Generation of solar radiation data in unmeasurable areas for photovoltaic power station planning

Yang, Chenxing; Xu, Qingshan; Xu, Xiaohui; Zeng, Pingliang; Yuan, Xiaodong

doi:10.1109/pesgm.2014.6939206

Cited by 3 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normalization was done to have the same range of values for each of the inputs to the models. This normalization procedure guarantees stable convergence of weight and biases [61,62].…”

Section: Data Preparation Feature Selection and Sensitivity Analysismentioning

confidence: 99%

Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction

et al. 2019

View full text Add to dashboard Cite

Solar energy predictive models designed to emulate the long-term (e.g., monthly) global solar radiation (GSR) trained with satellite-derived predictors can be employed as decision tenets in the exploration, installation and management of solar energy production systems in remote and inaccessible solar-powered sites. In spite of a plethora of models designed for GSR prediction, deep learning, representing a state-of-the-art intelligent tool, remains an attractive approach for renewable energy exploration, monitoring and forecasting. In this paper, algorithms based on deep belief networks and deep neural networks are designed to predict long-term GSR. Deep learning algorithms trained with publicly-accessible Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data are tested in Australia’s solar cities to predict the monthly GSR: single hidden layer and ensemble models. The monthly-scale MODIS-derived predictors (2003–2018) are adopted, with 15 diverse feature selection approaches including a Gaussian Emulation Machine for sensitivity analysis used to select optimal MODIS-predictor variables to simulate GSR against ground-truth values. Several statistical score metrics are adopted to comprehensively verify surface GSR simulations to ascertain the practicality of deep belief and deep neural networks. In the testing phase, deep learning models generate significantly lower absolute percentage bias (≤3%) and high Kling–Gupta efficiency (≥97.5%) values compared to the single hidden layer and ensemble model. This study ascertains that the optimal MODIS input variables employed in GSR prediction for solar energy applications can be relatively different for diverse sites, advocating a need for feature selection prior to the modelling of GSR. The proposed deep learning approach can be adopted to identify solar energy potential proactively in locations where it is impossible to install an environmental monitoring data acquisition instrument. Hence, MODIS and other related satellite-derived predictors can be incorporated for solar energy prediction as a strategy for long-term renewable energy exploration.

show abstract

“…Normalization was done to have the same range of values for each of the inputs to the models. This normalization procedure guarantees stable convergence of weight and biases [61,62].…”

Section: Data Preparation Feature Selection and Sensitivity Analysismentioning

confidence: 99%

Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Monthly global solar radiation (MGSR) data are estimated using the inverse distance weighting (IDW) and ANN methods. Yang et al, 24 proposed a typical MGSR, which is suitable for unmeasurable, relatively concentrated distributions and similar locations. Nandi et al 25 presented a multi‐layeredfeed‐forward ANN structure utilizing the back‐propagation algorithm to determine the monthly average insolation incident on horizontal surfaces.…”

Section: Introductionmentioning

confidence: 99%

Application of parallel Elman neural network to hourly area solar PV plant generation estimation

Cho

Chang

Huang

2020

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Summary Based on existing power generation data, an hourly area solar power estimation model using the parallel Elman neural network with solar radiation and system conversion efficiency is proposed. The accuracy and reliability of the assessment were verified using the information/data of solar photovoltaic power stations in various regions and timescales. Using the established appraisal algorithm involving K‐means evaluation and inverse distance weighting, regional forecasting of solar power generation was achieved. The prediction accuracy was also investigated using the actual details of the photovoltaic power stations. The results of the proposed model can assist the electricity dispatcher to not only precisely monitor the trend of solar power generation in different areas, but also coordinate with traditional power plants to meet the load demand more accurately. The proposed method can benefit power dispatching involving a larger scale of intermittent and unstable solar power electricity in the future.

show abstract

“…It consists of the months selected from the individual years and concatenated to form a complete year. Many methods have been made to produce such weather databases for different locations around the world [1, 4,5,6,7] Center for Solar Energy Research and Studies (CSERS) is the main renewable energy R&D organization in Libya. It was founded in 1978.…”

Section: Introductionmentioning

confidence: 99%

“…In the past, several methodologies for generating TMYs have been reported, such as Sandia method, Festa-Ratto method, Danish method and others [6], all targeted at selecting single months or years from a long-term typical weather condition. Among the different TMY generation methods, the Sandia method is widely adopted [1, [3][4][5][6][7][8][9][10][11][12][13][14] generated TMYs for different locations with different weather parameters and assigned weighting factors. This methodology has been adopted by different countries: for example, by date of publication, for Latvia [9], Athens, Greece [12], Oman [11], Damascus, Syria [6], Nigeria [3,[7][8], Chile [10], Istanbul, Turkey [14], and Armidale NSW, Australia [14].…”

Section: Introductionmentioning

confidence: 99%

The Generation of Typical Meteorological Year for Tripoli, Libya

Khalifa

Alargt

2018

jsesd

View full text Add to dashboard Cite

In this paper a Typical Meteorological Year (TMY) of Tripoli city, the capital of Libya was presented and analyzed using twelve year of measured weather data, this data includes global solar radiation, air temperature and wind speed. Th measurements were taken every ten minutes. Finkelstein–Schafer statistical method was applied to analyze the data and obtain the TMY for the site. Typical Meteorological Months (TMMs) were selected by choosing the one with the least deviation from the long-term cumulativedistribution function. A close-fi agreement is observed between the generated TMY and the long-term averages. Ths TMY generated will be very useful tool for optimal design and performance evaluation of solar energy conversion systems, heating, wind, and other renewable energy systems to be located in this part of Libya.

show abstract

Generation of solar radiation data in unmeasurable areas for photovoltaic power station planning

Cited by 3 publications

References 14 publications

Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction

Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction

Application of parallel Elman neural network to hourly area solar PV plant generation estimation

The Generation of Typical Meteorological Year for Tripoli, Libya

Contact Info

Product

Resources

About