Structural Assessment of High Temperature Solar Central Receiver Tubes

Abstract

In 2015 the Advanced Reactor Technologies (ART) program was formed to consolidate Department of Energy (DOE) programs in next generation nuclear plant, small modular reactors, and advanced reactor concepts. Some concepts explore temperatures up to 1000 ◦C, just as parallel DOE programs in concentrating solar power (CSP) such as CSP Gen3: Liquid-Phase Pathway to SunShot (Gen3L)— and closer to home the Australian Solar Thermal Research Institute (ASTRI)—are trying to reach temperatures as high as 750 ◦C to improve system efficiency. The ART program is currently developing unified constitutive material models for appendage to the American Society of Mechanical Engineers Boiler and Pressure Vessel Code (ASME B&PVC), to ensure that conservative yet competitive designs prevail. The constantly changing conditions a solar central receiver (SCR) is exposed to differs considerably from the gradual and infrequent changes a nuclear or other large thermal power plant experiences. While much knowledge concerning metals and their alloys at elevated temperature can be learnt from boiler codes and the research from whence they derive, vital experience is missing in the study of structural integrity in SCRs—a multitudinous design and erection of plants, successful long term commercial operation, the adjustment of prior predictions with real time operational data—and that gap only widens in the strive for advanced temperature systems. So while the boiler codes can serve as a welcome guide moving forward, the rules as they apply to historically slow-moving inertial machines are probably not universally applicable in the design of advanced solar central receiver (ASCR). While some conservatism might be removed if applying nuclear design methodology, knowing where and by how much is not clear. The work of this PhD followed closely the ASTRI high temperature receiver design group’s work on utilising liquid sodium as a heat transfer medium. Following a review of the current state of knowledge, three chapters follow the journey from classical analytical thermoelastic analysis, and through unified viscoplastic constitutive modelling, to arrive at predictions for high temperature and high flux density ASCR operation. Early work focused on a classical approach to elastic thermal stress in concentrating solar thermal (CST) receiver tubes. Such implementation is computationally fast for integration into plant optimisation software suites, however, once questions regarding cyclic plasticity and creep arise this technique becomes intractable, owing to a reliance on assumptions in simplification. Notwithstanding, early predictions of maximum allowable flux density (AFD) made extensive use of simple and conservative boiler code methodology. A second major study was made of the Bree archetype as it applies to CST service. The Bree archetype served as the basis for ASME B&PVC design against ratcheting, and some literature from last century claims that this failure mechanism need not be addressed in the case of SCRs. Other more recent evidence predicts that ratcheting becomes more pronounced at advanced temperatures, so much so that simple inelastic procedures lose some of their inherent conservatism. A unified constitutive model developed in the ART program for stainless steel UNS S31609 high-carbon content austenitic stainless steel (316H) is shown to be overly conservative, but modifications exist to correct for this, and one of them is used to indicate that ratcheting in liquid cooled ASCRs should not be a concern, maybe at temperatures as high as 750 ◦C. This has a bearing on the design of components such as heat exchangers where stainless steels are considered for their cost. The third and final contribution was an inelastic lifetime assessment of the Gen3L ASCR employing age-hardened superalloy 740H. It is shown to meet commercial lifetime requirements (30 years or more) by a considerable margin with a “best estimate” linear damage summation (LDS) creep-fatigue assessment, but if safety factors are considered to account for uncertainties in the heat of material, inelastic finite element modelling (FEM) model used, the non-idealised thermal and structural loads, and the actual as-built geometry, the lifetime of the receiver panels could be as low as 15 years. Whereas an analysis without safety factors predicted that the risk of fatigue was so low as to not exist, inclusion of safety factors meant that failure by creep-fatigue interaction becomes possible. Full panel analysis confirms that in the structural assessment of SCR tubes, the generalised plane-strain (GPS) simplification is most satisfactory in describing the state of the tube. Show more