Implementation of Internet of Things (IoT)-based Aquaculture System Using Machine Learning Approaches
Închide
Conţinutul numărului revistei
Articolul precedent
Articolul urmator
260 4
Ultima descărcare din IBN:
2023-07-03 16:25
Căutarea după subiecte
similare conform CZU
004.7:004.451 (1)
Comunicații între calculatoare. Rețele de calculatoare (514)
Programe. Software (295)
SM ISO690:2012
ABDUL KADAR, Muhammad Masum. Implementation of Internet of Things (IoT)-based Aquaculture System Using Machine Learning Approaches. In: Computer Science Journal of Moldova, 2021, nr. 3(87), pp. 320-339. ISSN 1561-4042.
EXPORT metadate:
Google Scholar
Crossref
CERIF

DataCite
Dublin Core
Computer Science Journal of Moldova
Numărul 3(87) / 2021 / ISSN 1561-4042 /ISSNe 2587-4330

Implementation of Internet of Things (IoT)-based Aquaculture System Using Machine Learning Approaches

CZU: 004.7:004.451

Pag. 320-339

Abdul Kadar Muhammad Masum
 
International Islamic University of Chittagong
 
 
Disponibil în IBN: 3 decembrie 2021


Rezumat

The protein demand on the planet is increased rapidly with respect to the fast growth of the population and can’t be met by meat. Fish is an amazing wellspring of protein that compares with meat even with minimal effort in production. The majority of consecutive fish farming and the subsisting intelligent system failed to produce anticipated amount of fish. The main aim of this research is to introduce an effective IoT-based low-cost fish farm which will also be high in the production of fish. A WEMOS D1 is actively involved in the determination of different basic water parameters such as temperature, oil layer, pH, water level, conductivity, oxygen level, turbidity, and fish behavior to anticipate their hunger status which significantly impacts on the quick fish growth. Via our built interactive smartphone application and Web interface, these pieces of information are transmitted to the end user. The D1 microcontroller works on the ESP8266 WIFI module through which the system sends the information to the mobile and web application. In the event of an abnormal situation, exceeding a predefined threshold value, the framework will inform the concerted authority to take immediate steps. In addition, the system is more remarkable and extraordinary as it can predict the amount of following day fish food and other constraints required for fish farmers to take a precautionary measure in advance. These characteristics would allow a fishery proprietor to grow a large number of fish via reducing the protein challenge.

Cuvinte-cheie
oil layer detection, pH sensor, Random Forest Regression (RFR), Support Vector Regression (SVR)