Open Access Open Access  Restricted Access Subscription Access

Adaptive AI-Based Control for Smart Grids with High Renewable Penetration

V. Rethash Dev Reddy

Abstract


Smart grids aren’t what they used to be. Not even close. Solar panels here, wind farms there, batteries everywhere—it’s kind of chaos, honestly.

The old systems? They were simple. Predictable. You set the rules and mostly forgot about them. That doesn’t work anymore. Not when energy supply keeps changing every hour, sometimes every minute.

Demand shifts. Prices jump. And still, the grid has to hold steady—voltage, frequency, costs. No excuses.

So this paper steps in with an AI-based approach. Not fancy for the sake of it, but necessary. The system pulls data from everywhere—grid sensors, weather feeds, market prices, battery levels. It cleans it,

reshapes it, learns from it. Then it predicts what’s coming next and adjusts schedules in real-time. No waiting around. It reacts.

We tried multiple models. Some basic, like linear regression. Others more complex—random forests, support vector machines, and LSTMs. And yeah, they all did something. But the sequence-aware ones? They stood out. LSTM especially. Lower errors. Better predictions. Just more reliable overall. But it’s not just about predicting numbers. That would be too easy.

The system actually decides—when to charge batteries, when to reduce renewable curtailment, how to balance imports. Real decisions, not just theory.

And the results? Pretty solid. Renewable usage goes up. Sudden fluctuations drop. Scheduling becomes… stable, finally.

In the end, it’s clear. You can’t manage modern grids with old thinking. You need systems that learn, adapt, and move fast. This one does that. Mostly.


Full Text:

PDF

References


M. Khalid et al., “Smart grids and renewable energy systems,” Renewable and Sustainable Energy Reviews, vol. 186, 2024.

P. Boopathy et al., “Deep learning for intelligent demand response and smart grids: A survey,” Renewable and Sustainable Energy Reviews, vol. 2024.

D. Aguilar et al., “Optimal scheduling of renewable energy microgrids,” Applied Energy, 2024.

T. Liang et al., “Deep reinforcement learning-based optimal scheduling of microgrids,” 2024.

V. Pramila et al., “Integrating hybrid intelligent algorithms for microgrid optimization,” 2024.

L. Geng et al., “A swarm intelligence and deep learning strategy for wind power and energy storage scheduling in smart grid,” 2024.

S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8,

pp. 1735–1780, 1997.

L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001.

V. Vapnik, “The nature of statistical learning theory,” Springer, 1995.

Goodfellow, Y. Bengio, and A. Courville, “Deep Learning,” MIT Press, 2016.


Refbacks

  • There are currently no refbacks.