你的私隱選擇

KiwiBee 使用 Google Analytics、Google Ads 及 Meta Pixel 的可選 Cookie,以了解成年訪客、統計已完成的註冊及改善推廣活動。必要 Cookie 會一直保持啟用。 OpenAI 的 ChatGPT 廣告像素亦會在您同意廣告追蹤後衡量頁面瀏覽及已提交的查詢,並可能使用經雜湊處理的聯絡資料作配對。 私隱政策

跳到主要內容

Gradient & Tangent Plane 3D

  • 數學
—

與此模擬實驗相關

按意思從遊戲、工作紙、模擬實驗、教師工具和影片中揀選。

關於這個模擬實驗

Gradient & Tangent Plane 3D brings the core geometry of multivariable differential calculus to life. As students drag the glowing probe sphere across the 3D terrain z = f(x, y), the simulation updates the translucent tangent plane and projects the gradient vector ∇f = ⟨∂f/∂x, ∂f/∂y⟩ onto the topographic contour map below. Red and green tangent arrows show the partial slopes along the x and y axes, while a rotatable cyan arrow shows the directional derivative D_u f = ∇f · û at any chosen angle θ. Clicking the probe releases a gradient-descent ball that rolls downhill along −∇f, tracing how optimization algorithms find local minima.

學生可以探索什麼

  • Drag the red probe sphere around the mountain peak and verify that the tangent plane becomes completely horizontal at the summit where ∇f = ⟨0, 0⟩.
  • Rotate the Direction angle θ slider from 0° to 360° and find the angle where the Directional derivative D_u f equals the full Gradient magnitude ‖∇f‖.
  • Switch to the Top Contour Map camera view and observe that the gold gradient arrow ∇f always points perpendicular to the level contour rings.
  • Click Drop descent ball or click the probe sphere in the scene to release the green ball and watch it roll downhill along −∇f.
  • Increase Surface steepness and observe how the tangent plane tilt and gradient arrow length scale proportionally.

學習目標

  1. 1Calculate and visualize partial derivatives ∂f/∂x and ∂f/∂y as slopes of traces parallel to the coordinate axes.
  2. 2Explain why the gradient vector ∇f points in the direction of steepest ascent and is orthogonal to contour lines.
  3. 3Relate the directional derivative D_u f = ∇f · û to the angle between the direction vector û and the gradient ∇f.
  4. 4Describe how the tangent plane provides a local linear approximation to a differentiable surface z = f(x, y).

教師常問的問題

What does the gradient vector ∇f represent geometrically?

The gradient ∇f = ⟨∂f/∂x, ∂f/∂y⟩ is a 2D vector in the domain plane whose direction points toward the steepest uphill climb on the surface z = f(x, y), and whose length ‖∇f‖ equals that maximum slope.

Why is the gradient always perpendicular to contour lines?

Along any contour curve, the surface height z = f(x, y) is constant, so the directional derivative tangent to the contour is zero. Because D_u f = ∇f · û = 0 along the contour tangent, the gradient vector ∇f must be perpendicular to the contour.

How is the directional derivative related to the partial derivatives?

For a unit direction vector û = ⟨cos θ, sin θ⟩, the directional derivative is the dot product D_u f = (∂f/∂x) cos θ + (∂f/∂y) sin θ, blending both partial slopes.

What is gradient descent and how is it shown here?

Gradient descent is an optimization algorithm that takes steps in the direction of −∇f (steepest descent) to minimize a function. In this simulation, the green ball follows −∇f downhill until it reaches a local valley.

嵌入這個模擬實驗

免費嵌入你的學習平台、網誌或班級網頁。

<iframe src="https://kiwibee.io/zh/simulations/gradient-directional-derivative-3d-3d-simulation/embed" title="Gradient &amp; Tangent Plane 3D | KiwiBee" loading="lazy" allowfullscreen style="width:100%;height:min(70vh,620px);min-height:420px;border:0;border-radius:16px"></iframe>

把互動科學帶進你的課堂

KiwiBee 的免費教師工作空間,把備課、工作紙、課堂工具和成績冊集中在同一處。

免費開始,建立你的第一個班別

或了解教師工作空間 →