Two-plant direct heat integration with safety redundancy based on a bilevel algorithm. (31st August 2020)

• Record Type: Journal Article
• Authors: Tian, Yitong; Wang, Shihao; Liu, Kai; Li, Shaojun
• Is Part Of: Chemical engineering science. Volume 222(2020)
• Publication Date: 2020-08-31
• Subjects: Interplant heat integration -- Safety redundancy -- MINLP model -- Simultaneous optimization -- Shutdown risk -- Shapley value; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2020.115662
• Languages: English
• ISSNs: 0009-2509

Abstract

Highlights: The proposed DHISR model incorporates safety redundancy to decrease the risk loss in case of unplanned shutdown in interplant heat integration. The HEN superstructure model is optimized with a bilevel meta-heuristic algorithm that combines AEA and PSO algorithm. The cost-sharing strategy of a two-plant HEN uses an improved risk-based Shapley value that considers contributions and shutdown risks simultaneously. Abstract: Heat exchanger network (HEN) synthesis is important for energy savings and cost reduction. Current research on interplant heat integration has mainly focused on the heat recovery aspect, while ignoring the operating risk due to the unplanned shutdown of other plants. In this paper, a new two-plant direct heat integration model that considers safety redundancy (DHISR) is proposed to decrease the risk loss as far as possible. A bilevel meta-heuristic algorithm that consists of an evolutionary algorithm of pattern extraction (ALOPEX) and a particle swarm optimization algorithm is implemented to optimize the mixed-integer non-linear programming superstructure model. The financial benefits allocated to the plants in the heat integration alliance are determined by the improved two-plant risk-based Shapley value. The results of three examples show that the proposed approach achieves a good performance in the two-plant HEN synthesis and provides good cost distribution.