Prethodno priopćenje

https://doi.org/10.17559/TV-20160419130812

The Determination of the Developments of Beehives via Artificial Neural Networks

Raif Bayir ; Karabuk University, Demir-Çelik Kampüsü, 78050 Karabuk / Turkey
Ahmet Albayrak ; Karadeniz Technical University, 61030 Trabzon / Turkey

Sažetak

Honeybees provide great benefits for people both with the foods they produce and as a pollinator. It is known that they pass the whole year with the foods they collect in spring and summer months. Beekeepers also benefit from the honey produced in these periods. Whether a beehive works adequately or not and its status of development can be understood through the observations by beekeepers. In this study, an Arduino-supported neural network model was developed in order to obtain information about the general situations of beehives. The three-input and three-output neural networks were embedded in a board after the training and testing stage. While temperature, humidity, and weight refer to inputs, good situation, stable situation, and bad situation represent outputs. The real-time model has an accuracy of 99.84%.

Ključne riječi

artificial intelligence; intelligent systems; backpropagation; neural networks; embedded systems

Hrčak ID:

199155

URI

https://hrcak.srce.hr/199155

Datum izdavanja:

21.4.2018.

Posjeta: 1.470 *