CHAPTER 3
METHODOLOGY
I. Materials:
Arduino Uno Optical sensor
Ultrasonic sensor Buck converter
Infrared sensor Inductive sensor
Stepper motor 10V water turbine
Breadboard or PCB Jumper wires
Conveyor belt or chute for the waste PVC end caps (2)
Different coloured bins for different types of waste PVC elbow joints (2)
PVC pipe (2" diameter, 5 feet long) Copper coil wire (14 gauge, 30 feet)
PVC T-joints (2) 1/8" drill bit
PVC reducer bushings (2) Blue LED light
Red LED light Yellow LED light
Rectifier bridge (1) 1/4" drill bit
Silicone caulk Waterproof wire connectors
II. Design
The prototype design was specifically made to be user friendly. The
collector contains an open/close mechanism which opens once the desired
rotation of the stepper motor is achieved. The prototype contains an infrared
sensor which detects whether waste is present or not. It also contains an
inductive sensor to detect metal wastes. Plastic wastes will be detected by the
built in ultrasonic sensor. There are 3 different colored LED lights found in the
breadboard that corresponds to different wastes. Each of the lights light up
depending on what type of waste is detected by the machine. The prototype
contains 3 plastic bins. 1 bin for metal, 1 bin for plastic and 1 bin for
biodegradable. Its size is compact and relatively small so that it would not
take up too much space. The segregator will be controlled using an Arduino
microcontroller. The segregator will be connected to a 10V hydropower
generator made by the researcher.
.
III. Procedure
A. Hydroelectricity Generator
The procedure starts by creating a scaled drawing plan in computer-
aided design (CAD) to ensure an accurate scaling of the prototype. This
provides the correct amount and sizes of the materials to avoid wastage. The
researcher starts by setting up the water source which consists of the
reservoir, return tank and water pump. After that, the researcher sets up the
turbine and installs it to the water source. Once the turbine is installed, it will
be connected to the generator to convert the mechanical energy into
electricity. Finally, the researcher will wire the generator in order to transfer
the electricity generated by the generator to the automated waste segregator.
B. Automated Waste Segregator
The researcher starts by building the framework of the machine. The
researcher sets up the bins, sensors and motor in place. The researcher then
sets up the control panel that consists of the Arduino Uno microcontroller and
connects the sensors, the motor and the flap for the open/close mechanism.
The researcher programs the Arduino and sets the angle of rotation of the
stepper motor depending on the type of waste detected. The angle of rotation
set by the researcher are the following: biodegradable = 0°, plastic = 90°,
metal = 180°. The researcher then uploads the code to the Arduino software.
Lastly, the researcher connects the output of the hydroelectricity generator to
the control panel of the automatic waste segregator.
IV. Data Gathering
The researcher will be testing the machine using 50 samples for each
waste types. The researcher will collect the value of the number of wastes
correctly identified and segregated accordingly by the machine (TP), the
number of wastes incorrectly identified but segregated accordingly by the
machine (FP), and the number of wastes incorrectly identified and incorrectly
segregated by the machine (FN). The researcher then consults 30 experts to
examine the machine’s functionality, durability, usefulness, impact on society
and safety.
After the data has been collected, the researcher will compute the
precision score of metal, plastic and biodegradable using the formula:
Precision = TP / (TP + FP). The researcher will then compute the recall score
of metal, plastic and biodegradable using the formula: Recall = TP / (TP +
FN). After computing both the precision and recall scores, the researcher will
then compute the F1-score of metal, plastic and biodegradable using the
formula: 2(precision*recall) / (precision + recall). Once computing for those
values is finished, the researcher will compute the machine’s overall
precision, recall and F1-score by finding the average of the value for each
waste types.
To determine whether there are statistically significant differences in
terms of the precision, recall and F1-scores in segregating each type of
wastes, the one-way Analysis of Variance (ANOVA) will be performed. After
this, an analysis of the results will be done.
