase of ice crystal size in cloud and this scattering effect was more important to UV-A radiation than Erythemal UV-B radiation. But scattering effects by altitude of cloud was not serious. The calculated cloud optical thicknesses in Erythemal UV-B and UV-A region were compared with CERES satellite data and the result by UV-A was more accurate than Erythemal UV-B region.

• Journal of Korean Society for Atmospheric Environment
• /
• v.33 no.5
• /
• pp.435-444
• /
• 2017

In this study, we investigated the effect of aerosol feedback on $PM_{10}$ simulation using a two-way coupled air quality model (WRF-CMAQ). $PM_{10}$ concentration over Korea in January 2014 was simulated, and the aerosol feedback effect on the simulated solar radiation was intensively examined. Two $PM_{10}$ simulations were conducted using the WRF-CMAQ model with (FB) and without(NFB) the aerosol feedback option. We find that the simulated solar radiation in the west part of Korea decreased by up to $-80MJ/m^2$ due to the aerosol feedback effect. The feedback effect was significant in the west part of Korea, showing high $PM_{10}$ estimates due to dense emissions and its long-range transport from China. The aerosol feedback effect contributed to the decreased rRMSE(relative Root Mean Square Error) for solar radiation (47.58% to 30.75%). Aerosol feedback effect on the simulated solar radiation was mainly affected by concentration of $PM_{10}$. Moreover, FB better matched the observed solar radiation and $PM_{10}$ concentration than NFB, implying that taking into account the aerosol direct effects resulted in the improved modeling performance. These results indicate that aerosol feedback effects can play an important role in the simulation of solar radiation over Korean Peninsula.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.479-482
• /
• 2015

The objective of this study is to develop adaptive neuro-fuzzy inference system (ANFIS) model for estimating daily solar radiation using limited weather variables at Champaign and Springfield stations in Illinois. The best input combinations (one, two, and three inputs) can be identified using ANFIS model. From the performance evaluation and scatter diagrams of ANFIS model, ANFIS 3 (three input) model produces the best results for both stations. Results obtained indicate that ANFIS model can successfully be used for the estimation of daily global solar radiation at Champaign and Springfield stations in Illinois. These results testify the generation capability of ANFIS model and its ability to produce accurate estimates in Illinois.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.483-486
• /
• 2015

The objective of this study is to develop generalized regression neural networks (GRNN) model for estimating daily solar radiation using limited weather variables at Champaign and Springfield stations in Illinois. The best input combinations (one, two, and three inputs) can be identified using GRNN model. From the performance evaluation and scatter diagrams of GRNN model, GRNN 3 (three input) model produces the best results for both stations. Results obtained indicate that GRNN model can successfully be used for the estimation of daily global solar radiation at Champaign and Springfield stations in Illinois. These results testify the generation capability of GRNN model and its ability to produce accurate estimates in Illinois.

• Journal of the Korean Solar Energy Society
• /
• v.31 no.4
• /
• pp.72-79
• /
• 2011

Solar radiation data are the best source of information for estimating average incident radiation. Lacking this or data from nearby locations of similar climate, it is possible to use empirical relationships to estimate radiation from days of hours of bright sunshine. It is necessary to estimate the regression coefficients in order to predict the daily global radiation on a horizontal surface. Therefore many different equations have proposed to evaluate them for certain areas. In this work a new correlation has been made to predict the solar radiation for 16 different areas over Korea by estimating the regression coefficients taking into account hours of bright sunshine. Particularly, the proposed straight line regression model shows reliable results for estimating the global radiation on a horizontal surface with monthly average deviation of -0.2 to +0.5% and each station annual average deviation of -1.6 to +1.7% from measured values.