PLASTIC WASTE CONVERSION TO LIQUID FUELS OVER MODIFIED-RESIDUAL CATALYTIC CRACKING CATALYSTS: MODELING AND OPTIMIZATION USING HYBRID ARTIFICIAL NEURAL NETWORK – GENETIC ALGORITHM

*Istadi Istadi -  Laboratory of Energy and Chemical Process Engineering, Department of Chemical Engineering, Faculty of Engineering, Diponegoro University, Jl. Prof. H. Sudharto, SH., Semarang, 50239 Indonesia, Telp: 024-7460058, Fax: 024-76480675, Indonesia
Luqman Buchori -  Laboratory of Energy and Chemical Process Engineering, Department of Chemical Engineering, Faculty of Engineering, Diponegoro University, Jl. Prof. H. Sudharto, SH., Semarang, 50239 Indonesia, Telp: 024-7460058, Fax: 024-76480675, Indonesia
Suherman Suherman -  Laboratory of Energy and Chemical Process Engineering, Department of Chemical Engineering, Faculty of Engineering, Diponegoro University, Jl. Prof. H. Sudharto, SH., Semarang, 50239 Indonesia, Telp: 024-7460058, Fax: 024-76480675, Indonesia
Published: 6 May 2011.
Open Access
Citation Format:
Article Info
Section: Research Article
Language: EN
Full Text:
Statistics: 923 1683
Abstract

The plastic waste utilization can be addressed toward different valuable products. A promising technology for the utilization is by converting it to fuels. Simultaneous modeling and optimization representing effect of reactor temperature, catalyst calcinations temperature, and plastic/catalyst weight ratio toward performance of liquid fuel production was studied over modified catalyst waste. The optimization was performed to find optimal operating conditions (reactor temperature, catalyst calcination temperature, and plastic/catalyst weight ratio) that maximize the liquid fuel product. A Hybrid Artificial Neural Network-Genetic Algorithm (ANN-GA) method was used for the modeling and optimization, respectively. The variable interaction between the reactor temperature, catalyst calcination temperature, as well as plastic/catalyst ratio is presented in surface plots. From the GC-MS characterization, the liquid fuels product was mainly composed of C4 to C13 hydrocarbons.

KONVERSI LIMBAH PLASTIK MENJADI BAHAN BAKAR CAIR DENGAN METODE PERENGKAHAN KATALITIK MENGGUNAKAN KATALIS BEKAS YANG TERMODIFIKASI: PEMODELAN DAN OPTIMASI MENGGUNAKAN GABUNGAN METODE ARTIFICIAL NEURAL NETWORK DAN GENETIC ALGOR

ITHM. Pemanfaatan limbah plastik dapat dilakukan untuk menghasilkan produk yang lebih bernilai tinggi. Salah satu teknologi yang menjanjikan adalah dengan mengkonversikannya menjadi bahan bakar. Permodelan, simulasi dan optimisasi simultan yang menggambarkan efek dari suhu reaktor, suhu kalsinasi katalis, dan rasio berat plastik/katalis terhadap kinerja produksi bahan bakar cair telah dipelajari menggunakan katalis bekas termodifikasi Optimisasi ini ditujukan untuk mencari kondisi operasi optimum (suhu reaktor, suhu kalsinasi katalis, dan rasio berat plastik/katalis) yang memaksimalkan produk bahan bakar cair. Metode Hybrid Artificial Neural Network-Genetic Algorithm (ANN-GA) telah digunakan untuk permodelan dan optimisasi simultan tersebut. Inetraksi antar variabel suhu reaktor, suhu kalsinasi katalis, dan rasio berat plastik/katalis digambarkan dalam bentuk plot surface. Berdasarkan karakterisasi GC-MS, produk bahan bakar yang diperoleh terdiri dari komponen-komponen hidrokarbon C4-C13.

Keywords
artificial neural network; central composite design; genetic algorithm; optimization; plastic waste; Residual Catalytic Cracking

Article Metrics: