You are on page 1of 131

# Page 1 of 2

uni kontakt
Universität Stuttgart

Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics
Boundary Element Methods

Lecture notes

Next: Introduction

Boundary Element Methods
Lothar Gaul and Matthias Fischer

Introduction
BE Formulation of Laplace's Equation
Weak formulation of the differential equation
Transformation on the boundary
Fundamental solution as weighting function
Boundary integral equation of the 2-D problem
Preparative example for the limit process
Calculation of the limit
Discretisation of the boundary
The collocation method
Example: Laplace problem of heat transfer
Numerical solution with the collocation method
Analytical solution

2007/05/23

Page 2 of 2

Computation of solution in the domain
Calculation of Dirichlet variable in the domain
Calculation of flux in the domain

BE formulation of Poisson's equation
Calculation of domain integrals by integration of cells
Calculation of domain integrals by transformation into a boundary integral
Calculation of the unknown boundary variables

Orthotropic constitutive behaviour in the domain
Indirect calculation of diagonal elements in
Concentrated source terms
Substructure technique
Example: Orthotropic heat transfer and subregion coupling
Fundamental solutions
Laplace equation
Fundamental solution of the 2D Laplace equation
Fundamental solution of the 3D Laplace equation
Helmholtz equations
Fundamental solution of the 3D Helmholtz equation

Next: Introduction

http://www.bem.uni-stuttgart.de/bem_pages/bem_script.html 2007/05/23

Introduction Page 1 of 2

uni kontakt
Universität Stuttgart

Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics
Boundary Element Methods

Introduction

Next: BE Formulation of Laplace's Equation
Up: Boundary Element Methods
Previous: Boundary Element Methods

Introduction
This manuscript accompanies the lecture Boundary Element Methods in Statics and Dynamics. However, the
material presented on the web cannot include all the aspects that are discussed in the class.

Focus point of the manuscript is the derivation of the standard boundary element method for Laplace's equation.
Starting from the differential equation, the BEM is formulated step by step. Simple examples are calculated and
compared to analytical solutions. The handling of domain integrals in the BEM is discussed on the example of
Poisson's equation. Some advanced techniques and the derivation of selected fundamental solutions conclude
the manuscript.

The lecture covers additional important aspects of boundary elements. For example the application of the
method to elastostatics and elastodynamics as well as to acoustics. Furthermore advanced formulations such as
the Dual Reciprocity BEM and variational BEM are presented.

2007/05/23

Introduction Page 2 of 2 Next: BE Formulation of Laplace's Equation Up: Boundary Element Methods Previous: Boundary Element Methods Last modified 02/11/2004 ( wm) | © Universität Stuttgart | Imprint http://www.html 2007/05/23 .bem.de/bem_pages/node1.uni-stuttgart.

BE Formulation of Laplace's Equation Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods BE Formulation of Laplace's Equation Next: Weak formulation of the differential equation Up: Boundary Element Methods Previous: Introduction Boundary Element Formulation of Laplace's Equation Subsections Weak formulation of the differential equation Transformation on the boundary Fundamental solution as weighting function Boundary integral equation of the 2-D problem Preparative example for the limit process Calculation of the limit Discretisation of the boundary 2007/05/23 .

BE Formulation of Laplace's Equation Page 2 of 2 The collocation method Example: Laplace problem of heat transfer Numerical solution with the collocation method Analytical solution Computation of solution in the domain Calculation of Dirichlet variable in the domain Calculation of flux in the domain Next: Weak formulation of the differential equation Up: Boundary Element Methods Previous: Introduction Last modified 02/13/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.de/bem_pages/node2.bem.html 2007/05/23 .

Weak formulation of the differential equation Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Weak formulation of the differential equation Next: Transformation on the boundary Up: BE Formulation of Laplace's Equation Previous: BE Formulation of Laplace's Equation Weak formulation of the differential equation Starting point for the boundary element formulation is the weighted residual (or weak) statement of the differential equation.de/bem_pages/node3. it is given by (1) with a test function. For Laplace's equation .uni-stuttgart.html 2007/05/23 .bem. Next: Transformation on the boundary Up: BE Formulation of Laplace's Equation Previous: BE Formulation of Laplace's Equation http://www.

html 2007/05/23 .bem.Weak formulation of the differential equation Page 2 of 2 Last modified 02/13/2004 ( wm) | © Universität Stuttgart | Imprint http://www.de/bem_pages/node3.uni-stuttgart.

In index notation this reads (2) 2007/05/23 . The transformation of the differential operator to the boundary is done by applying Green's theorem twice to the weighted residual statement.Transformation on the boundary Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Transformation on the boundary Next: Fundamental solution as weighting function Up: BE Formulation of Laplace's Equation Previous: Weak formulation of the differential equation Transformation on the boundary This step corresponds in 1-D to the partial integration of the differential operator. This is different from the 1-D case where integrals reduce to scalar quantities. It requires the application of special integral theorems depending on the problem dimension. These theorems reduce domain integrals in boundary integrals. In the following paragraphs the transformation on the boundary is treated for 2-D and 3-D by adopting Green's integral theorem in the plane and in space.

Transformation on the boundary Page 2 of 2 Next: Fundamental solution as weighting function Up: BE Formulation of Laplace's Equation Previous: Weak formulation of the differential equation Last modified 02/13/2004 ( wm) | © Universität Stuttgart | Imprint http://www.bem.uni-stuttgart.html 2007/05/23 .de/bem_pages/node4.

f. 2007/05/23 .Fundamental solution as weighting function Page 1 of 3 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Fundamental solution as weighting function Next: Boundary integral equation of the 2-D problem Up: BE Formulation of Laplace's Equation Previous: Transformation on the boundary Fundamental solution as weighting function The fundamental solution is the Green's function for the unbounded space and solves the differential equation (3) The minus sign of the Dirac distribution is introduced for convenience such that the obtained system matrices become positive. The 3-D case leads to (s. Appendix 9) is given by: (4) (5) with the common abbreviation of the Euclidean distance .f. In 2-D the fundamental solution (s.

http://www. that the definition (9) of deviates from the physical definition of the heat flux vector (10) In an actual heat transfer problem. physical constants such as the heat conductivity need to be taken into account.bem. Eq. After selecting . It has to be noticed.html 2007/05/23 .Fundamental solution as weighting function Page 2 of 3 Appendix 9) (6) (7) The functions and are denoted as single layer and double layer potentials.de/bem_pages/node5. (2) and associated with the sifting property of the Dirac distribution lead to (8) where (9) The common notations for the field point or receiver point (marked by the vector ) and for the load point or source point (marked by the vector ) have been used.uni-stuttgart. respectively.

or .Fundamental solution as weighting function Page 3 of 3 Other than in the 1-D case. By recalling all steps necessary to derive Eq. (8). (8) is an exact solution of Laplace's equation as well.over the boundary consisting of all field points on . It follows that statement (1) is only fulfilled if the differential equation is satisfied identically. the exact solution of (8) is an exact solution of the corresponding differential equation as well. whether an exact solution of Eq. The question arises. This seems not to be the case.bem. one recognizes.uni-stuttgart. Next: Boundary integral equation of the 2-D problem Up: BE Formulation of Laplace's Equation Previous: Transformation on the boundary Last modified 02/13/2004 ( wm) | © Universität Stuttgart | Imprint http://www. that the weighted residual statement Eq. (1) does not lead to an approximation. higher dimensional problems lead to integrals over the boundary of the domain .to be more precise . (8) which fulfills the integral pointwise represents a weighting with infinitely many linearly independent test functions in the residual statement (1).de/bem_pages/node5. An exact solution of Eq. since the weighted residual statement allows for local errors in the domain but averages them to zero by domain integration.html 2007/05/23 . Thus.

Boundary integral equation of the 2-D problem Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Boundary integral equation of the 2-D problem Next: Preparative example for the limit process Up: BE Formulation of Laplace's Equation Previous: Fundamental solution as weighting function Boundary integral equation of the 2-D problem If the load point in Eq. (8). (8) moves on the boundary. The singularities of fundamental solutions requires careful analysis when the load point is accommodated on the boundary. Calculation of the integrals by boundary discretisation leads to algebraic equations for solving unknown boundary data in terms of known boundary data. This equation is called a boundary integral equation. Subsections Preparative example for the limit process Calculation of the limit Discretisation of the boundary 2007/05/23 . only boundary data are present in Eq.

html 2007/05/23 .uni-stuttgart.Boundary integral equation of the 2-D problem Page 2 of 2 Next: Preparative example for the limit process Up: BE Formulation of Laplace's Equation Previous: Fundamental solution as weighting function Last modified 02/13/2004 ( wm) | © Universität Stuttgart | Imprint http://www.de/bem_pages/node6.bem.

(11) from both sides by the small quantities and . the correct result is obtained by chance but the integration 'with closed eyes' incorporates the singularity which is improper. one obtains 2007/05/23 .Preparative example for the limit process Page 1 of 3 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Preparative example for the limit process Next: Calculation of the limit Up: Boundary integral equation of the 2-D problem Previous: Boundary integral equation of the 2-D problem Preparative example for the limit process If the integral (11) is solved as shown. The singularity at becomes obvious when the integral is considered. If one approaches the singularity in Eq.

It is denoted by p.v.uni-stuttgart.bem. They are listed in Table 1. At the function is singular but the integral is continuous at according to the calculation with l'Hospital's rule (15) In the direct BEM. these two types of singularity dominate. (14) Besides strongly singular integrands.html 2007/05/23 . The function is an example.Preparative example for the limit process Page 2 of 3 (12) The result depends on and if they approach zero with different values. the integral exists and is smooth at the singularity.de/bem_pages/node7. http://www. The value of the integral calculated above is called Cauchy Principal Value. there exist weakly singular integrands. The choice leads to (13) The integrand is called strongly singular. Even though they are singular.

f. strong singularity Principal Value In stress calculations and special BEM formulations such as the hybrid BEM.uni-stuttgart.html 2007/05/23 .bem. weak singularity singularity Interpretation as Cauchy 2.Preparative example for the limit process Page 3 of 3 Table 1: Classification of singularities Type Property Example Integral is finite at 1. Next: Calculation of the limit Up: Boundary integral equation of the 2-D problem Previous: Boundary integral equation of the 2-D problem Last modified 02/13/2004 ( wm) | © Universität Stuttgart | Imprint http://www.de/bem_pages/node7. so called hyper singularities are encountered (s. Part III).

Calculation of the limit Page 1 of 5 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Calculation of the limit Next: Discretisation of the boundary Up: Boundary integral equation of the 2-D problem Previous: Preparative example for the limit process Calculation of the limit To locate the load point on the boundary. 2007/05/23 . (8) is still valid. 1 (16) Thus the point is inside the domain and Eq. we first adjust the boundary such that it contains the point inside a circle of radius according to Fig.

Furthermore holds (18) http://www. 2).f. Fig.Calculation of the limit Page 2 of 5 Figure 1: Boundary extension by a circle The integration along the -circle is parameterized by (17) (s.de/bem_pages/node8.uni-stuttgart.html 2007/05/23 .bem.

html 2007/05/23 . (8) on the boundary can now be calculated. (4) (19) In the limit. the first integral is weakly singular. For we get by Eq. With Eqs (17. the last integral in Eq.Calculation of the limit Page 3 of 5 Figure 2: Geometry for accommodating the load point on the boundary The limit value by taking Eq. 18) and l'Hospital's rule.de/bem_pages/node8.uni-stuttgart. (19) results in a vanishing contribution http://www.bem.

The second integral leads to (22) Summarizing these results and inserting in Eq.Calculation of the limit Page 4 of 5 (20) With Eq.uni-stuttgart. (5). (8) leads to the integral equation http://www. the integral leads to (21) The first integral in Eq. (21) is a strongly singular integral calculated by Cauchy's Principal Value.de/bem_pages/node8.html 2007/05/23 .bem.

uni-stuttgart.html 2007/05/23 .de/bem_pages/node8. (24) analytically is only possible for special cases. This approach is called discretisation. Next: Discretisation of the boundary Up: Boundary integral equation of the 2-D problem Previous: Preparative example for the limit process Last modified 02/13/2004 ( wm) | © Universität Stuttgart | Imprint http://www.Calculation of the limit Page 5 of 5 (23) respectively (24) The factor is called boundary factor and denotes the fraction of which is inside (25) Solving the integrals in Eq.bem. For a numerical integration the boundary is divided in segments with the interpolation of boundary data by piecewise continuous functions such as polynomials.

Discretisation of the boundary Page 1 of 8 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Discretisation of the boundary Next: The collocation method Up: Boundary integral equation of the 2-D problem Previous: Calculation of the limit Discretisation of the boundary 2007/05/23 .

html 2007/05/23 . At node of element the value of is and the value of is .uni-stuttgart. the shape of and are interpolated by (26) http://www. 3).Discretisation of the boundary Page 2 of 8 Figure 3: Discretisation of the boundary For an approximation of the geometry. the boundary of the domain is divided in boundary elements (Fig.de/bem_pages/node9. Every element has one or more nodes.bem. Shape functions describe the spatial distribution on the element. With nodes in element .

Discretisation of the boundary Page 3 of 8 and respectively. the values of and are constant throughout the element and have the value at the node.bem.uni-stuttgart. Or in matrix notation by and (27) where and are 1 x M row vectors and is a M x 1 column vector.de/bem_pages/node9. The simplest shape functions are constant and linear shape functions. Constant shape function Only one node exists per element.html 2007/05/23 . This means and and (28) http://www.

de/bem_pages/node9.html 2007/05/23 .Discretisation of the boundary Page 4 of 8 Figure 4: Constant element shape function and local coordinate Linear shape function http://www.bem.uni-stuttgart.

Discretisation of the boundary Page 5 of 8 Figure 5: Linear interpolation between two nodes The values and at the nodes and of the element with length in Fig.de/bem_pages/node9.bem. 5 lead to the linear interpolation (29) given in matrix notation http://www.html 2007/05/23 .uni-stuttgart.

The shape functions are depicted in Fig.uni-stuttgart.de/bem_pages/node9. (24) in 2-D leads to http://www.bem.Discretisation of the boundary Page 6 of 8 (30) with the local coordinate .html 2007/05/23 . 6. Figure 6: Linear shape functions Discretisation of Eq.

Fig.Discretisation of the boundary Page 7 of 8 (31) The nodal values and are constants and can be brought outside the integrals (32) If constant elements are used. and Eq.bem.html 2007/05/23 .uni-stuttgart. (25) lead to (33) The vector is perpendicular to if load point and field point are located on the same element. and therefore http://www. the node is usually located in the middle of the perfectly flat element (s. 4).f. Therefore.de/bem_pages/node9.

Discretisation of the boundary Page 8 of 8 (34) This simplifies the calculation and makes the numerical implementation easier. Next: The collocation method Up: Boundary integral equation of the 2-D problem Previous: Calculation of the limit Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.bem.de/bem_pages/node9.html 2007/05/23 .

Because linear and higher order polynomial shape functions lead to nodes which belong to more than one element. The simplest approach is to establish a system of equations with as many unknowns as equations. it is worthwhile to introduce a global node numbering ( ) which does not depend on the element.The collocation method Page 1 of 3 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods The collocation method Next: Example: Laplace problem of heat transfer Up: BE Formulation of Laplace's Equation Previous: Discretisation of the boundary The collocation method The collocation method allows to calculate the unknown boundary data from Eq. The principle of collocation means to locate the load point sequentially at all nodes of the discretisation such that the domain variable at the load point coincides with the nodal value. (32). If the load point is located on the first global node the first equation of the system reads 2007/05/23 .

html 2007/05/23 .bem.de/bem_pages/node10. The Eq. where is the corresponding shape function.The collocation method Page 2 of 3 (35) The notation means the sum of integrals contributed from those elements which contain the global node.uni-stuttgart. (35) is given in matrix notation by (36) where ( ) denotes the element which contains the boundary term . By collocating the load point with the nodes to the additional equations of the system (37) are obtained http://www.

The collocation method Page 3 of 3 (37) and read in matrix notation (38) The diagonal elements of the matrices and contain singular integrals because the distance vanishes at the nodes.html 2007/05/23 .uni-stuttgart. Since both vectors and in Eq. All other matrix elements contain regular integrals. (38) contain known as well as unknown boundary data.bem. Next: Example: Laplace problem of heat transfer Up: BE Formulation of Laplace's Equation Previous: Discretisation of the boundary Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www. it is necessary to rewrite the equations with all unknowns appearing in a vector on one side (39) A systematic way of doing this and solving the system of equations is demonstrated with a simple example which can be calculated by hand.de/bem_pages/node10.

the temperatures are prescribed. 7. At the horizontal boundary lines. Laplace's equation of heat transfer is considered in a 2-D rectangular domain with aspect ratio 1:2 as depicted in Fig.Example: Laplace problem of heat transfer Page 1 of 3 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Example: Laplace problem of heat transfer Next: Numerical solution with the collocation method Up: BE Formulation of Laplace's Equation Previous: The collocation method Example: Laplace problem of heat transfer Now. 2007/05/23 . At the vertical boundary lines the heat flux is given.

html 2007/05/23 .de/bem_pages/node11.uni-stuttgart. are unknown for a discretisation with four elements. and . Subsections http://www.Example: Laplace problem of heat transfer Page 2 of 3 Figure 7: Example: Heat transfer in rectangular domain The remaining boundary values .bem. The numerical results are afterwards compared to the analytical solution.

html 2007/05/23 .Example: Laplace problem of heat transfer Page 3 of 3 Numerical solution with the collocation method Analytical solution Next: Numerical solution with the collocation method Up: BE Formulation of Laplace's Equation Previous: The collocation method Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.bem.de/bem_pages/node11.

e. If Eq. . . constant elements are chosen in the example (i. (32) is written for the load point . . Each element has only one node located in the middle.Numerical solution with the collocation method Page 1 of 18 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Numerical solution with the collocation method Next: Analytical solution Up: Example: Laplace problem of heat transfer Previous: Example: Laplace problem of heat transfer Numerical solution with the collocation method For simplicity. one obtains (40) 2007/05/23 .

In matrix notation these equations are (41) Calculation of matrix elements and : .Numerical solution with the collocation method Page 2 of 18 Four equations for the four unknown boundary values are obtained if takes the values 1 to 4 and is located at the four nodes sequentially.de/bem_pages/node12. (40) for different values of and . The elements of the matrices are the integrals in Eq.uni-stuttgart. http://www.bem.html 2007/05/23 .

bem. 8 shows: http://www.uni-stuttgart.de/bem_pages/node12.Numerical solution with the collocation method Page 3 of 18 Figure 8: Calculation of matrix elements and Fig.html 2007/05/23 .

Numerical solution with the collocation method Page 4 of 18 Inserting leads to (42) and http://www.bem.html 2007/05/23 .de/bem_pages/node12.uni-stuttgart.

uni-stuttgart.de/bem_pages/node12.bem. and : .Numerical solution with the collocation method Page 5 of 18 (43) Spatial isotropy of the problem at hand leads to and (44) These symmetries do not hold in general for BEM.html 2007/05/23 . http://www.

Numerical solution with the collocation method Page 6 of 18 Figure 9: Calculation of matrix elements and Fig.html 2007/05/23 .de/bem_pages/node12. 9 shows: http://www.uni-stuttgart.bem.

bem.Numerical solution with the collocation method Page 7 of 18 Inserting leads to (45) and http://www.uni-stuttgart.html 2007/05/23 .de/bem_pages/node12.

uni-stuttgart.html 2007/05/23 .de/bem_pages/node12.Numerical solution with the collocation method Page 8 of 18 (46) Spatial isotropy leads to and (47) and : and The geometry for calculating and is obtained the same way http://www.bem.

Numerical solution with the collocation method Page 9 of 18 Inserting leads to (48) and http://www.html 2007/05/23 .uni-stuttgart.bem.de/bem_pages/node12.

uni-stuttgart.html 2007/05/23 .de/bem_pages/node12.Numerical solution with the collocation method Page 10 of 18 (49) and by virtue of symmetry and (50) and : and The geometry for calculating and is obtained the same way http://www.bem.

de/bem_pages/node12.Numerical solution with the collocation method Page 11 of 18 Inserting leads to (51) and http://www.html 2007/05/23 .bem.uni-stuttgart.

Numerical solution with the collocation method Page 12 of 18 (52) and by virtue of symmetry and (53) Diagonal terms: According to Eq. (34) the main diagonal of matrix vanishes (54) http://www.bem.html 2007/05/23 .de/bem_pages/node12.uni-stuttgart.

but the integral exists. Eq.f.bem.uni-stuttgart.Numerical solution with the collocation method Page 13 of 18 For one obtains And by inserting (55) The integrand in Eq. (15)) http://www. (55) is weakly singular.de/bem_pages/node12. With (s.html 2007/05/23 .

html 2007/05/23 .de/bem_pages/node12.Numerical solution with the collocation method Page 14 of 18 (56) is obtained and because of symmetry (57) holds.bem. The same way follows (58) and http://www.uni-stuttgart.

bem.uni-stuttgart.Numerical solution with the collocation method Page 15 of 18 (59) Assembling and solving the system of equations Rewriting of Eq.de/bem_pages/node12. (41) in index notation and summation of the terms leads to (60) Plugging the matrix elements and the known boundary data leads to (61) http://www.html 2007/05/23 .

html 2007/05/23 . the flux has to be multiplied by in order to obtain the heat flux (s.Numerical solution with the collocation method Page 16 of 18 and after rewriting the equations with all unknown boundary data appearing on the left side (62) Solving the equations leads to (63) In terms of physics.f.uni-stuttgart.de/bem_pages/node12. Eq.bem. (8) and http://www.

After having used the crudest form of discretisation. This discretisation still allows the calculation by hand.Numerical solution with the collocation method Page 17 of 18 Eq.bem. The system of equations is obtained as (64) Rearranging of known and unknown nodal data in the equations and solving the system of equations leads to the solution (65) http://www.html 2007/05/23 .de/bem_pages/node12. This means that the heat flux through element 1 has the direction of the outward normal vector and the heat flux through element 3 has the direction opposite to the outward normal vector. a finer boundary mesh with six constant elements of length 1 is used.uni-stuttgart. The matrix entries are calculated as shown for the four element mesh. After multiplication the nodal value is positive and is negative. (10)).

Numerical solution with the collocation method Page 18 of 18 where the nodes 1 and 4 of the six element discretisation coincide with the nodes 1 and 3 of the four element discretisation.uni-stuttgart. Next: Analytical solution Up: Example: Laplace problem of heat transfer Previous: Example: Laplace problem of heat transfer Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.html 2007/05/23 .bem.de/bem_pages/node12.

On the boundaries and the gradient of vanishes in direction. . This leads to the conclusion that the seeked solution is independent of and a trial function with unknown coefficients and depends linearly on 2007/05/23 .Analytical solution Page 1 of 3 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Analytical solution Next: Computation of solution in the domain Up: Example: Laplace problem of heat transfer Previous: Numerical solution with the collocation method Analytical solution The 2-D Laplacian is the field equation of the heat flux problem (66) The flux in direction vanishes on the boundaries and .

Analytical solution Page 2 of 3 (67) Fitting the boundary conditions (68) leads to the constants and (69) The analytical solution is thus given by and (70) The comparison between analytical and numerical solution shows.de/bem_pages/node13.uni-stuttgart.html 2007/05/23 . that for the Dirichlet variable even a coarse http://www.bem.

Analytical solution Page 3 of 3 discretisation leads to good accuracy. The finer discretisation by six elements already shows a considerable improvement of accuracy. Next: Computation of solution in the domain Up: Example: Laplace problem of heat transfer Previous: Numerical solution with the collocation method Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.html 2007/05/23 .de/bem_pages/node13. The larger error of the Neumann variable for the four element discretisation can be explained by the fact that the differentiated quantity requires finer discretisation because integration smoothes while differentiation creates roughness.bem.

The boundary factor is chosen according to Eq. Subsections Calculation of Dirichlet variable in the domain Calculation of flux in the domain 2007/05/23 .Computation of solution in the domain Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Computation of solution in the domain Next: Calculation of Dirichlet variable in the domain Up: BE Formulation of Laplace's Equation Previous: Analytical solution Computation of solution in the domain The solution of unknown data in the domain can only be obtained after the data on the boundary have been calculated. Integration of Eq. The load point is placed where the domain data shall be calculated. (25). The boundary data is completely known and consist of given boundary conditions and the values that were calculated using the collocation method. (32) along the boundary with a vector connecting each boundary point with the interior load point gives the value of the field variables.

html 2007/05/23 .uni-stuttgart.Computation of solution in the domain Page 2 of 2 Next: Calculation of Dirichlet variable in the domain Up: BE Formulation of Laplace's Equation Previous: Analytical solution Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.de/bem_pages/node14.bem.

(40) for the load point in the domain leads to (71) 2007/05/23 .Calculation of Dirichlet variable in the domain Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Calculation of Dirichlet variable in the domain Next: Calculation of flux in the domain Up: Computation of solution in the domain Previous: Computation of solution in the domain Calculation of Dirichlet variable in the domain The calculation of in the domain is demonstrated on the 2-D example. Rewriting Eq.

bem.html 2007/05/23 .uni-stuttgart.de/bem_pages/node15. For the solution of the domain variable all boundary integrals need to be evaluated. Next: Calculation of flux in the domain Up: Computation of solution in the domain Previous: Computation of solution in the domain Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www. (25) and the nodal data and .Calculation of Dirichlet variable in the domain Page 2 of 2 with the boundary factor according to Eq.

it is necessary to calculate the gradient of at the load point . (71) leads to (73) and 2007/05/23 . This leads to both flux coordinates and (72) With a matrix notation of scalar products. Eq.Calculation of flux in the domain Page 1 of 3 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Calculation of flux in the domain Next: BE formulation of Poisson's equation Up: Computation of solution in the domain Previous: Calculation of Dirichlet variable in the domain Calculation of flux in the domain For the calculation of the flux.

the corresponding integrals need to be solved and lead to the flux at the load point in the domain.uni-stuttgart. http://www.de/bem_pages/node16.html 2007/05/23 .Calculation of flux in the domain Page 2 of 3 (74) Exchanging differentiation and integration. the derivatives with respect to are obtained as (75) and (76) The derivatives of the matrix entries of and with respect to are (77) and (78) After this.bem.

bem.de/bem_pages/node16.Calculation of flux in the domain Page 3 of 3 Next: BE formulation of Poisson's equation Up: Computation of solution in the domain Previous: Calculation of Dirichlet variable in the domain Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.html 2007/05/23 .

BE formulation of Poisson's equation Page 1 of 4 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods BE formulation of Poisson's equation Next: Calculation of domain integrals by integration of cells Up: Boundary Element Methods Previous: Calculation of flux in the domain Boundary element formulation of Poisson's equation Poisson's equation with a non-homogeneous term in (79) describes for example the local heat conduction with sources in the domain or torsion of non-circular cross sections. The weighted residue statement (80) 2007/05/23 .

uni-stuttgart.BE formulation of Poisson's equation Page 2 of 4 and the inverse form with Green's theorem lead to the presence of a domain integral (81) With the domain integral is given in 2-D by (82) A mean to calculate the domain integral is to discretise the domain into integration cells and then using subsequent numerical integrations. Discretisation of the boundary with elements and of the domain with cells leads to (s. The cells look like a finite element mesh. (32)) (83) http://www. Eq. However.bem. the procedure has an essential difference because there are no unknowns in the domain.de/bem_pages/node17.f. The cells are used as integration regions over which analytical or Gaussian quadrature is performed.html 2007/05/23 .

uni-stuttgart. Subsections Calculation of domain integrals by integration of cells Calculation of domain integrals by transformation into a boundary integral Calculation of the unknown boundary variables Next: Calculation of domain integrals by integration of cells Up: Boundary Element Methods Previous: Calculation of flux in the domain http://www.html 2007/05/23 .BE formulation of Poisson's equation Page 3 of 4 The result is a system of equations (84) and after separating given boundary variables from unknown boundary variables (85) By adding the vectors and .bem.de/bem_pages/node17. the system of equations allows to calculate the vector containing the unknown boundary variables.

html 2007/05/23 .de/bem_pages/node17.bem.uni-stuttgart.BE formulation of Poisson's equation Page 4 of 4 Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.

If the complete domain is taken as integration cell and the boundary is dicretised with four constant elements. Eq. (83) leads to (86) with 2007/05/23 .Calculation of domain integrals by integration of cells Page 1 of 4 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Calculation of domain integrals by integration of cells Next: Calculation of domain integrals by transformation into a boundary integral Up: BE formulation of Poisson's equation Previous: BE formulation of Poisson's equation Calculation of domain integrals by integration of cells The example of Laplace's equation in a rectangular domain is now modified such that Poisson's equation (79) holds with const.

html 2007/05/23 .de/bem_pages/node18.bem.Calculation of domain integrals by integration of cells Page 2 of 4 The integration is carried out (87) Symmetry results in (88) The result for is obtained by http://www.uni-stuttgart.

Calculation of domain integrals by integration of cells Page 3 of 4 (89) Symmetry results in (90) Next: Calculation of domain integrals by transformation into a boundary integral Up: BE formulation of Poisson's equation Previous: BE formulation of Poisson's equation http://www.html 2007/05/23 .de/bem_pages/node18.uni-stuttgart.bem.

Calculation of domain integrals by integration of cells Page 4 of 4 Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.bem.uni-stuttgart.de/bem_pages/node18.html 2007/05/23 .

Calculation of domain integrals by transformation into a boundary integral Page 1 of 7 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Calculation of domain integrals by transformation into a boundary integral Next: Calculation of the unknown boundary variables Up: BE formulation of Poisson's equation Previous: Calculation of domain integrals by integration of cells Calculation of domain integrals by transformation into a boundary integral If is a harmonic function. After introducing a function defined by (91) and using Green's theorem (92) along with . if it satisfies . the domain integral may be transformed into a boundary integral. that is. one arrives at 2007/05/23 .

(94) is expressed by (95) If is assumed to have no dependence. Eq. Determination of With the 2-D fundamental solution. Eq.Calculation of domain integrals by transformation into a boundary integral Page 2 of 7 (93) with the fundamental solution of Laplace's equation.de/bem_pages/node19.uni-stuttgart. (91) is given by where (94) In polar coordinates .bem. Eq. (96) remains (96) A first integration http://www.html 2007/05/23 . (91). (93) are known. With its directional derivative all terms in the boundary representation Eq. A so called higher order fundamental solution can be calculated from Eq.

uni-stuttgart.bem.de/bem_pages/node19.html 2007/05/23 .Calculation of domain integrals by transformation into a boundary integral Page 3 of 7 (97) (98) and a second integration lead to (99) The choice of results in (100) Determination of The directional derivative is executed by http://www.

Calculation of domain integrals by transformation into a boundary integral Page 4 of 7 (101) With Eq. for (104) For an element which contains the load point as well as the field point. orthogonality leads to (105) http://www.html 2007/05/23 .uni-stuttgart.de/bem_pages/node19.g.bem. can be obtained from (102) The simple example where const leads to (103) The boundary integrals from to can be split into sub-integrals corresponding to each boundary element. This means e. (93).

Calculation of domain integrals by transformation into a boundary integral Page 5 of 7 The second integral leads to (106) and because of symmetry follows (107) The result for is (108) so that at the end is obtained as (109) http://www.de/bem_pages/node19.uni-stuttgart.html 2007/05/23 .bem.

Calculation of domain integrals by transformation into a boundary integral Page 6 of 7 The result for contains (110) The integral is now given by (111) and because of the problem. symmetry holds (112) The integral leads to (113) http://www.bem.de/bem_pages/node19.uni-stuttgart.html 2007/05/23 .

Calculation of domain integrals by transformation into a boundary integral Page 7 of 7

and finally

(114)

is obtained. One realizes that this result is identical with the one calculated by domain integration. The presented
approach allows to transform the domain integral onto the boundary But it has to be noticed that the approach
only applies for special functions of . For more general distributions other methods are available such as the
Multiple Reciprocity Method [#!nowak!#] which represents an extension of the approach presented in this
chapter, or the Dual Reciprocity Method [#!drm!#] with a slightly different approach.

Next: Calculation of the unknown boundary variables
Up: BE formulation of Poisson's equation
Previous: Calculation of domain integrals by integration of cells

http://www.bem.uni-stuttgart.de/bem_pages/node19.html 2007/05/23

Page 1 of 2

uni kontakt
Universität Stuttgart

Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics
Boundary Element Methods

Calculation of the unknown boundary variables

Next: Orthotropic constitutive behaviour in the domain
Up: BE formulation of Poisson's equation
Previous: Calculation of domain integrals by transformation into a boundary integral

Calculation of the unknown boundary variables
Inserting the results for in Eq. (85) leads to

(115)

As compared to the inhomogeneous set of equations (62), another known vector is added on the right hand side.
Solving Eq. (115) for a fixed value of leads to the unknown boundary variables. The calculation of the domain
variables proceeds analogue as shown for Laplace's equation.

Next: Orthotropic constitutive behaviour in the domain

2007/05/23

Page 2 of 2

Up: BE formulation of Poisson's equation
Previous: Calculation of domain integrals by transformation into a boundary integral

http://www.bem.uni-stuttgart.de/bem_pages/node20.html 2007/05/23

e. is associated with conduction coefficients and . 2007/05/23 . Orthotropic heat transfer.Orthotropic constitutive behaviour in the domain Page 1 of 6 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Orthotropic constitutive behaviour in the domain Next: Indirect calculation of diagonal elements in Up: Boundary Element Methods Previous: Calculation of the unknown boundary variables Orthotropic constitutive behaviour in the domain In an anisotropic domain the constitutive parameters depend on the direction.g. with coordinates and in the direction of orthotropy.

this leads to (117) http://www.html 2007/05/23 . In 2-D.de/bem_pages/node21.uni-stuttgart.Orthotropic constitutive behaviour in the domain Page 2 of 6 Figure 10: Direction of orthotropy The modified Fourier heat conduction equation in an orthotropic domain reads in index notation (116) where the brackets around the index exclude summation.bem.

uni-stuttgart.html 2007/05/23 .de/bem_pages/node21. (116) leads to the orthotropic heat transfer equation (119) Stationary heat transfer along with homogeneous orthotropic constants and lead to (120) http://www.bem.Orthotropic constitutive behaviour in the domain Page 3 of 6 and (118) Eq.

uni-stuttgart.bem.de/bem_pages/node21.Orthotropic constitutive behaviour in the domain Page 4 of 6 The associated fundamental solution is obtained from (121) by invoking a coordinate transformation and (122) such that the left hand side leads to the ordinary Laplacian operator (123) http://www.html 2007/05/23 .

bem.Orthotropic constitutive behaviour in the domain Page 5 of 6 Theorem 2 The -distribution has the property With this property.uni-stuttgart.f. (4)) (125) http://www. Eq.html 2007/05/23 . (123) is given by (124) and leads to the already known 2D fundamental solution (s. Eq.de/bem_pages/node21.

uni-stuttgart. The following calculation is now handled in a manner analogous to the case const. (122).Orthotropic constitutive behaviour in the domain Page 6 of 6 where and (126) can be transformed back to the physical space using Eq. Next: Indirect calculation of diagonal elements in Up: Boundary Element Methods Previous: Calculation of the unknown boundary variables Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.bem.de/bem_pages/node21.html 2007/05/23 .

In this homogeneous case.Indirect calculation of diagonal elements in Page 1 of 3 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Indirect calculation of diagonal elements in Next: Concentrated source terms Up: Boundary Element Methods Previous: Orthotropic constitutive behaviour in the domain Indirect calculation of diagonal elements in matrix from physical considerations Calculation of diagonal elements of matrix requires to determine the fractional boundary coefficients by integration. Different from the coefficient for a boundary point on a constant element. the determination is more complex for more complex elements. a simple way is discussed in which the diagonal elements can be computed regardless of the element complexity. In the following. With these boundary conditions and an arbitrary constant . The most simple solution to be described by the system matrices is a uniformly constant temperature on the boundary. there is no flux in the domain or on the boundary. the vectors and are given by 2007/05/23 .

This leads to the diagonal elements of by the negative sum of the off diagonal elements (129) As the matrix entries of in Eq. Both procedures lead to the same result.bem.de/bem_pages/node22. (61) show. the sum of the entries in a row does as well vanish when the boundary are determined explicitly.html 2007/05/23 . The sum of terms in any row of must vanish. http://www.uni-stuttgart. (60) leads to (128) with the singular matrix .Indirect calculation of diagonal elements in Page 2 of 3 and (127) Eq.

de/bem_pages/node22.uni-stuttgart.Indirect calculation of diagonal elements in Page 3 of 3 Next: Concentrated source terms Up: Boundary Element Methods Previous: Orthotropic constitutive behaviour in the domain Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.html 2007/05/23 .bem.

Concentrated source terms Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Concentrated source terms Next: Substructure technique Up: Boundary Element Methods Previous: Indirect calculation of diagonal elements in Concentrated source terms In the presence of concentrated source terms in the domain. the volume integral (130) can be simplified for 2007/05/23 .

html 2007/05/23 .bem.uni-stuttgart.de/bem_pages/node23. (130): (132) Next: Substructure technique Up: Boundary Element Methods Previous: Indirect calculation of diagonal elements in Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.Concentrated source terms Page 2 of 2 (131) By inserting in Eq.

Domains with piecewise non-homogeneity are now subdivided into homogeneous separate subregions.Substructure technique Page 1 of 7 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Substructure technique Next: Example: Orthotropic heat transfer and subregion coupling Up: Boundary Element Methods Previous: Concentrated source terms Substructure technique So far. only homogeneous domains have been treated in which the constitutive properties do not vary. Afterwards the formulations of the distinct regions are coupled by a substructure technique. 2007/05/23 .

uni-stuttgart.Substructure technique Page 2 of 7 Figure 11: Substructure technique: Division of non-homogeneous domain in piecewise homogeneous subregions Fig.bem. (84).html 2007/05/23 . the formulation for subdomain 1 is http://www. According to Eq. 11 illustrates homogeneous subregions 1 and 2 with different constitutive parameters.de/bem_pages/node24.

bem.de/bem_pages/node24.Substructure technique Page 3 of 7 (133) and for subdomain 2 is (134) Compatibility of at the interface ( compatibility) (135) as well as compatibility of ( compatibility) (136) http://www.uni-stuttgart.html 2007/05/23 .

de/bem_pages/node24. Method 1 After rearranging Eq. (133) for subregion 1 (137) and Eq. (134) for subregion 2 (138) http://www.html 2007/05/23 .Substructure technique Page 4 of 7 lead to a coupled system of equations which can be solved by two methods.bem.uni-stuttgart.

Substructure technique Page 5 of 7

the coupling with and compatibility according to the constraints in Eq. (135) and Eq. (136), respectively,

(139)

No inversion is necessary for setting up the coupled equations and the system matrix is banded which is an
advantage for the numerical treatment. Another advantage is that all unknowns in the interface, and , are
obtained at once.

Method 2

Multiplication of Eq. (84) with the inverse gives

(140)

http://www.bem.uni-stuttgart.de/bem_pages/node24.html 2007/05/23

Substructure technique Page 6 of 7

with matrix and vector .

The application to subregion 1 is

(141)

and correspondingly to subregion 2

(142)

Coupling with the constraints in Eq. (135) and Eq. (136) leads to

(143)

http://www.bem.uni-stuttgart.de/bem_pages/node24.html 2007/05/23

Substructure technique Page 7 of 7

As compared to Eq. (139), the smaller set of equations is an advantage. But this is obtained at the cost of an
inversion of and the necessity to calculate the flux in the interface from Eq. (141) or Eq. (142) after Eq. (143)
has been solved.

Next: Example: Orthotropic heat transfer and subregion coupling
Up: Boundary Element Methods
Previous: Concentrated source terms

http://www.bem.uni-stuttgart.de/bem_pages/node24.html 2007/05/23

Subregion 1 is isotropic and subregion 2 is orthotropic with conduction coefficients and .6 for the isotropic subregion are used. The matrices from the example in Chapter 2. 12 shows the rectangular subdomains with aspect ratio 2:1 which are discretised by 6 constant elements. 2007/05/23 . respectively. Fig. the coupling of an orthotropic and an isotropic subregion is treated.Example: Orthotropic heat transfer and subregion coupling Page 1 of 9 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Example: Orthotropic heat transfer and subregion coupling Next: Fundamental solutions Up: Boundary Element Methods Previous: Substructure technique Coupling of an orthotropic and an isotropic subregion For illustrating the methods outlined in the preceding chapters.

html 2007/05/23 .de/bem_pages/node25.uni-stuttgart.Example: Orthotropic heat transfer and subregion coupling Page 2 of 9 Figure 12: Example: Coupling of isotropic and orthotropic subregions The elements of matrices and of Eq.1 define the system of equations for subregion 1.bem. After rearranging. one obtains http://www.6. (64) in Chapter 2.

Example: Orthotropic heat transfer and subregion coupling Page 3 of 9 (144) The calculation of the system matrices for subregion 2 is demonstrated for an example of the matrix and an element of the matrix .bem.html 2007/05/23 . According to Eq.de/bem_pages/node25.uni-stuttgart. (122). new variables are introduced and (145) Calculation of matrix elements and ( ) http://www.

html 2007/05/23 .Example: Orthotropic heat transfer and subregion coupling Page 4 of 9 Figure 13: Calculation of matrix elements H78 and G78 According to Fig. 13 the following relations hold.bem.uni-stuttgart.de/bem_pages/node25. The variables and are replaced in the corresponding expressions http://www.

uni-stuttgart.Example: Orthotropic heat transfer and subregion coupling Page 5 of 9 (146) Inserting leads to http://www.de/bem_pages/node25.html 2007/05/23 .bem.

The following system of equations is obtained http://www.html 2007/05/23 .uni-stuttgart.bem.de/bem_pages/node25.Example: Orthotropic heat transfer and subregion coupling Page 6 of 9 and (147) The remaining elements can be calculated the same way.

Example: Orthotropic heat transfer and subregion coupling Page 7 of 9 (148) Coupling of Eq.html 2007/05/23 .bem.uni-stuttgart. (142) leads to sets of equations from which the unknown boundary variables and the interface variables can be solved. The solution is given by: (149) http://www. (139) or method 2 in Eq.de/bem_pages/node25. (144) and Eq. (148) according to either method 1 in Eq. (143) along with Eq. (141) or Eq.

uni-stuttgart.Example: Orthotropic heat transfer and subregion coupling Page 8 of 9 Analytical solution The field equations for the analytical solution are (150) where .de/bem_pages/node25. the variables at the nodes of the discretisation are obtained as (151) http://www.bem. According to the boundary and compatibility conditions. in subregion 2.html 2007/05/23 . in subregion 1 and .

bem. 12 should be taken into consideration when the numerical results in Eq. (151). (149) are compared to the analytical results in Eq. Next: Fundamental solutions Up: Boundary Element Methods Previous: Substructure technique Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.Example: Orthotropic heat transfer and subregion coupling Page 9 of 9 The crude discretisation of Fig. The Dirichlet data lead to very good accuracy while the Neumann data or fluxes show reasonable approximations.de/bem_pages/node25.html 2007/05/23 .

Potential problems are scalar field problems. thus the fundamental solution consists of a scalar function relating the effect of a source term at the load-point to its influence point .Fundamental solutions Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Fundamental solutions Next: Laplace equation Up: Boundary Element Methods Previous: Example: Orthotropic heat transfer and subregion coupling Fundamental solutions In this section fundamental solutions are derived for potential problems. Subsections Laplace equation Fundamental solution of the 2D Laplace equation Fundamental solution of the 3D Laplace equation Helmholtz equations Fundamental solution of the 3D Helmholtz equation 2007/05/23 . This point is usually called field-point.

bem.Fundamental solutions Page 2 of 2 Next: Laplace equation Up: Boundary Element Methods Previous: Example: Orthotropic heat transfer and subregion coupling Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.html 2007/05/23 .de/bem_pages/node26.uni-stuttgart.

Laplace equation Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Laplace equation Next: Fundamental solution of the 2D Laplace equation Up: Fundamental solutions Previous: Fundamental solutions Laplace equation The fundamental solution of the Laplace equation is a solution of the equation (152) Note that is the distance between the load.and field-point. This implies that a fundamental solution is a symmetric function (153) 2007/05/23 .

de/bem_pages/node27. Subsections Fundamental solution of the 2D Laplace equation Fundamental solution of the 3D Laplace equation Next: Fundamental solution of the 2D Laplace equation Up: Fundamental solutions Previous: Fundamental solutions Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.bem.uni-stuttgart.Laplace equation Page 2 of 2 The derivation of the fundamental solution of the Laplace equation in 2D and 3D is carried out here as an example for the general appraoch.html 2007/05/23 .

The derivation starts out by transforming the Laplace operator to polar coordinates (154) The excitation with the Dirac impulse is radial-symmetric and. too. The Dirac impulse in polar coordinates is stated as . Hence. This yields 2007/05/23 . it is implied that the fundamental solution is radial-symmetric. (154). since we are dealing with an infinite problem. (154) vanishes. a way to solve for is to integrate Eq. Thus the last term in Eq. there are no disturbances from the boundary.Fundamental solution of the 2D Laplace equation Page 1 of 4 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Fundamental solution of the 2D Laplace equation Next: Fundamental solution of the 3D Laplace equation Up: Laplace equation Previous: Laplace equation Fundamental solution of the 2D Laplace equation For simplicity the load-point is shifted in the origin.

(155) being a valid solution of Eq.Fundamental solution of the 2D Laplace equation Page 2 of 4 (155) with the integration constants and . (152).de/bem_pages/node28.html 2007/05/23 . This condition carries out the integral over the partial differential equation over an arbitrary volume enclosing the Dirac impulse (156) Application of Gauss' theorem transforms the volume integral on the left to a surface integral http://www.uni-stuttgart. Verification of the impulse condition The validity of a fundamental solution can be verified by evaluating the impulse condition.bem. It will be shown in the next section that for Eq. It is arbitrary and is generally set to zero. The constant introduces the notion of a constant potential.

uni-stuttgart.de/bem_pages/node28.html 2007/05/23 .bem. the outward normal on a circle is defined in polar coordinates as (159) Choosing the surface as a circle of arbitrary radius leads to the impulse condition (160) http://www.Fundamental solution of the 2D Laplace equation Page 3 of 4 (157) Since is radial-symmetric. In polar coordinates this reads as (158) Moreover. the gradient is also a pure function of the radius.

de/bem_pages/node28. (152).bem.uni-stuttgart. (155) is indeed a valid fundamental solution of Eq.Fundamental solution of the 2D Laplace equation Page 4 of 4 Since depends only on and is constant on a specific circle . Next: Fundamental solution of the 3D Laplace equation Up: Laplace equation Previous: Laplace equation Last modified 02/16/2004 ( wm) | © Universität Stuttgart | Imprint http://www.html 2007/05/23 . the impulse condition is reformulated as (161) which proves that Eq. Note that this condition also implies the must be zero as stated before because otherwise the terms would not cancel to -1.

Fundamental solution of the 3D Laplace equation Page 1 of 5 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Fundamental solution of the 3D Laplace equation Next: Helmholtz equations Up: Laplace equation Previous: Fundamental solution of the 2D Laplace equation Fundamental solution of the 3D Laplace equation In this case a transformation on spherical coordinates is carried out (162) Assuming radial symmetry yields (163) 2007/05/23 .

Fundamental solution of the 3D Laplace equation Page 2 of 5

The Dirac impulse in spherical coordinates is

(164)

since

(165)

and by equivalence

(166)

http://www.bem.uni-stuttgart.de/bem_pages/node29.html 2007/05/23

Fundamental solution of the 3D Laplace equation Page 3 of 5

The integration of the Laplace equation yields for the 3D case

(167)

Again, the impulse condition will show that . As before is arbitrary and is set to zero for convenience.

Verification of the impulse condition

The derivation is analogous to the 2D case. The integral over the partial differential equation is transformed to
the boundary. Since the solution is radial symmetric the gradient has only a component in the radial direction.

(168)

A sphere is chosen as arbitrary enclosing surface in the 3D case. The normal vector is a unit vector in spherical
coordinates

http://www.bem.uni-stuttgart.de/bem_pages/node29.html 2007/05/23

Fundamental solution of the 3D Laplace equation Page 4 of 5

(169)

With this the impulse condition is

(170)

Again, only depends on and thus is constant on a sphere of constant radius. It follows

(171)

As in the 2D case, this shows that must be set to zero so that fulfills this equation.

The solutions for the potential and the flux as the normal derivative of the potential in 2D and 3D are

http://www.bem.uni-stuttgart.de/bem_pages/node29.html 2007/05/23

uni-stuttgart.bem.html 2007/05/23 .de/bem_pages/node29. Table 2: Fundamental solutions of the Laplace equation 2D 3D Next: Helmholtz equations Up: Laplace equation Previous: Fundamental solution of the 2D Laplace equation Last modified 02/18/2004 ( wm) | © Universität Stuttgart | Imprint http://www.Fundamental solution of the 3D Laplace equation Page 5 of 5 summarized in Table 2.

the function is scalar.Helmholtz equations Page 1 of 2 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Helmholtz equations Next: Fundamental solution of the 3D Helmholtz equation Up: Fundamental solutions Previous: Fundamental solution of the 3D Laplace equation Helmholtz equations A fundamental solution for the Helmholtz equation is derived by solving (172) As in the case of the Laplace equation. Subsections Fundamental solution of the 3D Helmholtz equation 2007/05/23 .

html 2007/05/23 .de/bem_pages/node30.bem.Helmholtz equations Page 2 of 2 Next: Fundamental solution of the 3D Helmholtz equation Up: Fundamental solutions Previous: Fundamental solution of the 3D Laplace equation Last modified 02/18/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.

Up: Helmholtz equations Previous: Helmholtz equations Fundamental solution of the 3D Helmholtz equation Transformation on spherical coordinates and taking radial symmetry into account yields (173) A solution is obtained by choosing (174) 2007/05/23 ..Fundamental solution of the 3D Helmholtz equation Page 1 of 5 uni kontakt Universität Stuttgart Home | Lecture notes | Exercises | Institute of Applied and Experimental Mechanics Boundary Element Methods Fundamental solution of the 3D Helmholtz equation Next: About this document ..

(176) This yields http://www. Hence. which after comparison to Eq.uni-stuttgart. To verify the impulse condition the behavior for small is considered. ( ) for the 3D case. this second constant is free and is adjusted such that the ansatz in Eq.bem.html 2007/05/23 . (167) yields (175) The sine-term is again a homogeneous solution and does not contribute to the Dirac impulse. A series expansion of the cosine-function shows that the singularity behavior is .Fundamental solution of the 3D Helmholtz equation Page 2 of 5 where the term with the function is singular for while the term with the function remains regular.de/bem_pages/node31. (174) fulfills the Sommerfeld condition Eq.

Fundamental solution of the 3D Helmholtz equation Page 3 of 5 (177) and the complete solution is (178) For small the behavior is (179) This implies that also in 3D the fundamental solution behaves for small or like the fundamental solution of the Laplace equation.bem.uni-stuttgart.de/bem_pages/node31.html 2007/05/23 . The flux of the 3D solution is http://www.

uni-stuttgart. Up: Helmholtz equations Previous: Helmholtz equations http://www.bem.Fundamental solution of the 3D Helmholtz equation Page 4 of 5 (180) The solutions for the potential and the flux as the normal derivative of the potential in 2D and 3D are summarized in Table 3. Table 3: Fundamental solutions of the Helmholtz equation 2D 3D Next: About this document ..html 2007/05/23 ..de/bem_pages/node31.

bem.Fundamental solution of the 3D Helmholtz equation Page 5 of 5 Last modified 02/18/2004 ( wm) | © Universität Stuttgart | Imprint http://www.uni-stuttgart.html 2007/05/23 .de/bem_pages/node31.