Kvant
University Applicant's Workshop

Substitution method for solving equations and systems of equations

Сканави М. И.

Rough automatic draft, low quality: expect mistakes and awkward wording. It will be replaced by a better translation. Original →

43

not containing radicals of the second degree!): let’s compose the difference of the subdecimal expressions of the radicals located on the left, and we get that (t+3)−t=3(t+3) - t=3.

Now we factorize this difference (using the formula for the difference of squares) and find that (t+3+t)(t+3−t)=3(\sqrt{t+3} + \sqrt{t})(\sqrt{t+3} - \sqrt{t}) = 3.

From here, by dividing by the original equation (with respect to tt), we obtain that

t+3−t=32t+7.\sqrt{t+3} - \sqrt{t} = \frac{3}{\sqrt{2t+7}}.

Subtracting from the original equation what we now have, we find that

2t=2t+7−32t+7,2\sqrt{t} = \sqrt{2t+7} - \frac{3}{\sqrt{2t+7}},

2t=2t+42t+7,t=t+22t+7.2\sqrt{t} = \frac{2t+4}{\sqrt{2t+7}}, \sqrt{t} = \frac{t+2}{\sqrt{2t+7}}.

The procedure we have just carried out allows us to get rid of radicals by simply squaring the parts of the equation. Let’s do this:

t=t2+4t+42t+7,t2+3t−4=0.t = \frac{t^2+4t+4}{2t+7}, t^2+3t-4=0.

Hence t1=1,t2=−4t_1=1, t_2=-4. Next, note that we are only interested in positive values ​​of tt, since t=x2+x+1>0t = x^2+x+1 > 0

for all values ​​of xx. Therefore, we discard the unsuitable root t2t_2 and obtain one quadratic equation for finding xx

x2+x+1=1,x2+x=0x^2+x+1=1, \quad x^2+x=0

with roots x1=−1x_1=-1 and x2=0x_2=0. It is easy to check that both of them are roots of this irrational equation.

Example 5. Solve a system of equations

{x3y+xy=40,y3x+xy=10.\begin{cases} \frac{x^3}{y} + xy = 40, \\ \frac{y^3}{x} + xy = 10. \end{cases}

(Limit yourself to finding real solutions).

Solving a system of equations students practice a significant solution. At the same time, in these cases it often turns out to be very effective. Before going to the solution of the proposed system, I will allow myself to express some general messages about the substitution method when solving a system of two equations with two exceptions. If these latter are designated by the letters xx and yy, then it is often useful to move on to the new quantities ρ\rho and φ\varphi in accordance with the following equalities:

x=ρcos⁡φ,x = \rho \cos \varphi,

y=ρsin⁡φ.y = \rho \sin \varphi.

The reader who is sufficiently familiar with the method of coordinates on the plane, almost that if xx and yy serve as the Cartesian coordinates of some point, then ρ\rho and φ\varphi are the so-called polar coordinates of the same point. The geometric meaning of polar coordinates and their connection with Cartesian ones will become clear from figure, which we now recommend everyone to CAREFULLY consider:

44

We note that the origin OO has ρ=0\rho=0, and the angle φ\varphi has no defined value for it. All other points of the plane will be subordinated to a single pair of polar coordinates (ρ,φ\rho, \varphi), if we require that there be

0⩽ρ<+∞,0 \leqslant \rho < +\infty,
0⩽φ<2π.0 \leqslant \varphi < 2\pi.

We use equalities (!) to solve this system of equations, for which instead of xx and yy in the domain of the system equation we substitute their expressions through ρ\rho and φ\varphi; we get that

{ρ3cos⁡3φρsin⁡φ+ρcos⁡φρsin⁡φ=40,ρ3sin⁡3φρcos⁡φ+ρcos⁡φρsin⁡φ=10.\begin{cases} \frac{\rho^3 \cos^3 \varphi}{\rho \sin \varphi} + \rho \cos \varphi \rho \sin \varphi = 40, \\ \frac{\rho^3 \sin^3 \varphi}{\rho \cos \varphi} + \rho \cos \varphi \rho \sin \varphi = 10. \end{cases}

or after processing that

{ρ2ctgφ=40,ρ2tgφ=10.\begin{cases} \rho^2 \text{ctg} \varphi = 40, \\ \rho^2 \text{tg} \varphi = 10. \end{cases}

By dividing the first equation by the second we find that

ctg2φ=4.\text{ctg}^2 \varphi = 4.

Hence ctgφ=2\text{ctg} \varphi = 2 or ctgφ=−2\text{ctg} \varphi = -2. The second (negative) value ctgφ\text{ctg} \varphi is immediately discarded, since we have ctgφ\text{ctg} \varphi and ρ2\rho^2 having the same sign, and ρ2\rho^2 cannot be negative. Now, using the known cotangent, we need to find the sine and cosine of the angle φ\varphi. I follow this without words:

cos⁡2φ=1+ctg2φ=5,\cos^2 \varphi = 1 + \text{ctg}^2 \varphi = 5,
cos⁡φ=5,\cos \varphi = \sqrt{5},
cos⁡φ=−5,\cos \varphi = -\sqrt{5},

sin⁡φ=15,sin⁡φ=−15,\sin \varphi = \frac{1}{\sqrt{5}}, \quad \sin \varphi = -\frac{1}{\sqrt{5}},
cos⁡φ=25,cos⁡φ=−25.\cos \varphi = \frac{2}{\sqrt{5}}, \quad \cos \varphi = -\frac{2}{\sqrt{5}}.

Next we find ρ\rho from the fact that

ρ2ctgφ=ρ2⋅2=40,ρ2=20.\rho^2 \text{ctg} \varphi = \rho^2 \cdot 2 = 40, \quad \rho^2 = 20.

We get that ρ=25\rho = 2\sqrt{5}. Using equalities (!) using the known ρ\rho and φ\varphi, we find the following two pairs of numbers as a solution to this system of equations:

x1=25⋅25=4,x_1 = 2\sqrt{5} \cdot \frac{2}{\sqrt{5}} = 4,
y1=25⋅15=2;y_1 = 2\sqrt{5} \cdot \frac{1}{\sqrt{5}} = 2;
x2=25(−25)=−4,x_2 = 2\sqrt{5} \left(-\frac{2}{\sqrt{5}}\right) = -4,
y2=25(−15)=−2.y_2 = 2\sqrt{5} \left(-\frac{1}{\sqrt{5}}\right) = -2.

It is incorrect to assume that this system can be solved only by passing to polar coordinates (it can be solved, for example, also this way: transfer xyxy from the left sides of the equation to the right, then recombine the results and obtain a quadratic equation for xyxy; the rest is obvious). Meanwhile, this technique should be adopted. It allows us, in a certain sense, to regulate our transformations and makes it possible to use the entire apparatus of trigonometry. In what cases is it beneficial to use this technique, and in what cases is it not? Your experience helps answer this question, which you can enrich with your own work!

While there is time before the entrance exams, act!