Abstract
A significant reduction in energy consumption can be achieved by applying Process Integration methodology. More sustainable process solutions can be achieved both environmentally friendlier (reducing the emissions and pollution) and economically more attractive (reducing utility cost). An improved process design can be achieved by an appropriate trade-off between investment and operating cost. However, the trade-off can only be established correctly if it reflects the future variability, and even unpredictability, of prices for utilities, raw material and products. The objective of this work was to optimise process design for full process lifetime by considering some provisions for future price fluctuations.
Most process synthesis models are single-period models that only consider only fixed cost coefficients. However, prices are fluctuating rather quickly and an optimal process design obtained for one year can be different for another. This work focused on the separation processes as they consume a significant amount of energy. The synthesis of a distillation column sequence integrated with its heat exchanger network was used as a case study for the separation of a multi-component stream into pure component products. In order to consider future price fluctuations, a multi-period mixed-integer nonlinear (MINLP) model was developed. Different projections regarding utility prices based on past prices were derived at due to the uncertainties of forecasting. Maximisation of Net Present Value was chosen as an optimisation criterion, in order to account for future price fluctuations and the time-value of money over a full lifetime.
Trade-off between investment and operating cost, and their distribution between the heat exchanger network (HEN) and the distillation columns were evaluated in the following step. Significant utility savings and a reduction in operational cost can be achieved in this way when compared to those cases where empirical actual-to-minimal ratios within the range of 1.2 - 1.5 is used, especially when future utility prices are considered.
The solution obtained by this novel optimisation methodology seems to be more robust compared to the conventional single–period optimisation as it reflects any forecasted future price fluctuations. Consequently, the probability of meeting the optimal design for each year over a full lifetime of the separation process is thus higher.
