Evaluation of the linear and non-linear prediction models optimized with metaheuristics: application to anaerobic digestion processes.

摘要: Abstract This research represents an evaluation study of the linear and non-linear mathematical methods applied to predict biogas flow rate in anaerobic digestion processes. The model No.1 was used generate process data. For prediction partially least squares regression, locally weighted regression artificial neural networks were used. Two metaheuristic tools, here a genetic algorithm ant colony optimization improve models. They carried out variable selection procedure. implemented models could successfully perform rate. Nevertheless, more robust accurate done with help networks. Here error about 9% while coefficient determination reached 0.97.