Short-Term Photovoltaic Power Forecasting Based on VMD and ISSA-GRU

Pengyun Jia,Xianfu Gong,Xinmiao Liu,Haibo Zhang

doi:10.1109/access.2021.3099169

Pengyun Jia, Xianfu Gong + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3099169

Copy DOI

Abstract

Photovoltaic (PV) power generation is affected by many meteorological factors and environmental factors, which has obvious intermittent, random, and volatile characteristics. To improve the accuracy of short-term PV power prediction, a hybrid model (VMD-ISSA-GRU) based on variational mode decomposition (VMD), improved sparrow search algorithm (ISSA) and gated recurrent unit (GRU) is proposed. First of all, the PV time series is decomposed into a series of different subsequences by VMD to reduce the non-stationarity of the original data. Then, the main factors affecting PV power generation are obtained by using the correlation coefficients of Spearman and Pearson, which reduces the computational complexity of the model. Finally, the GRU network optimized by ISSA is used to predict all the subsequences and residual error of VMD, and the prediction results are reconstructed. The results show that the hybrid VMD-ISSA-GRU model has stronger adaptability and higher accuracy than other traditional models. The mean absolute error (MAE) in the whole test set is 1.0128 kW, the root mean square error (RMSE) is 1.5511 kW, and the R <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">adj</sub> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> can reach 0.9993.

Highlights

With the rapid development of the global new energy power generation industry, solar energy has been widely used because of its several advantages of safety, efficiency, and wide distribution [1]–[3]
Based on the previous research, this paper proposes a new hybrid model based on variational mode decomposition (VMD), improved sparrow search algorithm (ISSA), and gated recurrent unit (GRU) to enhance the accuracy of short-term PV power prediction
Compared with VMD-improved particle swarm optimization (IPSO)-GRU, ISSA-GRU and GRU, the mean absolute error (MAE) of VMD-ISSA-GRU is reduced by 13.95%, 52.76% and 86.13% respectively, and the R2adj can reach 0.9999, which indicates that the VMD-ISSA-GRU model can better fit the PV output when the weather conditions are relatively stable

Summary

INTRODUCTION

With the rapid development of the global new energy power generation industry, solar energy has been widely used because of its several advantages of safety, efficiency, and wide distribution [1]–[3]. Compared with the shallow machine learning algorithm, the accuracy of the above methods has been greatly improved, but there are still problems that can not fully mine the local features and internal hidden information of historical PV data. Based on the previous research, this paper proposes a new hybrid model based on VMD, improved sparrow search algorithm (ISSA), and GRU to enhance the accuracy of short-term PV power prediction. The mathematical theory of the VMD algorithm is defined in Section II.B. When forecasting short-term PV power, the meteorological factors with weak correlation will seriously affect the training efficiency and prediction accuracy of the model. Adam algorithm is used for local optimization of the GRU network, so that the training can adaptively calculate the learning efficiency of each parameter, reduce the influence of parameter selection on the accuracy of the model, and improve the convergence speed.

VARIATIONAL MODE DECOMPOSITION

IMPROVED SPARROW SEARCH ALGORITHM

GATED RECURRENT UNIT NETWORK

Findings

DISCUSSION

CONCLUSION