DWME: Don’t “waste” my energy

For far too long, Lebanon has suffered from the absence of necessary infrastructure which prevented its development specifically in the electricity and the waste management sectors. Sustainable low-cost solutions are crucially needed in times of economic crisis. Such solutions will allow for the development of multiple sectors in the Lebanese community providing the necessary elements for an economic rebound and recovery. In this project, we describe the design, and implementation ofa waste sorter based on a Convolutional Neural Network (CNN) which allows the separation of trashin such a way where the recyclable elements can be properly handled and the non recyclables are managed in the second part of our project. The latter part of our projectdeals with generating energy from non recyclable wastes. This is due to the waste-to-energy (WTE) technique in which we burn these types of waste,and the resulting temperature is channeled to one part of a temperature-to-energy device, the thermoelectric converter. Our scalable project is also complemented by a web application which provides a platform on further monitoring and enhancing the machine learning model whereby we can re-annotate the detected objects from the camera and provide them back for model retraining. The report describes the system’s overall architecture including both hardware and software components along with relevant analysis highlighting the effectiveness of our products.

Please click here to view the video of the project demonstration.

Project Details

[photo]