STUDY OF THE EFFECT PHYSICAL PARAMETERS ON BOK COY PLANTS (BRASSICA RAPA SUBSP. CHINENSIS) AT COMMERCIAL HYDROPONICS BASED ON INTERNET OF THINGS (IOT)

• Main Author: Partogi Nahotasi, Efraim
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/60604
• Institution: Institut Teknologi Bandung

Population growth causes the demand for food to increase. One solution that can be applied is agriculture with hydroponic technology, but this system has drawbacks such as being sensitive to changes in the environment or existing physical parameters so that vegetable production is not optimal. To increase production, one must be able to control the physical parameters that most influence the production process, the process can be modeled using machine learning from data obtained from the Internet of Things (IoT) system and the results of measuring growth manually. IoT is used in research as a data miner so that the measurement of physical parameters is accurate and precise so that it can be used as a dataset in machine learning. Physical parameters to be measured such as light intensity, humidity, air temperature, TDS (Total Dissolved Solids), and solution temperature. There are eight sensor nodes to measure the state of physical parameters in a hydroponic greenhouse with seven sensor nodes in the air and one sensor node in solution. The research was carried out for two harvests or about 56 days. Physical parameter data will be sent every three minutes from the sensor node to the server using a local WiFi router network so that a time-series graph is obtained in the database and then stored on the Raspberry Pi device. Leaf area growth data was measured every 3 days by photographing. The conclusion is that the pakcoy plant growth model is the most suitable and has good accuracy using the machine learning random forest regression algorithm with an R2 value of 0.933, RMSE 54.52 and an absolute error of 8.3%.The variables that most influence the growth model are TDS 67.11% and light intensity 32.89% so that they can be used as control variables for process production. The results of the TDS gradient graph will affect the leaf area gradient during growth when the light intensity is constant. From the results of this research, a dynamic growth model in pakcoy plants based on experimental data and the IoT system is obtained, which is expected to be applied to the TDS control system in future studies.