| Work Detail |
An international group of scientists has tested several machine learning algorithms to predict the optimal tilt angle (OTA) of solar projects in 37 Indian cities, leading to improvements of up to 90%.
An international group of researchers has used feature selection-based artificial neural networks (ANNs) with various machine learning algorithms to predict the OTA of PV systems. They conducted their experiments at 37 locations in India and based on the ANN evaluated, the accuracy improvements ranged from 38.59% to 90.72%.
“The OTA of solar panels is one of the crucial variables that determine the installation efficiency and operation of photovoltaic systems,” the group says. “An OTA allows the sun’s rays to be absorbed by the material to the greatest extent possible. When capturing solar energy, photovoltaic panels are affected by the angle at which light hits them from different directions. As a result, the annual energy production of the system is directly affected by the selection of the appropriate tilt angle.”
The experiment was based on data extracted from NASAs Power Data Access Viewer website and included the following parameters: global solar radiation (SR), diffuse SR, extraterrestrial SR, global SR on a tilted surface, and clarity index. They extracted data from 37 Indian cities, including New Delhi, Mumbai, Bangalore, and Kolkata.
“In machine learning and data analytics, feature selection is a crucial step that involves determining which variables and predictors in a dataset are most significant and contribute to the predictive potential of a model,” the academic team explains. “Selecting relevant features minimizes overfitting, increases interpretability, and increases model accuracy.” A list of feature selection methods is provided in the following subsections.”
The team used feature selection techniques such as the Pearson correlation coefficient to assess the robustness of data associations and the signal-to-noise ratio to simplify the process under noisy conditions. After applying these methods, they opted not to include extraterrestrial SR in the ANN prediction models.
“Global SR, diffuse SR, and monthly mean clarity index inputs showed a stronger negative relationship with OTA output compared to extraterrestrial radiation, which showed an insignificant relationship with OTA output,” they noted. “Global SR on tilted surfaces showed a moderate positive relationship. A negative correlation means that as global SR, diffuse SR, and monthly mean clarity index increase, OTA decreases. Extraterrestrial radiation input has a lower signal-to-noise value, implying that its correlation with OTA is low.”
Six ANN algorithms were then tested and their tilt angle predictions were compared to the actual OTA target values. The improvement in prediction accuracy (IPA) was then calculated to show how feature selection had improved the mean square error (MSE) in compression with the MSE of the full parameter calculation.
The lowest IPA was recorded for the scaled conjugate gradient (SCG) with an improvement of 38.59%. This was followed by an IPA of 53.33% for the Levenberg-Marquardt (LM) case and 66.93% for the Polak-Ribiere conjugate gradient (PRCG). The one-step secant (OSS) obtained an IPA of 86.88%, while the Broyden-Fletcher-Goldfarb-Shanno (BFGS) recorded 89.53%. The Elman neural network (ELM) provided the best improvement of 90.72%.
“The models developed in this study are used to optimize energy production, increase efficiency, and make informed judgments on tilt angles of solar panels,” the academics said. “The model helps industry participants achieve better results and promote the use of OTA by predicting at different sites. Future research work can focus on OTA prediction using high-precision measured values ??of solar radiation and incorporating other factors such as dust, pollution, and aerosols. Other region-specific OTA models will be developed and validated under different climatic conditions.”
The authors presented their results in “ Novel feature selection based ANN for optimal solar panels tilt angles prediction in micro grid ,” recently published in Case Studies in Thermal Engineering . The study was conducted by scientists from SR University in India, IIMT University, Government College Hamirpur, COER University, Gachon University in South Korea, and Eötvös Loránd University in Hungary. |
