ICA Research Website

Members

Denise Ang

Michico De Guzman Iya Dytiandu

Clair Llanos Dee

Frances Recto

Contact Information

llanosdeeclair@gmail.com, emvdeguzman@gmail.com

ACKNOWLEDGEMENT

The researchers would like to thank their research adviser, Ms. Guia, who supported and guided them, giving clear-minded advice to overcome challenges that arose throughout the project. Her feedback was vital in improving the quality of this research.

This research introduced an automated alternative and evaluated its efficiency in three categories: accuracy, precision, and the presence of visual and auditory cues. Following PAGASA’s standards, yellow, orange, and red flood warnings were set at 2, 4, and 6 inches respectively. Accuracy was assessed by comparing standard values to those collected by both systems, while precision was determined by measuring the deviation between the values. Activation of the visual and auditory cues was observed upon reaching orange and red flood heights. For all flood warnings, between the two systems, the Automated Robotics System (ARS) had a higher percentage accuracy of 98.26%, 99.22%, and 99.15% respectively, and these values are all higher than the averages for percentage accuracy of PAGASA’s manual staff gauge which are 91.79%, 96.48%, and 95.35% respectively. The ARS also demonstrated increased precision with collected data showing less deviation from the standard values as compared to the manual gauge. Furthermore, the ARS successfully activated visual and auditory cues when the orange and red flood heights were reached. The ARS outperformed the manual gauge in all efficiency tests, proving to be a potential alternative for flood detection and showcasing its ability to enhance flood monitoring and warning systems as a whole.

BASE TEMPLATE AND EDITING:

STEM C TECHNICAL COMMITTEE

INTRODUCTION

NDRRMC and PAGASA are responsible for the current flood detection systems; however, their system contains numerous flaws. PAGASA implemented a manual staff gauge system to monitor flood levels, while NDRRMC alerts the people and sends short message services about upcoming disasters. PAGASA (2008) states that the manual staff gauge system is cheap, but flawed as not only does it require repainting after every flood, but according to Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH (2012), measuring the gauges in the middle of a disaster can potentially endanger people. Despite the much-needed presence of alerts, NDRRMC tends to have a delay in the sending of messages (Zoleta, 2023). With the disadvantages of the current flood detection system, the goal was to create an alternative to the manual staff gauge: the automated robotics flood detection system, or ARS for short. The efficiency of the robotics system was tested through the means of determining the accuracy and precision of the obtained measurements of flood levels and testing the ability of the electronic buzzer and light-emitting diode (LED) of the robot to generate auditory and visual cues.

BASE TEMPLATE AND EDITING:

STEM C TECHNICAL COMMITTEE

METHODOLOGY

Figure 1

figure 3

FIGURE 2

A setup to simulate increasing levels of flood water was constructed as shown in Figure 1. The robotic components of the ARS are attached using the foam board. The ultrasonic sensor mount was attached 8 inches high or directly above the edge of the 8x8x8 glass box and the ultrasonic sensor rested on top of it. A laptop was used as a power source for the robot and ran the conditional code to measure the flood height and enable the presence of auditory and visual cues. A volume of water was dispensed from a height, simulating the yellow, orange, and red flood heights of 2.29, 5.63, and 6.58 inches respectively. The data of the ARS was collected through the Arduino serial monitor and the manual staff gauge was collected through the readings of the assigned researcher. Five trials were performed for each flood level warning and the water was manually drained out after each trial.

BASE TEMPLATE AND EDITING:

STEM C TECHNICAL COMMITTEE

FINDINGS

The percentage accuracy was taken from the raw data considering their type of system, trial, and type of flood warning whether yellow, orange, or red. The Automated Robotics System had a greater percentage accuracy than the manual staff gauge. The average percentage accuracies of the ARS for yellow, orange, and red flood warnings were 98.26%, 99.22%, and 99.15%. These values are higher than the average percentage accuracies of

the manual staff gauge which were 91.79%, 96.48%, and 95.35% respectively. All 15 trials, with 5 trials for each type of flood warning, were used to determine the precision of both ARS and manual gauge systems. It could be seen that the ARS had a higher precision than the manual staff gauge.

BASE TEMPLATE AND EDITING:

STEM C TECHNICAL COMMITTEE

CONCLUSION

From the data collected, it was proven that the ARS was more efficient than the manual gauge water system. The values collected by the ARS were both accurate and precise when compared to the standard flood heights of 2.29, 5.63, and 6.58. Moreover, it presented the ability to execute visual and auditory cues for orange and red rainfall alerts. This can be attributed to the background of using Robotics and Arduino having high levels of automation and accuracy, removing the need to manually observe and interpret data.

Despite the high levels of efficiency, possible errors may have occurred from the design of the physical setup, manual reading from the values from the Arduino serial monitor, and the distance of the ultrasonic sensor to the water level. As the researchers manually transferred water out after each trial, the ultrasonic sensor was removed and placed back, which may have possibly caused errors in measurements. Reading values manually from the Arduino serial monitor may have caused human error as well due to the fast delay time. Lastly, the distance of the ultrasonic

sensor might have caused slight errors as at times, it can have an error margin of three millimeters.

To alleviate possible sources of error and improve the scope of the ARS, several recommendations such as non-technical and technical elements could be applied to further improve the study. For non-technical recommendations, the addition of more trials, the use of a water pump for the transfer of water, and the use of a scale to measure the volume of water for every individual trial would contribute to further testing the systems. As for technical recommendations, the addition of databases, SMS, and additional parameters of the rate of increase of water level aside from flood height can be applied. With these, the automated robotics system’s base prototype would be upgraded to increase its range of operation, aiding more people within the community.

BASE TEMPLATE AND EDITING:

STEM C TECHNICAL COMMITTEE

REFERENCES

Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH. (2012, October). Local

Flood Early Warning System. PreventionWeb.

https://www.preventionweb.net/files/29230_29230gizphilippineslocalfloodearlyw.pdf

Indrasari, W., Iswanto, B. H., & Andayani, M. (2018). Early Warning System of Flood Disaster

Based on Ultrasonic Sensors and Wireless Technology.

https://iopscience.iop.org/article/10.1088/1757-899X/335/1/012005/pdf

Lim, M. B. B., Lim, H., & Piantanakulchai, M. (2013, December). (PDF) Factors Affecting Flood

Evacuation Decision and Its Implication to Transportation Planning. ResearchGate.

https://www.researchgate.net/publication/259562947_Factors_Affecting_Flood_Evacuation

_Decision_and_Its_Implication_to_Transportation_Planning

Official Gazette of the Republic of the Philippines. (n.d.). How to make sense of PAGASA's

color-coded rainfall advisories | GOVPH. Official Gazette of the Republic of the

Philippines.

https://www.officialgazette.gov.ph/how-to-make-sense-of-pagasas-color-coded-warning-sig

nals/

PAGASA. (2008, December). General Guidelines For Setting-Up a Community-Based Flood

Forecasting and Warning System (CBFFWS). Typhoon Committee.

https://typhooncommittee.org/docs/publications/WMO_TD1472_2008.pdf

Rahayu, F., Zuchriadi, A., Fauzi, A., & Dewantara, A. (2022). Prototype Flood Detection Water

Level Monitoring IoT Web Based With Ultrasonic Sensor HC-SR04. Jurnal Mantik, 6(2),

2006-2014.

Susila, M. T. A., Firdaus, I. N., Chuzairi, M. F., & Rahmawati, D. (2023, September). Flood Early

Warning System Prototype Based on Ultrasonic Sensor and Internet of Things. In (Vol. 19, No.

3, pp. 100–106). Jurnal Rekayasa Elektrika. 10.17529/jre.v19i3.33147

Zoleta, V. (2023, June 22). List of Flood-Prone Areas in the Philippines [Updated 2023].

Moneymax.

https://www.moneymax.ph/car-insurance/articles/flood-prone-areas-metro-manila