Track: Environmental Engineering

Abstract

Waste management and practices is a pervasive world problem due to the continuous rise in urbanization which comes along with a rise in waste generation. Even though proper waste management has a vital role to play in the ecological environment by greening through the recovery of energy from waste, its management is a menace. Energy from waste refers to the conversion of non-recyclable waste materials into usable ones in the form of electricity, heat and fuel by employing different processes such as, incineration, anaerobic digestion, gasification, pyrolization, and landfill gas recovery often referred to as waste to energy.

Reports in Ghana indicate that about 5 million tons of Municipal Solid Waste (MSW) is generated annually and about 60% is organic. Out of this, the non-recyclable components constitutes about 20%, which indicates that 80% can be recovered and recycled technically. Further, about 25% of the organic waste received at the material recovery and compost facility remains as compost for used in agricultural and other purposes.

Considering the population of Ghana pegged at 30 million in 2019, and daily solid waste production of about 0.45 kg per person (Amoah, 2006). Proper management and greening of MSW is very much essential with increasing demand of energy and that is what this paper seeks to tackle.

This paper mainly emphases on analyzing and classifying (segregating) solid waste using convolutional neural networks to productively process solid waste materials to enhance the separation process of converting waste to energy. Compared to other classification algorithms, convolutional neural networks use minimal pre-processing, meaning the network learns the filters that typically are hand engineered in other systems. Deep Machine Learning (DML) is a technique for processing large quantities of data since the performance of the machine improves as it analyses more data. As the amount of data increases, the machine becomes more adept at recognizing even hidden patterns among the data. Also, the potentials and prospects of organic waste to energy is exploited to reveal the technologies, socio-economic benefits as well as the challenges of implementing waste to energy plants in Ghana.

Thus, this project discusses the design and construction of a model that recognizes solid waste materials and segregates using Raspberry Pi Board, a camera, LEDs, an LCD screen and a buzzer as major components. Results indicate that, the system can effectively segregate solid waste that is recyclable and can be converted to energy. Feasibility studies of waste to energy also, indicates that, combustion and anaerobic process of conversion is mostly applied in Ghana and has improved on the greening and advocacy for clean environment. Again, the prospects of waste to energy was analysed by using smartPLS and results indicates that,  jobs, socio-economic, tourism , environmental cleanliness and reduction of communicable diseases are the benefits of installation of waste to energy plants in Ghana.