Engine Knock Detection Based on Computational Intelligence Methods

Artificial intelligence emerged from human thinking that has both logical and intuitive or subjective sides. The logical side has been developed and utilized, resulting advanced von Neumann type computers and expert systems, both constituting the hard computing domain. However, it is found that hard computing can’t give the solution of very complicated problems by itself. In order to cope with this difficulty, the intuitive and subjective thinking of human mind was explored, resulting the soft computing domain (also called computational intelligence). It includes neural networks, fuzzy logic and probabilistic reasoning, the last gathering evolutionary computation (including genetic algorithms with related efforts in genetic programming and classifier systems, evolution strategies and evolutionary programming), immune networks, chaos computing and parts of learning theory. In different kind of applications, all pure artificial intelligence methods mentioned above proved to be rather complementary than competitive, so that combined methods appeared in order to gather the advantages and to cope with the disadvantages of each pure method. The scope of this chapter is to study and finaly compare some representative classes of pure and combined computational intelligence methods applied in engine knock detection.