RESUMO
The present study compares the optimization using Artificial Neural Networks (ANN) and Adaptive Network-based Fuzzy Inference System (ANFIS) in the sugarcane bagasse delignification process using Alkaline Hydrogen Peroxide (AHP). Two variables were assessed experimentally: temperature (25-45⯰C) and hydrogen peroxide concentration (1.5-7.5%(w/v)). The Klason Method was used to measure the amount of insoluble lignin, the High Performance Liquid Chromatography (HPLC) was used to determine the glucose and xylose concentrations and the Fourier Transform Infrared Spectroscopy (FT-IR) was applied to identify oxidized lignin structure in the samples. The analytical results were used for training and testing of ANN and ANFIS models. The statistical quality of the models was significant due to the low values of the errors indices (RMSE) and determination coefficient R2 between experimental and calculated values.
Assuntos
Celulose , Peróxido de Hidrogênio/química , Saccharum , Espectroscopia de Infravermelho com Transformada de FourierRESUMO
The present study examines the use of Artificial Neural Networks (ANN) as prediction and fault detection tools for the delignification process of sugarcane bagasse via hydrogen peroxide (H2O2). Experimental conditions varied from 25 to 45°C for temperature and from 1.5% to 7.5% (v/v) for H2O2 concentrations. Analytical results for the delignification were obtained by Fourier Transform Infrared (FT-IR) analysis and used for the ANN training and testing steps, allowing for the development of ANN models. The condition experimentally identified as the most suitable for the delignification process was of 25°C with 4.5% (v/v) H2O2, oxidizing 54% of total lignin. An ANN topology was selected for each proposed model, whose performance was evaluated by the correlation coefficient (R2) and error indices (MSE and SSE). The values obtained for R2 and the error indices indicated good agreements of the theoretical and actual data, of close to 1 and close to 0, respectively.