IBDP Maths SL 5.8 Local maximum and minimum values AA HL Paper 2- Exam Style Questions- New Syllabus
Question
(ii) Explain why applying Euler’s method from the starting point \( (1, 0) \) does not produce an estimate for the negative part of the solution at \( x = 2 \).
Syllabus Topic Codes (IB Mathematics AA HL):
• AHL 5.18: Numerical solution of \( \frac{dy}{dx} = f(x,y) \) using Euler’s method — part (a)
▶️ Answer/Explanation
(a) Numerical Approximation using Euler’s Method
We use the iterative formula: \( x_{n+1} = x_n + h \) and \( y_{n+1} = y_n + h \cdot f(x_n, y_n) \), where \( h = 0.25 \) and \( f(x,y) = \frac{2x}{x^2 + y} \).
Step 1: Starting at \( (x_0, y_0) = (1, 0) \)
\( f(1, 0) = \frac{2(1)}{1^2 + 0} = 2 \)
\( y_1 = 0 + 0.25(2) = 0.5 \); \( x_1 = 1.25 \)
Step 2: At \( (1.25, 0.5) \)
\( f(1.25, 0.5) = \frac{2(1.25)}{(1.25)^2 + 0.5} = \frac{2.5}{1.5625 + 0.5} = \frac{2.5}{2.0625} \approx 1.21212 \dots \)
\( y_2 = 0.5 + 0.25(1.21212) \approx 0.80303 \); \( x_2 = 1.5 \)
Step 3: At \( (1.5, 0.80303) \)
\( f(1.5, 0.80303) = \frac{2(1.5)}{(1.5)^2 + 0.80303} = \frac{3}{2.25 + 0.80303} = \frac{3}{3.05303} \approx 0.98256 \dots \)
\( y_3 = 0.80303 + 0.25(0.98256) \approx 1.04867 \); \( x_3 = 1.75 \)
Step 4: At \( (1.75, 1.04867) \)
\( f(1.75, 1.04867) = \frac{2(1.75)}{(1.75)^2 + 1.04867} = \frac{3.5}{3.0625 + 1.04867} = \frac{3.5}{4.11117} \approx 0.85132 \dots \)
\( y_4 = 1.04867 + 0.25(0.85132) \approx 1.26150 \dots \); \( x_4 = 2.0 \)
Final Answer (3 s.f.): \( \boxed{y \approx 1.26} \)
(b) Justification of Results
(i) Analysis of Error:
From the diagram, the solution curve is concave down (\( f”(x) < 0 \)) over the interval \( [1, 2] \). In Euler’s method, the approximation is built using tangent line segments. Since tangents to a concave-down curve always lie above the curve, the numerical result is an overestimate.
(ii) Behaviour of the Method:
At the initial point \( (1, 0) \), the gradient is \( \frac{dy}{dx} = 2 \), which is positive. Euler’s method calculates the next point by following this positive slope in the direction of increasing \( x \). As long as \( x > 1 \), the calculated gradients remain positive, leading to positive \( y \)-values. The method is “trapped” in the upper branch and cannot jump to the negative branch of the relation without a change in the sign of the step or a singularity.
