Bibliography Of Turbulent Flows 1980-2002

by P. Bradshaw

SUMMARY

http://vonkarman.stanford.edu/tsd/resp_b.html

Original papers are not available!

1. INTRODUCTION

(i) File intro (this .HTML document). The Web site also contains the utility programs described in Section 3.

(ii) Annual ref files: "ref73" contains complex turbulent flow references (index categories 41 and above - see Section 2) up to 1972. Files "ref80", "ref81".... through "ref02" contain general references on turbulence and associated subjects for 1980 onward. Entries for a given year are generally in order of reading by this author, but reasonable care has been taken to keep periodicals in date order. ``ref73'' is sorted into index categories, entries in each category being in alphabetical order of senior author irrespective of date. ``ref80'' contains some pre-1980 references, in year order. All other files contain entries for the year indicated by the last two characters of the filename.

Entries are in plain ASCII text, and in absolutely standard format on the assumption that most searching will be done by computer. Where possible, "unpublished" versions of papers (especially conference papers) have been deleted in favor of the published version, unless the former is referred to by the latter or is obviously more comprehensive than the final publication. Total is over 2MB of information. The concatenation of all annual files "r" is available on the Web site. Note that although sorting annual files by name puts "ref00", etc. first, the utility that creates "r" produces proper year order.

(iii) refsnnn: the same entries as in the ref files, but sorted according to the index categories nnn listed below - one file per main category or important sub-category: e.g. refs116 contains references on combustion (sub-category 11.6). Entries in each file are generally in year order, many references appearing in more than one category.

Each entry has five lines, for

Some lines are longer than 80 characters, so a text editor which wraps, rather than truncates, long lines is needed.

A typical entry is as follows:

Each line ends with a "*", except the last which ends with an "open single quote", ASCII char. 96, ` ,(originating as the Pound Sterling sign on a British keyboard).

On the first line, the ".." marker is followed by the index categories - e.g. "13.0" is "transport-equation turbulence models" - arranged in decreasing order of interest to the present author at the time of indexing. The main categories were chosen in 1980 (categories from 40 upwards date from 1972). A completely logical system is impossible: longitudinal vortices usually have leading category 54, but wingtip vortices would have leading category 59.2 (airfoils), while flows around surface-mounted obstacles usually have leading category 45 (skew-induced secondary flows) rather than 54 - but should also appear in 59.2 in the case of wing-body junctions, in 44.2 (turbomachinery) in the case of blade-annulus junctions, and in 61 in the case of bluff obstacles. Many index numbers have one or two minor categories but these are not intended to be exhaustive - for example blade-annulus junctions are not usually indexed under 59.2 as well as 44.2 - so the user needs to use some initiative. Multiple categories are in the form "ij.k,lm.n" with no space after the comma, so that a search routine can distinguish an index category from a source like "J. Fluid Mech. 11, 97". Searching for "..ij." or "..ij.k" yields leading entries only, ",lm.n" yields secondary entries only, while searching for "ij.k" or "ij." yields both leading and secondary entries. The categories are separated by a colon, ":", from the present author's one-line abstract (if any). The abstract is intended to contain any useful keywords not present in the title. It is always informal, and occasionally mildly disrespectful - please remember that this bibliography was set up for private use, and is put on the Web as a public service with no guarantees.

Authors are separated by a semicolon, ";". ET AL. is used where the number of authors exceeds three, and ANON. for conference preprint collections and other works with no identifiable author or editor.

Where necessary, the title line is an English translation of the foreign language original, usually without indication of the original language (French, German or occasionally Russian), which should be obvious from the source.

The source line specifies journal, publisher, report series etc. The style is not completely uniform but journal title abbreviations are generally those of the World List of Scientific Periodicals. To prevent confusion with index categories, page numbers (at the end of the line) are not followed by a period, and other strings like "ij." have been avoided. Journals whose pagination is continuous throughout a volume are cited as "J. Fluid Mech. 123, 456" while monthlies appear in more explicit form as "BYTE vol. 10, no. 11, p. 12 - again note the spaces after commas and periods to avoid all confusion with index categories. In references to individual papers in conference proceedings, the editors' names are given in the source line after the proceedings title, e.g. "(T. Cebeci, ed.)" and the paper authors' names in the author line.

The final (date) line is in the form 19nn. ` " with a space between the period and the "open quote": index categories run up to 62., so dates up to 1962 are written with a space before the period so that a computer search will not pick up the last two digits of the date instead of an index category.

The index categories are broadly:

When a main entry category is divided into only two "decimal" sections, e.g. 18.1, 18.2, the first is generally on theoretical or computational work, while the second is more practically relevant or more directly related to experimental work. Note however that Direct Numerical Simulations are indexedin as "experiments" in the appropriate category (Large Eddy Simulations, which use empirical sub-grid-scale models, have their own subsection). Multiple sub-divisions, as in category 25 "Numerical Methods", indicate an expansion of the subject since the main categories were chosen, mostly in 1980. Partly for this reason, file sizes vary greatly, the largest being "refs250" at over 300kB.

The categorized files REFSnnn were extracted using the Unix "awk" utility, available for Microsoft Windows or MS-DOS in the MKS Toolkit from Mortice Kern Systems, www.mks.com - see for example the "sortweb" and "SORTALL.BAT" utilities (Section 3). Files for some important subsections (e.g. "refs256" for large-eddy simulation) are available as subsets of the main file.

Note that that some Unix-derived formatting programs use a "." at the beginning of a line to signal a formatting command and will therefore ignore the abstract line with its leading ".." - this is convenient if only the formal part of the reference is to be printed, but it is easy to change the ".." to some other block marker.

2. INDEX CATEGORIES

40+ - "Complex" flows. ".1" cats not listed below are for lesser, or purely computational, papers

3. UTILITY PROGRAMS

sortweb joins the annual files into one file called "r" and uses the awk language to generate the "refsnnn" sorted files. Details are in sortweb.txt.

Files pb and bp are Unix scripts (originally run under SGI Irix) using the awk language (a subset of nawk to search the files for keywords and output the complete entry for each hit. Details are in pb.txt and bp.txt.

(2) MS-DOS/Windows (change file names to suit your system)

SORTALL.BAT is the equivalent of sortweb, above, producing R.DOC.

RMAKE.BAT joins the annual files into R.DOC, as SORTALL.BAT does, but without generating sorted files - much faster, but useful mainly to generate the (2MB plus) R.DOC file if the other files are being transferred on 1.44MB diskette.

AWKKEYR.BAT is an MS-DOS file invoking the MKS Toolkit Windows version of awk. Typing ``AWKKEYR file key'' searches file for one keyword key, and outputs the complete entry for each hit, followed by a blank line for ease of reading, into a scratch file with extension .DOC. It is the equivalent of bp.