Derivatives and stationary points
Introducido| English | Significado |
|---|---|
| derivative/dɪˈrɪvətɪv/ | derivada |
| stationary point/ˈsteɪʃənəri pɔɪnt/ | A point on a differentiable curve where its derivative is zero. |
| optimisation/ˌɒptɪmaɪˈzeɪʃn/ | Finding the greatest or least attainable value of an objective while satisfying the stated constraints. |
Fermat's method for maxima, 1630s
- In the 1630s Pierre de Fermat found the largest and smallest values of curves by a trick: compare $f(x)$ with $f(x + e)$ for a tiny $e$, and at a maximum the two are almost equal. Decades before Newton and Leibniz, this is the idea that the gradient is zero at a peak.
- The derivative 导数 $f'(x)$ is the gradient of a curve. Where $f'(x) = 0$ the curve has a stationary point 驻点, and optimisation 最优化 uses this to find the best value.
At a smooth maximum the gradient of a curve is zero.
Dificultad 1/5 · 100 monedas + un pequeño bono en el primer intento de hoy
Fermat's insight.
The derivative
- $f'(x) = \displaystyle\lim_{h\to0}\frac{f(x + h) - f(x)}{h}$. Power rule: $\dfrac{d}{dx}x^n = nx^{n-1}$.
- The tangent at $x = a$ has gradient $f'(a)$; the normal is perpendicular to it.
Differentiate 4x³ − 5x.
Dificultad 1/5 · 100 monedas + un pequeño bono en el primer intento de hoy
nxⁿ⁻¹.
Find the gradient of y = x² + 3x at x = 2.
Dificultad 2/5 · 140 monedas + un pequeño bono en el primer intento de hoy
2x + 3 at x = 2.
Stationary points
- Solve $f'(x) = 0$. Classify with $f''(x)$: negative → maximum, positive → minimum, or check the sign of $f'$ on each side.
Derivatives and stationary points
For y=ax^n, dy/dx=anx^(n-1)
Compare the model with the worked case and explain one change.
Where are the stationary points of y = x³ − 3x?
Dificultad 1/5 · 100 monedas + un pequeño bono en el primer intento de hoy
3x² − 3 = 0.
f''(a) < 0 at a stationary point means
Dificultad 1/5 · 100 monedas + un pequeño bono en el primer intento de hoy
Concave down.
Increasing, decreasing and concavity
- $f'(x) > 0$: increasing. $f'(x) < 0$: decreasing. $f''(x) > 0$: concave up; a point of inflection is where concavity changes.
Worked example: the biggest box
Squares of side $x$ cm are cut from the corners of a $12 \times 12$ cm card, which is folded into an open box.
- Volume: $V = x(12 - 2x)^2$, with $0 < x < 6$.
- Derivative: $V' = (12 - 2x)^2 - 4x(12 - 2x) = (12 - 2x)(12 - 6x)$. $V' = 0$ at $x = 2$ (or $x = 6$, outside the domain).
- Check: $V'' = 24x - 96$, which is $-48$ at $x = 2$, so a maximum: $V = 2 \times 8^2 = 128\text{ cm}^3$.
What is the maximum volume of the box from the 12 × 12 card?
Dificultad 2/5 · 140 monedas + un pequeño bono en el primer intento de hoy
x = 2, V = 2 × 64.
Optimisation steps
- Write the quantity in one variable, state the domain, differentiate, solve $= 0$, classify, and check the endpoints.
$f'(x) = 0$ does not guarantee a maximum or minimum. $y = x^3$ has $y' = 0$ at $x = 0$ but keeps increasing — a stationary point of inflection. Always classify.
y = x³ has a maximum at x = 0.
Dificultad 1/5 · 100 monedas + un pequeño bono en el primer intento de hoy
It is a stationary point of inflection.
The derivative is the gradient: power rule $nx^{n-1}$. Stationary points satisfy $f'(x) = 0$; classify with $f''$ or sign changes. For optimisation, model in one variable, respect the domain, differentiate, classify and check endpoints.
A farmer has 100 m of fence for a rectangular pen against a wall (three sides fenced). Find the dimensions giving the largest area.
Model, differentiate, classify.