RT info:eu-repo/semantics/article
T1 Neural network modelling and prediction of an Anaerobic Filter Membrane Bioreactor
A1 Cámara Nebreda, José María
A1 Diez Blanco, Victorino
A1 Ramos Rodríguez, Cipriano
K1 AnMBR
K1 Filtration
K1 Neural networks
K1 Feed forward
K1 LSTM
K1 Preprocessing
K1 Retraining
K1 Prediction
K1 Ingeniería eléctrica
K1 Electric engineering
K1 Ingeniería química
K1 Chemical engineering
AB Anaerobic membrane bioreactors have become an environmentally friendly solution for wastewater treatment.The lack of sufficiently accurate mathematical procedures to model their behaviour and the fouling process ofthe membranes, poses a challenge when trying to optimise their energy consumption and maintenance costs.An accurate model of the fouling process of the membranes is critical to make the most of this technology. Thisis a perfect scenario in which to introduce neural networks (NN) as an alternative to mathematical modelling.However, the duration of the experiments and the difficulties in measuring some relevant variables, make ithard to collect high quality datasets to train the NN. Our goal is to obtain a good prediction of the foulingstatus of the membranes to enable an adjustment of operation conditions and maintenance procedures aheadin time. To do so we must obtain high quality datasets to train our neural networks. The combination of staticand dynamic networks enables us to leverage the best prediction capabilities of each one. This combinationrequires a preprocessing of the datasets that separates trends from oscillations. The outputs obtained need tobe put together to build up the predicted evolution of fouling. Accurate predictions are then extended from25 to up to 75 filtration cycles. To maintain and even extend accuracy after sudden changes in operatingconditions, retraining the NN every 25 cycles is proposed. AI based real time predictions open a new scopefor decision making, and optimisation in the field of anaerobic membrane reactors.
PB Elsevier
SN 0952-1976
YR 2023
FD 2023-02
LK http://hdl.handle.net/10259/7435
UL http://hdl.handle.net/10259/7435
LA eng
NO This work is part of the project TED2021-132393B-I00, funded by the MCIN/AEI/10.13039/501100011033 and the European Union ‘‘NextGenerationEU’’/PRTR.
DS Repositorio Institucional de la Universidad de Burgos
RD 03-may-2024