![](https://static.wixstatic.com/media/c5a8e0_8bd01b21a30749bba46afd90067c68c7f000.jpg/v1/fill/w_1920,h_1080,al_c,q_90,enc_avif,quality_auto/c5a8e0_8bd01b21a30749bba46afd90067c68c7f000.jpg)
Solar energy optimization and production
![example of solar energy optimization](https://static.wixstatic.com/media/c5a8e0_8d2ba6ffc21c42b394fdc8d562396c71~mv2.jpg/v1/fill/w_905,h_432,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/solarimg_edited.jpg)
![clock](https://static.wixstatic.com/media/c5a8e0_4c758907cf444c1682866edef6e90b77~mv2.png/v1/fill/w_78,h_78,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/c5a8e0_4c758907cf444c1682866edef6e90b77~mv2.png)
3' read
Problem
Challenges to demand plan for solar energy consumption due to anomalous and extreme weather conditions leading to blackouts, grid instability and inefficient energy storage optimization.
Solution
-
Collected large historical data sample and built live feed to Solar energy predictors such as temperature, humidity, solar irradiance.
-
Generative AI models identified additional features to account for emerging weather patterns and randomness.
-
Used Ensemble methods via SVM and Artificial Neural Networks to have multiple model outputs, handle randomness in data, outliers and noisy data.
-
Deployed on existing AWS Cloud seamlessly integrating into existing infrastructure, centralizing management / monitoring and connecting with existing processes.
Impact
-
Increased energy optimization: enabled to take advantage of high periods
-
Reduced reliance on non-renewable resources in low period
-
Reduction of blackouts & disruptions across grid
-
Demonstrated investment case for expansion of solar energy infrastructure