Linear Algebra and Its Applications · Problem 6.5.5
Computations with Matrices — Gilbert Strang
Mathematics · galerkin-method
Problem
Galerkin's method starts with the differential equation (say −u′′=f(x)) instead of the energy P. The trial solution is still u=y1V1+y2V2+⋯+ynVn, and the y's are chosen to make the difference between −u′′ and f orthogonal to every Vj:
Galerkin∫(−y1V1′′−y2V2′′−⋯−ynVn′′)Vjdx=∫f(x)Vj(x)dx.
Integrate the left side by parts to reach Ay=f, proving that Galerkin gives the same A and f as Rayleigh-Ritz for symmetric problems.