A Hierarchical Approach Using Machine Learning Methods in Solar Photovoltaic Energy Production Forecasting

Probabilistic forecasting accounts for the uncertainty in prediction that arises from inaccurate input data due to measurement errors, as well as the inherent inaccuracy of a prediction model. Because of the variable nature of renewable power generation depending on weather conditions, probabilistic forecasting is well suited to it. For a grid-tied solar farm, it is increasingly important to forecast the solar power generation several hours ahead. In this study, we propose three different methods for ensemble probabilistic forecasting, derived from seven individual machine learning models, to generate 24-h ahead solar power forecasts. We have shown that while all of the individual machine learning models are more accurate than the traditional benchmark models, like autoregressive integrated moving average (ARIMA), the ensemble models offer even more accurate results than any individual machine learning model alone does. Furthermore, it is observed that running separate models on the data belonging to the same hour of the day vastly improves the accuracy of the results. Getting more accurate forecasts will help the stakeholders come up with better decisions in resource planning and control when large-scale solar farms are integrated into the power grid.

Section: Machine Learning Methodsmentioning

confidence: 99%

Ensemble Learning Approach for Probabilistic Forecasting of Solar Power Generation

Mohammed¹,

Aung²

2016

“…The impacts on PV module temperature include internal and external aspects [31]. The internal aspect refers to the PV module physical characteristics related factors including material category, parameters and system-dependent properties [32], which are fixed and unique to those individual PV plants that already put into operation.…”

Section: Physical Description Of Photovoltaic Module Temperaturementioning

confidence: 99%

“…According to Equations (30) and (31), the values of II can measure the redundancy between two variables when they act on the same object, which is shown in Table 11. Furthermore, how much redundancy existed between each MIF is also addressed.…”

Section: Ratio Of Quantitative Influence Degreesmentioning

confidence: 99%

See 1 more Smart Citation

Correlation Feature Selection and Mutual Information Theory Based Quantitative Research on Meteorological Impact Factors of Module Temperature for Solar Photovoltaic Systems

Sun

Wang

et al. 2016

Abstract:The module temperature is the most important parameter influencing the output power of solar photovoltaic (PV) systems, aside from solar irradiance. In this paper, we focus on the interdisciplinary research that combines the correlation analysis, mutual information (MI) and heat transfer theory, which aims to figure out the correlative relations between different meteorological impact factors (MIFs) and PV module temperature from both quality and quantitative aspects. The identification and confirmation of primary MIFs of PV module temperature are investigated as the first step of this research from the perspective of physical meaning and mathematical analysis about electrical performance and thermal characteristic of PV modules based on PV effect and heat transfer theory. Furthermore, the quantitative description of the MIFs influence on PV module temperature is mathematically formulated as several indexes using correlation-based feature selection (CFS) and MI theory to explore the specific impact degrees under four different typical weather statuses named general weather classes (GWCs). Case studies for the proposed methods were conducted using actual measurement data of a 500 kW grid-connected solar PV plant in China. The results not only verified the knowledge about the main MIFs of PV module temperatures, more importantly, but also provide the specific ratio of quantitative impact degrees of these three MIFs respectively through CFS and MI based measures under four different GWCs.

“…According to the statistical data of the National Energy Administration of China, the phenomenon of abandoning PV power in northwestern China is intensely severe, where the abandonment rates in Gansu Province and in Xinjiang Province were 31% and 26% in 2015, respectively, corresponding to an abandoned capacity of 1.897 MW and 1.0556 MW, respectively. Therefore, the precise forecast of PV power plays a vital role in reducing the impacts of uncertainties of PV output, achieving a reasonable power dispatch, decreasing power generation costs, raising the utilization rate of PV units, promoting the development of PV enterprises and finally obtaining higher general social benefits [1][2][3][4][5][6][7].…”

Section: Introductionmentioning

confidence: 99%

Study of Short-Term Photovoltaic Power Forecast Based on Error Calibration under Typical Climate Categories

Gao

Zhu

et al. 2016

Abstract:With the increasing permeability of photovoltaic (PV) power production, the uncertainties and randomness of PV power have played a critical role in the operation and dispatch of the power grid and amplified the abandon rate of PV power. Consequently, the accuracy of PV power forecast urgently needs to be improved. Based on the amplitude and fluctuation characteristics of the PV power forecast error, a short-term PV output forecast method that considers the error calibration is proposed. Firstly, typical climate categories are defined to classify the historical PV power data. On the one hand, due to the non-negligible diversity of error amplitudes in different categories, the probability density distributions of relative error (RE) are generated for each category. Distribution fitting is performed to simulate probability density function (PDF) curves, and the RE samples are drawn from the fitted curves to obtain the sampling values of the RE. On the other hand, based on the fluctuation characteristic of RE, the recent RE data are utilized to analyze the error fluctuation conditions of the forecast points so as to obtain the compensation values of the RE. The compensation values are adopted to sequence the sampling values by choosing the sampling values closest to the compensation ones to be the fitted values of the RE. On this basis, the fitted values of the RE are employed to correct the forecast values of PV power and improve the forecast accuracy.