Multiple-Input Deep Convolutional Neural Network Model for Short-Term Photovoltaic Power Forecasting

Abstract

With the fast expansion of renewable energy system installed capacity in recent years, the availability, stability, and quality of smart grids have become increasingly important. The renewable energy output forecasting applications have also been developing rapidly in recent years, and such techniques have particularly been applied in the fields of wind and solar photovoltaic (PV). In the case of solar PV output forecasting, many applications have been performed with machine learning and hybrid techniques. In this paper, we propose a high-precision deep neural network model named PVPNet to forecast PV system output power. The methodology behind the proposed model is based on deep neural networks, and the model is able to generate a 24-h probabilistic and deterministic forecasting of PV power output based on meteorological information, such as temperature, solar radiation, and historical PV system output data. The forecasting accuracy of PVPNet is determined by the mean absolute error (MAE) and root mean square error (RMSE) values. The results from the experiments show that the MAE and RMSE of the proposed algorithm are 109.4845 and 163.1513, respectively. The results prove that the prediction accuracy of the PVPNet outperforms other benchmark models, and the algorithm also effectively predicts complex time series with a high degree of volatility and irregularity. INDEX TERMS Deep neural network, photovoltaic output power forecasting, photovoltaic system, renewable energy sources.

Short-Term Photovoltaic Power Forecasting Using an LSTM Neural Network and Synthetic Weather Forecast

“…An LSTM NN based PV power forecasting algorithm is proposed in [19] to predict intraday and 24-hour horizons using a time index as an additional input feature along with the relevant weather variables. A DNN is used in [20] to predict next 24-hour PV generation based on historical weather information and a rolling horizon strategy. The forecasting accuracy of the models proposed in [14], [16], [17], [19], [20] can still be enhanced by considering weather forecast data [18].…”

A Local Training Strategy-Based Artificial Neural Network for Predicting the Power Production of Solar Photovoltaic Systems

“…Huang et al, [35] proposed a high-precision PVPNet model-based Deep-Learning Neural Networks (DLNNs) for 1-day ahead power output prediction. The prediction results obtained by the proposed PVPNet model were evaluated (in terms of RMSE and MAE) and compared to other ML techniques of literature.…”

Designing Sustainable Smart Cities: Cooperative Energy Management Systems and Applications

“…Spatially distributed solar radiation information with a relatively high temporal resolution [90] obtained from meteorological satellites, for example, Himawari-8 [91], ‡ which have been undergoing remarkable technological innovation in recent years, contributes significantly to monitoring solar power generation for effective use. The methodology of PV power forecast using such information also has been actively discussed [92][93][94][95][96][97][98]; in particular, several recent studies have focused on the forecast of net-load that considers the effect of behind-the-meter PVs [99,100], which are necessary because of restrictions on the measurement location in real-world power systems. In addition, the situation is similar for wind power generation [101], although it is rare for largescale installations in cities; in the context of grasping wind power generation, the use of data via supervisory control and data acquisition (SCADA) system and the sophistication of numerical weather models [102] play important roles.…”