In this paper, in according to a lot of new achievements about Fuzzy Mathematics and ANN, we use Self Organizing Algorithm and Genetic Algorithm (GA) as the learning methods of FNN, hope to explore new way of information extraction from complicated data.
First,a group of initial fuzzy rules are produced using self organizing algorithm to cluster learning data. Then error back propagation algorithm is introduced to adjust the network parameters. The simulation results demonstrate that the fuzzy neural network has the advantages of simple structure and the high fitting accuracy.
To improve the nonlinear approximating ability of cerebellar model articulation controller(CMAC), by introducing the Gauss basis functions and the similarity measure based addressing scheme, a new kind of fuzzy CMAC with Gauss basis functions(GFCMAC) was presented. Moreover, based upon the improvement of the self organizing feature map algorithm of Kohonen, the structural self organizing algorithm for GFCMAC(SOGFCMAC) was proposed.
In the same time, we use Self Organizing Algorithm and GA to modify the structure and parameters of FNN, hope that the combination of these algorithms can be carried out in many kinds of complex informational problems, such as pattern recognition, quality evaluation, and data extraction and so on.
A automatic control system based on microcomputer for gas forming machine is presented. A fuzzy control self-organizing algorithm, which has been used in gas forming machine, and the design for hardware and software of control system are also described.
A distributed multi-channel self-organizing algorithm is first proposed.Next,a MCSON-II prototype system of a multi-channel self-organizing packet radio network with eight nodes using the algorithm is reported.Finally the system performance is evaluated by simulation.
A new modem technique for HF communication network with dynamic data rates is first given in the paper.Then a distributed self-organizing algorithm for the network is proposed.Finally its prototype system and performance are presented.