Integration by reduction formulae

From testwiki
Jump to navigation Jump to search

Script error: No such module "sidebar". Template:Short description In integral calculus, integration by reduction formulae is a method relying on recurrence relations. It is used when an expression containing an integer parameter, usually in the form of powers of elementary functions, or products of transcendental functions and polynomials of arbitrary degree, cannot be integrated directly. Using other methods of integration a reduction formula can be set up to obtain the integral of the same or similar expression with a lower integer parameter, progressively simplifying the integral until it can be evaluated. [1] This method of integration is one of the earliest used.Template:Citation needed

How to find the reduction formula

The reduction formula can be derived using any of the common methods of integration, like integration by substitution, integration by parts, integration by trigonometric substitution, integration by partial fractions, etc. The main idea is to express an integral involving an integer parameter (e.g. power) of a function, represented by In, in terms of an integral that involves a lower value of the parameter (lower power) of that function, for example In-1 or In-2. This makes the reduction formula a type of recurrence relation. In other words, the reduction formula expresses the integral

In=∫f(x,n)dx,

in terms of

Ik=∫f(x,k)dx,

where

k<n.

Reference works contain the general forms for recursive integration (see, for example Gradshteyn and Ryzhik).

How to compute the integral

To compute the integral, we set n to its value and use the reduction formula to express it in terms of the (n – 1) or (n – 2) integral. The lower index integral can be used to calculate the higher index ones; the process is continued repeatedly until we reach a point where the function to be integrated can be computed, usually when its index is 0 or 1. Then we back-substitute the previous results until we have computed In. [2]

Examples

Below are examples of the procedure.

Cosine integral

Typically, integrals like

∫cosnxdx,

can be evaluated by a reduction formula.

∫cosn(x)dx, for n = 1, 2 ... 30

Start by setting:

In=∫cosnxdx.

Now re-write as:

In=∫cosn−1xcos⁡xdx,

Integrating by this substitution:

cos⁡xdx=d(sin⁡x),
In=∫cosn−1xd(sin⁡x).

Now integrating by parts:

∫cosnxdx=∫cosn−1xd(sin⁡x)=cosn−1xsin⁡x−∫sin⁡xd(cosn−1x)=cosn−1xsin⁡x+(n−1)∫sin⁡xcosn−2xsin⁡xdx=cosn−1xsin⁡x+(n−1)∫cosn−2xsin2xdx=cosn−1xsin⁡x+(n−1)∫cosn−2x(1−cos2x)dx=cosn−1xsin⁡x+(n−1)∫cosn−2xdx−(n−1)∫cosnxdx=cosn−1xsin⁡x+(n−1)In−2−(n−1)In,

solving for In:

In +(n−1)In =cosn−1xsin⁡x + (n−1)In−2,
nIn =cosn−1(x)sin⁡x +(n−1)In−2,
In =1ncosn−1xsin⁡x +n−1nIn−2,

so the reduction formula is:

∫cosnxdx =1ncosn−1xsin⁡x+n−1n∫cosn−2xdx.

To supplement the example, the above can be used to evaluate the integral for (say) n = 5;

I5=∫cos5xdx.

Calculating lower indices:

n=5,I5=15cos4xsin⁡x+45I3,
n=3,I3=13cos2xsin⁡x+23I1,

back-substituting:

∵I1 =∫cos⁡xdx=sin⁡x+C1,
∴I3 =13cos2xsin⁡x+23sin⁡x+C2,C2 =23C1,
I5 =15cos4xsin⁡x+45[13cos2xsin⁡x+23sin⁡x]+C,

where C is a constant.

Exponential integral

Another typical example is:

∫xneaxdx.

Start by setting:

In=∫xneaxdx.

Integrating by substitution:

xndx=d(xn+1)n+1,
In=1n+1∫eaxd(xn+1),

Now integrating by parts:

∫eaxd(xn+1)=xn+1eax−∫xn+1d(eax)=xn+1eax−a∫xn+1eaxdx,
(n+1)In=xn+1eax−aIn+1,

shifting indices back by 1 (so n + 1 → n, n → n – 1):

nIn−1=xneax−aIn,

solving for In:

In=1a(xneax−nIn−1),

so the reduction formula is:

∫xneaxdx=1a(xneax−n∫xn−1eaxdx).

An alternative way in which the derivation could be done starts by substituting eax.

Integration by substitution:

eaxdx=d(eax)a,

In=1a∫xnd(eax),

Now integrating by parts:

∫xnd(eax)=xneax−∫eaxd(xn)=xneax−n∫eaxxn−1dx,

which gives the reduction formula when substituting back:

In=1a(xneax−nIn−1),

which is equivalent to:

∫xneaxdx=1a(xneax−n∫xn−1eaxdx).

Another alternative way in which the derivation could be done by integrating by parts:

In=∫xnxeaxdx,
u=xn ,  dv=eax,
dudx =nxn−1 ,  v=eaxa 
In=xneaxa −∫nxn−1 eaxa dx 
In=xneaxa −na ∫xn−1eax dx 

Remember:

In−1=∫xn−1eax dx 
∴ In=xneaxa −na In−1

which gives the reduction formula when substituting back:

In=1a(xneax−nIn−1),

which is equivalent to:

∫xneaxdx=1a(xneax−n∫xn−1eaxdx).

Tables of integral reduction formulas

Rational functions

The following integrals[3] contain:

  • Factors of the linear radical ax+b
  • Linear factors px+q and the linear radical ax+b
  • Quadratic factors x2+a2
  • Quadratic factors x2−a2, for x>a
  • Quadratic factors a2−x2, for x<a
  • (Irreducible) quadratic factors ax2+bx+c
  • Radicals of irreducible quadratic factors ax2+bx+c
Integral Reduction formula
In=∫xnax+bdx In=2xnax+ba(2n+1)−2nba(2n+1)In−1
In=∫dxxnax+b In=−ax+b(n−1)bxn−1−a(2n−3)2b(n−1)In−1
In=∫xnax+bdx In=2xn(ax+b)3a(2n+3)−2nba(2n+3)In−1
Im,n=∫dx(ax+b)m(px+q)n Im,n={−1(n−1)(bp−aq)[1(ax+b)m−1(px+q)n−1+a(m+n−2)Im,n−1]1(m−1)(bp−aq)[1(ax+b)m−1(px+q)n−1+p(m+n−2)Im−1,n]
Im,n=∫(ax+b)m(px+q)ndx Im,n={−1(n−1)(bp−aq)[(ax+b)m+1(px+q)n−1+a(n−m−2)Im,n−1]−1(n−m−1)p[(ax+b)m(px+q)n−1+m(bp−aq)Im−1,n]−1(n−1)p[(ax+b)m(px+q)n−1−amIm−1,n−1]
Integral Reduction formula
In=∫(px+q)nax+bdx ∫(px+q)nax+bdx=2(px+q)n+1ax+bp(2n+3)+bp−aqp(2n+3)In

In=2(px+q)nax+ba(2n+1)+2n(aq−bp)a(2n+1)In−1

In=∫dx(px+q)nax+b ∫ax+b(px+q)ndx=−ax+bp(n−1)(px+q)n−1+a2p(n−1)In

In=−ax+b(n−1)(aq−bp)(px+q)n−1+a(2n−3)2(n−1)(aq−bp)In−1

Integral Reduction formula
In=∫dx(x2+a2)n In=x2a2(n−1)(x2+a2)n−1+2n−32a2(n−1)In−1
In,m=∫dxxm(x2+a2)n a2In,m=Im,n−1−Im−2,n
In,m=∫xm(x2+a2)ndx In,m=Im−2,n−1−a2Im−2,n
Integral Reduction formula
In=∫dx(x2−a2)n In=−x2a2(n−1)(x2−a2)n−1−2n−32a2(n−1)In−1
In,m=∫dxxm(x2−a2)n a2In,m=Im−2,n−Im,n−1
In,m=∫xm(x2−a2)ndx In,m=Im−2,n−1+a2Im−2,n
Integral Reduction formula
In=∫dx(a2−x2)n In=x2a2(n−1)(a2−x2)n−1+2n−32a2(n−1)In−1
In,m=∫dxxm(a2−x2)n a2In,m=Im,n−1+Im−2,n
In,m=∫xm(a2−x2)ndx In,m=a2Im−2,n−Im−2,n−1
Integral Reduction formula
In=∫dxxn(ax2+bx+c) −cIn=1xn−1(n−1)+bIn−1+aIn−2
Im,n=∫xmdx(ax2+bx+c)n Im,n=−xm−1a(2n−m−1)(ax2+bx+c)n−1−b(n−m)a(2n−m−1)Im−1,n+c(m−1)a(2n−m−1)Im−2,n
Im,n=∫dxxm(ax2+bx+c)n −c(m−1)Im,n=1xm−1(ax2+bx+c)n−1+a(m+2n−3)Im−2,n+b(m+n−2)Im−1,n
Integral Reduction formula
In=∫(ax2+bx+c)ndx 8a(n+1)In+12=2(2ax+b)(ax2+bx+c)n+12+(2n+1)(4ac−b2)In−12
In=∫1(ax2+bx+c)ndx (2n−1)(4ac−b2)In+12=2(2ax+b)(ax2+bx+c)n−12+8a(n−1)In−12

note that by the laws of indices:

In+12=I2n+12=∫1(ax2+bx+c)2n+12dx=∫1(ax2+bx+c)2n+1dx

Transcendental functions

Template:Main article

The following integrals[4] contain:

  • Factors of sine
  • Factors of cosine
  • Factors of sine and cosine products and quotients
  • Products/quotients of exponential factors and powers of x
  • Products of exponential and sine/cosine factors
Integral Reduction formula
In=∫xnsin⁡axdx a2In=−axncos⁡ax+nxn−1sin⁡ax−n(n−1)In−2
Jn=∫xncos⁡axdx a2Jn=axnsin⁡ax+nxn−1cos⁡ax−n(n−1)Jn−2
In=∫sin⁡axxndx

Jn=∫cos⁡axxndx

In=−sin⁡ax(n−1)xn−1+an−1Jn−1

Jn=−cos⁡ax(n−1)xn−1−an−1In−1

the formulae can be combined to obtain separate equations in In:

Jn−1=−cos⁡ax(n−2)xn−2−an−2In−2

In=−sin⁡ax(n−1)xn−1−an−1[cos⁡ax(n−2)xn−2+an−2In−2]

∴In=−sin⁡ax(n−1)xn−1−a(n−1)(n−2)(cos⁡axxn−2+aIn−2)

and Jn:

In−1=−sin⁡ax(n−2)xn−2+an−2Jn−2

Jn=−cos⁡ax(n−1)xn−1−an−1[−sin⁡ax(n−2)xn−2+an−2Jn−2]

∴Jn=−cos⁡ax(n−1)xn−1−a(n−1)(n−2)(−sin⁡axxn−2+aJn−2)

In=∫sinnaxdx anIn=−sinn−1axcos⁡ax+a(n−1)In−2
Jn=∫cosnaxdx anJn=sin⁡axcosn−1ax+a(n−1)Jn−2
In=∫dxsinnax (n−1)In=−cos⁡axasinn−1ax+(n−2)In−2
Jn=∫dxcosnax (n−1)Jn=sin⁡axacosn−1ax+(n−2)Jn−2
Integral Reduction formula
Im,n=∫sinmaxcosnaxdx Im,n={−sinm−1axcosn+1axa(m+n)+m−1m+nIm−2,nsinm+1axcosn−1axa(m+n)+n−1m+nIm,n−2
Im,n=∫dxsinmaxcosnax Im,n={1a(n−1)sinm−1axcosn−1ax+m+n−2n−1Im,n−2−1a(m−1)sinm−1axcosn−1ax+m+n−2m−1Im−2,n
Im,n=∫sinmaxcosnaxdx Im,n={sinm−1axa(n−1)cosn−1ax−m−1n−1Im−2,n−2sinm+1axa(n−1)cosn−1ax−m−n+2n−1Im,n−2−sinm−1axa(m−n)cosn−1ax+m−1m−nIm−2,n
Im,n=∫cosmaxsinnaxdx Im,n={−cosm−1axa(n−1)sinn−1ax−m−1n−1Im−2,n−2−cosm+1axa(n−1)sinn−1ax−m−n+2n−1Im,n−2cosm−1axa(m−n)sinn−1ax+m−1m−nIm−2,n
Integral Reduction formula
In=∫xneaxdx

n>0

In=xneaxa−naIn−1
In=∫x−neaxdx

n>0

n≠1

In=−eax(n−1)xn−1+an−1In−1
In=∫eaxsinnbxdx In=eaxsinn−1bxa2+(bn)2(asin⁡bx−bncos⁡bx)+n(n−1)b2a2+(bn)2In−2
In=∫eaxcosnbxdx In=eaxcosn−1bxa2+(bn)2(acos⁡bx+bnsin⁡bx)+n(n−1)b2a2+(bn)2In−2

References

Template:Reflist

Bibliography

Template:Wikibooks

  • Anton, Bivens, Davis, Calculus, 7th edition.

Template:Integrals

  1. ↑ Mathematical methods for physics and engineering, K.F. Riley, M.P. Hobson, S.J. Bence, Cambridge University Press, 2010, Template:ISBN
  2. ↑ Further Elementary Analysis, R.I. Porter, G. Bell & Sons Ltd, 1978, Template:ISBN
  3. ↑ http://www.sosmath.com/tables/tables.html -> Indefinite integrals list
  4. ↑ http://www.sosmath.com/tables/tables.html -> Indefinite integrals list