The Design of an Arduino Based Low-Cost Ultrasonic Tide Gauge With the Internet of Things (Iot) System

• Published in: IOP conference series. Earth and environmental science 2021-03, Vol.698 (1), p.12004
• Main Authors: Haq, Nabil Amirul, Khomsin, Pratomo, Danar Guruh
• Format: Article
• Language: English
• Subjects: Accuracy; Control equipment; Data analysis; Filtration; Global positioning systems; GPS; Human resources; Hydrography; Internet of Things; Low cost; Microcontrollers; Outliers (statistics); Real time; Root-mean-square errors; Tides; Ultrasonic testing
• ISSN: 1755-1307; 1755-1315
• DOI: 10.1088/1755-1315/698/1/012004

Ocean tides play essential roles in the field of hydrography and navigation. The conventional method to measure tides is by observing the vertical variation of sea surface using a tide staff. However, this method has several disadvantages. It consumes more human resources and time to observe and record the take. Moreover, the recorded data tends to be subjective, depending on the observer. To overcome these disadvantages is by using an automatic tide gauge. This instrument can also provide almost continuous and near real-time tide data. The price of automatic tide gauge in the Indonesia market is costly. This research attempted to build a low-cost ultrasonic tide gauge by utilizing an Arduino microcontroller, ultrasonic sensor, GPS module, which is embedded with the Internet of Things (IoT) so that the real-time data can be uploaded and monitored on a webserver. The quality of the data collected from the instrument is controlled using the 3 σ rule to detect and remove outliers. Moving Average and Moving Median filtering methods are applied in the system to eliminate the noise. The data accuracy-test was performed using a relative error and Root Mean Square Error (RMSE) methods. The test compares the data collected with the ultrasonic tide gauge and the direct observation method. The result shows the relative error, and RMSE values 0.226% and 6.629mm for raw data, 0.636% and 18.542mm for moving average data, 0.437% and 13.242mm for the moving median data, with the best filtering method is moving median. The instrument accuracy test results show that this instrument has excellent accuracy, with a 2,240,000 IDR instrument production budget or 7.5% of the lowest tide gauge prices.