3rd Workshop of Pivnet task 5 - ERCOFTAC SIG 32

The aim of this series of Workshop is to enhance cooperative activity and exchange of information between the participating teams around the validation and the improvement of PIV.

For that purpose, a list of topics of interest has been identified by the organizing committee and a questionnaire was distributed. The first workshop took place in Lisbon in 1998. It was mostly organized as round table discussions to discuss topics of cooperation. The interest was demonstrated by the fact that 65 participants were registered at this Workshop.

The second Workshop took place in Rome University la Sapienza in September 1999. This second Workshop was organized in two parts. In the first part, two invited lectures of general interest were given. In the second part, four sessions were run on various subjects of interest concerning both the development and the application of the method. In this workshop, a proposal was raised by Dr Okamoto from Tokio University (Japan) to organize a competition workshop on some up to date PIV problems. The participation to this workshop was of 48 people from both research and industry.

The third workshop was organized on the same basis as the second one. Two invited lectures were presented. The first one by Dr Coltuc from ICPE Bucharest on interpolation and super resolution and the second one by Mr Kähler from DLR Göttingen on advanced stereoscopic PIV set-ups. Five sessions were devoted to the different topics identified in the first workshop. They showed that PIV is now a well established technique and that significant work is going on in several European teams on advanced analysis algorithms. Developments goes on now on stereo PIV and on combined measurements with other non intrusive techniques (PIV & LIF, PIV & DGV ...)

Two round table discussions were run during the workshop. One on coherent structure assessment with PIV and the second on the organization of the PIV challenge proposed by Dr Okamoto in Rome.

Concerning the assessment of coherent structures, the main conclusions are :

• PIV is a unique tool to assess coherent structures in real turbulent flows as it is a spatial technique,

• accuracy and dynamic range has to be checked carefully and improved, as the full spectral range of a real turbulent flow cannot generally be covered by a single PIV record at relevant Reynolds numbers,
• each type of coherent structure has to be precisely defined mathematically and the definitions and criteria used in numerical simulations have to be adapted, as PIV is presently 2D2C or at best 2D3C,
• Proper Orthogonal Decomposition and wavelets analysis appear as promising tools for coherent structure extraction and should be investigated further.

Concerning the PIV challenge, which will take place from September 2000 to September 2001, it was agreed that Europe will participate through PivNet Task 5 and ERCOFTAC SIG 32. Pr Stanislas from LML and Mr Kähler from DLR Göttingen will take in charge the organization on the European side. The workshop concluding this challenge will take place in DLR Göttingen on September 14^th and 15^th 2001. It will be the next workshop of the ERCOFTAC SIG 32 and the final workshop of the PivNet Task 5. It will be linked to PIV'01, the 4^th International Symposium on Particle Image Velocimetry, which previous issue was in Santa Barbara (USA) in 1999. Information about the challenge will be distributed through the Pivnet web site, the PivNet network and the ERCOFTAC association and bulletin.