Optimal methods for ill-posed problems. With applications to heat conduction (Q1704614)

The textbook is concerned with the design and theoretical analysis of some optimal numerical solvers for ill-posed problems, with direct applications in inverse heat exchange problems. It is based on the lecture notes delivered by the authors to the students involved in the Applied Mathematics and Informatics master programe at Chelyabinsk and South Ural State Universities over the past decade. It is structured in five chapters. The first four chapters are concerned with basic aspects from the theory of inverse problems and regularization techniques. The authors first present the Lavrent'ev method for approximate solution of first kind linear equations, together with its accuracy level for the Lavrent'ev scheme regularization parameter choice. Then, they describe the Tikhonov regularization method and its error estimate for the residual principle choice of the regularization parameter. Chapter 4 is dedicated to projection-regularization method for unbounded operator values. The last chapter of the book is concerned with applications to inverse heat exchange problems. The authors analyse the accuracy estimation of approximate solutions for both constant coefficient and variable coefficient heat conduction equation. All the applications are on a pure theoretical level, without numerical experiments. The book can serve as a course in the corresponding topics for master and PhD students, but it can be also used by researchers working in the field of inverse problems.