Structural monitoring using engineer–computer interaction
Artificial Intelligence for Engineering Design, Analysis and Manufacturing
| Hi-index | 0.01 |
Four decision-support systems: Mistral, Damsafe, Kaleidos, and Igor--provide powerful, AI-based tools for evaluating structural data. Safety managers, engineers, and authorities are using the systems to handle safety problems in structures.Civil engineers need significant resources to ensure the safety of structures. They collect data about structural behavior through tests and visual inspections, while using automatic instrumentation and data acquisition systems for real-time monitoring. Interpreting such data is not easy. Several factors are involved, such as the large amount of data; the uncertainty and incompleteness of information; and the need for engineering judgment, knowledge of the particular structure, experience with the behavior of structures in general, and general engineering knowledge to interpret the data.AI concepts and technologies can assist engineers in safety management by providing new software components to the existing information systems, such as real-time interpretation systems linked to the data acquisition units, qualitative models, and reasoning agents supporting the off-line management of information and interpretation.In this article, we describe four decision-support systems that use such concepts and technologies to better manage the safety of civil engineering structures. During the last six years, the software development unit of ISMES--an R&D company involved in structural and mechanical engineering, environmental and land-use engineering, and information and communication technologies--has focused on AI applications to structural safety. We have addressed two main problems: the safety management of dams and monuments, and the seismic risk assessment of buildings. This led to the development of the four systems: Mistral, Damsafe, Kaleidos, and Igor.We exploit AI tools for designing intelligent modules of our support systems, including causal networks of processes, qualitative modeling, model-based reasoning, and hierarchical object-oriented representations. The systems also employ AI techniques such as rule-based systems, pattern matching, and neural networks, in conjunction with conventional techniques, to implement these representation and reasoning schemes.