Chapter
stringclasses
18 values
sentence_range
stringlengths
3
9
Text
stringlengths
7
7.34k
1
3318-3321
But these integrals are very similar geometrically Thus, y = x2 + C, where C is arbitrary constant, represents a family of integrals By assigning different values to C, we get different members of the family These together constitute the indefinite integral
1
3319-3322
Thus, y = x2 + C, where C is arbitrary constant, represents a family of integrals By assigning different values to C, we get different members of the family These together constitute the indefinite integral In this case, each integral represents a parabola with its axis along y-axis
1
3320-3323
By assigning different values to C, we get different members of the family These together constitute the indefinite integral In this case, each integral represents a parabola with its axis along y-axis Clearly, for C = 0, we obtain y = x2, a parabola with its vertex on the origin
1
3321-3324
These together constitute the indefinite integral In this case, each integral represents a parabola with its axis along y-axis Clearly, for C = 0, we obtain y = x2, a parabola with its vertex on the origin The curve y = x2 + 1 for C = 1 is obtained by shifting the parabola y = x2 one unit along y-axis in positive direction
1
3322-3325
In this case, each integral represents a parabola with its axis along y-axis Clearly, for C = 0, we obtain y = x2, a parabola with its vertex on the origin The curve y = x2 + 1 for C = 1 is obtained by shifting the parabola y = x2 one unit along y-axis in positive direction For C = – 1, y = x2 – 1 is obtained by shifting the parabola y = x2 one unit along y-axis in the negative direction
1
3323-3326
Clearly, for C = 0, we obtain y = x2, a parabola with its vertex on the origin The curve y = x2 + 1 for C = 1 is obtained by shifting the parabola y = x2 one unit along y-axis in positive direction For C = – 1, y = x2 – 1 is obtained by shifting the parabola y = x2 one unit along y-axis in the negative direction Thus, for each positive value of C, each parabola of the family has its vertex on the positive side of the y-axis and for negative values of C, each has its vertex along the negative side of the y-axis
1
3324-3327
The curve y = x2 + 1 for C = 1 is obtained by shifting the parabola y = x2 one unit along y-axis in positive direction For C = – 1, y = x2 – 1 is obtained by shifting the parabola y = x2 one unit along y-axis in the negative direction Thus, for each positive value of C, each parabola of the family has its vertex on the positive side of the y-axis and for negative values of C, each has its vertex along the negative side of the y-axis Some of these have been shown in the Fig 7
1
3325-3328
For C = – 1, y = x2 – 1 is obtained by shifting the parabola y = x2 one unit along y-axis in the negative direction Thus, for each positive value of C, each parabola of the family has its vertex on the positive side of the y-axis and for negative values of C, each has its vertex along the negative side of the y-axis Some of these have been shown in the Fig 7 1
1
3326-3329
Thus, for each positive value of C, each parabola of the family has its vertex on the positive side of the y-axis and for negative values of C, each has its vertex along the negative side of the y-axis Some of these have been shown in the Fig 7 1 Let us consider the intersection of all these parabolas by a line x = a
1
3327-3330
Some of these have been shown in the Fig 7 1 Let us consider the intersection of all these parabolas by a line x = a In the Fig 7
1
3328-3331
1 Let us consider the intersection of all these parabolas by a line x = a In the Fig 7 1, we have taken a > 0
1
3329-3332
Let us consider the intersection of all these parabolas by a line x = a In the Fig 7 1, we have taken a > 0 The same is true when a < 0
1
3330-3333
In the Fig 7 1, we have taken a > 0 The same is true when a < 0 If the line x = a intersects the parabolas y = x2, y = x2 + 1, y = x2 + 2, y = x2 – 1, y = x2 – 2 at P0, P1, P2, P–1, P–2 etc
1
3331-3334
1, we have taken a > 0 The same is true when a < 0 If the line x = a intersects the parabolas y = x2, y = x2 + 1, y = x2 + 2, y = x2 – 1, y = x2 – 2 at P0, P1, P2, P–1, P–2 etc , respectively, then dy dx at these points equals 2a
1
3332-3335
The same is true when a < 0 If the line x = a intersects the parabolas y = x2, y = x2 + 1, y = x2 + 2, y = x2 – 1, y = x2 – 2 at P0, P1, P2, P–1, P–2 etc , respectively, then dy dx at these points equals 2a This indicates that the tangents to the curves at these points are parallel
1
3333-3336
If the line x = a intersects the parabolas y = x2, y = x2 + 1, y = x2 + 2, y = x2 – 1, y = x2 – 2 at P0, P1, P2, P–1, P–2 etc , respectively, then dy dx at these points equals 2a This indicates that the tangents to the curves at these points are parallel Thus, 2 C 2 C F ( ) x dx x x = + = ∫ (say), implies that 292 MATHEMATICS the tangents to all the curves y = F C (x), C ∈ R, at the points of intersection of the curves by the line x = a, (a ∈ R), are parallel
1
3334-3337
, respectively, then dy dx at these points equals 2a This indicates that the tangents to the curves at these points are parallel Thus, 2 C 2 C F ( ) x dx x x = + = ∫ (say), implies that 292 MATHEMATICS the tangents to all the curves y = F C (x), C ∈ R, at the points of intersection of the curves by the line x = a, (a ∈ R), are parallel Further, the following equation (statement) ( ) F ( ) C (say) f x dx x y = + = ∫ , represents a family of curves
1
3335-3338
This indicates that the tangents to the curves at these points are parallel Thus, 2 C 2 C F ( ) x dx x x = + = ∫ (say), implies that 292 MATHEMATICS the tangents to all the curves y = F C (x), C ∈ R, at the points of intersection of the curves by the line x = a, (a ∈ R), are parallel Further, the following equation (statement) ( ) F ( ) C (say) f x dx x y = + = ∫ , represents a family of curves The different values of C will correspond to different members of this family and these members can be obtained by shifting any one of the curves parallel to itself
1
3336-3339
Thus, 2 C 2 C F ( ) x dx x x = + = ∫ (say), implies that 292 MATHEMATICS the tangents to all the curves y = F C (x), C ∈ R, at the points of intersection of the curves by the line x = a, (a ∈ R), are parallel Further, the following equation (statement) ( ) F ( ) C (say) f x dx x y = + = ∫ , represents a family of curves The different values of C will correspond to different members of this family and these members can be obtained by shifting any one of the curves parallel to itself This is the geometrical interpretation of indefinite integral
1
3337-3340
Further, the following equation (statement) ( ) F ( ) C (say) f x dx x y = + = ∫ , represents a family of curves The different values of C will correspond to different members of this family and these members can be obtained by shifting any one of the curves parallel to itself This is the geometrical interpretation of indefinite integral 7
1
3338-3341
The different values of C will correspond to different members of this family and these members can be obtained by shifting any one of the curves parallel to itself This is the geometrical interpretation of indefinite integral 7 2
1
3339-3342
This is the geometrical interpretation of indefinite integral 7 2 2 Some properties of indefinite integral In this sub section, we shall derive some properties of indefinite integrals
1
3340-3343
7 2 2 Some properties of indefinite integral In this sub section, we shall derive some properties of indefinite integrals (I) The process of differentiation and integration are inverses of each other in the sense of the following results : ( ) d f x dx dx ∫ = f (x) and f( ) ∫′x dx = f (x) + C, where C is any arbitrary constant
1
3341-3344
2 2 Some properties of indefinite integral In this sub section, we shall derive some properties of indefinite integrals (I) The process of differentiation and integration are inverses of each other in the sense of the following results : ( ) d f x dx dx ∫ = f (x) and f( ) ∫′x dx = f (x) + C, where C is any arbitrary constant Fig 7
1
3342-3345
2 Some properties of indefinite integral In this sub section, we shall derive some properties of indefinite integrals (I) The process of differentiation and integration are inverses of each other in the sense of the following results : ( ) d f x dx dx ∫ = f (x) and f( ) ∫′x dx = f (x) + C, where C is any arbitrary constant Fig 7 1 INTEGRALS 293 Proof Let F be any anti derivative of f, i
1
3343-3346
(I) The process of differentiation and integration are inverses of each other in the sense of the following results : ( ) d f x dx dx ∫ = f (x) and f( ) ∫′x dx = f (x) + C, where C is any arbitrary constant Fig 7 1 INTEGRALS 293 Proof Let F be any anti derivative of f, i e
1
3344-3347
Fig 7 1 INTEGRALS 293 Proof Let F be any anti derivative of f, i e , dF( ) x dx = f (x) Then ( ) ∫f x dx = F(x) + C Therefore ( ) d dx ∫f x dx = ( F ( ) + C) d x dx = F ( ) = ( ) d x f x dx Similarly, we note that f ′(x) = ( ) d f x dx and hence f( ) ∫′x dx = f (x) + C where C is arbitrary constant called constant of integration
1
3345-3348
1 INTEGRALS 293 Proof Let F be any anti derivative of f, i e , dF( ) x dx = f (x) Then ( ) ∫f x dx = F(x) + C Therefore ( ) d dx ∫f x dx = ( F ( ) + C) d x dx = F ( ) = ( ) d x f x dx Similarly, we note that f ′(x) = ( ) d f x dx and hence f( ) ∫′x dx = f (x) + C where C is arbitrary constant called constant of integration (II) Two indefinite integrals with the same derivative lead to the same family of curves and so they are equivalent
1
3346-3349
e , dF( ) x dx = f (x) Then ( ) ∫f x dx = F(x) + C Therefore ( ) d dx ∫f x dx = ( F ( ) + C) d x dx = F ( ) = ( ) d x f x dx Similarly, we note that f ′(x) = ( ) d f x dx and hence f( ) ∫′x dx = f (x) + C where C is arbitrary constant called constant of integration (II) Two indefinite integrals with the same derivative lead to the same family of curves and so they are equivalent Proof Let f and g be two functions such that ( ) d f x dx dx ∫ = ( ) d g x dx dx ∫ or ( ) ( ) d f x dx – g x dx dx     ∫ ∫ = 0 Hence ( ) ( ) f x dx – g x dx ∫ ∫ = C, where C is any real number (Why
1
3347-3350
, dF( ) x dx = f (x) Then ( ) ∫f x dx = F(x) + C Therefore ( ) d dx ∫f x dx = ( F ( ) + C) d x dx = F ( ) = ( ) d x f x dx Similarly, we note that f ′(x) = ( ) d f x dx and hence f( ) ∫′x dx = f (x) + C where C is arbitrary constant called constant of integration (II) Two indefinite integrals with the same derivative lead to the same family of curves and so they are equivalent Proof Let f and g be two functions such that ( ) d f x dx dx ∫ = ( ) d g x dx dx ∫ or ( ) ( ) d f x dx – g x dx dx     ∫ ∫ = 0 Hence ( ) ( ) f x dx – g x dx ∫ ∫ = C, where C is any real number (Why ) or ( ) ∫f x dx = ( ) g x dx +C ∫ So the families of curves { } 1 1 ( ) C ,C R f x dx + ∈ ∫ and { } 2 2 ( ) C , C R g x dx + ∈ ∫ are identical
1
3348-3351
(II) Two indefinite integrals with the same derivative lead to the same family of curves and so they are equivalent Proof Let f and g be two functions such that ( ) d f x dx dx ∫ = ( ) d g x dx dx ∫ or ( ) ( ) d f x dx – g x dx dx     ∫ ∫ = 0 Hence ( ) ( ) f x dx – g x dx ∫ ∫ = C, where C is any real number (Why ) or ( ) ∫f x dx = ( ) g x dx +C ∫ So the families of curves { } 1 1 ( ) C ,C R f x dx + ∈ ∫ and { } 2 2 ( ) C , C R g x dx + ∈ ∫ are identical Hence, in this sense, ( ) and ( ) f x dx g x dx ∫ ∫ are equivalent
1
3349-3352
Proof Let f and g be two functions such that ( ) d f x dx dx ∫ = ( ) d g x dx dx ∫ or ( ) ( ) d f x dx – g x dx dx     ∫ ∫ = 0 Hence ( ) ( ) f x dx – g x dx ∫ ∫ = C, where C is any real number (Why ) or ( ) ∫f x dx = ( ) g x dx +C ∫ So the families of curves { } 1 1 ( ) C ,C R f x dx + ∈ ∫ and { } 2 2 ( ) C , C R g x dx + ∈ ∫ are identical Hence, in this sense, ( ) and ( ) f x dx g x dx ∫ ∫ are equivalent 294 MATHEMATICS � Note The equivalence of the families { } 1 1 ( ) f x dx+ C ,C ∈ ∫ R and { } 2 2 ( ) g x dx+ C ,C ∈ ∫ R is customarily expressed by writing ( ) = ( ) f x dx g x dx ∫ ∫ , without mentioning the parameter
1
3350-3353
) or ( ) ∫f x dx = ( ) g x dx +C ∫ So the families of curves { } 1 1 ( ) C ,C R f x dx + ∈ ∫ and { } 2 2 ( ) C , C R g x dx + ∈ ∫ are identical Hence, in this sense, ( ) and ( ) f x dx g x dx ∫ ∫ are equivalent 294 MATHEMATICS � Note The equivalence of the families { } 1 1 ( ) f x dx+ C ,C ∈ ∫ R and { } 2 2 ( ) g x dx+ C ,C ∈ ∫ R is customarily expressed by writing ( ) = ( ) f x dx g x dx ∫ ∫ , without mentioning the parameter (III) [ ] ( ) + ( ) ( ) + ( ) f x g x dx f x dx g x dx = ∫ ∫ ∫ Proof By Property (I), we have [ ( ) + ( )] d f x g x dx dx     ∫ = f (x) + g(x)
1
3351-3354
Hence, in this sense, ( ) and ( ) f x dx g x dx ∫ ∫ are equivalent 294 MATHEMATICS � Note The equivalence of the families { } 1 1 ( ) f x dx+ C ,C ∈ ∫ R and { } 2 2 ( ) g x dx+ C ,C ∈ ∫ R is customarily expressed by writing ( ) = ( ) f x dx g x dx ∫ ∫ , without mentioning the parameter (III) [ ] ( ) + ( ) ( ) + ( ) f x g x dx f x dx g x dx = ∫ ∫ ∫ Proof By Property (I), we have [ ( ) + ( )] d f x g x dx dx     ∫ = f (x) + g(x) (1) On the otherhand, we find that ( ) + ( ) d f x dx g x dx dx     ∫ ∫ = ( ) + ( ) d d f x dx g x dx dx dx ∫ ∫ = f (x) + g(x)
1
3352-3355
294 MATHEMATICS � Note The equivalence of the families { } 1 1 ( ) f x dx+ C ,C ∈ ∫ R and { } 2 2 ( ) g x dx+ C ,C ∈ ∫ R is customarily expressed by writing ( ) = ( ) f x dx g x dx ∫ ∫ , without mentioning the parameter (III) [ ] ( ) + ( ) ( ) + ( ) f x g x dx f x dx g x dx = ∫ ∫ ∫ Proof By Property (I), we have [ ( ) + ( )] d f x g x dx dx     ∫ = f (x) + g(x) (1) On the otherhand, we find that ( ) + ( ) d f x dx g x dx dx     ∫ ∫ = ( ) + ( ) d d f x dx g x dx dx dx ∫ ∫ = f (x) + g(x) (2) Thus, in view of Property (II), it follows by (1) and (2) that ( ) ( ) ( ) f x g x dx + ∫ = ( ) ( ) f x dx g x dx + ∫ ∫
1
3353-3356
(III) [ ] ( ) + ( ) ( ) + ( ) f x g x dx f x dx g x dx = ∫ ∫ ∫ Proof By Property (I), we have [ ( ) + ( )] d f x g x dx dx     ∫ = f (x) + g(x) (1) On the otherhand, we find that ( ) + ( ) d f x dx g x dx dx     ∫ ∫ = ( ) + ( ) d d f x dx g x dx dx dx ∫ ∫ = f (x) + g(x) (2) Thus, in view of Property (II), it follows by (1) and (2) that ( ) ( ) ( ) f x g x dx + ∫ = ( ) ( ) f x dx g x dx + ∫ ∫ (IV) For any real number k, ( ) ( ) k f x dx k f x dx = ∫ ∫ Proof By the Property (I), ( ) ( ) d k f x dx k f x dx = ∫
1
3354-3357
(1) On the otherhand, we find that ( ) + ( ) d f x dx g x dx dx     ∫ ∫ = ( ) + ( ) d d f x dx g x dx dx dx ∫ ∫ = f (x) + g(x) (2) Thus, in view of Property (II), it follows by (1) and (2) that ( ) ( ) ( ) f x g x dx + ∫ = ( ) ( ) f x dx g x dx + ∫ ∫ (IV) For any real number k, ( ) ( ) k f x dx k f x dx = ∫ ∫ Proof By the Property (I), ( ) ( ) d k f x dx k f x dx = ∫ Also ( ) d k f x dx dx     ∫ = ( ) = ( ) kd f x dx k f x dx ∫ Therefore, using the Property (II), we have ( ) ( ) k f x dx k f x dx = ∫ ∫
1
3355-3358
(2) Thus, in view of Property (II), it follows by (1) and (2) that ( ) ( ) ( ) f x g x dx + ∫ = ( ) ( ) f x dx g x dx + ∫ ∫ (IV) For any real number k, ( ) ( ) k f x dx k f x dx = ∫ ∫ Proof By the Property (I), ( ) ( ) d k f x dx k f x dx = ∫ Also ( ) d k f x dx dx     ∫ = ( ) = ( ) kd f x dx k f x dx ∫ Therefore, using the Property (II), we have ( ) ( ) k f x dx k f x dx = ∫ ∫ (V) Properties (III) and (IV) can be generalised to a finite number of functions f1, f2,
1
3356-3359
(IV) For any real number k, ( ) ( ) k f x dx k f x dx = ∫ ∫ Proof By the Property (I), ( ) ( ) d k f x dx k f x dx = ∫ Also ( ) d k f x dx dx     ∫ = ( ) = ( ) kd f x dx k f x dx ∫ Therefore, using the Property (II), we have ( ) ( ) k f x dx k f x dx = ∫ ∫ (V) Properties (III) and (IV) can be generalised to a finite number of functions f1, f2, , fn and the real numbers, k1, k2,
1
3357-3360
Also ( ) d k f x dx dx     ∫ = ( ) = ( ) kd f x dx k f x dx ∫ Therefore, using the Property (II), we have ( ) ( ) k f x dx k f x dx = ∫ ∫ (V) Properties (III) and (IV) can be generalised to a finite number of functions f1, f2, , fn and the real numbers, k1, k2, , kn giving [ ] 1 1 2 2 ( ) ( ) ( ) n n k f x k f x
1
3358-3361
(V) Properties (III) and (IV) can be generalised to a finite number of functions f1, f2, , fn and the real numbers, k1, k2, , kn giving [ ] 1 1 2 2 ( ) ( ) ( ) n n k f x k f x k f x dx + + + ∫ = 1 1 2 2 ( ) ( ) ( ) n n k f x dx k f x dx
1
3359-3362
, fn and the real numbers, k1, k2, , kn giving [ ] 1 1 2 2 ( ) ( ) ( ) n n k f x k f x k f x dx + + + ∫ = 1 1 2 2 ( ) ( ) ( ) n n k f x dx k f x dx k f x dx + + + ∫ ∫ ∫
1
3360-3363
, kn giving [ ] 1 1 2 2 ( ) ( ) ( ) n n k f x k f x k f x dx + + + ∫ = 1 1 2 2 ( ) ( ) ( ) n n k f x dx k f x dx k f x dx + + + ∫ ∫ ∫ To find an anti derivative of a given function, we search intuitively for a function whose derivative is the given function
1
3361-3364
k f x dx + + + ∫ = 1 1 2 2 ( ) ( ) ( ) n n k f x dx k f x dx k f x dx + + + ∫ ∫ ∫ To find an anti derivative of a given function, we search intuitively for a function whose derivative is the given function The search for the requisite function for finding an anti derivative is known as integration by the method of inspection
1
3362-3365
k f x dx + + + ∫ ∫ ∫ To find an anti derivative of a given function, we search intuitively for a function whose derivative is the given function The search for the requisite function for finding an anti derivative is known as integration by the method of inspection We illustrate it through some examples
1
3363-3366
To find an anti derivative of a given function, we search intuitively for a function whose derivative is the given function The search for the requisite function for finding an anti derivative is known as integration by the method of inspection We illustrate it through some examples INTEGRALS 295 Example 1 Write an anti derivative for each of the following functions using the method of inspection: (i) cos 2x (ii) 3x2 + 4x3 (iii) 1 x , x ≠ 0 Solution (i) We look for a function whose derivative is cos 2x
1
3364-3367
The search for the requisite function for finding an anti derivative is known as integration by the method of inspection We illustrate it through some examples INTEGRALS 295 Example 1 Write an anti derivative for each of the following functions using the method of inspection: (i) cos 2x (ii) 3x2 + 4x3 (iii) 1 x , x ≠ 0 Solution (i) We look for a function whose derivative is cos 2x Recall that d dx sin 2x = 2 cos 2x or cos 2x = 1 2 d dx (sin 2x) = 21 sin 2 d x dx       Therefore, an anti derivative of cos 2x is 1 sin 2 2 x
1
3365-3368
We illustrate it through some examples INTEGRALS 295 Example 1 Write an anti derivative for each of the following functions using the method of inspection: (i) cos 2x (ii) 3x2 + 4x3 (iii) 1 x , x ≠ 0 Solution (i) We look for a function whose derivative is cos 2x Recall that d dx sin 2x = 2 cos 2x or cos 2x = 1 2 d dx (sin 2x) = 21 sin 2 d x dx       Therefore, an anti derivative of cos 2x is 1 sin 2 2 x (ii) We look for a function whose derivative is 3x2 + 4x3
1
3366-3369
INTEGRALS 295 Example 1 Write an anti derivative for each of the following functions using the method of inspection: (i) cos 2x (ii) 3x2 + 4x3 (iii) 1 x , x ≠ 0 Solution (i) We look for a function whose derivative is cos 2x Recall that d dx sin 2x = 2 cos 2x or cos 2x = 1 2 d dx (sin 2x) = 21 sin 2 d x dx       Therefore, an anti derivative of cos 2x is 1 sin 2 2 x (ii) We look for a function whose derivative is 3x2 + 4x3 Note that ( ) 3 4 d x x dx + = 3x2 + 4x3
1
3367-3370
Recall that d dx sin 2x = 2 cos 2x or cos 2x = 1 2 d dx (sin 2x) = 21 sin 2 d x dx       Therefore, an anti derivative of cos 2x is 1 sin 2 2 x (ii) We look for a function whose derivative is 3x2 + 4x3 Note that ( ) 3 4 d x x dx + = 3x2 + 4x3 Therefore, an anti derivative of 3x2 + 4x3 is x3 + x4
1
3368-3371
(ii) We look for a function whose derivative is 3x2 + 4x3 Note that ( ) 3 4 d x x dx + = 3x2 + 4x3 Therefore, an anti derivative of 3x2 + 4x3 is x3 + x4 (iii) We know that 1 1 1 (log ) 0 and [log ( )] ( 1) 0 d d x ,x – x – ,x dx x dx – x x = > = = < Combining above, we get ( ) 1 log 0 d x , x dx =x ≠ Therefore, 1 log dx x x = ∫ is one of the anti derivatives of 1 x
1
3369-3372
Note that ( ) 3 4 d x x dx + = 3x2 + 4x3 Therefore, an anti derivative of 3x2 + 4x3 is x3 + x4 (iii) We know that 1 1 1 (log ) 0 and [log ( )] ( 1) 0 d d x ,x – x – ,x dx x dx – x x = > = = < Combining above, we get ( ) 1 log 0 d x , x dx =x ≠ Therefore, 1 log dx x x = ∫ is one of the anti derivatives of 1 x Example 2 Find the following integrals: (i) 3 2 1 x – dx ∫x (ii) (32 1) x dx + ∫ (iii) ∫ 23 1 ( 2 – ) + ∫ x x e dx x Solution (i) We have 3 2 2 1 – x – dx x dx – x dx x = ∫ ∫ ∫ (by Property V) 296 MATHEMATICS = 1 1 2 1 1 2 C C 1 1 2 1 – x x – – + +     + +     + +     ; C1, C2 are constants of integration = 2 1 1 2 C C 2 1 – x –x – – + = 2 1 2 1 + C C x2 – x + = 2 1 + C 2 x x + , where C = C 1 – C2 is another constant of integration
1
3370-3373
Therefore, an anti derivative of 3x2 + 4x3 is x3 + x4 (iii) We know that 1 1 1 (log ) 0 and [log ( )] ( 1) 0 d d x ,x – x – ,x dx x dx – x x = > = = < Combining above, we get ( ) 1 log 0 d x , x dx =x ≠ Therefore, 1 log dx x x = ∫ is one of the anti derivatives of 1 x Example 2 Find the following integrals: (i) 3 2 1 x – dx ∫x (ii) (32 1) x dx + ∫ (iii) ∫ 23 1 ( 2 – ) + ∫ x x e dx x Solution (i) We have 3 2 2 1 – x – dx x dx – x dx x = ∫ ∫ ∫ (by Property V) 296 MATHEMATICS = 1 1 2 1 1 2 C C 1 1 2 1 – x x – – + +     + +     + +     ; C1, C2 are constants of integration = 2 1 1 2 C C 2 1 – x –x – – + = 2 1 2 1 + C C x2 – x + = 2 1 + C 2 x x + , where C = C 1 – C2 is another constant of integration �Note From now onwards, we shall write only one constant of integration in the final answer
1
3371-3374
(iii) We know that 1 1 1 (log ) 0 and [log ( )] ( 1) 0 d d x ,x – x – ,x dx x dx – x x = > = = < Combining above, we get ( ) 1 log 0 d x , x dx =x ≠ Therefore, 1 log dx x x = ∫ is one of the anti derivatives of 1 x Example 2 Find the following integrals: (i) 3 2 1 x – dx ∫x (ii) (32 1) x dx + ∫ (iii) ∫ 23 1 ( 2 – ) + ∫ x x e dx x Solution (i) We have 3 2 2 1 – x – dx x dx – x dx x = ∫ ∫ ∫ (by Property V) 296 MATHEMATICS = 1 1 2 1 1 2 C C 1 1 2 1 – x x – – + +     + +     + +     ; C1, C2 are constants of integration = 2 1 1 2 C C 2 1 – x –x – – + = 2 1 2 1 + C C x2 – x + = 2 1 + C 2 x x + , where C = C 1 – C2 is another constant of integration �Note From now onwards, we shall write only one constant of integration in the final answer (ii) We have 2 2 3 3 ( 1) x dx x dx dx + = + ∫ ∫ ∫ = 2 1 3 C 2 1 3 x x + + + + = 335 C 5 x +x + (iii) We have 3 3 2 2 1 1 ( 2 ) 2 x x x e – dx x dx e dx – dx x x + = + ∫ ∫ ∫ ∫ = 3 1 2 2 – log + C 3 1 2 x x e x + + + = 25 2 2 – log + C 5 x x e x + Example 3 Find the following integrals: (i) (sin cos ) x x dx + ∫ (ii) cosec (cosec cot ) x x x dx + ∫ (iii) 2 1 –cossin xx dx ∫ Solution (i) We have (sin cos ) sin cos x x dx x dx x dx + = + ∫ ∫ ∫ = – cos sin C x x + + INTEGRALS 297 (ii) We have 2 (cosec (cosec + cot ) cosec cosec cot x x x dx x dx x x dx = + ∫ ∫ ∫ = – cot cosec C x – x + (iii) We have 2 2 2 1 sin 1 sin cos cos cos – x x dx dx – dx x x x = ∫ ∫ ∫ = sec2 tan sec x dx – x x dx ∫ ∫ = tan sec C x – x + Example 4 Find the anti derivative F of f defined by f (x) = 4x3 – 6, where F (0) = 3 Solution One anti derivative of f (x) is x4 – 6x since (4 6 ) d x – x dx = 4x3 – 6 Therefore, the anti derivative F is given by F(x) = x4 – 6x + C, where C is constant
1
3372-3375
Example 2 Find the following integrals: (i) 3 2 1 x – dx ∫x (ii) (32 1) x dx + ∫ (iii) ∫ 23 1 ( 2 – ) + ∫ x x e dx x Solution (i) We have 3 2 2 1 – x – dx x dx – x dx x = ∫ ∫ ∫ (by Property V) 296 MATHEMATICS = 1 1 2 1 1 2 C C 1 1 2 1 – x x – – + +     + +     + +     ; C1, C2 are constants of integration = 2 1 1 2 C C 2 1 – x –x – – + = 2 1 2 1 + C C x2 – x + = 2 1 + C 2 x x + , where C = C 1 – C2 is another constant of integration �Note From now onwards, we shall write only one constant of integration in the final answer (ii) We have 2 2 3 3 ( 1) x dx x dx dx + = + ∫ ∫ ∫ = 2 1 3 C 2 1 3 x x + + + + = 335 C 5 x +x + (iii) We have 3 3 2 2 1 1 ( 2 ) 2 x x x e – dx x dx e dx – dx x x + = + ∫ ∫ ∫ ∫ = 3 1 2 2 – log + C 3 1 2 x x e x + + + = 25 2 2 – log + C 5 x x e x + Example 3 Find the following integrals: (i) (sin cos ) x x dx + ∫ (ii) cosec (cosec cot ) x x x dx + ∫ (iii) 2 1 –cossin xx dx ∫ Solution (i) We have (sin cos ) sin cos x x dx x dx x dx + = + ∫ ∫ ∫ = – cos sin C x x + + INTEGRALS 297 (ii) We have 2 (cosec (cosec + cot ) cosec cosec cot x x x dx x dx x x dx = + ∫ ∫ ∫ = – cot cosec C x – x + (iii) We have 2 2 2 1 sin 1 sin cos cos cos – x x dx dx – dx x x x = ∫ ∫ ∫ = sec2 tan sec x dx – x x dx ∫ ∫ = tan sec C x – x + Example 4 Find the anti derivative F of f defined by f (x) = 4x3 – 6, where F (0) = 3 Solution One anti derivative of f (x) is x4 – 6x since (4 6 ) d x – x dx = 4x3 – 6 Therefore, the anti derivative F is given by F(x) = x4 – 6x + C, where C is constant Given that F(0) = 3, which gives, 3 = 0 – 6 × 0 + C or C = 3 Hence, the required anti derivative is the unique function F defined by F(x) = x4 – 6x + 3
1
3373-3376
�Note From now onwards, we shall write only one constant of integration in the final answer (ii) We have 2 2 3 3 ( 1) x dx x dx dx + = + ∫ ∫ ∫ = 2 1 3 C 2 1 3 x x + + + + = 335 C 5 x +x + (iii) We have 3 3 2 2 1 1 ( 2 ) 2 x x x e – dx x dx e dx – dx x x + = + ∫ ∫ ∫ ∫ = 3 1 2 2 – log + C 3 1 2 x x e x + + + = 25 2 2 – log + C 5 x x e x + Example 3 Find the following integrals: (i) (sin cos ) x x dx + ∫ (ii) cosec (cosec cot ) x x x dx + ∫ (iii) 2 1 –cossin xx dx ∫ Solution (i) We have (sin cos ) sin cos x x dx x dx x dx + = + ∫ ∫ ∫ = – cos sin C x x + + INTEGRALS 297 (ii) We have 2 (cosec (cosec + cot ) cosec cosec cot x x x dx x dx x x dx = + ∫ ∫ ∫ = – cot cosec C x – x + (iii) We have 2 2 2 1 sin 1 sin cos cos cos – x x dx dx – dx x x x = ∫ ∫ ∫ = sec2 tan sec x dx – x x dx ∫ ∫ = tan sec C x – x + Example 4 Find the anti derivative F of f defined by f (x) = 4x3 – 6, where F (0) = 3 Solution One anti derivative of f (x) is x4 – 6x since (4 6 ) d x – x dx = 4x3 – 6 Therefore, the anti derivative F is given by F(x) = x4 – 6x + C, where C is constant Given that F(0) = 3, which gives, 3 = 0 – 6 × 0 + C or C = 3 Hence, the required anti derivative is the unique function F defined by F(x) = x4 – 6x + 3 Remarks (i) We see that if F is an anti derivative of f, then so is F + C, where C is any constant
1
3374-3377
(ii) We have 2 2 3 3 ( 1) x dx x dx dx + = + ∫ ∫ ∫ = 2 1 3 C 2 1 3 x x + + + + = 335 C 5 x +x + (iii) We have 3 3 2 2 1 1 ( 2 ) 2 x x x e – dx x dx e dx – dx x x + = + ∫ ∫ ∫ ∫ = 3 1 2 2 – log + C 3 1 2 x x e x + + + = 25 2 2 – log + C 5 x x e x + Example 3 Find the following integrals: (i) (sin cos ) x x dx + ∫ (ii) cosec (cosec cot ) x x x dx + ∫ (iii) 2 1 –cossin xx dx ∫ Solution (i) We have (sin cos ) sin cos x x dx x dx x dx + = + ∫ ∫ ∫ = – cos sin C x x + + INTEGRALS 297 (ii) We have 2 (cosec (cosec + cot ) cosec cosec cot x x x dx x dx x x dx = + ∫ ∫ ∫ = – cot cosec C x – x + (iii) We have 2 2 2 1 sin 1 sin cos cos cos – x x dx dx – dx x x x = ∫ ∫ ∫ = sec2 tan sec x dx – x x dx ∫ ∫ = tan sec C x – x + Example 4 Find the anti derivative F of f defined by f (x) = 4x3 – 6, where F (0) = 3 Solution One anti derivative of f (x) is x4 – 6x since (4 6 ) d x – x dx = 4x3 – 6 Therefore, the anti derivative F is given by F(x) = x4 – 6x + C, where C is constant Given that F(0) = 3, which gives, 3 = 0 – 6 × 0 + C or C = 3 Hence, the required anti derivative is the unique function F defined by F(x) = x4 – 6x + 3 Remarks (i) We see that if F is an anti derivative of f, then so is F + C, where C is any constant Thus, if we know one anti derivative F of a function f, we can write down an infinite number of anti derivatives of f by adding any constant to F expressed by F(x) + C, C ∈ R
1
3375-3378
Given that F(0) = 3, which gives, 3 = 0 – 6 × 0 + C or C = 3 Hence, the required anti derivative is the unique function F defined by F(x) = x4 – 6x + 3 Remarks (i) We see that if F is an anti derivative of f, then so is F + C, where C is any constant Thus, if we know one anti derivative F of a function f, we can write down an infinite number of anti derivatives of f by adding any constant to F expressed by F(x) + C, C ∈ R In applications, it is often necessary to satisfy an additional condition which then determines a specific value of C giving unique anti derivative of the given function
1
3376-3379
Remarks (i) We see that if F is an anti derivative of f, then so is F + C, where C is any constant Thus, if we know one anti derivative F of a function f, we can write down an infinite number of anti derivatives of f by adding any constant to F expressed by F(x) + C, C ∈ R In applications, it is often necessary to satisfy an additional condition which then determines a specific value of C giving unique anti derivative of the given function (ii) Sometimes, F is not expressible in terms of elementary functions viz
1
3377-3380
Thus, if we know one anti derivative F of a function f, we can write down an infinite number of anti derivatives of f by adding any constant to F expressed by F(x) + C, C ∈ R In applications, it is often necessary to satisfy an additional condition which then determines a specific value of C giving unique anti derivative of the given function (ii) Sometimes, F is not expressible in terms of elementary functions viz , polynomial, logarithmic, exponential, trigonometric functions and their inverses etc
1
3378-3381
In applications, it is often necessary to satisfy an additional condition which then determines a specific value of C giving unique anti derivative of the given function (ii) Sometimes, F is not expressible in terms of elementary functions viz , polynomial, logarithmic, exponential, trigonometric functions and their inverses etc We are therefore blocked for finding ( ) ∫f x dx
1
3379-3382
(ii) Sometimes, F is not expressible in terms of elementary functions viz , polynomial, logarithmic, exponential, trigonometric functions and their inverses etc We are therefore blocked for finding ( ) ∫f x dx For example, it is not possible to find e– x2 dx ∫ by inspection since we can not find a function whose derivative is e– x2 298 MATHEMATICS (iii) When the variable of integration is denoted by a variable other than x, the integral formulae are modified accordingly
1
3380-3383
, polynomial, logarithmic, exponential, trigonometric functions and their inverses etc We are therefore blocked for finding ( ) ∫f x dx For example, it is not possible to find e– x2 dx ∫ by inspection since we can not find a function whose derivative is e– x2 298 MATHEMATICS (iii) When the variable of integration is denoted by a variable other than x, the integral formulae are modified accordingly For instance 4 1 4 5 1 C C 4 1 5 y y dy y + = + = + + ∫ 7
1
3381-3384
We are therefore blocked for finding ( ) ∫f x dx For example, it is not possible to find e– x2 dx ∫ by inspection since we can not find a function whose derivative is e– x2 298 MATHEMATICS (iii) When the variable of integration is denoted by a variable other than x, the integral formulae are modified accordingly For instance 4 1 4 5 1 C C 4 1 5 y y dy y + = + = + + ∫ 7 2
1
3382-3385
For example, it is not possible to find e– x2 dx ∫ by inspection since we can not find a function whose derivative is e– x2 298 MATHEMATICS (iii) When the variable of integration is denoted by a variable other than x, the integral formulae are modified accordingly For instance 4 1 4 5 1 C C 4 1 5 y y dy y + = + = + + ∫ 7 2 3 Comparison between differentiation and integration 1
1
3383-3386
For instance 4 1 4 5 1 C C 4 1 5 y y dy y + = + = + + ∫ 7 2 3 Comparison between differentiation and integration 1 Both are operations on functions
1
3384-3387
2 3 Comparison between differentiation and integration 1 Both are operations on functions 2
1
3385-3388
3 Comparison between differentiation and integration 1 Both are operations on functions 2 Both satisfy the property of linearity, i
1
3386-3389
Both are operations on functions 2 Both satisfy the property of linearity, i e
1
3387-3390
2 Both satisfy the property of linearity, i e , (i) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) d d d k f x k f x k f x k f x dx dx dx + = + (ii) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) k f x k f x dx k f x dx k f x dx + = + ∫ ∫ ∫ Here k1 and k2 are constants
1
3388-3391
Both satisfy the property of linearity, i e , (i) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) d d d k f x k f x k f x k f x dx dx dx + = + (ii) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) k f x k f x dx k f x dx k f x dx + = + ∫ ∫ ∫ Here k1 and k2 are constants 3
1
3389-3392
e , (i) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) d d d k f x k f x k f x k f x dx dx dx + = + (ii) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) k f x k f x dx k f x dx k f x dx + = + ∫ ∫ ∫ Here k1 and k2 are constants 3 We have already seen that all functions are not differentiable
1
3390-3393
, (i) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) d d d k f x k f x k f x k f x dx dx dx + = + (ii) [ ] 1 1 2 2 1 1 2 2 ( ) ( ) ( ) ( ) k f x k f x dx k f x dx k f x dx + = + ∫ ∫ ∫ Here k1 and k2 are constants 3 We have already seen that all functions are not differentiable Similarly, all functions are not integrable
1
3391-3394
3 We have already seen that all functions are not differentiable Similarly, all functions are not integrable We will learn more about nondifferentiable functions and nonintegrable functions in higher classes
1
3392-3395
We have already seen that all functions are not differentiable Similarly, all functions are not integrable We will learn more about nondifferentiable functions and nonintegrable functions in higher classes 4
1
3393-3396
Similarly, all functions are not integrable We will learn more about nondifferentiable functions and nonintegrable functions in higher classes 4 The derivative of a function, when it exists, is a unique function
1
3394-3397
We will learn more about nondifferentiable functions and nonintegrable functions in higher classes 4 The derivative of a function, when it exists, is a unique function The integral of a function is not so
1
3395-3398
4 The derivative of a function, when it exists, is a unique function The integral of a function is not so However, they are unique upto an additive constant, i
1
3396-3399
The derivative of a function, when it exists, is a unique function The integral of a function is not so However, they are unique upto an additive constant, i e
1
3397-3400
The integral of a function is not so However, they are unique upto an additive constant, i e , any two integrals of a function differ by a constant
1
3398-3401
However, they are unique upto an additive constant, i e , any two integrals of a function differ by a constant 5
1
3399-3402
e , any two integrals of a function differ by a constant 5 When a polynomial function P is differentiated, the result is a polynomial whose degree is 1 less than the degree of P
1
3400-3403
, any two integrals of a function differ by a constant 5 When a polynomial function P is differentiated, the result is a polynomial whose degree is 1 less than the degree of P When a polynomial function P is integrated, the result is a polynomial whose degree is 1 more than that of P
1
3401-3404
5 When a polynomial function P is differentiated, the result is a polynomial whose degree is 1 less than the degree of P When a polynomial function P is integrated, the result is a polynomial whose degree is 1 more than that of P 6
1
3402-3405
When a polynomial function P is differentiated, the result is a polynomial whose degree is 1 less than the degree of P When a polynomial function P is integrated, the result is a polynomial whose degree is 1 more than that of P 6 We can speak of the derivative at a point
1
3403-3406
When a polynomial function P is integrated, the result is a polynomial whose degree is 1 more than that of P 6 We can speak of the derivative at a point We never speak of the integral at a point, we speak of the integral of a function over an interval on which the integral is defined as will be seen in Section 7
1
3404-3407
6 We can speak of the derivative at a point We never speak of the integral at a point, we speak of the integral of a function over an interval on which the integral is defined as will be seen in Section 7 7
1
3405-3408
We can speak of the derivative at a point We never speak of the integral at a point, we speak of the integral of a function over an interval on which the integral is defined as will be seen in Section 7 7 7
1
3406-3409
We never speak of the integral at a point, we speak of the integral of a function over an interval on which the integral is defined as will be seen in Section 7 7 7 The derivative of a function has a geometrical meaning, namely, the slope of the tangent to the corresponding curve at a point
1
3407-3410
7 7 The derivative of a function has a geometrical meaning, namely, the slope of the tangent to the corresponding curve at a point Similarly, the indefinite integral of a function represents geometrically, a family of curves placed parallel to each other having parallel tangents at the points of intersection of the curves of the family with the lines orthogonal (perpendicular) to the axis representing the variable of integration
1
3408-3411
7 The derivative of a function has a geometrical meaning, namely, the slope of the tangent to the corresponding curve at a point Similarly, the indefinite integral of a function represents geometrically, a family of curves placed parallel to each other having parallel tangents at the points of intersection of the curves of the family with the lines orthogonal (perpendicular) to the axis representing the variable of integration 8
1
3409-3412
The derivative of a function has a geometrical meaning, namely, the slope of the tangent to the corresponding curve at a point Similarly, the indefinite integral of a function represents geometrically, a family of curves placed parallel to each other having parallel tangents at the points of intersection of the curves of the family with the lines orthogonal (perpendicular) to the axis representing the variable of integration 8 The derivative is used for finding some physical quantities like the velocity of a moving particle, when the distance traversed at any time t is known
1
3410-3413
Similarly, the indefinite integral of a function represents geometrically, a family of curves placed parallel to each other having parallel tangents at the points of intersection of the curves of the family with the lines orthogonal (perpendicular) to the axis representing the variable of integration 8 The derivative is used for finding some physical quantities like the velocity of a moving particle, when the distance traversed at any time t is known Similarly, the integral is used in calculating the distance traversed when the velocity at time t is known
1
3411-3414
8 The derivative is used for finding some physical quantities like the velocity of a moving particle, when the distance traversed at any time t is known Similarly, the integral is used in calculating the distance traversed when the velocity at time t is known 9
1
3412-3415
The derivative is used for finding some physical quantities like the velocity of a moving particle, when the distance traversed at any time t is known Similarly, the integral is used in calculating the distance traversed when the velocity at time t is known 9 Differentiation is a process involving limits
1
3413-3416
Similarly, the integral is used in calculating the distance traversed when the velocity at time t is known 9 Differentiation is a process involving limits So is integration, as will be seen in Section 7
1
3414-3417
9 Differentiation is a process involving limits So is integration, as will be seen in Section 7 7
1
3415-3418
Differentiation is a process involving limits So is integration, as will be seen in Section 7 7 INTEGRALS 299 10
1
3416-3419
So is integration, as will be seen in Section 7 7 INTEGRALS 299 10 The process of differentiation and integration are inverses of each other as discussed in Section 7
1
3417-3420
7 INTEGRALS 299 10 The process of differentiation and integration are inverses of each other as discussed in Section 7 2