Özet:

Agriculture has been one of our most important needs in the world since the first ages. Nowadays, human knowledge and experience, especially in agriculture, are still lacking in achieving the most productivity. For a plant to grow close to 100% yield, multiple variables must be in optimal condition. In the Agriculture Tracking System, people are able to control the environment required for growing plants, i.e. optimal levels of their variables. Therefore, in this study, we have designed a hardware system with an Internet of Things (IoT) device and 2 sensors. The solar energy to power this hardware system has been used. Also, these sensors are a digital humidity and temperature sensor (DHT11) and Soil Moisture Sensors. Values of incoming from sensors are read with an IP Address. Moreover, these values are written to SQLite database and displayed last 5 records with bar charts. The users can save plants with optimal values. With these values, can make predictions. We have studied the effect of the sun angle on temperature and humidity in the last 5 years with months. We have compared this data with the plants added by the user and presented the most appropriate and closest months as a warning to the user. Another prediction is possible with instant records. When we researched, we saw that the condition of the plant can be categorized according to temperature and humidity. By checking the instant data, a warning message to the user according to these rates is sent by using the K-Nearest Neighbour classification algorithm. As a result, in the tests, the results have shown that this approach can increase productivity.

Anahtar Kelimeler: