1Departemen Teknik Elektro, Fakultas Teknik, Universitas Diponegoro, Indonesia
2Departemen Ilmu Kelautan, Fakultas Perikanan dan Ilmu Kelautan, Universitas Diponegoro, Indonesia
BibTex Citation Data :
@article{JKT14256, author = {Erwin Adriono and Maman Somantri and Chrisna Adhi Suryono}, title = {Model Prediksi Jumlah Pakan menggunakan Algoritma Evolusi Pikiran - Jaringan Syaraf Tiruan Rambatan Balik untuk Budidaya Udang}, journal = {Jurnal Kelautan Tropis}, volume = {25}, number = {2}, year = {2022}, keywords = {Litopenaeus Vannamei; Model Prediksi; BPNN; MEA}, abstract = { Menentukan jumlah pakan yang sesuai merupakan hal penting dalam kegiatan budidaya udang berjenis Litopenaeus Vannamei . Jumlah pakan dapat dipengaruhi oleh banyak faktor antara lain Jumlah Udang, Umur udang, DO, Salinitas, Alkalinitas, Suhu dan PH. Hubungan antar faktor tersebut dengan jumlah pakan sulit dibuatkan dalam persamaan matematis maupun dengan metode statisik. Permasalahan tersebut dapat diselesaikan menggunakan Neural network . Neural network menjadi solusi untuk memodelkan hubungan input dan output yang kompleks. Hubungan Jumlah pakan dan faktorlainnya akan dimodelkan menggunakan metode Backpropagation NN yang dikombinasikan dengan algoritma optimasi seperti Genetic Algoritm dan Mind Evotionary Algoritm . Model BPNN, BPNN – GA dan BPNN MEA akan dibandingkan performa menggunakan MSE, RSME, MAE dan MAPE. Dari ketiga metode yang digunakan didapatkan hasil paling baik adalah pada BPNN MEA yaitu nilai MSE, RSME, MAE dan MAPE berturut-turut adalah 40,92; 6,39; 6,51 dan 20,29 . Determining the appropriate amount of feed is important in the aquaculture of Litopenaeus Vannamei shrimp. The amount of feed can be influenced by many factors including the number of shrimp, shrimp age, DO, salinity, alkalinity, temperature and PH. The relationship between these factors and the amount of feed is difficult to make in mathematical equations or with statistical methods. These problems can be solved using a neural network. Neural network is a solution for modeling complex input and output relationships. The relationship between the amount of feed and other factors will be modeled using the Backpropagation NN method combined with optimization algorithms such as Genetic Algorithm and Mind Evotionary Algorithm. The BPNN, BPNN – GA and BPNN MEA models will be compared using MSE, RSME, MAE and MAPE. Of the three methods used, the best results were obtained on BPNN MEA, with values of MSE, RSME, MAE and MAPE respectively 40,92; 6,39; 6,51 and 20,29. }, issn = {2528-3111}, pages = {266--278} doi = {10.14710/jkt.v25i2.14256}, url = {https://ejournal2.undip.ac.id/index.php/jkt/article/view/14256} }
Refworks Citation Data :
Menentukan jumlah pakan yang sesuai merupakan hal penting dalam kegiatan budidaya udang berjenis Litopenaeus Vannamei. Jumlah pakan dapat dipengaruhi oleh banyak faktor antara lain Jumlah Udang, Umur udang, DO, Salinitas, Alkalinitas, Suhu dan PH. Hubungan antar faktor tersebut dengan jumlah pakan sulit dibuatkan dalam persamaan matematis maupun dengan metode statisik. Permasalahan tersebut dapat diselesaikan menggunakan Neural network. Neural network menjadi solusi untuk memodelkan hubungan input dan output yang kompleks. Hubungan Jumlah pakan dan faktorlainnya akan dimodelkan menggunakan metode Backpropagation NN yang dikombinasikan dengan algoritma optimasi seperti Genetic Algoritm dan Mind Evotionary Algoritm. Model BPNN, BPNN – GA dan BPNN MEA akan dibandingkan performa menggunakan MSE, RSME, MAE dan MAPE. Dari ketiga metode yang digunakan didapatkan hasil paling baik adalah pada BPNN MEA yaitu nilai MSE, RSME, MAE dan MAPE berturut-turut adalah 40,92; 6,39; 6,51 dan 20,29.
Determining the appropriate amount of feed is important in the aquaculture of Litopenaeus Vannamei shrimp. The amount of feed can be influenced by many factors including the number of shrimp, shrimp age, DO, salinity, alkalinity, temperature and PH. The relationship between these factors and the amount of feed is difficult to make in mathematical equations or with statistical methods. These problems can be solved using a neural network. Neural network is a solution for modeling complex input and output relationships. The relationship between the amount of feed and other factors will be modeled using the Backpropagation NN method combined with optimization algorithms such as Genetic Algorithm and Mind Evotionary Algorithm. The BPNN, BPNN – GA and BPNN MEA models will be compared using MSE, RSME, MAE and MAPE. Of the three methods used, the best results were obtained on BPNN MEA, with values of MSE, RSME, MAE and MAPE respectively 40,92; 6,39; 6,51 and 20,29.
Article Metrics:
Last update:
Upon acceptance for publication, authors agree to transfer the copyright of their article to Jurnal Kelautan Tropis, while retaining the right to reuse their work under the terms of the open license applied.
From the date of publication, the copyright for each article is held by Jurnal Kelautan Tropis. This transfer allows the journal to manage, disseminate, and preserve scholarly content in accordance with international standards and open access best practices.
Although copyright is held by the journal, all published articles are made available under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0). Under this license, anyone may:
Copy and redistribute the material in any medium or format
Remix, transform, and build upon the material for any purpose, even commercially
provided that:
Appropriate credit is given to the original author(s) and the source
Indications are made of any changes that were made
Derivative works are distributed under the same license (CC BY-SA 4.0)
While copyright is held by the journal, authors retain important reuse rights. Authors may:
Reuse the published version of their article in future works, including books, compilations, and lectures
Deposit the published version in institutional or subject repositories
Share the article freely, including on personal websites or academic networks
as long as the original publication in Jurnal Kelautan Tropis is cited and the CC BY-SA 4.0 license terms are respected.
Authors must ensure that any third-party content included in the article (e.g., figures, images, datasets) is either original, in the public domain, or licensed for reuse under compatible terms. If specific permissions are required, authors must obtain them prior to submission.
For questions regarding copyright or licensing, please contact the editorial office at: j.kelautantropis@gmail.com
View My Stats
Jurnal Kelautan Tropis is published by Departement of Marine Science, Faculty of Fisheries and Marine Science, Universitas Diponegoro under a Creative Commons Attribution-ShareAlike 4.0 International License.
