Modeling of photovoltaic soiling loss as a function of environmental variables

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

In this study, an artificial neural network (ANN) approach was applied for modeling the relation between environmental variables and daily change in the Cleanness Index (ΔCI, a measure of performance loss due to soiling) of photovoltaic modules in the field in Doha, Qatar. The daily ΔCI was examined among a number of three-dimensional intervals of the daily mean of the environmental variables (i.e., the intervals of two environmental variables were presented on x and y dimensions, and average values of daily ΔCI on the third dimension), in order to qualitatively establish the relations that might help to develop improved PV soiling prediction models. Then, an ANN-based model was set up to simulate the relationship between daily ΔCI and environmental variables and compared with a linear regression model, both models using the same input variables, including present day and previous day environmental conditions, and cumulative exposure time. Strong interactions were observed among environmental variables PM10, relative humidity (RH) and wind speed (WS) regarding their effect on the daily ΔCI. Overall, higher PM10 resulted in more negative daily ΔCI (i.e. the module became more soiled), and this effect was more visible at low WS and RH levels, but at high WS (>4 m s−1) and high RH (>65%) levels, PM10 had no significant (p > 0.05, two tailed t-test) effect on daily ΔCI. Mostly, WS and RH determined how much airborne dust accumulates on the module surfaces and thereby affects the output of the PV modules. Higher WS typically favored more positive daily ΔCI when RH was low, but at higher RH levels (>50%) daily ΔCI was more likely to be negative with increasing WS. In fact, high RH levels were related to negative daily ΔCI only at higher WS levels (>2 m s−1); at lower WS levels RH had no significant effect on daily ΔCI. These effects were apparently due to the deposition-resuspension mechanisms of dust accumulation on the PV panel surfaces. The ANN model performed significantly better in predicting daily ΔCI as well as cumulative CI than the linear model in term of R2 values and statistical error indexes. The previous day environmental conditions had a significant effect on the modeling outcome. The inclusion of the wind gustiness and cumulative exposure time also considerably improved the model prediction capability. The advantage of the ANN-based model is its simplicity, ease of data fitting and no requirement of an accurate mathematical model.

Original languageEnglish
Pages (from-to)397-407
Number of pages11
JournalSolar Energy
Volume157
DOIs
Publication statusPublished - 15 Nov 2017

Fingerprint

Atmospheric humidity
Neural networks
Dust
Linear regression
Mathematical models

Keywords

  • Artificial neural network
  • Cleanness index
  • Multilinear regression
  • PV soiling prediction

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment
  • Materials Science(all)

Cite this

Modeling of photovoltaic soiling loss as a function of environmental variables. / Javed, Wasim; Guo, Bing; Figgis, Benjamin.

In: Solar Energy, Vol. 157, 15.11.2017, p. 397-407.

Research output: Contribution to journalArticle

@article{9a2434c0205d4355b6b7ac0758d2b53d,
title = "Modeling of photovoltaic soiling loss as a function of environmental variables",
abstract = "In this study, an artificial neural network (ANN) approach was applied for modeling the relation between environmental variables and daily change in the Cleanness Index (ΔCI, a measure of performance loss due to soiling) of photovoltaic modules in the field in Doha, Qatar. The daily ΔCI was examined among a number of three-dimensional intervals of the daily mean of the environmental variables (i.e., the intervals of two environmental variables were presented on x and y dimensions, and average values of daily ΔCI on the third dimension), in order to qualitatively establish the relations that might help to develop improved PV soiling prediction models. Then, an ANN-based model was set up to simulate the relationship between daily ΔCI and environmental variables and compared with a linear regression model, both models using the same input variables, including present day and previous day environmental conditions, and cumulative exposure time. Strong interactions were observed among environmental variables PM10, relative humidity (RH) and wind speed (WS) regarding their effect on the daily ΔCI. Overall, higher PM10 resulted in more negative daily ΔCI (i.e. the module became more soiled), and this effect was more visible at low WS and RH levels, but at high WS (>4 m s−1) and high RH (>65{\%}) levels, PM10 had no significant (p > 0.05, two tailed t-test) effect on daily ΔCI. Mostly, WS and RH determined how much airborne dust accumulates on the module surfaces and thereby affects the output of the PV modules. Higher WS typically favored more positive daily ΔCI when RH was low, but at higher RH levels (>50{\%}) daily ΔCI was more likely to be negative with increasing WS. In fact, high RH levels were related to negative daily ΔCI only at higher WS levels (>2 m s−1); at lower WS levels RH had no significant effect on daily ΔCI. These effects were apparently due to the deposition-resuspension mechanisms of dust accumulation on the PV panel surfaces. The ANN model performed significantly better in predicting daily ΔCI as well as cumulative CI than the linear model in term of R2 values and statistical error indexes. The previous day environmental conditions had a significant effect on the modeling outcome. The inclusion of the wind gustiness and cumulative exposure time also considerably improved the model prediction capability. The advantage of the ANN-based model is its simplicity, ease of data fitting and no requirement of an accurate mathematical model.",
keywords = "Artificial neural network, Cleanness index, Multilinear regression, PV soiling prediction",
author = "Wasim Javed and Bing Guo and Benjamin Figgis",
year = "2017",
month = "11",
day = "15",
doi = "10.1016/j.solener.2017.08.046",
language = "English",
volume = "157",
pages = "397--407",
journal = "Solar Energy",
issn = "0038-092X",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Modeling of photovoltaic soiling loss as a function of environmental variables

AU - Javed, Wasim

AU - Guo, Bing

AU - Figgis, Benjamin

PY - 2017/11/15

Y1 - 2017/11/15

N2 - In this study, an artificial neural network (ANN) approach was applied for modeling the relation between environmental variables and daily change in the Cleanness Index (ΔCI, a measure of performance loss due to soiling) of photovoltaic modules in the field in Doha, Qatar. The daily ΔCI was examined among a number of three-dimensional intervals of the daily mean of the environmental variables (i.e., the intervals of two environmental variables were presented on x and y dimensions, and average values of daily ΔCI on the third dimension), in order to qualitatively establish the relations that might help to develop improved PV soiling prediction models. Then, an ANN-based model was set up to simulate the relationship between daily ΔCI and environmental variables and compared with a linear regression model, both models using the same input variables, including present day and previous day environmental conditions, and cumulative exposure time. Strong interactions were observed among environmental variables PM10, relative humidity (RH) and wind speed (WS) regarding their effect on the daily ΔCI. Overall, higher PM10 resulted in more negative daily ΔCI (i.e. the module became more soiled), and this effect was more visible at low WS and RH levels, but at high WS (>4 m s−1) and high RH (>65%) levels, PM10 had no significant (p > 0.05, two tailed t-test) effect on daily ΔCI. Mostly, WS and RH determined how much airborne dust accumulates on the module surfaces and thereby affects the output of the PV modules. Higher WS typically favored more positive daily ΔCI when RH was low, but at higher RH levels (>50%) daily ΔCI was more likely to be negative with increasing WS. In fact, high RH levels were related to negative daily ΔCI only at higher WS levels (>2 m s−1); at lower WS levels RH had no significant effect on daily ΔCI. These effects were apparently due to the deposition-resuspension mechanisms of dust accumulation on the PV panel surfaces. The ANN model performed significantly better in predicting daily ΔCI as well as cumulative CI than the linear model in term of R2 values and statistical error indexes. The previous day environmental conditions had a significant effect on the modeling outcome. The inclusion of the wind gustiness and cumulative exposure time also considerably improved the model prediction capability. The advantage of the ANN-based model is its simplicity, ease of data fitting and no requirement of an accurate mathematical model.

AB - In this study, an artificial neural network (ANN) approach was applied for modeling the relation between environmental variables and daily change in the Cleanness Index (ΔCI, a measure of performance loss due to soiling) of photovoltaic modules in the field in Doha, Qatar. The daily ΔCI was examined among a number of three-dimensional intervals of the daily mean of the environmental variables (i.e., the intervals of two environmental variables were presented on x and y dimensions, and average values of daily ΔCI on the third dimension), in order to qualitatively establish the relations that might help to develop improved PV soiling prediction models. Then, an ANN-based model was set up to simulate the relationship between daily ΔCI and environmental variables and compared with a linear regression model, both models using the same input variables, including present day and previous day environmental conditions, and cumulative exposure time. Strong interactions were observed among environmental variables PM10, relative humidity (RH) and wind speed (WS) regarding their effect on the daily ΔCI. Overall, higher PM10 resulted in more negative daily ΔCI (i.e. the module became more soiled), and this effect was more visible at low WS and RH levels, but at high WS (>4 m s−1) and high RH (>65%) levels, PM10 had no significant (p > 0.05, two tailed t-test) effect on daily ΔCI. Mostly, WS and RH determined how much airborne dust accumulates on the module surfaces and thereby affects the output of the PV modules. Higher WS typically favored more positive daily ΔCI when RH was low, but at higher RH levels (>50%) daily ΔCI was more likely to be negative with increasing WS. In fact, high RH levels were related to negative daily ΔCI only at higher WS levels (>2 m s−1); at lower WS levels RH had no significant effect on daily ΔCI. These effects were apparently due to the deposition-resuspension mechanisms of dust accumulation on the PV panel surfaces. The ANN model performed significantly better in predicting daily ΔCI as well as cumulative CI than the linear model in term of R2 values and statistical error indexes. The previous day environmental conditions had a significant effect on the modeling outcome. The inclusion of the wind gustiness and cumulative exposure time also considerably improved the model prediction capability. The advantage of the ANN-based model is its simplicity, ease of data fitting and no requirement of an accurate mathematical model.

KW - Artificial neural network

KW - Cleanness index

KW - Multilinear regression

KW - PV soiling prediction

UR - http://www.scopus.com/inward/record.url?scp=85028015057&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028015057&partnerID=8YFLogxK

U2 - 10.1016/j.solener.2017.08.046

DO - 10.1016/j.solener.2017.08.046

M3 - Article

VL - 157

SP - 397

EP - 407

JO - Solar Energy

JF - Solar Energy

SN - 0038-092X

ER -