Path: news.mathworks.com!not-for-mail From: <HIDDEN> Newsgroups: comp.soft-sys.matlab Subject: Re: symbolic integration Date: Sat, 17 Jan 2009 18:26:01 +0000 (UTC) Organization: The MathWorks, Inc. Lines: 39 Message-ID: <gkt7rp$9o2$1@fred.mathworks.com> References: <9842254.1232212065386.JavaMail.jakarta@nitrogen.mathforum.org> Reply-To: <HIDDEN> NNTP-Posting-Host: webapp-02-blr.mathworks.com Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 8bit X-Trace: fred.mathworks.com 1232216761 9986 172.30.248.37 (17 Jan 2009 18:26:01 GMT) X-Complaints-To: news@mathworks.com NNTP-Posting-Date: Sat, 17 Jan 2009 18:26:01 +0000 (UTC) X-Newsreader: MATLAB Central Newsreader 1187260 Xref: news.mathworks.com comp.soft-sys.matlab:512238 Serena <darkbluee@gmail.com> wrote in message <9842254.1232212065386.JavaMail.jakarta@nitrogen.mathforum.org>... > When I run the following code: > > D=30;L=152;m=39600;w1=1.5;ro=1.225;damp=0.01;xo=10; > lamda=1.875;sigma=0.7341; > > a=lamda/L > syms Uo x > U=x^0.4 > Fix=cosh(a*x)-cos(a*x)-sigma*(sinh(a*x)-sin(a*x)) > > A=U*Fix^2 > AA=int(A,0,L) > > > This message refers to the following lines: > > Warning: Explicit integral could not be found. > > In sym.int at 58 > In velocita_critica at 19 > > AA = > > int(x^(2/5)*(cosh(15/1216*x)-cos(15/1216*x)-7341/10000*sinh(15/1216*x)+7341/10000*sin(15/1216*x))^2,x = 0 .. 152) > > > What exactly does this mean and is there any way around the problem? > > Thanks, > Serena It's the 0.4 exponent on x that is the barrier to solving this integral. If it had been an integer, matlab would likely have succeeded. Mine found it with x^4 for example. You shouldn't blame matlab for failing on this. It is very likely that no explicit solution is known to the entire mathematical world. I learned very early in my mathematical career that it is very easy to invent functions for which no explicit integral is known. A friend of mine and I in college wasted about a week one time trying to find the integral of f(x) = x^x without any success whatever. To solve your integral you will undoubtedly have to resort to numerical methods. Roger Stafford