Evaluation of IoT Sensor Utilization for River Water

Study Case Development of an IoT-Based Environmental Monitoring System with the Application of a Multi-Sensor Wireless Sensor Network Michael Kurniawan (2018123456) University of Cincinnati, 2600 Clifton Ave, Cincinnati, OH 45221, United States kurniaml@mail.uc.edu Abstract This paper discusses the application of the Internet of Things (IoT) based on a Wireless Sensor Network (WSN) for an environmental monitoring system with the integration of a Multi Sensor Network (MSN) that can be accessed in real time. The system is designed using various electronic sensor modules, namely DHT22 as a temperature and humidity sensor, MQ135 as an air quality sensor, Flame Sensor as a fire detector, SW420 to detect vibrations, Raindrops Sensor as a rain sensor, and ESP32 microcontroller board as the main controller. The research method used is Research and Development (R&D), by designing a WSN prototype as an environmental monitoring device capable of performing real-time measurements via the internet. The measurement data is collected by the sensor, sent to the microcontroller, and then displayed through the Blynk IoT application on a smartphone. This research is expected to improve the efficiency and effectiveness of environmental condition monitoring, while providing accurate and reliable data for further analysis. Keywords: IoT, Wireless Sensor Network, Multi Sensor Network, monitoring, Blynk INTRODUCTION Environmental conditions are an important factor that affects people's daily activities. A clean and healthy environment contributes to physical and mental wellbeing, while preventing the spread of disease. Conversely, pollution can cause health problems such as respiratory and digestive disorders. Extreme weather changes, both hot and cold, can also pose risks such as dehydration, heatstroke, or hypothermia. MICHAEL KURNIAWAN 1 Study Case Unpredictable environmental conditions, such as fires, landslides, or earthquakes, cause particular concern. Therefore, environmental monitoring is important so that the community can obtain accurate information and increase awareness in facing potential disasters. The development of digital technology has brought humans into the era of the Internet of Things (IoT), a concept that allows devices to connect via the internet to share data and perform remote control. One of its implementations is the Wireless Sensor Network (WSN), which is a wireless network consisting of several sensor nodes that collect environmental data in real-time. A further development of WSN is the Multi Sensor Network (MSN), where several sensors are combined in one node to improve measurement accuracy. Previous studies have discussed the application of WSN and MSN in environmental monitoring, but they are generally limited to specific parameters or have limitations in data transmission methods. Based on previous studies, this research designs an IoT system that can monitor various environmental parameters simultaneously, including temperature, humidity, air quality, vibration (indication of earthquakes), rain, and fire potential. Data from these sensors is sent in real-time to the Blynk IoT application on a smartphone. By using sensors that are proven to be reliable yet economical, this study offers an effective and affordable environmental monitoring solution. RESEARCH METHOD The method used is Research and Development (R&D), which is a systematic process to produce and develop a product to be more optimal. The stages of this research consist of: 1. Design and Development Creating a preliminary system design to ensure that functional requirements are met. 2. Hardware Design The main components used include: • ESP32: a dual-core microcontroller with power-efficient WiFi and Bluetooth. • DHT22: a high-accuracy temperature and humidity sensor. MICHAEL KURNIAWAN 2 Study Case • MQ135: air quality sensor to detect pollutant gases such as CO₂, ammonia, nitrogen oxide, and smoke. • SW420: vibration/shock sensor. • Flame Sensor: infrared-based fire detector. • Raindrops Sensor: rain sensor based on digital/analog output. • 18650 Battery: main power source. • Switch: circuit breaker from the battery. • Protective box: container for all components. 3. Software Design The software system is built using Arduino IDE for sensor programming, as well as Blynk IoT as a visual interface application on smartphones. 4. Testing All nodes are tested to ensure that the sensors can read and transmit data according to the design. 5. Evaluation The device's performance is assessed to ensure the effectiveness and reliability of the system. 6. Revision and Improvement If obstacles are found, improvements and iterations are made to optimize the device. 7. Dissemination of Results The research results are published in scientific journals to contribute academically and practically. MICHAEL KURNIAWAN 3 Study Case Figure 1. Flowchart of the research method RESULTS AND DISCUSSION Design and Development The initial stage of this research involved the design of various sensor modules, including DHT22 for temperature and humidity, MQ135 for air quality, SW420 for vibration, Flame Sensor for fire detection, and Raindrops Sensor for rainfall. These sensors were chosen because they are relatively affordable, have been proven to be stable in previous studies, and are effective for environmental monitoring. This system was developed to expand the scope of environmental monitoring while adding an automatic notification feature to the user's smartphone. The data obtained from all sensors is processed by the ESP32 microcontroller, programmed through the Arduino IDE, and displayed through the Blynk IoT application. The system was thoroughly tested to ensure there were no malfunctions before the full implementation stage. If there were any deficiencies, adjustments were made so that the system functioned according to specifications before entering the results publication stage. Hardware Design The main hardware consists of ESP32, DHT22, MQ135, SW420, Flame Sensor, Raindrops Sensor, battery, and switch. Each sensor is connected to the ESP32 pin according to its configuration. The assembly process requires precision because pin errors can cause damage to components. MICHAEL KURNIAWAN 4 Study Case Figure 2. Device Circuit In addition, two identical circuits (Node 1 and Node 2) were created to improve monitoring capabilities. All components are placed in an enclosure box to protect them from mechanical damage, dust, and moisture, while also making the system neater and more robust. Figure 3. Installation of Node 1 and Node 2 Software Design The software is designed using Blynk IoT, which displays data from both nodes in real-time on the user's smartphone. This platform allows the hardware to be controlled wirelessly without additional cables. Figure 4. Blynk Console Web Display The initial configuration includes creating an account with an email address, creating two devices (Environmental Monitoring 1 and 2), and five data streams: temperature, humidity, pollution, vibration, and fire. MICHAEL KURNIAWAN 5 Study Case Figure 5. Creating Data Streams in the Blynk Console The web dashboard display was created using Gauge and Chart Widgets. In addition, four automatic notification events were set up for rain, sunshine, fire, and vibration. Figure 6. Blynk Console Dashboard Design Figure 7. Creating Events in the Blynk Console After that, the dashboard was accessed through the Blynk IoT application on a smartphone with the same email account. Figure 8. Blynk IoT Node 1 and Node 2 Dashboard Display MICHAEL KURNIAWAN 6 Study Case For programming, the Arduino IDE with C/C++ language was used as a platform to write, upload, and test code on ESP32. The program allows ESP32 to read sensor data, process it, and then send the results to Blynk IoT. Figure 9. Arduino IDE Display with Program Code System Testing Testing was conducted to ensure that all sensors were working properly. The following are the test results for each sensor: 1. Temperature and Humidity Test The DHT22 sensor was covered with a finger, causing the temperature and humidity values to increase due to body heat. • Node 1: temperature rose from 30.2°C → 31.6°C, humidity from 65% → 90%. • Node 2: temperature rose from 29.6°C → 30.3°C, humidity from 65% → 91%. Figure 10. Temperature and Humidity Test Node 1 MICHAEL KURNIAWAN 7 Study Case Figure 11. Temperature and Humidity Test Node 2 When the finger is removed, the temperature remains relatively stable but the humidity decreases more quickly. • Node 1: humidity drops from 90% → 64%. • Node 2: humidity drops from 91% → 64%. Figure 12. Temperature and Humidity Test Decrease Node 1 Figure 13. Temperature and Humidity Test Decrease Node 2 Results: the DHT22 sensor proved to be responsive and accurate. 2. Pollution Test (MQ135) The test was conducted by spraying butane gas from a lighter. MICHAEL KURNIAWAN 8 Study Case • Node 1: concentration increased from 519 to 2310. • Node 2: concentration increased from 830 to 3014. Figure 14. Pollution Test Node 1 Figure 15. Pollution Test Node 2 The MQ135 sensor successfully detected the increase in pollutants. 3. Vibration Test (SW420) The node was shaken to test vibration detection. • Node 1: value increased from 0 → 2184 → 10000. • Node 2: value increased from 0 → 1974 → 10000. Figure 16. Vibration Test Node 1 MICHAEL KURNIAWAN 9 Study Case Figure 17. Vibration Test Node 2 However, the SW420 sensor was not accurate in displaying values, only able to detect the presence of vibration. 4. Fire Test (Flame Sensor) The node was placed close to a lighter flame. The system successfully sent a fire notification to a smartphone. Figure 18. Flame Test Node 1 Figure 19. Flame Test Node 2 5. Rain Test (Raindrops Sensor) Water was sprayed onto the sensor → a rain notification was sent. The sensor was cleaned → a clear weather notification was sent. MICHAEL KURNIAWAN 10 Study Case Figure 20. Rain Test Node 1 Figure 21. Rain Test Node 2 Figure 22. Clear Test Node 1 MICHAEL KURNIAWAN 11 Study Case Figure 23. Clear Test Node 2 Evaluation All sensors functioned properly except for SW420, which was unable to provide accurate quantitative values. Solution: vibration values were removed from the dashboard and replaced with notifications only to keep the system informative. Revision and Improvement The revision was made by removing the vibration value display and adding a poor air quality notification feature when the MQ135 value is > 2500. Figure 24. Removal of Vibration Display Figure 25. Addition of Pollution Sensor Event MICHAEL KURNIAWAN 12 Study Case Program Code 1. Additional Pollution Detection Script The revision results are shown on the final dashboard: Figure 26. Final Display of Node 1 Figure 27. Final Display of Node 2 CONCLUSION AND RECOMMENDATIONS Conclusion This study successfully designed and implemented an IoT-based WSN system with MSN for real-time environmental monitoring. All sensors functioned as intended, except for SW420, which could only detect vibrations without displaying numerical values. Revisions to the system, including the addition of air pollution notifications and improvements to the dashboard display, resulted in a more stable, accurate, and comprehensive monitoring system. This system has proven capable of supporting environmental monitoring with high efficiency. Recommendations This system can still be improved by adding additional sensors, such as light, wind direction, sound, and GPS modules. These additions will expand the scope of monitoring, resulting in more detailed data for more comprehensive environmental analysis. MICHAEL KURNIAWAN 13 Study Case REFERENCE MICHAEL KURNIAWAN 14 Study Case MICHAEL KURNIAWAN 15