西班牙公共工程研究与实验中心水动力学实验室主任Luis Balairon Perez博士做《The role of physical and numerical modeling in experimental hydraulics in Spain and Europe》的学术报告

Abstract: Hydraulic studies have been traditionally undertaken with physical models (which reproduce flow phenomena at reduced scale with dynamic similarity) and with empirical simplified assumptions. Physical modelling is expensive and time consuming process and certain assumptions derived from empirical studies are not appropriate for specific case problems in hydraulic.

There is a long experience in physical scale modelling along the last 100 years (in Spain, the Hydraulics Laboratory of CEDEX has been building and testing physical scale models since its commissioning in 1928, having developed among these period more than 300 studies of dams), but as computing power is increasing, the more recent numerical models are offering an alternative. However for large complex problems, some physical phenomena cannot be accurately predicted through CFD.

Advances in computer processing power enabled important improvements over the recent decades and a range of numerical approaches have been proposed to model free surface flows over complex structures. Such numerical models can determine data such as pressure, velocity, flow depth, and streamlines at any and all locations in the flow domain. The data can also be archived for later use and analysis.

Numerical models come in a wide range of shapes and flavours: one, two or three dimensions, steady or unsteady flow conditions, hydrodynamic and sedimentation models, etc. There are also commercial models, open source models and models developed by research centres for its proper use. Models can be classified also into four approaches: the finite difference method, the finite volume method, the finite element method and the boundary element method.

The credibility of the results of the numerical models depends on the extent of their validations with field or laboratory data. In most cases, the collection of field data is quite costly and time consuming, which makes the use of laboratory data a more attractive option for model validation.

Physical and numerical modelling are two efficient analysis approaches in hydraulic engineering that may be combined in a unique technique called Composite or Hybrid modelling. This strategy combines the benefits of physical modelling with the flexibility of numerical modelling to create a valuable tool for engineers designing new hydraulic structures or improving existing ones. There are different types of composite or hybrid modelling, such as the next:  Model Nesting; Design of physical model; Physical model representation of one element of the system or Modelling the model.

Although the use of numerical models in the design of hydraulic structures has increased very much in the last years, there are also many specific types of hydraulic structures that are also difficult to be modelled properly exclusively with CFDs methods. For complex or particular flow analysis, CFDs still often have lacks in representativeness as they are directly dependent of the mathematical model used to idealize the physical processes occurring in real flow, or their computation costs are prohibitive for practical case studies when fully representative mathematical models are solved.

For instance, one of the difficulties in solving flow numerically over a weir is the presence of a free surface that tends to be transient in nature (changing with time). This is especially difficult when the water surface is rapidly changing with a high degree of curvature, such as when the flow changes from subcritical flow to supercritical flow.

So, in today’s world of computational fluid dynamics (CFD), there are some applications where physical models remain as the only analysis tool that can provide reliable engineering design results. Some examples of hydraulic projects that are difficult to evaluate with computational methods are the next:

-  Spillways (including Stepped spillways, Labyrinth and Piano Key spillways and in general spillways with a complex geometry.

-  Stilling basin performance, including scour downstream of stilling basins.

-  Pump intakes.

-  Vortex drop shafts structures.

-  Lock filling and emptying systems