Adopting IoT-Based machine learning for Smart Municipal solid Waste Management and Municipal solid Plastic Waste Treatment via pyrolysis

Project Title

Adopting IoT-Based machine learning for Smart Municipal solid Waste Management and Municipal solid Plastic Waste Treatment via pyrolysis

Project Area of Specialization Mechanical EngineeringProject Summary

Recycling (Waste Treatment) is vital for a sustainable and clean environment. Solid waste management and its recycling issues are issues of high concern around the globe. Pakistan is one of the developing countries facing the issue of solid waste management and recycling problems.

Pakistan is also facing an energy shortage and needs to supply clean and cheap energy on an ongoing basis. Compared to its rapid urbanization, Pakistan also experiences massive energy shortages and, as a result, residents use conventional methods to adjust the need for energy and strive to achieve their respective goals.

Renewable energy is a resourceful tool for solving energy crises in Pakistan and resolving energy challenges under natural conditions. The most beneficial feature of renewable energy is that it is natural that these sources are replenished over time. We can name many natural sources, such as sunlight, wind, rain, tides, waves, and geothermal heat. Energy extraction is not only limited to the sources listed, but it can also be extracted from synthetic products (e. g: Plastic).

This is an axiomatic truth that rapid urbanization is the primary cause of the rise in urban waste (municipal, commercial, industrial, construction and demolition waste), worldwide. Waste used to be any country's least desired product until the beginning of the Cultural Revolution. Researchers have developed methods for producing energy using the least preferred material.

Proper management and classification of waste is a safe way to distinguish waste from recycled materials. We propose solid waste management adopting IoT-Based machine learning in this work and especially on waste treatment (via pyrolysis) of non-degradable material (i.e. municipal solid plastic waste). To enhance the accuracy of the classification for optimized management, various methods defined in the literature will be used, such as data augmentation and hyper-parameter tuning.

Project Objectives

The objectives of this thesis project are:

1) To propose solid waste management classification algorithm adopting IoT-Based machine learning to

• Identify and classify recyclables
• Optimize waste collection
• Improve sorting efficiency
• Reduce human drudgery and drive a waste data revolution.

The method of building an IoT-Based machine-learning model will be divided into three stages: 1) data collection, 2) modeling, and model 3) validation. This data will be compared with the experimental data.

2) Municipal solid plastic waste (MSpW) treatment

• via pyrolysis

Project Implementation Method

This work will be carried out in three stages to design a plastic treatment reactor and waste segregation machine.

1) Related literature review,

Literature review on various ways of

i) Smart municipal solid waste (MSW) management and its classification using

• Conventional techniques and
• IoT-Based machine learning, and

ii) Municipal solid Plastic Waste (MSpW) Treatment via various treatment techniques with the main focus on pyrolysis.

2) Waste segregation,

In this stage of the project, we will sort at least three types of different wastes i.e. Plastic, Metal, and Glass. The waste segregation will be implemented as follows:

• 1st phase: We will work on all the desired components and hardware to design the prototype and system
• 2nd phase: Development of the algorithm
• 3rd phase: We will perform tests on the prototype in a real environment to perform different operations
• 4th phase: Finally, we will work on conclusions and will present our results.

3) Plastic treatment,

There are two types of waste in nature:

i) Degradable waste

ii) Non-degradable waste

We will utilize the non-degradable waste which is plastic and convert it into energy that can be used further. There are many methods of incineration but we will be focusing on the pyrolysis process to break down the solid plastic chains into hydro-carbon fumes and then collect the fumes via condensing processes to collect syn-oil (crude oil).

Benefits of the Project

In urban environments, waste management and its classification is a regular activity requiring a significant amount of labor resources and impacting natural, budgetary, productivity, and social aspects. Several approaches to optimize waste management or its classification have been developed, such as using the nearest neighbor search, colony optimization, genetic algorithm, and particle swarm optimization methods. The findings, however, are still too ambiguous and cannot be implemented in actual environments.

We will focus on development to combine optimal techniques for waste management, its classification, and recycling with low-cost IoT architectures and its treatment to produce useful energy.

Expected Outcome

Finally, this work will focus to design low cost, ease of use, and replaceability system that saves time by finding the best route in the management, classification, and recycling of waste.

• IoT-Based machine learning techniques is a modern and innovative technology to potentially replace the conventional MSW management techniques globally to replace human interaction, which will change human life in a positive way. Further, MSW classification by ML and its treatment via pyrolysis is an interesting option for reducing greenhouse gases (GHG) emissions and local air pollutants as well as energy dependence, without sacrificing performance.

Value of the Research Work

• Adopting IoT-Based ML for smart MSW Management and MSpW treatment via pyrolysis technique has the potential to contribute to an increasingly sustainable Pakistan’s energy sector. Smart MSpW management and its treatment may, depending on how electricity is generated, cut greenhouse gas emissions while reducing Pakistan’s exposure to rising whole prices.
• Achieving to reduce landfill site area in urban areas, greater fuel economy, and lower emission through IoT-Based ML for smart MSW Management and MSpW treatment.
• Improving the management, classification, and treatment experience for MSW management and MSpW by implementing an IoT-Based ML and treatment with the following benefits are expected.

1. Reduces waste volume: It will help to make the environment clean and the area required for the installation is small
2. Reduced emissions due to smarter scheduling of garbage collection vehicles: It will reduce hazardous gases generated by vehicles and burning of open waste hence will be good for human health
3. Generates hydro-carbon flammable contents which can be used as fuel
4. No missed pickups: Waste segregation will help to sort out waste and collect them accordingly
5. Reduction in collection cost: Segregation would be done automatically due to which large labor won’t be required.
6. Reduces unnecessary fuel consumption
7. Reduced overflows.

Technical Details of Final Deliverable

1) The traditional waste management system today cannot cope with the tons of trash generated every day.

• At the completion of this thesis project using IoT-Based Machine Learning for smart, urban waste (municipal, commercial, industrial, construction, and demolition waste) can be managed and recycled through automated disposal or sorting process, which is supposed to be a safer recycling system for waste disposal.

2) A prototype model will be developed for energy production via pyrolysis of the MSpW.

• The worldwide production of plastics has also increased with little focus on recycling. Around 80 percent of plastics are either disposed of in landfills or burn up in the open atmosphere, creating a large amount of waste.
• Pyrolysis has been explored as an appealing alternative to municipal solid waste (MSW) disposal incineration that allows recovery of energy and resources; however, it has rarely been used as end products independently of the production of pyrolysis products. Municipal solid plastic waste (MSpW) Pyrolysis treatment would have more environmental benefits, such as reducing emissions of GHGs and fossil fuel consumption.
• In this study, the effect on pyrolysis behaviors and products of important operating parameters such as final temperature, heating rate (HR), and residence time in the reaction zone will be examined first; then the pyrolysis technologies and reactors available in the literature and scale-up plants will be reviewed for optimized assessment. In the third instance, the yields and key properties of pyrolytic products will be summarized from the individual components of MSW, MSW-based refuse-derived fuel (RDF), and MSW. In addition to emissions from pyrolysis processes such as HCl, SO2, and NH3, the fourth phase will observe the pollutants in the products and will discuss available steps to enhance the environmental effect of pyrolysis. It can be assumed that a successful waste-to-energy converter is the single pyrolysis mechanism but is not a guaranteed clean solution for the disposal of MSW. The prospects of applying pyrolysis technologies to deal with MSW will be assessed and suggested based on this knowledge.

Final Deliverable of the Project HW/SW integrated systemCore Industry Energy Other Industries IT , Petroleum Core Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT), Clean TechSustainable Development Goals Good Health and Well-Being for People, Affordable and Clean Energy, Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Climate ActionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	79980
Arduino(uno)	Equipment	1	3000	3000
Motor driver module	Equipment	3	400	1200
Pump	Equipment	1	5000	5000
Cylinder	Equipment	1	3000	3000
Arduino (NANO KIT)	Equipment	1	1600	1600
Stove	Equipment	1	5000	5000
Steel Pipes	Equipment	10	85	850
Aluminum Pipes	Equipment	10	650	6500
Copper Pipes	Equipment	8	700	5600
Reactor	Equipment	1	8000	8000
Hosepipe	Equipment	10	70	700
Inductive sensor	Equipment	1	5230	5230
Capacitive sensor	Equipment	1	13300	13300
Electric Cable	Equipment	1	1000	1000
Machine case	Equipment	1	3000	3000
Industrial Thermometer	Equipment	1	7000	7000
Stationary, printing	Miscellaneous	1	10000	10000