Team Hawks

https://drive.google.com/drive/folders/1iI9GYXS2kYsuCu3_myHSzVpwj4iGre3b?usp=sharing

https://github.com/kd100100/NASA-Space-Apps-Challenge-2021

Hexolar is an application that is developed by Team Hawks on the topic You Are My Sunshine. Our project has 4 different features on which we have done various analysis in order to get the desired information.

1. Check Solar Panel Feasibility:-

This feature helps the user to automatically detect their location by enabling his/her permission.

The user can also manually enter the location using their latitude/longitude.

Once the location is entered the calculation process takes place.

The user gets to know whether their location is feasible or not to install Solar Panel. 

2. Solar Irradiance Visualizer:-

The solar irradiance visualizer shows a interactive 3D earth globe which shows the annual solar irradiance on various parts of the globe.

3. Solar Panel Power Predictor:-

This feature helps the user to automatically detect their location by enabling his/her permission.

The user can also manually enter the location using their latitude/longitude. Once location is given they are asked about the wattage rating of their panel.

If selected Yes, the wattage rating is entered and the calculations are done in order to calculate the power generated by the solar panel in the future. The user can check the power output on Daily/Monthly/Yearly basis.

If selected No, then they have to enter the area of panel and the efficiency of the panel. Once it’s entered, the calculations are done in a similar way.

4. Compare Solar Panel Efficiency:-

This feature helps the user to automatically detect their location by enabling his/her permission.

The user can also manually enter the location using their latitude/longitude.

Once location is captured then the output power of the panels are entered on yearly/monthly basis. Once the power output is entered, the user’s power rating are compared with the NASA’s data. So that the users can check the efficiency of solar panel by comparing with NASA’s data. 

An additional features is used to perform 1 year analysis in order to see the performance of user’s solar panel. Once the power output is entered for analysis, a graphical representation is given to the user to verify the performance of their panel when compared with that of NASA’s data.

In our project, the frontend of the application is done using ReactJS and the backend runs on Flask. For frontend and backend, the software used was visual studio code and for Machine learning, we used Google collab and Jupyter notebook. The ML model we used to predict Solar irradiance and temperature data is Random Forest Regressor. 

Github Link:- https://github.com/kd100100/NASA-Space-Apps-Challenge-2021

PPT Link:- https://docs.google.com/presentation/d/1I9IAYzVxoGKFhAv0RVfS8HozsTyBmp3DRH7ntCrLFM0/edit?usp=sharing

30 Second Demo Link:- https://youtu.be/FJ5ijRl5q9g

Full Explanation Video Link:- https://youtu.be/76lYjs9cOJs

We used two datasets from the NASA power web portal.

Link:- https://power.larc.nasa.gov/data-access-viewer/

1) All Sky Surface Shortwave Downward Irradiance

2) Temperature at 2 meters

For our project, the datasets are to be dynamic since the data changes according to the user-specified location. We utilized the power portal's API to fetch data and use the same for training the ML model.

The space apps challenge made us push the boundaries of our knowledge and skills in the field of Solar Energy and Web development. The hackathon made us develop teamwork though we are present virtually. The need for energy has been increasing day by day and generating Renewable energy has been the ultimate aim for all. As solar power can be installed almost in every house, analyzing the flexibility of the system is a crucial part. The way we have made our project is to give a user-friendly interface to analyze the solar power at our location, and much more. The main idea is to make the installation of Solar at a place more effective. We have used data from NASA’s database and used it to train an ML model to predict the temperature and Irradiance at any given coordinates.

1) Dataset:- https://power.larc.nasa.gov/data-access-viewer/

2) Highcharts:- https://www.highcharts.com/demo

3)Three.js:- https://threejs.org/

4)Google Map API:- https://cloud.google.com/maps-platform

5) ML:- https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html

#solarpower, #irradiance, #sunshine, #NASA's power, #analytics, #ML, #DataScience, #Electrical, #solarpanel

This project has been submitted for consideration during the Judging process.