Page 315 - Sustainability in the Process Industry Integration and Optimization

292 Cha p te r T w e l v e

many publications and projects where this rule was not followed.) It
is better to use the anticipated average energy price for the life span
of the plant or, in the case of a retrofit, for the payback period. The
problem then becomes one of estimating this future energy price for
periods that may be as long as five or ten years. It has been shown
(see, e.g., Klemeš and Bulatov, 2001; Donnelly, Klemeš, and Perry,
2005) that even the forecasts of highly qualified experts are frequently
inaccurate. One of the obvious potential approaches is to use the
scenarios and target the most flexible design that would provide a
balanced optimum for various situations.

12.2 Integration of Renewables: Fluctuating
Demand and Supply

The integration of renewable energy sources was discussed in
Chapter 6. Renewable resources are usually available on a smaller
scale and are often distributed over a certain region. Their availability
(with the exception of biomass) varies significantly with time and
location. This variability is due to changing weather and geographic
conditions. The energy demands (heating, cooling, and power) of
sites also vary significantly with time of the day and period of the
year. These variations in the supply and demand of renewables can
be predicted in part, and some of the variation is fairly regular—for
instance, day versus night in predominantly cloudless areas for solar
energy. The availability of wind-generated energy can be less
predictable. One approach to dealing with these problems is the
advanced PI technique that employs time as an additional problem
dimension. A basic methodology along these lines (involving “Time
Slices” and Time Average Composite Curves) was developed for the
Heat Integration of batch processes (Klemeš et al., 1994; Kemp and
Deakin, 1998). This methodology was recently revisited by Foo,
Chew, and Lee (2008).
This methodology has also been extended to the Heat Integration
of renewables. Important steps in this direction were reported by
Perry, Klemeš, and Bulatov (2008) and Varbanov and Klemeš (2010).
Dealing with variation and fluctuation brought another complexity
into data extraction. Data should be collected for all time slices that
increase the complexity. Especially, for each case, it is necessary to
choose the time horizon for the analysis and number of time slices.
This is a fast-developing field, so it is advisable to monitor recently
published research papers and conference presentations.

12.3 Steady-State and Dynamic Performance

It has been assumed that all analyzed and optimized processes
operate in a steady state. Many industrial processes do operate in this