Robert Israel beat me to it, but I'll let the following stay.

Assuming that the values x[i] are sorted increasingly, you can do

piecewise(t < x[1], hz[1], t < x[2], hz[2], t< x[3], hz[3], t < x[4], hz[4], t < x[5], hz[5],t < x[6], hz[6], t < x[7], hz[7], t< x[8], hz[8], t < x[9], hz[9], t <= x[10], hz[10]);
 

since piecewise checks the inequalities from the left and stops if the inequality is satisfied.

Even shorter is

piecewise(op(map(op,[seq([t<x[i],h[i]],i=1..9)])),t<=x[10],hz[10]);

where I have singled out the last one since you want <= (if you really do).

Notice that I have used t instead of x. It is not a good idea to use x as a variable name if x[i] is in use.

Preben Alsholm

I don't think you are missing anything.

I tried

F1:=combine(Fder):

solve(F1=0, h);
                                      h
i.e. output is h (!). However, the expression F1 is huge. But certainly this answer is strange.
If you plot F1 you run into numerical problems, which is not surprising.

plot(F1,h=0..Pi,-1.5..1.5);
 

Preben Alsholm

Try executing the two sums

sum(f,i=-infinity..infinity);
sum(f(i),i=-infinity..infinity);

and ask yourself if those sums really are what you want to plot. 

Preben Alsholm

Just one thing.

In your example the first term in the sum is p*ln(0), which prevents solve from doing anything in solve( eqn[21,2], s);

Preben Alsholm

Since

eval(rhs(eqn13),{f(z)=1,e(z)=1,z=0});
evaluates to

-6210.378246*I

i.e. an imaginary number, you may want to rethink the mathematics of the problem. Is there a wrong sign under the squareroot in eqn13?

Preben Alsholm

You are trying to assign to a sequence of sums.

In the help page for assign it says that you can assign to names or functions, where 'function' means an unevalated  function call, like f(8).

Preben Alsholm

I have not changed much of your original code.

restart;
with(plots):
S1:=unapply(u*FresnelS(u)+((1/Pi)*cos((1/2)*Pi*u^2))-(1/Pi),u);
C1:=unapply(u*FresnelC(u)-((1/Pi)*sin((1/2)*Pi*u^2)),u);
R1:=1;
u1:=evalf(sqrt(0.4));
E1:=evalf(Pi*R1*S1(u1)+R1);
D1:=evalf(Pi*R1*C1(u1)+1);
a:=evalf(Pi*R1*(u1));
p1:=plot([1+a*FresnelC(u),a*FresnelS(u)-E1,u=0..u1]):
p2:=plot(-1.064877386,x=0..1):
p3:=plot(1.064877386,x=0..1):
p4:=plot([1+a*FresnelC(u),-a*FresnelS(u)+E1,u=0..u1]):
p5:=plot(-1.064877386,x=-1..0):
p6:=plot(1.064877386,x=-1..0):
p7:=plot([-1+(-a*FresnelC(u)),a*FresnelS(u)-E1,u=0..u1]):
p8:=plot([-1+(-a*FresnelC(u)),-a*FresnelS(u)+E1,u=0..u1]):
X:=1+a*FresnelC(u1);
t1:=fsolve(D1+R1*cos(t)=X,t);
C1:=plot([D1+R1*cos(t),R1*sin(t),t=-t1..t1]):
C2:=plot([-D1-R1*cos(t),R1*sin(t),t=-t1..t1]):
display(p7,p1,p2,p3,p4,p5,p6,p8,C1,C2,scaling=constrained);
 

Preben Alsholm

You already got some excellent solutions.

Here is another.

eq:=diff(y(x),x,x)+y(x)^2*x^2=x^2;
a:=[-0.6,-0.4,2.4,3.4];
sol:=y1->subs(dsolve({eq,y(0)=0,D(y)(0)=y1},numeric,output=listprocedure),y(x)):
#sol(y1) is the numerical procedure for evaluating y(x) (with D(y)(0)=y1), so that
sol(-0.6)(0.12345);
#will return the value of y(0.12345)
#and
plot([seq(sol(a[i]),i=1..4)],0..2);
# will plot the 4 different solutions in one plot.

Preben Alsholm
 

Do you expect us to do your homework?

(Excuse me if I was mistaken).

Preben Alsholm

Two sums inside each other are using the same summation index. This leads to questions of interpretation.

If you change the expression to

f := Sum( exp(-beta*(2*x2-1)-alpha*(2*x2-1))
   / Sum( Sum( exp(-beta*(4*x0*x1-2*x0-2*x1+1)-alpha*(2*x0-1)), x1=0..1 ),
               x0=0..1 ),
          x2=0..1 );
where I have replaced x0 in three places by x2 the interpretation is clear and there is no problem.

Preben Alsholm

subs(solutions,P);

will give you -6.960403739*10^(-13).

Preben Alsholm

You could use dsolve/numeric with infinity = 20 (or whatever):

m := 1;
de := diff(y(x), x, x, x)+((m+1)*(1/2))*y(x)*(diff(y(x), x, x))+m*(1-(diff(y(x), x))^2) = 0;
L:=dsolve({de,y(0) = 0, D(y)(0) = 0, D(y)(20) = 1},numeric);
plots:-odeplot(L,[x,y(x)],x=0..8);

The graphs on your photo are not graphs of y(x), but you can use odeplot to plot functions of y(x) also.

Preben Alsholm
 

You need 'and' between the boolean expressions. Also a few commas in the print statements were forgotten.

Here is the corrected version:

quad:=proc(a,b,c)
local det;
det:=b^2-4*a*c;
if a=0 and b=0 and c=0 then print( "All x is solution")
 elif a=0 and b=0 and c<>0 then print ("No solution")
 elif a=0 and b<>0 and c<>0 then print("One solution:",x=-c/b)
 elif det>0 then print("Two real solutions:",x1=(-b+sqrt(det))/(2*a),x2=(-b-sqrt(det))/(2*a))
 elif (det<0) then print("Two complex solutions:",x1=(-b+sqrt(det))/(2*a),x2=(-b-sqrt(det))/(2*a))
 else print("Two identical solutions:",x=-b/(2*a))
end if;
end proc;
 

Be aware that your procedure is designed for numerical input only. Thus quad(r,s,t); doesn't work unless r, s, and t are assigned numerical values.

Preben Alsholm

Here is one way,

with(LinearAlgebra):
G:=RandomTools:-Generate(complex(distribution(Normal(0,.5))), makeproc=true);
Gd:=RandomTools:-Generate(distribution(Normal(0,.5)), makeproc=true);
A:=RandomMatrix(3,3,generator=G,outputoptions=[shape=hermitian]);
B:=RandomMatrix(3,3,generator=Gd,outputoptions=[shape=diagonal]);
C:=A+B;
Eigenvalues(C);
 

Preben Alsholm

Looks like an exercise I could have given my first semester students.
Maple sees xysin as a variable name, xysin(x-y) is an application of the function xysin to x-y.
Maple doesn't know any function by that name, so cannot plot it.

You need two multiplication signs.

Why grid = [2,2]?

Preben Alsholm