Artificial neural networks applied to manage the variable operation of a simple seawater reverse osmosis plant

Smart renewable energy penetration strategies on islands: The case of Gran Canaria

“…Therefore, electricity is also the main driver of the island's water supply. Electricity consumption for desalination, with a high flexibility potential [49,50], represents approximately 10% of total electricity demand [48]. Overall, the most notable conclusions that can be drawn from the analysis of the current energy system of Gran Canaria, taking into account the particularities of the different agents involved and with a special focus on the unexploited flexibility potential of each sector, are as follows:…”

“…However, given that the number of vehicles in use in Gran Canaria is very high [43] and given the greater efficiency of electric vehicles, it is assumed that the transition from fossil-fuelled to electric vehicles (eV) in the transport sector is potentially beneficial. A considerable percentage of the electricity demand is for water desalination (51%), fundamentally using reverse osmosis processes which have a high flexibility potential [50,51] that could allow the implementation of smart strategies to adapt this electricity demand to the intermittent nature of renewable energy generation. With 517 people/km 2 [43], the population density of Gran Canaria is high.…”

“…As previously discussed in section 3.1, the desalination sector is one of the most important sectors in the island of Gran Canaria. Some authors have demonstrated the flexible capacity of reverse osmosis desalination systems when sophisticated control techniques are implemented [49,50]. This part of the analysis concentrates on the effects of managing all desalination plants as flexible demand.…”

A review of the artificial neural network based modelling and simulation approaches applied to optimize reverse osmosis desalination techniques

“…The mean square error for total flux (0.00009 and 0.00012), for separation factor (0.00009 and 0.00016) and for pure solvent flux (0.00013 and 0.00013) were observed. Cabrera et al [31] presented models for handling the process of a small-scale prototype SWRO desalination plant (using ANNs as control strategy) located at Gran Canaria, Spain. The work helps them manage varying available electrical power.…”

“…They included various controllers such as proportional integral derivative (PID) and fuzzy-logic based systems to improvise the process control. Cabrera et al [31] well defined the use of control strategy in their research, managing the operation using ANNs for a small scale SWRO desalination plant located at Gran Canaria, Spain.…”