Numerical optimization of combined heat and power Organic Rankine Cycles – Part B: Simultaneous design & part-load optimization. (October 2015)

• Record Type: Journal Article
• Authors: Capra, Federico; Martelli, Emanuele
• Is Part Of: Energy. Volume 90:Part 1(2015)
• Page Start: 329
• Page End: 343
• Publication Date: 2015-10
• Subjects: Control strategy -- Off-design model -- Optimal operation -- SQP -- Black-box optimization; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2015.06.113
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: This two-part paper proposes an approach based on state-of-the-art numerical optimization methods for simultaneously determining the most profitable design solution and part-load operation of Combined Heat and Power Organic Rankine Cycles. Compared to the usual design practice, the important advantage of such an approach is to consider the part-load performance of the Organic Rankine Cycle at the design stage. In this second part (Part B), first the part-load model and optimization algorithm are presented and tested. Then, the part-load optimization algorithm is combined with the design optimization algorithm proposed in Part A, and the simultaneous approach is applied to a real-world test case. Computational results show that, with respect to a design optimized for full-load equivalent hours, the solution optimized for part-load operation has a lower investment cost, better part-load efficiencies, and thus up to 22% higher annual profit. The optimized part-load control strategy uses the sliding-pressure mode in the high-load range, and a combination of sliding-pressure and throttling in the low-load range. Highlights: Simultaneous optimization of design and part-load control of Organic Rankine Cycles. Part-load models formulated as a nonlinear optimization problem. Part-load optimization problem tackled with Sequential Quadratic Programming. The optimal part-load control strategy using sliding-pressure and throttling is found. The design & part-load optimization algorithm leads to considerable gains of profit.