Deep Learning Enhanced Solar Energy Forecasting with AI‐Driven IoT

Abstract

Short-term photovoltaic (PV) energy generation forecasting models are important, stabilizing the power integration between the PV and the smart grid for artificial intelligence- (AI-) driven internet of things (IoT) modeling of smart cities. With the recent development of AI and IoT technologies, it is possible for deep learning techniques to achieve more accurate energy generation forecasting results for the PV systems. Difficulties exist for the traditional PV energy generation forecasting method considering external feature variables, such as the seasonality. In this study, we propose a hybrid deep learning method that combines the clustering techniques, convolutional neural network (CNN), long short-term memory (LSTM), and attention mechanism with the wireless sensor network to overcome the existing difficulties of the PV energy generation forecasting problem. The overall proposed method is divided into three stages, namely, clustering, training, and forecasting. In the clustering stage, correlation analysis and self-organizing mapping are employed to select the highest relevant factors in historical data. In the training stage, a convolutional neural network, long short-term memory neural network, and attention mechanism are combined to construct a hybrid deep learning model to perform the forecasting task. In the testing stage, the most appropriate training model is selected based on the month of the testing data. The experimental results showed significantly higher prediction accuracy rates for all time intervals compared to existing methods, including traditional artificial neural networks, long short-term memory neural networks, and an algorithm combining long short-term memory neural network and attention mechanism.

Wind, Solar, and Photovoltaic Renewable Energy Systems with and without Energy Storage Optimization: A Survey of Advanced Machine Learning and Deep Learning Techniques

“…A hybrid deep learning system is proposed by integrating clustering algorithms, convolution neural networks, long short-term memory, and attention mechanisms with a wireless sensor network to tackle the current PV electrical generation estimation problems. Clustering, training, and forecasting are the three steps of the overall suggested strategy [73]. In contrast to previous methods, such as computational models, long-short attention span deep learning, and an algorithm combining a long defeatist mentality neural network and an attention mechanism, the experimental results indicated massively better prediction test accuracy for all frequency ranges.…”

Review on maintenance of photovoltaic systems based on deep learning and internet of things

“…From a technical perspective, almost all of the research papers used the widely-used CNN or LTSM algorithms [55], [62], [64], [66]. Besides some developed their own variants of the CNN or LTSM models [54], [55], [58]- [60], and the rest of them worked on the traditional ANN and MLP [56], [57], [63], and one paper worked with a GAN variant named it "DA-DCGAN" [65].…”

“…From a technical perspective, almost all of the research papers used the widely-used CNN or LTSM algorithms [55], [62], [64], [66]. Besides some developed their own variants of the CNN or LTSM models [54], [55], [58]- [60], and the rest of them worked on the traditional ANN and MLP [56], [57], [63], and one paper worked with a GAN variant named it "DA-DCGAN" [65]. The authors applied classification classes ranging from 2 (binary anomaly detection) [56], [57], [64]- [66] through to 3 for the [54] (multiclass classification), other authors used regression methods to predict the output power of the solar plant [55], [58]- [60], [62], [63].…”

“…In accordance with the main objectives of these research papers (the maintenance of photovoltaic solar panels), some are attempting to build a model able to detect any default occurring while the solar plant is running in order to prevent any breakdown. The first thing that stands out is that the majority of the papers are dealing with the power prediction [55], [58]- [60], [62], [63]. The connection between the system output prediction and his maintenance is not immediately apparent at first glance, but in fact, PV maintenance can be effectively aided by forecasting power generation: it is considered as a reference for alert thresholds, and more important is the stability of the electrical network, when our plant is ongrid.…”

Machine Learning Approach on Time Series for PV-Solar Energy

“…e photovoltaic (PV) panel is the primary component in the solar energy system responsible for generating electricity [4][5][6]. Solar energy will lead to a large reduction in the amount of energy saved.…”