In this paper, an attempt is made to develop a method of selecting an appropriate repair or reinforcement method by achieving useful rules throughout all the process from inspection to completion of repair. The selection of repair methods is carried out based upon the quality of material, bridge form, degree of damage and so on. Because of the variety of damage and environmental conditions, rules for repair and reinforcement should be modified to be suitable for each bridge. In order to acquire available rules, classifier system is introduced, which uses Genetic Algorithm (GA) as a framework of machine learning. The rules generated are applicable to the search for the optimal repair schedule and making a robot for inspection of each bridge so as to reduce the time and budget required for inspection. When realizing the inspection robot, it is possible to avoid the time-consuming and troublesome works by an autonomy type control system.

The classifier system consists of rule base, effecter, and detector; the rule base accumulates a set of rules, the effecter sends signals to the environment, and the detector changes the rule base by use of the credit assignment system. The classifier system can finally realize the refinement or generation mechanism. The structure of classifier system is shown in Figure 1. In this research, the classifier system obtains the rewards when attaining the purpose. The Bucket Brigade Algorithm is used to strengthen the intensity of rules. Several numerical examples are presented to demonstrate the applicability and efficiency of the proposed method.

Figure 1: Architecture of classifier system.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £95 +P&P)