Prognoses in geotechnical engineering require both adequate material models and identification of the involved material constants. Scatter of the properties of geomaterials renders parameter identification a challenging task.
Herein, this problem is treated in an integrated manner. The first part of the book is devoted to methodical aspects: based on soft computing, an iterative parameter identification method is developed. Artificial neural networks are trained to approximate the underlying direct problem. Subsequently, a genetic algorithm solves the inverse problem. The method is applied to tunneling according to the New Austrian Tunneling Method and to ground improvement by means of jet-grouting. The second part of the book deals with conceptual aspects of parameter identification. To this end, the problem of rockfall protection of a gravel-buried steel pipeline is considered. It is highlighted that identification of material parameters and verification of a structural model must be based on two independent sets of experiments.
The book will appeal to the scientific community as well as to practical engineers faced with the task of parameter identification.