Research And Development Of Elevator Electrical Control Fault Diagnosis System Based On Fuzzy Neural Network

Under the condition of the large number of elevators and the shortage of maintenance and management personnel,it is impossible to timely detect and diagnose the elevator faults.In view of the contradiction in the quality inspection work,on January 26,2010,the state administration of quality supervision,inspection and quarantine officially released the national special equipment safety development strategy outline requirements thoroughly change the traditional rely mainly on human on-site inspection,emergency rescue process and the present situation of the accident appraisal is given priority to,to support for special equipment safety science and technology.Therefore,it is in line with the development trend of elevator safety work to develop the elevator remote online fault efficiency monitoring system and establish the long-term mechanism of the fault efficiency monitoring.Because of the dynamic and random nature of elevator traffic flow,the same scheduling algorithm cannot be used to realize the full day elevator scheduling.Generally,the scheduling algorithm is selected according to the results of traffic pattern recognition.Therefore,the study of elevator traffic flow and scheduling algorithm has been the focus and difficulty of elevator group control system.(1)On the basis access to relevant literature,research progress on comprehensive analysis,the theory of fuzzy fault detection and diagnosis of electrical control system of the elevator involved in the development of neural network,BP neural network,BP network and its learning methods,genetic algorithm The related theory and elevator system,elevator group control system,etc.are introduced.The common faults such as inverter,traction machine,safety circuit,door machine,elevator controller and three-phase power input unit in the components and control components of the elevator control system are analyzed.On this basis,the elevator fault diagnosis model is built.(2)The establishment of a fault diagnosis neural network model.In the collection of various state parameters,the fuzzy system and the neural network are combined to establish a fuzzy neural network model.Then the input parameters are fuzzified,the membership functions of different parameters are determined,and finally the BP model is designed.In the process of building the model,the fuzzy information is used to process the elevator fault information,which can improve the accuracy of fault diagnosis and reduce the fault diagnosis error.(3)The design and simulation of fuzzy neural network algorithm are deeply analyzed,and the method of using genetic algorithm to optimize neural network is proposed.Firstly,the weights and thresholds of the neural network are optimized,and the weights and thresholds obtained by the optimization are used to replace the original random weights and thresholds.Finally,Matlab is used to simulate and compare the algorithm to prove the feasibility and advantages of the improved method.(4)The elevator remote monitoring and fault diagnosis system was designed with the community as the research object.The system development environment was designed as a whole,which can clearly show the advantages of flexible data storage and beautiful interface with Flex,Java and mySQL.The physical structure and functional structure of the system are designed.Finally,the elevator remote monitoring and fault online diagnosis are realized.