Bu kaydın yasal hükümlere uygun olmadığını düşünüyorsanız lütfen sayfa sonundaki Hata Bildir bağlantısını takip ederek bildirimde bulununuz. Kayıtlar ilgili üniversite yöneticileri tarafından eklenmektedir. Nadiren de olsa kayıtlarla ilgili hatalar oluşabilmektedir. MİTOS internet üzerindeki herhangi bir ödev sitesi değildir!

THE GLOBAL NEURAL NETWORK LEARNING USING GENETIC ALGORITHM FOR ORE GRADE PREDICTION OF AN IRON ORE DEPOSIT

Oluşturulma Tarihi: 2013

Niteleme Bilgileri

Tür: Makale

Yayınlanma Durumu: Yayınlanmış

Dosya Biçimi: PDF

Dil: İngilizce

Konu(lar): Cevher yatakları ve belirli metallerin çıkarılması,

Yazar(lar): CHATTERJEE, SNEHAMOY (Yazar), BANDOPADHYAY, SUKUMAR (Yazar),

Emeği Geçen(ler):


Yayınlayan: Atılım Üniversitesi Yayın Adı: The International Journal of Mineral Resources Engineering Yayın Yeri: Ankara Yayın Tarihi: 2007 Sayı: 4 Cilt: 12 Yayınlandığı Sayfalar: 217-240


Dosya:
file show file
Görüntüle
download file
Kaydet

Anahtar Kelimeler
Özet

In this paper, genetic algorithm is used for training the neural network model for ore

grade prediction of an iron ore mine located in India. The network model incorporated

the spatial locations as well as the lithological information as input parameters for

the neural network model. A genetic algorithm was incorporated to find a neural

network solution close to the global optimum. A pre-established topology was used

for this case study. The network topology constituted three layers: an input, an output

and a hidden layer. The input layer consisted of three spatial coordinates (x, y and z)

and seven litho-types. The output layer is comprised of the Fe grade. The optimum

connection weights and biased terms were obtained by a genetic algorithm search

process. The numbers of hidden nodes were selected by searching with all possible

hidden nodes size. The optimum number of hidden nodes, however, was obtained

based on the minimum error of training data. To justify the use of GA-based neural

network modeling, a comparative evaluation between Levenberg-Marquardt neural

network learning, ordinary kriging, and lognormal kriging method was performed.

The comparative evaluation of these algorithms suggests that the GA-based neural

network model determinedly outperformed the other models. The grade tonnage curve

was then developed using the GA neural network model. Before developing the grade

tonnage curve, it was required to develop the lithological maps of the deposit at the

unknown grid points. Indicator kriging was used for developing the lithological maps

of the deposit.


İçindekiler
Açıklamalar
Haklar
Notlar

Kaynakça


Atıf Yapanlar

Gözat Sayfasına Dön

 

Sosyal Medya ve Araçlar

İstatistikler

  • Kayıt
    • Bu ay: 3
    • Toplam: 2417
  • Online
    • Ziyaretçi: 12
    • Üye: 0
    • Toplam: 12

Detaylı İstatistikler