Vertical extrapolation of wind speed using artificial neural network hybrid system

Islam, Md. Saiful; Mohandes, M.; Rehman, Shafiqur

doi:10.1007/s00521-016-2373-x

Cited by 34 publications

(19 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Such models offer added opportunity to forecast future PAR on very short-and long-term scales, without needing extensive atmospheric model initialization conditions. Yet, even though much research has been performed for solar, wind, and atmospheric modeling applications (e.g., (Deo & Sahin, 2017;Deo et al, 2016;Dokur, Kurban, & Ceyhan, 2015;Ghorbani, Khatibi, Hosseini, & Bilgili, 2013;Islam, Mohandes, & Rehman, 2016;Mohammadi et al, 2015;Salcedo-Sanz, Deo, Cornejo-Bueno, Camacho-Gómez, & Ghimire, 2018)), modeling of very short-term (i.e., near realtime) PAR in sub-tropical regions remains scarce. These regions have less cloud cover than tropical or temperate climate areas and therefore enhance the suitability of sites for solar and bio-energy generation.…”

Section: Background Reviewmentioning

confidence: 99%

Adaptive Neuro‐Fuzzy Inference System integrated with solar zenith angle for forecasting sub‐tropical Photosynthetically Active Radiation

Deo

Downs

Adamowski

et al. 2018

Food and Energy Security

View full text Add to dashboard Cite

Advocacy for climate mitigation aims to minimize the use of fossil fuel and to support clean energy adaptation. While alternative energies (e.g., biofuels) extracted from feedstock (e.g., micro‐algae) represent a promising role, their production requires reliably modeled photosynthetically active radiation (PAR). PAR models predict energy parameters (e.g., algal carbon fixation) to aid in decision‐making at PAR sites. Here, we model very short‐term (5‐min scale), sub‐tropical region's PAR with an Adaptive Neuro‐Fuzzy Inference System model with a Centroid‐Mean (ANFIS‐CM) trained with a non‐climate input (i.e., only the solar angle, θZ). Accuracy is benchmarked against genetic programming (GP), M5Tree, Random Forest (RF), and multiple linear regression (MLR). ANFIS‐CM integrates fuzzy and neural network algorithms, whereas GP adopts an evolutionary approach, M5Tree employs binary decision, RF employs a bootstrapped ensemble, and MLR uses statistical tools to link PAR with θZ. To design the ANFIS‐CM model, 5‐min θZ (01–31 December 2012; 0500H–1900H) for sub‐tropical, Toowoomba are utilized to extract predictive features, and the testing accuracy (i.e., differences between measurements and forecasts) is evaluated with correlation (r), root‐mean‐square error (RMSE), mean absolute error (MAE), Willmott (WI), Nash–Sutcliffe (ENS), and Legates & McCabes (ELM) Index. ANFIS‐CM and GP are equivalent for 5‐min forecasts, yielding the lowest RMSE (233.45 and 233.01μ mol m−2s−1) and MAE (186.59 and 186.23 μmol m−2s−1). In contrast, MLR, M5Tree, and RF yields higher RMSE and MAE [(RMSE = 322.25 μmol m−2s−1, MAE = 275.32 μmol m−2s−1), (RMSE = 287.70 μmol m−2s−1, MAE = 234.78 μmol m−2s−1), and (RMSE = 359.91 μmol m−2s−1, MAE = 324.52 μmol m−2s−1)]. Based on normalized error, ANFIS‐CM is considerably superior (MAE = 17.18% versus 19.78%, 34.37%, 26.39%, and 30.60% for GP, MLR, M5Tree, and RF models, respectively). For hourly forecasts, ANFIS‐CM outperforms all other methods (WI = 0.964 vs. 0.942, 0.955, 0.933 & 0.893, and ELM = 0.741 versus 0.701, 0.728, 0.619 & 0.490 for GP, MLR, M5Tree, and RF, respectively). Descriptive errors support the versatile predictive skills of the ANFIS‐CM model and its role in real‐time prediction of the photosynthetic‐active energy to explore biofuel generation from micro‐algae, studying food chains, and supporting agricultural precision.

show abstract

Section: Background Reviewmentioning

confidence: 99%

Adaptive Neuro‐Fuzzy Inference System integrated with solar zenith angle for forecasting sub‐tropical Photosynthetically Active Radiation

Deo

Downs

Adamowski

et al. 2018

Food and Energy Security

View full text Add to dashboard Cite

show abstract

“…Similarly, capacity factors (CFs) were found to be about 2-3% higher. Another such work has been presented in Islam et al (2017).…”

Section: Introductionmentioning

confidence: 99%

Investigation of wind shear coefficients and their effect on annual energy yields along the coastal sites of Pakistan

Sumair

Aized

Gardezi

et al. 2020

Energy Exploration & Exploitation

View full text Add to dashboard Cite

Ten-year hourly recorded wind meteorological data at six sites along the coastline of Pakistan at two heights (10 and 50 m) were extrapolated to two higher heights (80 and 100 m). Monthly and seasonal analysis of variation in air density (ρ), wind speed, Weibull parameters ( K and C), wind power density, and wind energy density with height was investigated. Analysis shows that wind shear coefficient is highest in winter and lowest in summer. ρ, wind speed, wind power density, and wind energy density all increase with increasing hub height, with the most prominent increment in winter and the lowest in summer. With increasing height, K has been found to decrease slightly while C increases. Techno-economic feasibility analysis of annual energy production using 15 turbines was carried out which shows that capacity factor alone cannot render a turbine feasible but also economic assessment is mandatory to evaluate the feasibility of turbines. G1 and G2 turbines have been found the best options while B5 and V2 as the worst. Comparison among sites shows that Karachi is the most potential site with cost of energy of $0.017/kW h while Jiwani is the worst site with cost of energy of $0.039/kW h both at 100 m height.

show abstract

“…Regrettably, to date, no guidelines exist to define both features of the network. It is feasible to find the ideal and optimal network just by utilizing a trial and error procedure [45][46][47][48][49].…”

Section: Introductionmentioning

confidence: 99%

A Hybrid Analytic Network Process and Artificial Neural Network (ANP-ANN) Model for Urban Earthquake Vulnerability Assessment

et al. 2018

View full text Add to dashboard Cite

Vulnerability assessment is one of the prerequisites for risk analysis in disaster management. Vulnerability to earthquakes, especially in urban areas, has increased over the years due to the presence of complex urban structures and rapid development. Urban vulnerability is a result of human behavior which describes the extent of susceptibility or resilience of social, economic, and physical assets to natural disasters. The main aim of this paper is to develop a new hybrid framework using Analytic Network Process (ANP) and Artificial Neural Network (ANN) models for constructing a composite social, economic, environmental, and physical vulnerability index. This index was then applied to Tabriz City, which is a seismic-prone province in the northwestern part of Iran with recurring devastating earthquakes and consequent heavy casualties and damages. A Geographical Information Systems (GIS) analysis was used to identify and evaluate quantitative vulnerability indicators for generating an earthquake vulnerability map. The classified and standardized indicators were subsequently weighed and ranked using an ANP model to construct the training database. Then, standardized maps coupled with the training site maps were presented as input to a Multilayer Perceptron (MLP) neural network for producing an Earthquake Vulnerability Map (EVM). Finally, an EVM was produced for Tabriz City and the level of vulnerability in various zones was obtained. South and southeast regions of Tabriz City indicate low to moderate vulnerability, while some zones of the northeastern tract are under critical vulnerability conditions. Furthermore, the impact of the vulnerability of Tabriz City on population during an earthquake was included in this analysis for risk estimation. A comparison of the result produced by EVM and the Population Vulnerability (PV) of Tabriz City corroborated the validity of the results obtained by ANP-ANN. The findings of this paper are useful for decision-makers and government authorities to obtain a better knowledge of a city's vulnerability dimensions, and to adopt preparedness strategies in the future for Tabriz City. The developed hybrid framework of ANP and ANN Models can easily be replicated and applied to other urban regions around the world for sustainability and environmental management.

show abstract

Vertical extrapolation of wind speed using artificial neural network hybrid system

Cited by 34 publications

References 20 publications

Adaptive Neuro‐Fuzzy Inference System integrated with solar zenith angle for forecasting sub‐tropical Photosynthetically Active Radiation

Adaptive Neuro‐Fuzzy Inference System integrated with solar zenith angle for forecasting sub‐tropical Photosynthetically Active Radiation

Investigation of wind shear coefficients and their effect on annual energy yields along the coastal sites of Pakistan

A Hybrid Analytic Network Process and Artificial Neural Network (ANP-ANN) Model for Urban Earthquake Vulnerability Assessment

Contact Info

Product

Resources

About