OVERVIEW
Allocation of power (to meet real time demand) between traditional power and alternate power has become a challenge worldwide. The aim of this project was to predict how much power is available from alternative energy based on geo location, temperature etc. This helped the client to make better decisions on utilization of other sources of energy. The forecast weather data was collected from external sites using Robotic Process Automation (RPA) and used by the Minsky generated models to predict realtime solar power generation.
Key Benefits (Minsky)
User-Friendly, cloud-based AI platform
No coding skills are required for results or predictions.
Provides you a list of % dependency features that can be used to optimize your business
Ability to fine-tune or optimize the models by trying different algorithms/prediction attributes
Easy integration with other third-party solutions such as TABLEAU for data visualization
Results
Better planning to allocate power from available sources of energy.
Efficiency in maintenance
Increase in ROI
Reduced power cuts or last minute re-allocations
Balance between demand and supply of energy
Executive Summary
The client is a large Power Company that generates and distributes power from traditional as well as alternative sources such as Solar, Wind, etc. The main challenge for them was how to balance Power scheduling / allocation between traditional energy and alternative energy. This was due to the daily unpredictable variations from their solar Power generation. This required accurate solar power generation prediction based on geographical parameters like longitude, latitude, weather, sun intensity, cloud cover, relative humidity etc. After a detailed evaluation of their operations, AI Labs (www.ailabsinc.com) used its proprietary Minsky AI Engine to optimize the models by using a combination of AI algorithms and prediction attributes. In this case, Minsky used historical power generation data along with other weather related attributes for model creation. Weather forecast data was then used in conjunction with the Minsky models to predict Solar Power Generation for future dates. This solution was optimized and implemented in less than a week.
Typical Alternative Energy Forecast Challenges:
Getting Accurate Long term Weather forecasts
Getting Accurate Sun Intensity and Solar angle forecasts.
Variations of solar generations based on manufacturer/ technology
Variations in AI Models based on Geo locations.
Interruption in daily activities.
Solution
After comprehensively evaluating the client‘s challenges and data, we used Minsky to accurately model historical data of power generation. This process uses geographical parameters along with historical power generations, weather data such as relative humidity, sun intensity, temperature, cloud cover, etc to generate AI Models. Once the AI Models were generated by Minsky for the selected Algorithms, Solar Generation prediction data was produced using weather Forecast for that Geo location.. The forecast weather data was collected using Robotic Process Automation (RPA) from external sites. Prediction data from Minsky was also integrated with 3rd Party data visualization application like Tableau. In addition, RPA was also used to email power allocation forecast report generated by Minsky to desired personnel.