Research And Implementation Of Monitoring And Warning System For Elevator Operation

For several decades since the reform and opening up,China has reached the world advanced level in the development of economic,political and cultural aspects.Under the policy of rapid urbanization and countryside into community,a large number of bungalows change to tall-buildings.As a vehicle for running up and down inside tall-buildings,the elevator's existence can't be ignored,whether in shopping malls,hospitals or residential areas.With the rapid increase in the use of elevators,the imperfect maintenance measures and unseasonal warning will lead to frequent occurrence of elevator accidents.Even now the elevator accident also wounding still occurs.Therefore,the safety operation monitoring and fault prediction of the elevator has become an important issue in the process of modern social development.Firstly the paper introduces the situation of elevator field in China,and then analysis on the system structure,technical specifications and common faults of elevator.With investigation on the current research on the elevator monitoring system at home and abroad,this paper pointed out the advantages and disadvantages of the elevator monitoring system,and emphasizes the high demand of fault prediction technology in elevator field.At the same time,this paper introduced the development of fault prediction technology and RBF neural network.Aim at questions above,the main works are as followed:First,a new elevator operating monitoring and warning system with improved openness and versatility has been established.Equipped with a number of external sensors,the system monitors the running status of the elevator all day.The functions of the system include the monitoring and collection of elevator operation parameters,the storage of elevator maintenance and inspection record,the information management of all units(such as property units,maintenance units,inspection units),map visualization,the prompt of the alarm message and statistical analysis of relevant information.The innovations of this system are added database management function and the prompt of the alarm message function,so that the level of information management of traditional monitoring system improves and it is possible to predict and prompt the types of faults that may occur after a period of time.This can reduce the frequency of elevator fault by maintained the elevator in advance by the relevant personnel.Aim at the fault that can't eliminate advanced,the system also can prompt the alarm message to the users in the shortest possible time to provide conditions for the timely release of the fault.Second,combining RBF neural network with elevator fault prediction,this paper established a neural network model that can predict the fault types may occur after a certain period of time based on the elevator running state parameters.Meanwhile,because the disadvantage of the random selection of initial hidden layer nodes of the standard radial basis function neural network,and it has a great impact on the results of the model,so that PSO algorithm is used to optimize the model parameters in this paper.Above the data collected from the elevator operating monitoring and warning system,a contrast experiment is carried out on the standard RBF neural network prediction model and the optimized model.Through the analysis of experimental results,it can be concluded that RBF neural network is feasible in the field of elevator fault prediction,and the result shows that the PSO algorithm can effectively improve the prediction accuracy and training speed of RBF neural network.