Decomposition And Implementation Of Large-scale Interplant Heat Integration

With the continuous improvement of energy-saving and emission-reduction requirements,the scale of the object systems faced and handled by energy-saving work continues to expand.The mathematical programming method which continues to develop with the continuous improvement of computer solving capabilities,and the wellestablished Pinch technology,provide two powerful tools for the integration and optimization of process systems including energy system.For the interplant heat integration(IPHI)problem,the approaches for extracting the hot and cold streams participated in IPHI,selecting the specific implementation way for IPHI,and the objective function and mathematical model under different scenarios,are worthy of further study.In addition,for a large-scale IPHI problem,how to simplify the problem or decompose it to reduce the solving difficulty while making the final IPHI scheme feasible,is an issue that needs in-depth research.In the previous studies,the three methods for extracting hot and cold streams participated in IPHI have their own pros and cons.However,the quality of streams actually affects the potential of the interplant heat recovery.This study proposed a simple but effective and new method for extracting streams participated in IPHI from each heat source plant and heat sink plant.Compared to the direct use of the waste heat sources/sinks from the existing heat exchanger network(HEN)of each plant,the feasible heat sources/sinks extracted by the new method are of higher grade while the existing HEN design is almost kept unchanged.The new method also avoids complex retrofits or new design of the HEN when using the Grand Composite Curve(GCC)of each plant for analysis to extract surplus heat sources/ sinks.It is difficult to solve a large-scale IPHI problem because of its huge amount of calculation.The complexity of a single large-scale IPHI scheme is also a huge challenge for its control and safety.Most of the previous studies used a step-wise strategy or directly used a very small number of participating streams to reduce the difficulty of solving a large-scale IPHI problem.The step-wise strategy usually pre-targets some potentials and corresponding parameters,followed by a specific network design.This study proposed a new screening algorithm named "NLQSA".By using this screening algorithm,a largescale IPHI problem including multiple plants is divided into several smaller sections with each section including two or three plants.The result obtained by using NLQSA is equal to or close to the theoretical maximum interplant heat recovery potential.Furthermore,NLQSA greatly reduces the difficulty of solving large-scale problems.The final solution is also more practical and feasible than the previous ones.Based on the previously proposed solution strategies and steps,a new stage-wise superstructure and a mathematical model were also presented for the design of interplant HEN.It is a Mixed Integer Nonlinear Programming(MINLP)model that includes effects of various aspects such as interplant distances,pipelines,pumps,heat exchangers,and multi-level utility etc.The interplant HEN configuration with minimum cost can be obtained by using this MINLP model.The interplant connection pattern affects the interplant energy recovery potential and the total annual cost(TAC).Different connection patterns have different performances in the stability and reliability of the interplant HEN.However,the previous research in this area is still relatively few.This study analyzed the advantages of parallel connection pattern over other connection patterns and proposed a comprehensive framework to determine the indirect IPHI with parallel connection pattern.This method can identify the real maximum interplant heat recovery potential while minimizing the corresponding intermedium flow rates.