Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/125635| Title: | Identifikasi Penggunaan Koagulan dan Flokulan pada Instalasi Pengolahan Air Limbah Tambang Emas Berbasis Jaringan Saraf Tiruan |
| Other Titles: | Identification of Coagulant and Flocculant Usage in Gold Mining Wastewater Treatment Plant Based on Artificial Neural Network |
| Authors: | Saptomo, Satyanto Krido Habibie, Rosihanudin Yusuf |
| Issue Date: | 2023 |
| Publisher: | IPB University |
| Abstract: | Salah satu resiko kerusakan lingkungan karena kegiatan pertambangan emas,
yaitu air limbah. Instalasi pengolahan air limbah (IPAL) perusahaan tambang emas
belum memiliki teknologi maju untuk menangani air limbah. Jaringan saraf tiruan
(JST) sebagai hasil perkembangan teknologi 4.0 digunakan sebagai solusi
permasalahan IPAL tersebut. Penelitian ini bertujuan untuk mengidentifikasi
penggunaan koagulan dan flokulan pada IPAL tambang emas menggunakan basis
jaringan saraf tiruan. Algoritma yang digunakan JST merupakan model
feedforward neural network untuk menganalisis data sianida (CN), derajat
keasaman (pH), dan total suspended solid (TSS), serta reagen koagulan flokulan
IPAL perusahaan tambang. Model JST menggunakan 6 parameter input, 2
parameter output, dan 6 hidden layer sebanyak 512, 256, 128,64, 32, 2 serta
menghasilkan nilai evaluasi MSE 0,0076. Evaluasi hasil prediksi dibandingkan data
aktualnya menunjukan hasil yang baik ditinjau dari persamaan regresi dan koefisien
determinasi. Namun masih ada titik poin yang berada jauh dari garis trendline,
sehingga menunjukan hasil belum bagus. Hasil Program JST tidak dapat
direkomendasikan untuk digunakan secara langsung pada operasional IPAL
tambang emas. One of the risks of environmental damage due to gold mining activities is wastewater. Gold mining companies' wastewater treatment plants (IPAL) do not yet have advanced technology to handle waste water. Artificial neural networks (ANN) as a result of the development of technology 4.0 are used as a solution to the IPAL problem. This research aims to identify the use of coagulants and flocculants in gold mining WWTPs using an artificial neural network basis. The algorithm used by JST is a feedforward neural network model to analyze data on cyanide (CN), degree of acidity (pH), and total suspended solids (TSS), as well as flocculant coagulant reagents for mining companies' WWTPs. The ANN model uses 6 input parameters, 2 output parameters, and 6 hidden layers totaling 512, 256, 128.64, 32, 2 and produces an MSE evaluation value of 0.0076. Evaluation of the prediction results compared to the actual data shows good results in terms of the regression equation and coefficient of determination. However, there are still points that are far from the trendline, so the results are not good. The results of the JST Program cannot be recommended for direct use in gold mining IPAL operations. |
| URI: | http://repository.ipb.ac.id/handle/123456789/125635 |
| Appears in Collections: | UT - Civil and Environmental Engineering |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Cover.pdf Restricted Access | Cover | 224.6 kB | Adobe PDF | View/Open |
| F44190051_Rosihanudin Yusuf Habibie.pdf Restricted Access | Fulltext | 3.63 MB | Adobe PDF | View/Open |
| Lampiran.pdf Restricted Access | Lampiran | 3.28 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.