How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case

Abstract

The use of models capable of forecasting the production of photovoltaic (PV) energy is essential to guarantee the best possible integration of this energy source into traditional distribution grids. Long Short-Term Memory networks (LSTMs) are commonly used for this purpose, but their use may not be the better option due to their great computational complexity and slower inference and training time. Thus, in this work, we seek to evaluate the use of neural networks MLPs (Multilayer Perceptron), Recurrent Neural Networks (RNNs), and LSTMs, for the forecast of 5 min of photovoltaic energy production. Each iteration of the predictions uses the last 120 min of data collected from the PV system (power, irradiation, and PV cell temperature), measured from 2019 to mid-2022 in Maceió (Brazil). In addition, Bayesian hyperparameters optimization was used to obtain the best of each model and compare them on an equal footing. Results showed that the MLP performs satisfactorily, requiring much less time to train and forecast, indicating that they can be a better option when dealing with a very short-term forecast in specific contexts, for example, in systems with little computational resources.

Time Series Analysis and Forecasting of Solar Generation in Spain Using eXtreme Gradient Boosting: A Machine Learning Approach

“…This section delves into the most recent research literature within the area of study, highlighting the extensive range of solar power models and techniques that have been proposed. These models encompass various mathematical functions, both linear and nonlinear, applied across diverse contexts, including projects in Saudi Arabia [9], Malaysia [10], Brazil [11], Israel [12], Australia [13,14], Turkey [15], India [16], the United States [17], Scotland [18], South Korea [19], Nigeria [20], Italy [21], and Algeria [22]. Moreover, non-linear functions have been employed for daily diffuse solar energy radiation calculations [23], irradiation simulations [24], and unrestricted methods [25].…”

“…Numerous prior studies in [11,[46][47][48] have conventionally focused on a multitude of meteorological parameters. These encompass, but are not limited to, irradiance, temperature, humidity, air pressure, wind speed, wind direction, precipitation, dust deposition, and cloud cover.…”

Comparative Performance Analysis of a Grid-Connected Photovoltaic Plant in Central Greece after Several Years of Operation Using Neural Networks

“…Andrade et al used an MLP, RNN, and LSTM to forecast photovoltaic energy from data collected from the PV system in Brazil. The MLP performed adequately, requiring less training time [23]. Kim et al proposed a combination of a two-step NN bi-directional long short-term memory (BD-LSTM) model with an ANN model using exponential moving average (EMA) preprocessing of historical hourly input data of horizontal radiation, ambient temperature, and surface temperature [24].…”

Photovoltaic Power Prediction Based on Hybrid Deep Learning Networks and Meteorological Data

“…In recent years, deep learning methods have garnered significant attention from researchers due to their exceptional feature extraction and transformation capabilities, leading to remarkable achievements in PV power prediction [ 4 ]. Long short-term memory (LSTM), as a classical deep learning approach, with its unique architecture facilitating the transfer of available information from previous states to the current state through memory units, is well suited for PV power forecasting [ 5 , 6 , 7 ]. The aforementioned studies did not succeed in enhancing the forecasting accuracy through improvements to the LSTM structure.…”