Optimization of a photovoltaic/wind/battery energy-based microgrid in distribution network using machine learning and fuzzy multi-objective improved Kepler optimizer algorithms
2024

Optimizing a Hybrid Microgrid with Machine Learning

publication 10 minutes Evidence: moderate

Author Information

Author(s): Duan Fude, Eslami Mahdiyeh, Khajehzadeh Mohammad, Basem Ali, Jasim Dheyaa J., Palani Sivaprakasam

Hypothesis

Can a fuzzy multi-objective framework optimize a hybrid microgrid to minimize energy losses, voltage oscillations, and power costs?

Conclusion

The study found that using forecasted data improved the optimization of a hybrid microgrid, but also led to increased annual energy losses and costs compared to real data.

Supporting Evidence

  • The optimization reduced annual energy losses from 666,052 kW to 423,227 kW.
  • Voltage deviations decreased from 0.0147 p.u to 0.0129 p.u.
  • The cost of power purchased from the hybrid microgrid was reduced to $340,630.

Takeaway

This study shows how to make a power grid that uses solar and wind energy work better by using smart computer programs to predict energy needs.

Methodology

The study used a fuzzy multi-objective improved Kepler optimization algorithm (MOIKOA) and a multilayer perceptron artificial neural network (MLP-ANN) for forecasting.

Limitations

The study did not account for all possible uncertainties in renewable energy generation.

Digital Object Identifier (DOI)

10.1038/s41598-024-64234-x

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication