A method for elevator fault identification using multilayer perceptron neural network

Abstract

The present invention discloses a method for elevator fault identification by employing multilayer perceptrons. The method comprises the steps of: collecting related data of elevator faults, and establishing a database; establishing multilayer perceptrons for features of elevator fault identification; reading elevator fault data for training of a neural network from the database to perform training and learning of the neural network, and after the training of the neural network is completed, using the neural network to perform fault identification of the actual data of the elevator; and finally, outputting one or more than one fault codes with large a probability for maintenance of field staff according to a total quantity control principle. Improvement of safety can be achieved through remote update of a weight when part of the fault identification algorithm needs upgrading. The method provided by the invention can remote grade the fault identification algorithm through adoption of elevator state parameters and analysis of fault reasons so as to maintain rapidly determined faults and have a great meaning for guaranteeing the masses' life safety.

Description

technical field [0001] The invention relates to a method for identifying elevator faults by using a multilayer perceptron neural network, belonging to the field of elevator fault identification. Background technique [0002] With the rapid development of urbanization in our country, there are more and more high-rise buildings, and the popularity of household elevators is also increasing rapidly. Elevators have also become necessary transportation equipment in people's daily life. [0003] According to news reports, as of the end of 2014, the number of elevators in use in my country has exceeded 3.5 million, and the number of new elevators is close to 500,000 each year. The number of old elevators that have been in operation for many years has increased rapidly, and the number of elevators that have been used for more than ten years has exceeded 1 million. Although the reliability of elevators has been greatly improved in recent years, in actual operation, abnormal operating c...

AI Technical Summary

Problems solved by technology

However, with the continuous development of elevator fault identification technology and the advancement of elevator state sensor technology, the number of elevator state sensors continues to increase, and the range of fault codes continues to expand. The traditional method of program writing encounters great difficulties, and the program becomes more and more difficult. The more complicated it is, the more difficult it is to solve in terms of design methods and iterative upgrades

[0005] In particular, the original judgment program that could be easily written needs to completely modify the logic relationship due to the addition of sensors. If something goes wrong, engineers must write the program on site or recall products.

And because the fault classification logic is too complicated, there is a problem of logical overlap. The same input signal may produce different outputs, which will mislead the maintenance personnel. When the output fault code is wrong, you can only try one by one. The efficiency of maintenance is very low

Images