= ( ) ( ), ( ) f x f x f x = '( ) min '( ), '( ) ,max '( ), '( ) f x f x f x f x f x What is is th the defini initio ion n of the solu lutio ion n of dif iffer ferentia ential l equation? ation?
( ) f x ( ) ( ) ( ) v , x f x f x ( ) f x ( ) ( ) ( ) ( ) ( ) v' min ' , ' ,max ' , ' x f x f x f x f x
Dubois D., Prade H., 1987, On Several Definition of the Differentiation of Fuzzy Mapping, Fuzzy Sets and Systems, Vol.24, pp.117-120 How about integral equations?
Modal interval arithmetic Affine arithmetic Constrain interval arithmetic Ellipsoidal arithmetic Convex models (equations with the ellipsoidal parameters) General set valued arithmetic Fuzzy relational equations …. Etc.
y y ( ) ( ) ( ) ( ) + − + + − ,..., ,..., ... y p p y p p p p p p 1 10 0 1 10 0 m m m m p p 1 m ( ) = ,..., y y p p 0 10 0 m y y + + ... y p p 1 m p p 1 m − + , , y y y y y y y 0 0
Gradient descent Interior point method Sequential quadratic programming Genetic algorithms …
Endpoint combination method Interval Gauss elimination Interval Gauss-Seidel method Linear programming method Rohn method Jiri Rohn, "A Handbook of Results on Interval Linear Problems“,2006 http://www.cs.cas.cz/~rohn/publist/handboo k.zip
( ) ( ) ( ) ( ) ( ) − + ( ) x A,b mid A x rad b rad A x rad b , A A b b W. Oettli, W. Prager. Compatibility of approximate solution of linear equations with given error bounds for coefficients and right- hand sides. Numer. Math. 6: 405-409, 1964.
min max x x i i ( ) ( ) ( ) ( ) − − ( ) ( ) mid A D rad A x b mid A D rad A x b s s ( ) ( ) ( ) ( ) + + ( ) ( ) mid A D rad A x b mid A D rad A x b s s 0 0 x x H.U. Koyluoglu, A. Çakmak, S.R.K. Nielsen. Interval mapping in structural mechanics. In: Spanos, ed. Computational Stochastic Mechanics. 125-133. Balkema, Rotterdam 1995.
( ) ( ) = − A mid A D rad A D rc r c ( ) ( ) = + b mid b D rad b r r ( ) = − T 1, 1,1,...,1 r J ( ) ( ) = A,b A ,b conv conv rc rc ( ) = = : , , conv A ,b conv x A x b r c J rc rc rc r = 2 + n n n 2 2 2 2 Rohn’s method Combinatoric solution 2 n n
For every r J ( ) ( ) − = 1 Select recommended c sign mid A b c J r Solve = A x b rc r ( ) If then register x and go to = sign x c next r ( ) Otherwise find = min : k j sign x c j j Set and go to step 1 . = − c c k k
VERSO SOFT: FT: Veri rifi fica cati tion on software are in MATLA LAB B / I INTLAB LAB http://uivtx.cs.cas.cz/~rohn/matlab/index.html -Real data only: Linear systems (rectangular) -Verified description of all solutions of a system of linear equations -Verified description of all linear squares solutions of a system of linear equations -Verified nonnegative solution of a system of linear inequalities -VERLINPROG for verified nonnegative solution of a system of linear equations -Real data only: Matrix equations (rectangular) -See VERMATREQN for verified solution of the matrix equation A*X*B+C*X*D=F (in particular, of the Sylvester or Lyapunov equation) Etc.
Find the solution of Ax = b ◦ Transform into fixed point equation g ( x ) = x g ( x ) = x – R ( Ax – b ) = Rb+ ( I – RA ) x ( R nonsingular) ◦ Brouwer’s fixed point theorem If Rb + ( I – R A) X int (X) then x X, Ax = b
Solve AX AX=b ◦ Brouwer’s fixed point theorem w/ Krawczyk’s operator If R b + ( I – R A) X int (X) then (A, b) X ◦ Iteration X n +1 = R b + ( I – R A) εX n (for n = 0, 1, 2,…) Stopping criteria: X n +1 int( X n ) Enclosure: (A, b) X n +1
+ − 0 k k k u = 1 2 2 1 − k k u p 2 2 2
k 1 = [0.9, 1.1], k 2 = [1.8, 2.2] , p = 1.0 1 1 = k = = [ 0 . 91 , 1 . 11 ] u 1 [ 0 . 9 , 1 . 1 ] 1 + + [ 0 . 9 , 1 . 1 ] [ 1 . 8 , 2 . 2 ] k k = = = 1 2 [ 1 . 12 , 2 . 04 ] u 2 [ 0 . 9 , 1 . 1 ] [ 1 . 8 , 2 . 2 ] k k 1 2 ( overestima tion ) 1 1 1 1 = + = + = ' [ 1 . 36 1 . 67 ] u , 2 [ 0 . 9 , 1 . 1 ] [ 1 . 8 , 2 . 2 ] k k 1 2 ( exact solution)
+ k k = 1 2 u 2 k k 1 2 Two k 1 : the same physical quantity Interval arithmetic: treat two k 1 as two independent interval quantities having same bounds
Replace floating point arithmetic by interval arithmetic Over-pessimistic result due to dependency − − [ 2 . 7 , 3 . 3 ] [ 2 . 2 , 1 . 8 ] 0 u = 1 − − [ 2 . 2 , 1 . 8 ] [ 1 . 8 , 2 . 2 ] 1 u 2 Naïve solution Exact solution − [ 110 , 112 ] [ 0 . 91 , 1 . 11 ] u u = = 1 1 − [ 134 . 5 , 137 . 5 ] u [ 1 . 36 , 1 . 67 ] u 2 2
How to reduce overestimation? ◦ Manipulate the expression to reduce multiple occurrence ◦ Trace the sources of dependency + − 0 k k k u = 1 2 2 1 − k k u p 2 2 2
Element-by-Element ◦ K : diagonal matrix, singular p L 2, E 2, A 2 L 1, E 1, A 1 p
Element-by-element method = ◦ Element stiffness: + ( ) K d K I i i i E A E A E A E A − − 1 1 1 1 1 1 1 1 α 1 0 0 L L L L = = + 1 1 1 1 1 K 1 α E A E A 0 1 0 E A E A − − 1 1 1 1 1 1 1 1 1 L L L L 1 1 1 1 ◦ System stiffness: = + ( ) K d K I
Lagrange Multiplier method ◦ With the constraints: CU – t = 0 ◦ Lagrange multipliers: λ T T u p u p K C K C = = λ 0 0 t 0 C C
System equation: Ax Ax = b b C T u p K = λ 0 0 C rewrite as: + = ( ) A S D x b T 0 d u p 0 k C k + = λ 0 0 0 0 0 0 C
x x = [u, λ ] T , u is the displacement vector Calculate element forces ◦ Conventional FEM: F=k u ( overestimation) Ku = P – C T λ ◦ Present formulation: Ku λ = = L x, p = N b P – C T λ = = p – C T L (x * n+ 1 + x 0 ) P – C T λ = = N b – C T L ( R b – RS M n δ ) P – C T λ = = ( N – C T LR )b + C T LRS M n δ
Click here
http://andrze tp://andrzej.po j.pownuk.com/ wnuk.com/ Click here
Monotone solution + u p p 1 1 = 1 1 2 − 1 1 u p 2 2 p p = + = u 1 p , u 1 1 2 2 2 2 − = 2 4 u p 0 Non-monotone solution = = − 2 2 u p u , p 1 2
u then = = If m i n m ax p p p , p 0 p u then = = If m i n m ax p p p , p 0 p = = m i n m ax u u p ( ) , u u p ( )
analysis_type linear_static_functional_derivative parameter 1 [210E9,212E9] # E parameter 2 [0.2,0.4] # Poisson number parameter 3 0.1 # thickness parameter 4 [-3,-1] sensitivity # fy point 1 x 0 y 0 point 2 x 1 y 0 point 3 x 1 y 1 point 4 x 0 y 1 point 5 x -1 y 0 point 6 x -1 y 1 rectangle 1 points 1 2 3 4 parameters 1 2 3 rectangle 2 points 5 1 4 6 parameters 1 2 3 load constant_distributed_in_y_local_direction geometrical_object 1 fy 4 load constant_distributed_in_y_local_direction geometrical_object 2 fy 4 boundary_condition fixed point 1 ux boundary_condition fixed point 1 uy boundary_condition fixed point 2 ux boundary_condition fixed point 2 uy boundary_condition fixed point 5 ux boundary_condition fixed point 5 uy #Mesh
( ) = ( ) = min u u min u u or ( ) = 0 C , ,
A B A B = = Ω , :
Recommend
More recommend