Pythagorean
triples are sets of three integers that satisfy the equation a2 + b2
= c2. Examples of Pythagorean
triples include (3, 4, 5) and (5, 12, 13), because 32 + 42
= 52 and 52 + 122 = 132.
The most common approach to generate Pythagorean triples is to use Euclid’s method, in which you choose two integers m and n and generate a, b, and c by using the formulas a = m2 – n2, b = 2mn, and c = m2 + n2. For example, if you choose m = 2 and n = 1, then a = 22 – 12 = 3, b = 2∙2∙1 = 4, and c = 22 + 12 = 5, which is the (3, 4, 5) triple. This can be proven through substitution: a2 + b2 = (m2 – n2)2 + (2mn)2 = m4 – 2m2n2 + n4 + 4m2n2 = m4 + 2m2n2 + n4 = (m2 + n2)2 = c2.
But here’s a
different approach. We can take a known
Pythagorean triple, 32 + 42 = 52, but let x = 0,
which produces a quadratic equation (x + 3)2 + (x + 4)2 =
(x + 5)2. Since many
quadratic equations have two real answers, there may be one other real solution
for x that satisfies this equation, and in this case there is:
(x + 3)2
+ (x + 4)2 = (x + 5)2
(x2 + 6x
+ 9) + (x2 + 8x + 16) = (x2 + 10x + 25)
x2 + (6 +
8 – 10)x + 9 + 16 – 25 = 0
x2 + 4x =
0
x(x + 4) = 0
So x = 0
(which we already knew) and x = -4.
Plugging x = -4 back into the original quadratic equation (x + 3)2
+ (x + 4)2 = (x + 5)2, we get (-4 + 3)2 + (-4 +
4)2 = (-4 + 5)2 or (-1)2 + (0)2 =
(1)2, which means we generated a new Pythagorean triple (-1, 0, 1).
In other words, there are two sets of Pythagorean triples whose integers are
consecutive: (3, 4, 5) and (-1, 0, 1).
Now, you
might recognize that the triple (-1, 0, 1) is in the same family as (1, 0, 1)
or (0, 1, 1), because (-1)2 and (1)2 are both equal to 1,
and addition is commutative. The same
principle applies with the Pythagorean triple (3, 4, 5) we used in our above
example, so we can use (-3, 4, 5) instead.
In fact, we can use any combination of (±3, ±4, ±5). Because there are so many combinations, it is
helpful generalize our method with (±a, ±b, ±c), where (a, b, c) represent a
Pythagorean triple. So:
(x ± a)2
+ (x ± b)2 = (x ± c)2
(x2 ± 2ax
+ a2) + (x2 ± 2bx + b2) = (x2 ± 2cx
+ c2)
x2 + 2(±a
± b ∓ c)x + a2 + b2 –
c2 = 0
x2 + 2(±a
± b ∓ c)x = 0
x(x + 2(±a ± b ∓ c)) = 0
So x = 0 or
x = -2(±a ± b ∓ c) = ∓2a ∓ 2b ± 2c.
Plugging x = ∓2a ∓ 2b
± 2c back into the original quadratic equation (x ± a)2 + (x ± b)2
= (x ± c)2, we get (∓2a ∓ 2b ± 2c ± a)2 + (∓2a ∓ 2b ± 2c ± b)2 = (∓2a ∓ 2b ± 2c ± c)2 or (∓a ∓ 2b ± 2c)2 + (∓2a ∓ b ± 2c)2 = (∓2a ∓ 2b ± 3c)2. Therefore,
if (a, b, c) is a Pythagorean triple, then (∓a ∓ 2b ± 2c, ∓2a ∓ b ± 2c, ∓2a ∓ 2b ± 3c) are also Pythagorean triples.
For example,
since (3, 4, 5) is a Pythagorean triple, then (∓3 ∓ 2∙4 ± 2∙5, ∓2∙3 ∓ 4 ± 2∙5, ∓2∙3 ∓ 2∙4 ± 3∙5) are also Pythagorean triples. All the possibilities include:
(-3 – 2∙4 + 2∙5, -2∙3 – 4 + 2∙5,
-2∙3 – 2∙4 + 3∙5) = (-1, 0, 1) → (0, 1, 1)
(3 – 2∙4 + 2∙5, 2∙3 – 4 + 2∙5, 2∙3
– 2∙4 + 3∙5) = (5, 12, 13) → (5, 12, 13)
(-3 + 2∙4 + 2∙5, -2∙3 + 4 + 2∙5,
-2∙3 + 2∙4 + 3∙5) = (15, 8, 17) → (8, 15, 17)
(-3 – 2∙4 – 2∙5, -2∙3 – 4 – 2∙5,
-2∙3 – 2∙4 – 3∙5) = (-21, -20, -29) → (20, 21, 29)
(3 + 2∙4 – 2∙5, 2∙3 + 4 – 2∙5, 2∙3
+ 2∙4 – 3∙5) = (1, 0, -1) → (0, 1, 1)
(-3 + 2∙4 – 2∙5, -2∙3 + 4 – 2∙5,
-2∙3 + 2∙4 – 3∙5) = (-5, -12, -13) → (5, 12, 13)
(3 – 2∙4 – 2∙5, 2∙3 – 4 – 2∙5, 2∙3
– 2∙4 – 3∙5) = (-15, -8, -17) → (8, 15, 17)
(3 + 2∙4 + 2∙5, 2∙3 + 4 + 2∙5, 2∙3
+ 2∙4 + 3∙5) = (21, 20, 29) → (20, 21, 29)
So using this
method, the Pythagorean triple (3, 4, 5) generates four new families of
Pythagorean triples which include (0, 1, 1), (5, 12, 13), (8, 15, 17), and (20,
21, 29).
This Pythagorean
triple generator can be proved by expanding (∓a ∓ 2b ± 2c)2 + (∓2a ∓ b ± 2c)2 = (∓2a ∓ 2b ± 3c)2 and simplifying it to a2
+ b2 = c2:
(∓a ∓ 2b ± 2c)2 +
(∓2a ∓ b ±
2c)2 = (∓2a ∓ 2b
± 3c)2
(a2 + 4b2
+ 4c2 ± 4ab ∓ 4ac ∓
8bc) + (4a2 + b2 +
4c2 ± 4ab ∓ 8ac ∓
4bc) = 4a2 + 4b2 + 9c2 ± 8ab ∓ 12ac∓12bc
5a2 + 5b2
+ 8c2 ± 8ab ∓ 12ac ∓
12bc = 4a2 + 4b2 + 9c2 ± 8ab ∓ 12ac∓12bc
5a2 + 5b2
+ 8c2 = 4a2 + 4b2 + 9c2
a2 + b2
= c2
A quick Google
search shows that this Pythagorean triple generator was first published by a
Swedish mathematician named Berggen in 1934.
I am not sure if he arrived to it the same way as I did.
No comments:
Post a Comment