# Comparative Analysis of Machine Learning Algorithms for Solar Irradiance Forecasting in Smart Grids

arXiv: https://arxiv.org/abs/2310.13791 · HTML: https://mlsanity.com/paper/2310.13791/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 31 references that are in the map.
- First posted on arXiv 20 Oct 2023.

## Builds on (most cited first, 5 of 5)

- [Normalization Techniques in Training DNNs: Methodology, Analysis and Application](https://mlsanity.com/paper/2009.12836/) (2020)
- [Optimization of Residential Demand Response Program Cost with Consideration for Occupants Thermal Comfort and Privacy](https://mlsanity.com/paper/2305.08077/) (2023)
- [Sizing and Allocation of Distributed Energy Resources for Loss Reduction using Heuristic Algorithms](https://mlsanity.com/paper/1911.07443/) (2019)
- [Dead-time Compensation Method for Bus-clamping Modulated Voltage Source Inverter](https://mlsanity.com/paper/2307.11868/) (2023)
- [A High Frequency Active Clamp Forward Converter with Coreless Transformer](https://mlsanity.com/paper/2307.12804/) (2023)

## Abstract (arXiv)

The increasing global demand for clean and environmentally friendly energy resources has caused increased interest in harnessing solar power through photovoltaic (PV) systems for smart grids and homes. However, the inherent unpredictability of PV generation poses problems associated with smart grid planning and management, energy trading and market participation, demand response, reliability, etc. Therefore, solar irradiance forecasting is essential for optimizing PV system utilization. This study proposes the next-generation machine learning algorithms such as random forests, Extreme Gradient Boosting (XGBoost), Light Gradient Boosted Machine (lightGBM) ensemble, CatBoost, and Multilayer Perceptron Artificial Neural Networks (MLP-ANNs) to forecast solar irradiance. Besides, Bayesian optimization is applied to hyperparameter tuning. Unlike tree-based ensemble algorithms that select the features intrinsically, MLP-ANN needs feature selection as a separate step. The simulation results indicate that the performance of the MLP-ANNs improves when feature selection is applied. Besides, the random forest outperforms the other learning algorithms.