Are the solutions of the diophantine equations right?

  • MHB
  • Thread starter evinda
  • Start date
In summary, the diophantine equations $18x+5y=48$ and $158x-57y=7$ have infinite solutions, given by the formulas $x=48x_0+5k, y=48y_0-18k$ and $x=7x_0+57k, y=-7y_0+158k$, respectively. The solutions are subject to the constraints $k>-\frac{48x_0}{5}, k<\frac{48y_0}{18}$ and $k>\max \{\frac{-7x_0}{57},\frac{7y_0}{158}
  • #1
evinda
Gold Member
MHB
3,836
0
Hi! (Sun)

I have to solve the diophantine equations:

  • $18x+5y=48$
  • $158x-57y=7$

for $x,y >0$

That's what I have tried:

  • $$gcd(18,5)=1, \text{ so } 18x+5y=48 \text{ has infinite solutions.}$$

    $$\text{ As } gcd(18,15)=1, \exists x_0,y_0 \text{ such that } 18x_0+5y_0=1 \Rightarrow 18 (48x_0)+5(48y_0)=48$$

    $$\text{So, } x_1=48x_0 , y_1=48y_0 \text{ is a solution of the diophantine equation}$$

    The solutions are given by the formulas:

    $$x=48x_0+18k , y=48y_0+5k, k \in \mathbb{Z}$$

    $$x>0 \Rightarrow 48x_0+18k>0 \Rightarrow k>-\frac{48x_0}{18}$$

    $$y>0 \Rightarrow 48y_0+5k>0 \Rightarrow k>-\frac{48y_0}{5}$$

    $$ \text{ Therefore, }k> \max \{ -\frac{48x_0}{18}, -\frac{48y_0}{5}\}$$
  • $$gcd(158,57)=1, \text{ so the diophantine equation } 158x-57y=7 \text{ has infinite solutions.}$$

    Since, $gcd(158,57)=1$, $\exists x_0,y_0 \text{ such that } 158x_0+57y_0=1 \Rightarrow 158(7 x_0)-57(-7y_0)=7$

    So,$x_1=7x_0 \text{ and } y_1=-7y_0$ is a solution of the diophantine equation.

    So,the solutions from $ 158x-57y=7$ are given by the formulas:

    $$x=7x_0+158k, y=-7y_0+57k , k \in \mathbb{Z}$$

    $$x>0 \Rightarrow k > \frac{-7x_0}{158}$$

    $$y>0 \Rightarrow k>\frac{7y_0}{57}$$

    Therefore, $k>\max \{ \frac{-7x_0}{158},\frac{7y_0}{57} \}$

Could you tell me if it is right or if I have done something wrong? (Thinking)
 
Mathematics news on Phys.org
  • #2
I thought about it again and now I think that it is like that:

The solutions of the diophantine equation $18x+5y=48$ are given by the formulas:

$$x=48x_0+5k, y=48y_0+18k, \text{ with } k> \max \{ \frac{-48x_0}{18}, \frac{-48y_0}{5}\}$$

and the solutions of the diophantine equation $158x-57y=7$ are given by the formulas:

$$x=7x_0+57k, y=-7y_0+158k, \text{ with } k> \max \{ \frac{-7x_0}{158},\frac{7y_0}{57} \}$$

Am I right? (Blush)
 
  • #3
Equation: \(\displaystyle 18x+5y=48\)

Has the solution:

\(\displaystyle x=5k+1\)

\(\displaystyle y=6-18k\)

equation: \(\displaystyle 158x-57y=7\)

Decisions no.
 
  • #4
Hey! (Smile)

evinda said:
I thought about it again and now I think that it is like that:

The solutions of the diophantine equation $18x+5y=48$ are given by the formulas:

$$x=48x_0+5k, y=48y_0+18k, \text{ with } k> \max \{ \frac{-48x_0}{18}, \frac{-48y_0}{5}\}$$

Let's substitute your solution:
$$18(48x_0+5k)+5(48y_0+18k)=48$$
$$48(18x_0+5y_0)+18\cdot 5k +18\cdot 5k=48$$
Hmm... that left hand side becomes bigger whenever we pick $k$ bigger.
That can't be right! (Worried)

And what are $x_0$ and $y_0$?
Can you apply the Euclidean algorithm? (Thinking)
 
  • #5
I like Serena said:
Hey! (Smile)
Let's substitute your solution:
$$18(48x_0+5k)+5(48y_0+18k)=48$$
$$48(18x_0+5y_0)+18\cdot 5k +18\cdot 5k=48$$
Hmm... that left hand side becomes bigger whenever we pick $k$ bigger.
That can't be right! (Worried)

And what are $x_0$ and $y_0$?
Can you apply the Euclidean algorithm? (Thinking)

I found it like that:

$$18=5 \cdot 3+3$$
$$5=3 \cdot 1+ 2$$
$$3= 2 \cdot 1 + 1$$
$$2=2 \cdot 1+0$$

So, $gcd(18,5)=1$

So, $\exists x_0,y_0$ such that:
$$18x_0+5y_0=1 \Rightarrow 18(48x_0)+5(48y_0)=48$$

Therefore $\displaystyle{x_1=48x_0 , y_1=48y_0} \text{ is a solution of the diophantine equation } 18x+5y=48 $

So,the solutions of $18x+5y=48$ are given by the formulas:

$$x=x_1+kb_1 ,y=y_1-ka_1$$

$$a_1=18, b_1=5$$

So, $x=48x_0+5k ,y=48y_0-18k$

$$x>0 \Rightarrow k>\frac{-48x_0}{5} , y>0 \Rightarrow k< \frac{48y_0}{18} \Rightarrow \frac{-48x_0}{5}<k<\frac{48y_0}{18}$$

Have I done something wrong? (Thinking) (Thinking) (Thinking)
 
  • #6
One way of arriving at a result:
$$18x + 5y = 48 ~ ~ \iff ~ ~ 18(x - 1) + 5y = 30 ~ ~ \iff ~ ~ 18(x - 1) + 5(y - 6) = 0 ~ ~ \iff ~ ~ 18(x - 1) = 5(6 - y)$$
Substitute $u = x - 1$, $v = 6 - y$, then we must solve $18u = 5v$. The intersection of the set $\{ 18u \mid u \in \mathbb{Z} \}$ and $\{ 5v \mid v \in \mathbb{Z} \}$ is the set of integers that are multiples of both $18$ and $5$. Now $\mathrm{lcm}(18, 5) = 18 \times 5 = 90$ since they are coprime, so the intersection consists of the multiples of $90$, that is, the solutions are $(u, v) = (5n, 18n)$ for all $n \in \mathbb{Z}$. Substituting back, $(x, y) = (5n + 1, 6 - 18n)$. And finally, adding the range constraints:
$$5n + 1 > 0 ~ ~ \implies ~ ~ 5n > 1 ~ ~ \implies ~ ~ n \geq 0$$
$$6 - 18n > 0 ~ ~ \implies ~ ~ 18n < 6 ~ ~ \implies n < 1$$
Therefore $0 \leq n < 1$, and so there is a unique solution at $n = 0$, which is $(u, v) = (0, 0)$, so $(x, y) = (1, 6)$.

This is what the first few solutions look like (note no other solution has both $x$ and $y$ positive):

Code:
(-9, 42)
(-4, 24)
(1, 6)
(6, -12)
(11, -30)
(16, -48)

Here it's the range constraint that is killing all the solutions except the smallest one: without it, you would indeed have infinitely many solutions. Remember: Bézout's identity considers all integers, both positive and negative (since it operates on a principal ideal domain, in this case the ring of integers) so it does not apply directly to natural numbers - you need to check that separately.

Once you work out a systematic technique to solve a diophantine equation of the form $ax + by = c$ with $\gcd(a, b) = 1$ (and generalized it to handle the case where $\gcd(a, b) > 1$, with constraints on $c$ for the existence of a solution) you will have rediscovered the extended Euclidean algorithm :)

For this problem it was a bit simpler because the linear combination of $18$ and $5$ that gave $48$ was obvious - in the general case you would work out that linear combination via extended Euclid, as said above).​
 
Last edited:
  • #7
And..that's how I found the solutions for the second diophantine equation:

$$158=57 \cdot 2+44$$

$$57=44 \cdot 1+13$$

$$44=13 \cdot 3+5$$

$$13=5 \cdot 2+3$$
$$5=3 \cdot 1+2$$

$$3=2 \cdot 1+1$$
$$2=2 \cdot 1+ 0$$

So, $gcd(158,57)=1$

$$\text{So , }\exists x_0, y_0 \text{ such that : } 158x_0+57y_0=1 \Rightarrow 158(7x_0)-57(-7y_0)=7 $$

$$\text{So, } x_1=7x_0 , y_1=-7y_0 \text{ is a solution of the diophantine equation: } 158x-57y=7$$

$$a_1=158, b_1=57$$

$$\text{So, } x=7x_0+57k , y=-7y_0+158k \text{ are the solutions of the diophantine equation } 158x-57y=7$$

$$x>0 \Rightarrow 7x_0+57k>0 \Rightarrow k>\frac{-7x_0}{57}, y>0 \Rightarrow -7y_0+158k>0 \Rightarrow k>-\frac{7y_0}{158} \Rightarrow k>\max \{\frac{-7x_0}{57},\frac{7y_0}{158} \}$$
 
  • #8
evinda said:
I found it like that:

$$18=5 \cdot 3+3$$
$$5=3 \cdot 1+ 2$$
$$3= 2 \cdot 1 + 1$$
$$2=2 \cdot 1+0$$

So, $gcd(18,5)=1$

So, $\exists x_0,y_0$ such that:
$$18x_0+5y_0=1 \Rightarrow 18(48x_0)+5(48y_0)=48$$

Sorry. (Blush)

I meant the Extended Euclidean algorithm.
It means you keep track of a linear combination from step to step.

Let me show you:
\begin{array}{| l | r c r l l |}
\hline
\text{Euclidean algorithm} & \text{Extended} \\
\hline
18=5 \cdot 3+3 & 18&=&1 \cdot 18 \\
5=3 \cdot 1+ 2 & 5 &= & && 1 \cdot 5 \\
3= 2 \cdot 1 + 1 & 3 &=& 1\cdot 18 &-& 3 \cdot 5 \\
2=2 \cdot 1+0 & 2 &=& -1 \cdot 18 &+& 4 \cdot 5 \\
& 1 &=& 2 \cdot 18 &-&7 \cdot 5 \\
\hline
\end{array}

Can you see what $x_0$ and $y_0$ should be now? (Thinking)
evinda said:
So,the solutions of $18x+5y=48$ are given by the formulas:

$$x=x_1+kb_1 ,y=y_1-ka_1$$

$$a_1=18, b_1=5$$

So, $x=48x_0+5k ,y=48y_0-18k$

$$x>0 \Rightarrow k>\frac{-48x_0}{5} , y>0 \Rightarrow k< \frac{48y_0}{18} \Rightarrow \frac{-48x_0}{5}<k<\frac{48y_0}{18}$$

Have I done something wrong? (Thinking) (Thinking) (Thinking)

You fixed the minus sign! Much better! (Angel)
 
  • #9
I like Serena said:
Sorry. (Blush)

I meant the Extended Euclidean algorithm.
It means you keep track of a linear combination from step to step.

Let me show you:
\begin{array}{| l | r c r l l |}
\hline
\text{Euclidean algorithm} & \text{Extended} \\
\hline
18=5 \cdot 3+3 & 18&=&1 \cdot 18 \\
5=3 \cdot 1+ 2 & 5 &= & && 1 \cdot 5 \\
3= 2 \cdot 1 + 1 & 3 &=& 1\cdot 18 &-& 3 \cdot 5 \\
2=2 \cdot 1+0 & 2 &=& -1 \cdot 18 &+& 4 \cdot 5 \\
& 1 &=& 2 \cdot 18 &-&7 \cdot 5 \\
\hline
\end{array}

Can you see what $x_0$ and $y_0$ should be now? (Thinking)

Oh,yes! Sorry! I forgot it... (Blush) $x_0=2, y_0=-7$

I like Serena said:
You fixed the minus sign! Much better! (Angel)

So,now it is
$$ \frac{-48x_0}{18}<k<\frac{48y_0}{18} \Rightarrow \frac{-48}{9} < k< -\frac{336}{18}$$

So,there are no solutions,or am I wrong?? (Thinking) :confused: (Thinking)
 
  • #10
evinda said:
Oh,yes! Sorry! I forgot it... (Blush) $x_0=2, y_0=-7$

Good!
So,now it is
$$ \frac{-48x_0}{18}<k<\frac{48y_0}{18} \Rightarrow \frac{-48}{9} < k< -\frac{336}{18}$$

So,there are no solutions,or am I wrong?? (Thinking) :confused: (Thinking)

Well... (Thinking)
We have:
$$18x+5y=48$$
If we pick $x=1$ and $y=6$ we have a solution! (Doh)
 
  • #11
I don't understand. Why such a simple equation so long to decide?
The solution was recorded immediately!
 
  • #12
individ said:
I don't understand. Why such a simple equation so long to decide?
The solution was recorded immediately!

I think the idea is to convey an understanding of why and how this solution was determined to evinda. Just giving the solution with no explanation is not particularly useful to furthering that understanding, you know, the whole give a man a fish thing (and ILS seems to have taken a liking to tutoring evinda :p)

edit: this isn't the challenge questions subforum
 
  • #13
I like Serena said:
Good!

Well... (Thinking)
We have:
$$18x+5y=48$$
If we pick $x=1$ and $y=6$ we have a solution! (Doh)

But..how can this be? :eek: (Thinking) Isn't it $\frac{-48}{9}<k<\frac{-16}{3}$ ? So,do we take $k=\frac{-16}{3}$, and this will be the only $k$, for which we have a solution?? :confused:
 
  • #14
evinda said:
But..how can this be? :eek: (Thinking) Isn't it $\frac{-48}{9}<k<\frac{-16}{3}$ ? So,do we take $k=\frac{-16}{3}$, and this will be the only $k$, for which we have a solution?? :confused:

There must be a mistake somewhere! (Sweating)

What would the value of $k$ be that would yield $x=1$ and $y=6$?

What would you get if you fill in $k=\frac{-16}3$?
 
  • #15
I like Serena said:
There must be a mistake somewhere! (Sweating)

What would the value of $k$ be that would yield $x=1$ and $y=6$?

What would you get if you fill in $k=\frac{-16}3$?

I tried it again and I found:

$$x=2 \cdot 48+5k, y=-7 \cdot 48-18k$$

$$x>0 \Rightarrow k> \frac{-96}{5}$$

$$y>0 \Rightarrow k<\frac{-56}{3}$$

So,it is:

$$\frac{-96}{5}<k< \frac{-56}{3} \Rightarrow k=-19$$

$$\text{For } k=-19, \text{ we get : } x=1,y=6 \text{ ! } $$

(Happy)
 
  • #16
I tried also again to solve the second diophantine equation.
That's what I did:

$$158x-57y=7$$

$$gcd(158,57)=1, \text{ so } \exists x_0,y_0 \text{ such that } 158x_0+57y_0=1$$

From the Extended Euclidean algorithm,it is like that:

$$(-22) \cdot 158-57 \cdot (-61)=1 \Rightarrow x_0=-22 , y_0=-61$$

$$(-22 \cdot 7) \cdot 158-57 \cdot (-7 \cdot 61)=7$$

So, $\displaystyle{x_1=-154 , y_1=-427} \text{ is a solution of the diophantine equation.}$

Therefore,the solutions are given by the formulas:

$$x=-154+57k , y=-427+158k $$

$$x>0 \Rightarrow k>\frac{154}{57} \Rightarrow k \geq 3$$

$$y>0 \Rightarrow k>\frac{427}{158} \Rightarrow k \geq 3$$

So,it must be $k \geq 3$.

Is it right or have I done something wrong? (Thinking)
 
  • #17
evinda said:
$$\text{For } k=-19, \text{ we get : } x=1,y=6 \text{ ! } $$

(Happy)

Good! (Nod)
evinda said:
I tried also again to solve the second diophantine equation.
That's what I did:

$$158x-57y=7$$

$$gcd(158,57)=1, \text{ so } \exists x_0,y_0 \text{ such that } 158x_0+57y_0=1$$

From the Extended Euclidean algorithm,it is like that:

$$(-22) \cdot 158-57 \cdot (-61)=1 \Rightarrow x_0=-22 , y_0=-61$$

$$(-22 \cdot 7) \cdot 158-57 \cdot (-7 \cdot 61)=7$$

So, $\displaystyle{x_1=-154 , y_1=-427} \text{ is a solution of the diophantine equation.}$

Therefore,the solutions are given by the formulas:

$$x=-154+57k , y=-427+158k $$

$$x>0 \Rightarrow k>\frac{154}{57} \Rightarrow k \geq 3$$

$$y>0 \Rightarrow k>\frac{427}{158} \Rightarrow k \geq 3$$

So,it must be $k \geq 3$.

Is it right or have I done something wrong? (Thinking)

Looks good... (Sweating)

Which solution do you get if you fill in $k=3$?
Does the solution fit the original equation? (Wondering)

How about $k=4$? (Wondering)
 
  • #18
I like Serena said:
Good! (Nod)

Looks good... (Sweating)

Which solution do you get if you fill in $k=3$?
Does the solution fit the original equation? (Wondering)

How about $k=4$? (Wondering)

For $k=3$,we get:

$$x=17, y=47$$

$$158x-57y=158 \cdot 17-57 \cdot 47=7 \checkmark$$

For $k=4$,we get:

$$x=74, y=205$$

$$158 \cdot 74-57 \cdot 205=7 \checkmark $$

So,is it right? (Nerd)
 
  • #19
Yep!
You have just verified that you did not make mistakes. (Mmm)
 
  • #20
I like Serena said:
Yep!
You have just verified that you did not make mistakes. (Mmm)

Great! (Clapping)(Clapping) Thank you very much! (Smile)
 

FAQ: Are the solutions of the diophantine equations right?

What are diophantine equations?

Diophantine equations are polynomial equations with integer coefficients, where the solutions must also be integers. They are named after the ancient Greek mathematician Diophantus.

How do you know if a solution to a diophantine equation is correct?

The solutions to diophantine equations can be verified by plugging them back into the equation and checking if they satisfy it. Additionally, there are algorithms and methods that can be used to find all the solutions to a diophantine equation.

Can diophantine equations have infinitely many solutions?

Yes, some diophantine equations can have infinitely many solutions. For example, the equation x^2 + y^2 = z^2, known as the Pythagorean equation, has infinitely many solutions.

How are diophantine equations used in real-world applications?

Diophantine equations have various applications in fields such as cryptography, coding theory, and number theory. They are also used in problems involving discrete quantities, such as finding the number of ways to distribute objects or arranging objects in patterns.

Are there any unsolved diophantine equations?

Yes, there are many unsolved diophantine equations, some of which have been open for centuries. One famous example is Fermat's Last Theorem, which states that there are no positive integer solutions to the equation x^n + y^n = z^n for n>2. This was famously proven by Andrew Wiles in 1995.

Back
Top