Hybrid prediction method for solar photovoltaic power generation using normal cloud parrot optimization algorithm integrated with extreme learning machine | JoVE Visualize

Area of Science:

Renewable Energy Systems
Artificial Intelligence in Energy
Meteorological Forecasting

Background:

The increasing demand for sustainable energy highlights the critical role of renewable sources like photovoltaics (PV).
Seasonal variability and intermittent nature of solar power present significant challenges for accurate energy forecasting and grid stability.
Existing forecasting methods often struggle to capture complex spatial and temporal dependencies in meteorological data.

Purpose of the Study:

To propose a novel hybrid forecasting method, NCPO-ELM, for enhanced prediction of renewable energy generation.
To develop an optimization algorithm, Normal Cloud Parrot Optimization (NCPO), for improving Extreme Learning Machine (ELM) performance.
To address the limitations of standard ELM in handling noisy and unstable data due to random initialization.

Main Methods:

Development of the Normal Cloud Parrot Optimization (NCPO) algorithm, inspired by parrot flocking behavior and cloud model theory, featuring five search strategies and a random structure.
Integration of the normal cloud model within NCPO to generate specific random samples, improving exploration of the solution space.
Optimization of Extreme Learning Machine (ELM) hyperparameters, including Single-Layer Feedforward Network (SLFN) hidden layer weights and biases, using the developed NCPO algorithm.

Main Results:

The proposed NCPO-ELM method demonstrated superior prediction accuracy and performance compared to existing hybrid forecasting techniques.
The method effectively captured spatial and temporal dependencies within meteorological data for more reliable PV power forecasting.
NCPO-ELM showed improved robustness against noise and instability, outperforming benchmark ELM models.

Conclusions:

The NCPO-ELM hybrid forecasting method offers a significant advancement in predicting renewable energy output, particularly for photovoltaic power.
The novel NCPO algorithm effectively optimizes ELM, leading to enhanced accuracy and stability in energy forecasting.
This approach is well-suited for time series data with diverse characteristics and seasonal variations, contributing to more reliable grid integration of renewable energy sources.