This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: MECH201 Engineering Analysis (2009) p. 50 8 PARTIAL DIFFERENTIATION A REVISON Consider an analytic function which has two depend variables: x and y . That is z = f(x,y). Figure 8.1 A 3D profile. If z is the vertical axis, the xaxis is in the EastWest direction and yaxis in the NorthSouth direction, then the partial differentiation of z with respect to x while holding y constant: y z x , or simply z x is the slope of the surface represented by z = f(x,y) in the EW direction (along the xaxis). Similarly z y represents the slope in the NSouth direction. A B C y x z = f(x,y) ( f/ x)dx ( f/ y)dy MECH201 Engineering Analysis (2009) p. 51 The increment of height (that is increment of z) for a small increase x in the EW direction (from point A to C as shown in the Figure) is given by ( ) EW z z x x = . Further, the increase in height from C to E, that is moving North (holding x constant) is given by: ( ) NS z z y y = . Thus the total increase in height z after moving from A to C and then to E is: z z z x y x y = + For infinitesimal increment, we write d for and the above may be replaced by: z z dz dx dy x y = + If one travels from A to E along the direction which has a bearing of (measured anticlockwise from the East), the increment of height is given by the same formula, namely: except that we must have ( ) tan y x = . Let the projection of the path AE on the horizontal plane (the xyplane) is s . the average slope of the path from A to E is given by: tan AE z z x z y s x s y s = = + Taking the limit of s b 0 we arrived at a formula for the total differentiation of z: dz z dx z dy ds x ds y ds = + note that: cos and sin dx dy ds ds = = . In general for z = f(x 1 , x 2 , x 3 , x 4 , x 5 , x i ,) z z z x y x y = + MECH201 Engineering Analysis (2009) p. 52 we have i i i z dz dx x = and the slope is given by: i i i dx dz z ds x ds = MECH201 Engineering Analysis (2009) p. 53 9 THE ELLIPTIC PARTIAL DIFFERENTIAL EQUATION 9.1 Steady State Heat Conduction Section 29.1 Here we will develop a PDE for the computation of temperature distribution on a 2D. The distribution of temperature is represented by the isotherms lines of constant temperature. A 2dimensional plate as shown below has a prescribed temperature on its sides: C on side ABCD and 100 C on side EF. The temperature varies linearly from C to 100 C on sides AF and DE. Figure 9.1 A 2D conduction domain As this is a steady state situation, the net heat flow into an element of the plate (size x by y ) is zero at all time....
View
Full Document
 Three '09
 WeihuaLi
 Boundary value problem, TI, Boundary conditions, Dirichlet boundary condition

Click to edit the document details