Development of expeditious process integration methods for retrofit of non-energy-intensive industries

Abstract

Industry is expected to play a prominent role in decarbonising the energy system towards reaching the EU climate targets. This requires retrofitting existing process plants with the goal of reducing their energy use. Process integration methods proved to be highly effective in analysing the energy utilisation of industrial facilities and identifying possible actions for increasing their energy efficiency. However, they are far from constituting the industrial practice. A major barrier to their use is the large time and resources required for performing the analysis. This issue is especially felt in non-energy-intensive industries, which are deemed to hide a large potential for energy savings. In fact, the low cost savings deriving from energy-efficiency projects in individual plants do not justify lengthy and expensive investigations. In this way, a large potential for energy saving which lays in non-obvious solutions is missed. This thesis aims at lowering this barrier by developing expeditious process integration retrofit methods. The most time-consuming activities of available methods were identified, and two novel methods were proposed, named “Required Data Reduction Analysis” (RDRA) and “Energy-Saving Decomposition” (ESD) method. They respectively aim at reducing the time consumption of the “data acquisition” and of the “design” phases of process integration retrofit projects. Their performance was tested and validated by applying them to nine case studies belonging to three different industrial sectors. This allowed to investigate their major merits and limitations and propose future development activities. The RDRA bases on the idea that measurements are performed to increase our knowledge, and this knowledge is quantifiable in terms of uncertainty. It employs uncertainty analysis, sensitivity analysis, and mathematical optimisation techniques in a systematic setting, to identify (i) a limited number of process parameters to measure, and (ii) the maximum acceptable uncertainty in their measurement. The results of its application to five case studies testify that it can significantly decrease the amount of parameters to measure, compared to what traditionally recommended. In all the cases, the parameters to measure were more than halved, arriving at a maximum reduction of 86 %. Moreover, the method proved to be robust with respect to the assumptions required, and flexible in the scope of the analysis. It showed the potential to be employed combined to other energy analysis tools, and for designing industrial energy monitoring systems. The ESD method builds on the idea that most of the energy-saving potential achievable by heat integration in existing industrial plants resides in a limited number of the pro- cess streams and of the “pinch violations”. It aims at reducing the solution space before embarking in the time-consuming design activities, avoiding to waste time in inspecting unfruitful solutions. This is achieved by two consecutive simplifications. The first identifies and eliminates useless process streams based on their energy-saving potential. The latter disregards heat exchangers responsible for cross-pinch heat transfer based on economic considerations. The application of the method to nine case studies proved that it can significantly reduce the size of the problem and the time employed in formulating profitable design proposals. The reduction of considered process streams ranged from 33 % to 78 % compared to the total plant, and the reduction in “pinch violations” from 66 % to 88 %. Moreover, the comparison with six state-of-the-art process integration methods showed that this faster analysis allowed to identify the same or even a better solution of other tools, paying no price in terms of rigorousness. All in all, the novel methods showed a high potential to lower the barriers to the use of process integration tools in the industry, potentially providing access to energy-saving opportunities today hidden.