Application Research Of Fuzzy Neural Network In Elevator Group Control System

With the development of high-rise buildings, elevator group is playing a more important role in high-rise buildings and intelligent buildings, so the elevator group control system (EGCS) is now the focus of the researchers at home and aboard. In this paper suitable intelligent algorithms are applied in this complicated system according to the theory of ECGS and research focus.Firstly the history and status quo are reviewed, and the basic theory of EGCS is introduced, then two common methods in EGCS:Fuzzy Logic and Neural Network, are put forward and their advantages and Shortcoming are analyzed.Then the combine of Fuzzy Logic and Neural Network:Fuzzy Neural Network, is put forward, and also the structure and reasoning process are analyzed. Then the MATLAB program is made use of to verify the effectiveness and reliability of the Fuzzy Neural Network model introduced in this paper.After the preparation of theory, a mixed learn and train method is used in the Fuzzy Neural Network model. Firstly k-means algorithm is used to get the preliminary centers and widths of the membership functions, then the fuzzy rules are abstracted by the means of sequence clustering, at last, PSO with dynamic inertia weight is applied in the optimization and adjustment of centers and widths of the membership functions. When the mixed learn and train method is finished, a Fuzzy Neural Network model is built. The means used here to build and train a Fuzzy Neural Network will be applied in the following pattern recognition of traffic module and elevator scheduling module.At first this means is used to build a FNN for the pattern recognition of traffic according to the request and feature of this module, then a result of pattern recognition of traffic through analysing the feature data of traffic is got, so the corresponding strategy will be desighed according to the feature of traffic.The the same means is used to built a FNN for the elevator scheduling module, according to the pattern recognition of traffic and the corresponding strategy, with the help of FNN, the elevator scheduling plan is got through analyzing the state data of elevatorsAt last, the EGCS is simulated and verified with experimental data, the result proves the effectiveness of the algorithm and model put forward and built in this paper.