Genetic Algorithm Layout Optimisation for Small Modular Reactors Balance of Plant
摘要
Background. Off-Site Modular Construction (OSMC) can decrease direct costs by 20–80% (Mignacca et al. in 26th International conference on nuclear engineering. ASME, London, 2018 [1]). Utilizing OSMC presents exponentially more options for plant designers, along with the configurations presented by cogeneration options such as desalination, residential/ industrial heat, cogeneration, hydrogen production. Objective. It has therefore been identified that a tool to help designers navigate these challenges could be beneficial. The aims are to demonstrate the tools can effectively: (1) Reduce pipe costs. (2) Reduce time to find a viable concept design. (3) Explore concepts not previously considered. Methodology. This study outlines a genetic algorithm model developed in MATLAB. Using data from a Westinghouse 3 loop plant. (1) Plant equipment sizes are first assessed for off site transport. (2) A Design Structure Matrix is developed for the plant systems network flow diagrams and coefficients assigned for cost. (3) A model is outlined for reduction of pipe costs including non overlapping constraints. (4) The resulting output is assessed by design engineers for feasibility. (5) Further constraints can then be added to the model. Main findings and their potential applications. The method assessed configurations in 3 and 5 m in both latitudinally and longitudinally stacked module configurations and showed up to a 58.09% reduction in piping costs. With over 50 SMRs in development worldwide, it has therefore been identified as possibly increasing the efficiency and effectiveness of the design process, especially in the concept design phase. However, the disadvantage is that more upfront work is required to convert engineering knowledge into mathematical terms and relationships.