Sari, Adelia Wahyu Puspita (2019) Rancang Bangun Prediktor Sebaran Ikan Menggunakan Metode Jaringan Syaraf Tiruan Dengan Optimasi Particle Swarm Optimization. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
02311540000016-Undergraduate_Theses.pdf Download (3MB) | Preview |
Abstract
Badan Informasi Geospasial (BIG) menyatakan bahwa garis pantai Indonesia memiliki total panjang 99,093 kilometer. Hal ini menunjukkan bahwa Indonesia juga memiliki kekayaan laut yang sangat banyak terutama ikan. Sampai saat ini masih terdapat permasalahan dimana kurangnya informasi mengenai keberadaan ikan di perairan Indonesia, sehingga perlu adanya suatu perancangan orediksi untuk menginformasikan adanya keberadaan ikan di suatu wilayah. Tujuan penelitian ini adalah untuk memperoleh algoritma prediktor terbaik untuk menentukan keberadaan ikan dengan menghitung prediksi jumlah tangkapan ikan untuk ikan biji nangka, kerapu dan kuniran. Metode yang diusulkan pana penelitian ini yaitu JST, dan JST-PSO. Data sekunder didapatkan dari NOAA, BMKG, dan PPN Brondong. Data sekunder dependen dan independen diuji missing data , outlier, dan korelasi. Kemudian JST dan JST-PSO dirancang arsitektur nya. Arsitektur terbaik yang digunakan untuk memprediksi real time di pantai kenjeran . Arsitektur terbaik yang didapatkan pada JST-PSO dengan input berupa kecepatan angin, suhu permukaan laut, dan curah hujan. Gabungan JST-PSO dengan hidden node yang dipasang sebanyak 5, 9 dan 10 unit mampu menurunkan RMSE masing-masing sebesar 26 kg, 228,7 kg, dan 121,9 kg.
========================================================================================================================
The Geospatial Information Agency (BIG) states that Indonesia's coastline has a total length of 99,093 kilometers. This shows that Indonesia also has a wealth of marine resources, especially fish. Until now there are still problems where there is a lack of information about the presence of fish in Indonesian waters, so there needs to be an edictional design to inform the existence of fish in an area. The purpose of this study was to obtain the best predictor algorithm for determining the presence of fish by calculating the prediction of the number of fish catches for fish of jackfruit seeds, groupers and turmeric. The method proposed for this research is ANN, and JST-PSO. Secondary data was obtained from NOAA, BMKG, and Brondong PPN. Dependent and independent secondary data were tested for missing data, outliers, and correlations. Then the JST and JST-PSO are designed by the architecture. The best architecture that is used to predict real time on kenjeran beach. The best architecture obtained on ANN-PSO with input in the form of wind speed, sea surface temperature and rainfall. Combined JST-PSO with hidden nodes installed 5, 9 and 10 units were able to reduce RMSE by 26 kg, 228.7 kg and 121.9 kg respectively.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Particle Swarm Optimization, hidden node, neural network, realtime, RMSE |
| Subjects: | T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.62 Simulation |
| Divisions: | Faculty of Industrial Technology > Physics Engineering > 30201-(S1) Undergraduate Thesis |
| Depositing User: | Adelia Wahyu Puspita Sari |
| Date Deposited: | 13 Nov 2024 01:08 |
| Last Modified: | 13 Nov 2024 01:08 |
| URI: | http://repository.its.ac.id/id/eprint/69958 |
Actions (login required)
 |
View Item |