To be truly effective, a program must address the four key areas of lifecycle energy management — design, operations, automation and maintenance (Figure 1). It must aim to:
• design the plant for efficiency;
• operate intelligently to keep efficiency high;
• control processes to reduce energy consumption; and
• maintain equipment at design performance.
These activities involve different people within the organization and different time frames. Design efforts generally occur only during the initial engineering phase or during subsequent capital projects. Operations and control activities are on-going; best-managed sites continually strive to improve efficiency. Maintenance takes place periodically, often during plant shutdowns but sometimes, depending on the equipment, while the plant is running. Table 1 lists some resources processors should provide in these four areas.
Decisions made during the design phase to a large extent determine the ultimate energy efficiency of a process unit. Choices, e.g., of technology to license, major equipment, heat recovery operations, etc., will have lasting effects throughout the life of the unit and typically constrain the maximum efficiency achievable once the unit is operating.
Evaluation of various options will consume time and budget; so among the most important steps during the design phase is to set aside resources so energy efficiency can be considered early in the process. Waiting until late in the design stage after most major decisions have been made often only results in a list of changes that would have been beneficial but can't be implemented without significant delays in the overall project schedule.
A common technique employed in process design for improving energy efficiency is pinch analysis. Typically a resulting solution will require additional heat exchange surface area, which normally will increase pressure drop. However, a creative and thorough analysis may yield improved hydraulics as well as better heat recovery. Consider Figures 2 and 3, which depict a crude unit preheat train. The original design (Figure 2) has a higher overall pressure drop for the crude feed path than the revised design (Figure 3), even though the latter includes new exchanger shells in the path. The changes allowed the unit to increase maximum crude throughput by 10% without changes to major equipment; the capital required was paid back in less than three years based solely on the energy savings.
Finally, even in relatively small projects such as turnarounds, it can be worthwhile to conduct an engineering review of the process during the planning stage to search for ways to save energy. An energy loss analysis is one assessment that process engineers not trained in pinch analysis can perform relatively quickly. It simply involves cataloging the energy lost to either cooling utilities or to surroundings in the unit, and heuristically identifying easy ways to capture some of that heat. Typical energy loss points are air or water coolers, furnace stacks and even insulation.
MONITORING OPERATIONS BETTER
Process operators are tasked with running their units safely while meeting production targets for throughput and quality. They also can address and impact energy efficiency. A necessary first step is to measure and report energy consumption in detail and in real time — however, in many cases, this alone doesn't suffice. When throughput rates, feed properties or product qualities highly vary, it's critical to provide energy consumption targets that account for current process conditions so operators can spot opportunities to improve. Figures 4 and 5 illustrate the importance of this point.
Figure 4 tracks steam consumption and charge rate for a refinery's fluid catalytic cracking unit. Ovals highlight several events of interest. In the first, steam consumption increased for a short time with no change in unit throughput. In the second, steam consumption rose with feed rate, as would be expected. Monitoring steam consumption alone doesn't provide guidance to operators about which changes are opportunities for improvement.
Adding intelligent targeting information for the unit (Figure 5) makes the picture clearer. The steam consumption targets, shown in blue, depend upon throughput, feed quality and reaction conversion levels. The excursion in steam consumption during the first event is worse than indicated when only considering throughput because the target energy actually decreased during the event due to process changes other than the throughput. In the second, although the steam consumption is expected to rise with feed rate, it increased more than it should have because the operation wasn't executed carefully. The situation was corrected shortly after the feed rate increase to bring the actual consumption again in line with the target rate.
Real-time monitoring and reporting of energy consumption will allow operators to make moves that result in energy conservation only if the following conditions are met:
• The energy consumption must be tracked in detail — for each energy/utility type, at least down to the process unit level.
• Tracking and reporting must take place in real time so responses can be made as soon as a change in performance occurs.
• The energy must be put into context by comparing actual consumption to a target that accounts for process conditions such as throughput, feed and product quality, conversion levels, etc.
USING AUTOMATION MORE EFFECTIVELY
One of the simplest ways to address energy efficiency through automation is to routinely review the performance of regulatory control loops that affect energy consumption. Examples are steam/oil ratio controls in oil-fired furnaces, steam pressure controls for sub-headers or individual heat exchangers, plant air header pressure controls, etc. Often, operations doesn't flag these loops for attention from process control staff as long as their performance doesn't impact production rate or final quality. Identifying such loops and periodically focusing on them frequently can improve energy efficiency without affecting the overall process. Routinely checking for poor controller tuning, valve stiction and transmitter problems can yield benefits for many plants.
Figure 6 shows a capture from an automated performance evaluation of proportional-integral-derivative loops at one site. The program identified several loops (highlighted with a red background) as having poor performance. Prior to the automated analysis, neither operations nor controls personnel had flagged any of the loops because their poor performance wasn't causing any noticeable problems in the unit. Addressing only one of those loops, PDIC-1318, to improve its performance, reduced steam consumption by over 1,000 kg/hr.
Another activity that can achieve lasting benefits is to review the control structure for ways to reduce energy consumption while meeting all processing objectives. For example, many sites now tie distillation-column reflux flow rate to the feed flow rate instead of holding reflux flow constant. A similar control strategy for the reboiler duty, varying heat input with column throughput, may suit some columns. Operations and process control staff can work together to identify such opportunities at most sites.
Keeping assets, especially rotating equipment and heat transfer units, in peak condition is crucial for energy efficiency. Often it's possible to perform important cleaning activities while the plant is in operation rather than waiting for the next shutdown or turnaround. The question then becomes when is the right time?
Changing stream properties and flow rates can make it difficult to know the underlying performance of exchangers. Simply tracking the overall heat transfer coefficient isn't sufficient, as the trend chart from an operating process (Figure 7) shows. Relatively sharp changes in the UActual value didn't indicate rapid fouling, a conclusion borne out by comparing the changes during the same period of the UExpected value, which compensates for the process variations, and the calculated fouling factor, which displays the ultimate performance of the exchanger on the right-hand y-axis.
Many installations now rely on real-time monitoring to track in detail the underlying condition of critical heat exchangers in fouling services. This very effectively identifies when cleaning is cost-effective.
Tracking allowed for cleaning operations to be scheduled at the right times, as evident from the sharp drops in fouling factor at the beginning and again at the end of the time period shown in Figure 7.
AN INDIVIDUAL ROLE
Just as each person and job function at a process plant plays a part in ensuring production targets are achieved safely, everyone has a similar opportunity to impact energy efficiency. However, people must understand their roles, get objectives and instructions on how to achieve them, along with the required time and resources.
A comprehensive energy management program will address efficiency in all stages of the plant's lifecycle and for each role. It can provide significant and lasting improvements.JESUS VALLEJO is performance service director for Honeywell Process Solutions in Madrid, Spain. E-mail him at[email protected].