Analysis of daily solar power prediction with data-driven approaches

Long, Huan; Zhang, Zijun; Su, Yan

doi:10.1016/j.apenergy.2014.03.084

Cited by 131 publications

(40 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another method based on Markov chain [7]- [9] also relied on a large number of raw data, but it has shown higher accuracy even if some weather data are omitted. Furthermore, there are other machine learning algorithms such as kNN regression algorithms, the K-nearest neighbor technology has been validated to be a good way for future PV power prediction [10], [11]. Support vector machine (SVM) was also applied to predict PV power outputs [12].…”

Section: Introductionmentioning

confidence: 99%

An Improved Photovoltaic Power Forecasting Model With the Assistance of Aerosol Index Data

Liu

Fang

Zhang

et al. 2015

IEEE Trans. Sustain. Energy

305

121

View full text Add to dashboard Cite

Due to the intermittency and randomness of solar photovoltaic (PV) power, it is difficult for system operators to dispatch PV power stations. In order to find a precise expectation for day-ahead PV power generation, conventional models have taken into consideration the temperature, humidity, and wind speed data for forecasting, but these predictions were always not accurate enough under extreme weather conditions. Aerosol index (AI), which indicates the particulate matter in the atmosphere, has been found to have strong linear correlation with solar radiation attenuation, and might have potential influence on the power generated by PV panels. A novel PV power forecasting model is proposed in this paper, considering AI data as an additional input parameter. Based on seasonal weather classification, the back propagation (BP) artificial neural network (ANN) approach is utilized to forecast the next 24-h PV power outputs. The estimated results of the proposed PV power forecasting model coincide well with measurement data, and the proposed model has shown the ability of improving prediction accuracy, compared with conventional methods using ANN.Index Terms-Aerosol index (AI), artificial neural network (ANN) method, back propagation (BP) network, maximum absolute prediction error, photovoltaic (PV) power forecasting.

show abstract

Section: Introductionmentioning

confidence: 99%

An Improved Photovoltaic Power Forecasting Model With the Assistance of Aerosol Index Data

Liu

Fang

Zhang

et al. 2015

IEEE Trans. Sustain. Energy

305

121

View full text Add to dashboard Cite

show abstract

“…Hybrid systems incorporating more than one renewable energy source are proposed to predict capacity of electricity generation, using solar and wind power models in [27], and solar and hydro power models in [28]. An ANN using classified weather information based on the past metrological information in [29], and there are other research results that suggest the better performance of DNN-based models with one more hidden layers using weather data [30][31][32]. Similarity, as an approximation method, the regularized partial functional linear regression model (PFLRM) and multivariate adaptive regression splines (MAR) [33,34] were developed.…”

Section: Introductionmentioning

confidence: 99%

Recurrent Neural Network-Based Hourly Prediction of Photovoltaic Power Output Using Meteorological Information

2019

View full text Add to dashboard Cite

Recently, the prediction of photovoltaic (PV) power has become of paramount importance to improve the expected revenue of PV operators and the effective operations of PV facility systems.Additionally, the precise PV power output prediction in an hourly manner enables more sophisticated strategies for PV operators and markets as the electricity price in a renewable energy market is continuously changing. However, the hourly prediction of PV power outputs is considered as a challenging problem due to the dynamic natures of meteorological information not only in a day but also across days. Therefore, in this paper, we suggest three PV power output prediction methods such as artificial neural network (ANN)-, deep neural network (DNN)-, and long and short term memory (LSTM)-based models that are capable to understand the hidden relationships between meteorological information and actual PV power outputs. In particular, the proposed LSTM based model is designed to capture both hourly patterns in a day and seasonal patterns across days. We conducted the experiments by using a real-world dataset. The experimental results show that the proposed ANN based model fails to yield satisfactory results, and the proposed LSTM based model successfully better performs more than 50% compared to the conventional statistical models in terms of mean absolute error.Energies 2019, 12, 215 2 of 22 By extension, for the more sophisticated decisions of PV operators and the effective operations of electricity markets, the prediction of PV power outputs in a hourly level is crucially necessary [9]. Since the prices of electricity sold in the most electricity markets for renewable energy dynamically change over time according to the expected amount of generated electricity, the ability to predict PV power outputs for each particular hour in a day enables the adaptive adjustment of the transaction volumes with tailored schedules. Figure 1 shows typical examples of PV power outputs during two consecutive days in summer and winter in South Korea obtained from the PV power system considered in this research. Here, easy and hard seasons divide according to fluctuation levels of the PV power output. For example, as shown Figure 1, more fluctuation of the PV power output occurs in winter compared to summer, meaning it is harder to predict the PV power output in winter. Naturally, PV power output starts to be generated when the sun rises

show abstract

“…In some previous works, comparative analysis of ANN, SVM and linear regression has been carried out [18], yet in most of the previous studies such an extensive test period has not been considered (1796 days at 98 sites i.e. a matrix of size 1796 Â 98) and a comparison of regularized linear and nonlinear models has not been performed.…”

Section: Problem Formulation and Research Motivationmentioning

confidence: 99%

“…Each GEFS data file consists of the forecast values of one variable, e.g. precipitation or temperature, by the each of the 11 ensemble members of NWP model on each day, at 5 different time steps of the next day (12,15,18,21,24 h), on specified latitudes and longitudes (16 Â 9 grid). As observed in Section 2.2.2, all these weather variables are strongly correlated with the target variable i.e.…”

Section: Feature Segregationmentioning

confidence: 99%

Solar energy prediction using linear and non-linear regularization models: A study on AMS (American Meteorological Society) 2013–14 Solar Energy Prediction Contest

Aggarwal¹,

Saini

2014

Energy

View full text Add to dashboard Cite

Analysis of daily solar power prediction with data-driven approaches

Cited by 131 publications

References 34 publications

An Improved Photovoltaic Power Forecasting Model With the Assistance of Aerosol Index Data

An Improved Photovoltaic Power Forecasting Model With the Assistance of Aerosol Index Data

Recurrent Neural Network-Based Hourly Prediction of Photovoltaic Power Output Using Meteorological Information

Solar energy prediction using linear and non-linear regularization models: A study on AMS (American Meteorological Society) 2013–14 Solar Energy Prediction Contest

Contact Info

Product

Resources

About