The Application And Research Of Granular Computing In Hierarchical Fuzzy Control Based On Fuzzy Set

The idea of granular computing emerged in the late 1970's. It imitates the manner of human thinking, just as Zhang and Zhang said: A well known feature human intelligence is that human can not only observe and analyze a problem at different grain-sizes but also translate from one granule world to the others with no difficulty. As a new tool dealing with incomplete, uncertain, imprecise and inconsistent knowledge, granular computing is a big umbrella which covers all the research of the theories, methodologies, technologies, and tools about granules. It is an important base of artificial intelligence and now becomes a hot research topic domestically and abroad which includes fuzzy set theory, rough set theory and quotient space theory. The thesis covers the following aspects:1. The background, status and main theories of granular computing are introduced. Thus explain the research background and significance of the thesis. Then we introduce the model of granule and discuss its main properties. We place emphasis on the measurement of granules, universes and attribute of granular model, and the method of choosing the granule world.2. Introduce the method that how numerical variables are given to linguistic variables and the extent of the fuzzy thought, which leads to a fuzzy .set. Then a fuzzy set is the basic definition of basic concepts and algorithms, and make a detailed description, so that people had a fuzzy set of general awareness. Finally, give the relationship among the fuzzy sets, computing with words and granule.3. Introduce the relationship between the fuzzy thought and hierarchy, namely, research the solution which is from local to global. On this basis, put forward a new hierarchical fuzzy model, whose property and structure methods described in detail. Then try to use this method to solve the problem that number of fuzzy rules with fuzzy system input variables and the increase in the number of index rose to the "Dimension Disaster".4. Put forward a new algorithm based on hierarchical fuzzy neural networks that combines hierarchical evolutionary programming and mixed-coded genetic algorithm. The networks use the mixed-coded which is improved so as to get the optimal topology and network parameters at the same time. The algorithm also can be used to reduce the number of further optimization of fuzzy rules in the system.