Optimal time-dependent operation of seawater reverse osmosis

The operation of seawater reverse osmosis (SWRO) plants is systematically optimized with the objective to reduce the electricity charges, which constitute the largest portion of the operation costs. Variable operating conditions and a time-of-use electricity tariff rate schedule are considered. A well-established model is extended to model a reverse osmosis (RO) system equipped with highly efficient pressure exchangers (PXs) and variable frequency drives (VFDs). All major performance factors are considered, including membrane's permeability, concentration polarization, temperature effects, membrane fouling, and efficiency dependency of the VFD on the loading conditions. The proposed formulation minimizes the electricity cost per day, given a desired daily production rate. The problem is formulated as a mixed-integer nonlinear program (MINLP), allowing for periods without operation. The time domain (24 h) is discretized into 0.5 h increments, and a pseudo steady-state assumption is used. The optimal operation plan resulting in maximum electricity cost-savings is established with standard global optimization tools. The results show significant electricity and production cost-saving potentials. More savings in electricity can be achieved by oversizing the plant, thus allowing shutting-off at periods of high electricity cost. The cost-saving potentials are particularly promising in electricity markets in which the electricity rate schedule is widely distributed.