Standard

ASME VVUQ 30.1

Publisert
For bestilling og priser av dette produktet ta kontakt med salg@standard.no

Omfang

This Standard is focused on the scaling analysis that is used to evaluate the effects of differences (e.g., distortions) in the phenomenological behavior of experimental facilities compared to the phenomenological behavior of the real-world system. This includes scaling analysis methodologies for supporting the design of facilities and experiments capable of generating data that characterize the phenomena present in an entire system [such facilities are known as integral effects test (IET) facilities] and in components of the system (e.g., the nuclear core or the steam generator) [such facilities are known as separate effects test (SET) facilities]. Although this best-practice Standard is focused on nuclear system applications, many portions of the methods and techniques discussed here can be applied to other engineering systems such as in chemical processing, oil and gas production, and power generation systems based on other fuel sources. Purpose When determining the credibility of a model, a key question is what the accuracy of the computational model is for the real-world conditions where the system will operate. This accuracy is called predicative capability and is often based on the model validation. To estimate the model’s predictive capability, first the error of the model needs to be determined under conditions where empirical data is available. This is referred to as the validation error. Often, based on the similarity of the test facilities and real-world systems, the validation error is used as an estimate of the model’s error when making predictions on the real-world system. Thus, a key assumption is that the model’s predictive capability of the real-world system is similar to the model’s accuracy in predicting the empirical (experimental) data. If both systems have similar physical behavior, it is expected that the model’s accuracy will be similar in both systems (the real-world system and the experiment). There can be many reasons why the model’s validation error may be very different from the model’s predictive capability. While experimentalists strive to ensure that the experiment is similar to the real-world system, some sacrifices often need to be made. For example, due to the large size and inherent complexity, experimental facilities used to provide data to validate models for nuclear power plant scenarios often must be scaled down from the true nuclear power plant dimensions and operational conditions (such as pressure, temperature, and flow rates). This may include operating the experiment at lower powers and pressures, at a reduced size, or using other fluid. While these changes may not directly impact the model validation (since validation is based on the comparison of the empirical data to the model’s predictions), these changes certainly impact the applicability of the model for the real-world system. For example, if a specific system was influenced by behavior that was sensitive to a characteristic length (e.g., hydraulic diameter), area (e.g., flow area), and volume, the scaled system (e.g., experiment) could not be scaled in all three values at once. Consider liquid flow through a tube. If the diameter is reduced by a factor of 2, the flow cross-section area and volume must be reduced by a factor of 4 while the wall heat transfer area is still decreased (as diameter) by a factor of 2. Thus, a phenomenon such as boiling, in which all of these geometry factors could be important, requires a method to determine if the scaled system can provide useful data, or if the scaled system is not similar to the particular scenario in the real-world system. In nuclear thermal fluid systems, the relevance of the empirical (experimental) data to the real-world system is determined through scaling analysis. Scaling is not focused on how well the computational model predicts the empirical data (i.e., validation). Instead, scaling is focused on if a model validated with the empirical data will be relevant to the real-world system. In other words, scaling formalizes the connection between the test facility and real-world system. This Standard provides practices and procedures for determining if experimental data (used to validate models) is applicable to the real-world system. Historically, such analysis has been unique for nuclear reactor applications where conditions of fluid, both single- and two-phase, are highly size dependent due to surface-to-volume ratio, size-dependent interfacial shape (flow regimes), and interfacial area density. However, it is hoped that the presented scaling analyses methodologies developed for the nuclear community can be used to benefit other fields of engineering and science or combined with other methodologies already developed.

Dokumentinformasjon

  • Standard fra ASME_AC
  • Publisert:
  • Versjon: 0
  • Varetype: IS