ODIN PRESS
Technical › The Resolvent Method for PDEs
The Resolvent Method for PDEs, front cover

The Resolvent Method for PDEs

Solving Evolutionary Equations Where the Semigroup Cannot

by T. Zamrik

Python358 pages14 chapters6 × 9 inPaperbackHardcoverDust jacket

Contents

  1. Preface
  2. 1The Resolvent and the Semigroup: A Laplace Duality
  3. 2Sectorial Operators and Holomorphic Semigroups
  4. 3The Dunford–Riesz Functional Calculus
  5. 4Resolvent Estimates and Maximal Regularity
  6. 5Non-Self-Adjoint Generators and the Failure of Eigenexpansion
  7. 6Pseudospectra: When Eigenvalues Mislead
  8. 7Degenerate and Non-Densely-Defined Generators
  9. 8Evolution Families and the Nonautonomous Cauchy Problem
  10. 9Constructing the Propagator from Frozen Resolvents
  11. 10Volterra and Fractional-in-Time Equations: Resolvent Families
  12. 11Accretive Operators and the Nonlinear Resolvent
  13. 12Viscosity Solutions and the Hamilton-Jacobi-Bellman Equation
  14. 13Policy Iteration as Iterated Linear Resolvents
  15. 14Time-Inhomogeneous HJB via the Nonautonomous Nonlinear Resolvent
  16. A Source Code
  17. Bibliography
  18. Index
  19. About the Author

About this book

The standard theory of evolution equations puts the semigroup — the exponential of the generator — at its centre and treats the resolvent, the inverse of the shifted generator, as scaffolding. This book reverses the order. The resolvent is the Laplace transform of the semigroup: prior to it, and, unlike it, available even where no semigroup exists.

That is not a matter of taste. There is a ladder of problems on which the semigroup degrades or vanishes while the resolvent still carries the solution: • the spectrum betrays you — non-self-adjoint generators, where pseudospectra tell the truth the eigenvalues cannot; • time refuses to factor — non-autonomous and fractional-in-time problems, where the propagator is built from frozen resolvents; • nonlinearity — where the nonlinear resolvent solves the Hamilton–Jacobi–Bellman equation.

The book carries one thesis across all three, to the doubly-hard summit: a time-inhomogeneous control problem — correcting circadian misalignment by timed light — solved as a convergent product of frozen-time nonlinear resolvents. Rigorous throughout and exact about where the method stops, it includes the Python code that reproduces every table and figure.

The covers

The Resolvent Method for PDEs, back coverThe Resolvent Method for PDEs, front cover