Menu Close

I-have-no-formal-background-in-number-theory-but-I-m-curious-of-how-to-find-positive-integer-solutions-x-y-z-to-the-equation-x-n-y-n-z-n-for-n-Z-Fermat-s-last-theorem-led-me-to-this-Tell




Question Number 4203 by Yozzii last updated on 01/Jan/16
I have no formal background in   number theory, but I′m curious  of how to find positive integer solutions   (x,y,z) to the equation x^n +y^n =z^n  for   n∈Z^− . Fermat′s last theorem led  me to this. Tell me about the cases  of n=−1,n=−2 and n=−3.
Ihavenoformalbackgroundinnumbertheory,butImcuriousofhowtofindpositiveintegersolutions(x,y,z)totheequationxn+yn=znfornZ.Fermatslasttheoremledmetothis.Tellmeaboutthecasesofn=1,n=2andn=3.
Commented by 123456 last updated on 01/Jan/16
what if n∈Q?
whatifnQ?
Commented by Rasheed Soomro last updated on 02/Jan/16
9^(1/2)  +16^(1/2) =49^(1/2)   64^(−1/2) +64^(−1/2) =16^(−1/2)
91/2+161/2=491/2641/2+641/2=161/2
Answered by RasheedSindhi last updated on 01/Jan/16
Perhaps you are interesting in  general solution,not in individual   solutions!  I′ve attempt for individual ones.  x^(−1) +y^(−1) =z^(−1)   (1/x_ )+(1/y)=(1/z)⇒yz+xz=xy  z=((xy)/(x+y))  x=y=8  z=((64)/(16))=4  −−−−−−−−  x^(−2) +y^(−2) =z^(−2)   (1/x^2 )+(1/y^2 )=(1/z^2 )  y^2 z^2 +x^2 z^2 =x^2 y^2   z^2 =((x^2 y^2 )/(x^2 +y^2 ))  x=y=4,z=8
Perhapsyouareinterestingingeneralsolution,notinindividualsolutions!Iveattemptforindividualones.x1+y1=z11x+1y=1zyz+xz=xyz=xyx+yx=y=8z=6416=4x2+y2=z21x2+1y2=1z2y2z2+x2z2=x2y2z2=x2y2x2+y2x=y=4,z=8
Commented by Yozzii last updated on 01/Jan/16
Yes, general solutions.
Yes,generalsolutions.
Commented by Rasheed Soomro last updated on 01/Jan/16
One general solution for n=−1:  x=y=2^m ,z=2^(m−1)  ,∀m≥1∧ m∈ Z  Verification:  (2^m )^(−1) +(2^m )^(−1) =(2^(m−1) )^(−1)   2.2^(−m) =2^(−m+1)   2^(1−m) =2^(1−m)
Onegeneralsolutionforn=1:x=y=2m,z=2m1,m1mZVerification:(2m)1+(2m)1=(2m1)12.2m=2m+121m=21m
Commented by Yozzii last updated on 01/Jan/16
How do you prove this answer?  If you guessed it, what led you  to this guess?
Howdoyouprovethisanswer?Ifyouguessedit,whatledyoutothisguess?
Commented by Rasheed Soomro last updated on 01/Jan/16
x^(−1) +y^(−1) =z^(−1) ⇒z=((xy)/(x+y))  z will be integer in case x+y ∣ xy  Assuming special case x=y  z=(x^2 /(x+x))=(x^2 /(2x))=(x/2)  As z is integer so 2∣x⇒ x∈E  x=y=2k and z=k ,∀k∈N (More simpler condition)  Verification:  (2k)^(−1) +(2k)^(−1) =k^(−1)   2(2k)^(−1) =k^(−1)   (2.2^(−1) ).k^(−1) =k^−   k^(−1) =k^(−1)
x1+y1=z1z=xyx+yzwillbeintegerincasex+yxyAssumingspecialcasex=yz=x2x+x=x22x=x2Aszisintegerso2xxEx=y=2kandz=k,kN(Moresimplercondition)Verification:(2k)1+(2k)1=k12(2k)1=k1(2.21).k1=kk1=k1
Commented by Yozzii last updated on 01/Jan/16
Thank you.
Thankyou.
Commented by Rasheed Soomro last updated on 01/Jan/16
Assuming special case  y=2x  z=((xy)/(x+y))⇒z=((x.2x)/(x+2x))=((2x^2 )/(3x))=((2x)/3)  z is integer ⇒3∣x  Hence Let x=3k  y=2(3k)=6k  z=((2x)/3)=((2(3k))/3)=2k  (x,y,z)=(3k,6k,2k)  Verification:  (3k)^(−1) +(6k)^(−1) =(2k)^(−1)   (1/(3k))+(1/(6k))=(1/(2k))⇒(1/3)+(1/6)=(1/2)  ⇒((2+1)/6)=(1/2)  ⇒(1/2)=(1/2)
Assumingspecialcasey=2xz=xyx+yz=x.2xx+2x=2x23x=2x3zisinteger3xHenceLetx=3ky=2(3k)=6kz=2x3=2(3k)3=2k(x,y,z)=(3k,6k,2k)Verification:(3k)1+(6k)1=(2k)113k+16k=12k13+16=122+16=1212=12
Commented by Rasheed Soomro last updated on 03/Jan/16
Assuming y=mx  z=((xy)/(x+y))=((x.mx)/(x+mx))=((mx^2 )/((1+m)x))=((mx)/(1+m))  z is integer ⇒ 1+m ∣ x  Let x=(1+m)k  y=m(1+m)k  z=((m(1+m)k)/(1+m))=mk  (x,y,z)=( m+1, m(m+1), m ) ∀ m∈Z^+
Assumingy=mxz=xyx+y=x.mxx+mx=mx2(1+m)x=mx1+mzisinteger1+mxLetx=(1+m)ky=m(1+m)kz=m(1+m)k1+m=mk(x,y,z)=(m+1,m(m+1),m)mZ+
Commented by Rasheed Soomro last updated on 03/Jan/16
Assume y=(p/q)x,(p/q) is in reduced form  z=((xy)/(x+y))=((x((p/q)x))/(x+(p/q)x))=(((p/q)x^2 )/(((p+q)/q)x))=((px)/(p+q))  z is integer ⇒p+q ∣ x  Let x=(p+q)k  y=((p(p+q)k)/q)  z=(p/(p+q))×(p+q)k=pk  Now consider y=((p(p+q)k)/q)  y is integer ⇒q ∣ p  ∨  q ∣ p+q ∨ q ∣ k  Since (p,q)=1 so q ∤ p ∧ q ∤ p+q  Hence q ∣ k  Let k  is replaced by qk  x=(p+q)qk  y=p(p+q)k  z=pqk  Or  x=q(p+q)  y=p(p+q)  z=pq  Where p , q ∈N
Assumey=pqx,pqisinreducedformz=xyx+y=x(pqx)x+pqx=pqx2p+qqx=pxp+qzisintegerp+qxLetx=(p+q)ky=p(p+q)kqz=pp+q×(p+q)k=pkNowconsidery=p(p+q)kqyisintegerqpqp+qqkSince(p,q)=1soqpqp+qHenceqkLetkisreplacedbyqkx=(p+q)qky=p(p+q)kz=pqkOrx=q(p+q)y=p(p+q)z=pqWherep,qN
Answered by 123456 last updated on 02/Jan/16
i think we can use tha fermat theorem  to that  by fermat last theorem there no other  solution than the trivials one to  x^n +y^n =z^n ,n∈N,n>2  so by this there no other rational solution to  x^n +y^n =z^n   because if it was  x=(p_x /q_x ),y=(p_y /q_y ),z=(p_z /q_z )  x^n +y^n =z^n   (p_x ^n /q_x ^n )+(p_y ^n /q_y ^n )=(p_z ^n /q_z ^n )  (p_x q_y q_z )^n +(q_x p_y q_z )^n =(q_x q_y p_z )^n   a^n +b^n =c^n   and note that  x^(−n) =(1/x^n )=((1/x))^n =a^n  (rational solution)
ithinkwecanusethafermattheoremtothatbyfermatlasttheoremtherenoothersolutionthanthetrivialsonetoxn+yn=zn,nN,n>2sobythistherenootherrationalsolutiontoxn+yn=znbecauseifitwasx=pxqx,y=pyqy,z=pzqzxn+yn=znpxnqxn+pynqyn=pznqzn(pxqyqz)n+(qxpyqz)n=(qxqypz)nan+bn=cnandnotethatxn=1xn=(1x)n=an(rationalsolution)

Leave a Reply

Your email address will not be published. Required fields are marked *