| dc.description.abstract | Amidst global research on IoT-based water level monitoring devices, questions persist
about sensor dependability and accuracy, especially in real-world scenarios. The
research main objective was to design, develop and validate the Arduino-based Global
System for Mobile Communication (GSM) prototype that utilizes a JSN-SR04T
waterproof ultrasonic sensor in remote water level monitoring that meets the
requirements of accuracy and reliability. Two theoretical frameworks were used in this
research study to bring out the relationship between the predictor and outcome
variables: The ultrasonic wave reflection principle and the Piezoelectric Effect. The
research method employed in this research was mixed methods research comprising
of explanatory and predictive statistical approaches of gathering and analyzing data to
ascertain whether there was a significant difference between the means of data
obtained from the IoT-based monitoring system and those from the manual gauge.
Linear and Multiple regression were used to analyze the effect of air temperature
variations on ultrasonic sensors under different locations. One (1) Arduino-based
GSM water level monitoring prototype that utilized a JSN-SR04T waterproof sensor
was developed and used to collect sensor datasets. The research study used stratified
sampling techniques and had the following sample sizes: 48 manual gauge readings,
48 sensor readings and 15 tests of GSM efficiency. Data was visualized in the Arduino
serial monitor, the ThingSpeak Server application and in Excel using graphs and
charts. The research study found that the developed and validated Arduino-based GSM
water level monitoring prototype was very accurate and reliable for application in
water reservoir level monitoring. The IoT prototype achieved a water level accuracy
of 98.92 percent in a lab setting environment and an accuracy of 95.083 percent in an
outdoor simulation. The ANOVA test confirms the lack of significance for the location
effect on accuracy with a p-value of 0.09099. While the sensor demonstrates high
accuracy, reaching 100% in certain temperature ranges during outdoor testing, there
are instances of lower accuracy, particularly noticeable at 23.44°C and 42.97°C. The
impact of temperature on accuracy is evident, with fluctuations noted, and a distinct
decrease in accuracy observed at higher temperatures. The research study concludes
that while temperature significantly affects the accuracy of outdoor readings, the
location of the sensor, whether indoors or outdoors, does not significantly influence
accuracy. These findings provide valuable insights for optimizing sensor use in various
environmental conditions. The significance of this research study is that it adds to the
existing knowledge on effective water resource management through the
implementation of IoT water level monitoring systems. | en_US |
