An electronic irrigation system using IoT and neural networks

Abstract

One of the approaches that fall under the alternative application of water on earth or soil is electronic irrigation. It is aware of the need to irrigate crops, restore the vegetation of difficult soil in arid areas, and because of dry spells, as our state has experienced in recent years. Other issues, such as increasing plant growth while lowering the value of agriculture, necessitate installing an irrigation system that cuts back effort, reduces farm and field employees, and minimizes monetary matters within the construction of agricultural comes is crucial. Soil wetness measure is incredibly tough; thus the economic maintaining of its target levels. The answer to this drawback is an automatic irrigation system. This analysis proposed an electronic irrigation system that reduces users' effort to plant care. The system kernel is the self-learning Kohonen Neural Network, which depends on the reading of the detector of soil wetness, plant type, and forecast data. The soil wetness detector indicates the soil wetness level. Also, the system is mechanically started once the wetness level is not up to the extent necessary for the plant's growth. When the system reaches the soil wetness level, it is mechanically stopped for a defined period of morning and evening. As a soil wetness level differs from one plant kind to a different, 3 plant varieties area unit used during this analysis. Beginning the system littered with the weather data, is saving time and effort for the employees.