Brazil leads the world in sugarcane production, consequently produces also the greatest amount of sugarcane bagasse. Currently, this sugarcane bagasse is leveraged for power generation in the mills, but this biomass still has a potential for production of others value-added compounds such as the second-generation ethanol, lactic acid, butanediol and etc. The present work was carried out in order to study the efficiency of the delignification process of sugarcane bagasse with alkaline hydrogen peroxide. Two variable were assessed experimentally: temperature (25 Celsius degrees - 45 Celsius degrees) and H2O2 concentration (1.5 percent -7.5 percent) at pH 11.5 for 1 h in an orbital shaker at 100 rpm. The Klason Method was used to measure concentration of extracted lignin, HPLC was used to measure the concentration of glucose and xylose and FT-IR analysis was applied to identify lignin structure in the samples. The Klason analysis indicated the 45 Celsius degrees/7,5 percent as the optimum condition with 75,4 percent of the lignin solubilidized, the glucose analysis indicated 45 Celsius degrees/7,5 percent as the optimum condition with a concentration of 1,66 g/L, the xylose analysis indicated 25 Celsius degrees/7,5 percent as the optimum condition with a concentration of 0,82 g/L, and the FT-IR analysis indicated 25 Celsius degrees/1,5 percent as the optimum condition with 66,9 percent of the lignin oxidized. For each analysis, an ANN model was proposed. The network of the Klason analysis had a trainlm/logsig/4 topology with SSE 0,00723 and R2 0,995, the network of the glucose analysis had a trainlm/logsig/4 topology with SSE 0,0328 and R2 0,97384, the network of the xylose analysis had a trainlm/logsig/5 topology with SSE 0,289 and R2 0,87441, the network of the FT-IR analysis had a trainlm/logsig/5 topology with SSE 0,0316 and R2 0,98414.