Design And Implementation Of Elevator Fault Prediction System

In recent years,with the sustained and rapid increase in the number of elevators in China,malignant accidents happen frequently,such as drop,out of control.The elevator accidents pose an enormous threat to passengers'safety,so the safety issues of elevator are widely concerned.The ownership,management and use of the elevator involves multiple subjects,each negligence may lead to a safety incident of elevator.In fact,before the most severe accidents,elevators would stay in different degrees of abnormal conditions.The fault rate of elevators will be' greatly reduced through predictive maintenance by the means of monitoring the running status of the elevator.Hence,it will possess significant research consequence and practical value to design and implement an elevator fault prediction system.From the perspective of elevator's safety,I designed and developed an elevator fault prediction system which realizes some functions such as elevator running status monitoring,fault alarming,fault prediction,urgent video call and passengers' identities verification,and it provides the corresponding information communication channels forpassengers,property companies and maintenance companies by the cloud automatic message sending function.The main contents of this paper are listed as follows:Firstly,in this paper,we proposed an approach of monitoring the running status of elevators,identifying passengers'identities and urgent video call,and a data acquisition terminal was designed and developed.The data acquisition terminal realizes the functions of elevator running status monitoring,passengers' identities verification,edge computing and data transmission,etc.The micro-controller on the data acquisition terminal collects sensors' data and sends it to the RT5350 module on the data acquisition terminal which through the serial port.The RT5350 module can continuously process the data in real time to determine the safety status of elevator and send the preprocessed data to the server through the HTTP protocol.The identification module on the data acquisition terminal can acquire the MAC address of the passenger's mobile device through the WiFi/Bluetooth probe,then the elevator fault prediction system can identify the passengers' identities according to the obtained MAC address.When the elevator malfunctions,the RT5350 module would start the WiFi usage right,and the trapped passengers can make video calls with the property staffs through the browser of their mobile phones after they connect the mobile phones to WiFi.The data acquisition terminal is equipped with simulation software which can simulate the running state and fault state of elevator.Secondly,a set of system management platform software was designed and developed.The system management platform can receive the elevator's running data which is sent by the data collection terminal,and the real-time running status,fault alarm information,and fault prediction results of the elevator are displayed through an interface of Human-Computer Interaction,and it can manage and configure the data collection terminal remotely.The system management platform mainly provides services for the property companies,and the staffs in property companies can uniformly supervise the state of elevators after sign-up in the system.The passengers of elevators can use four functions provided by the system:message sending,fault diagnosis,identity verification and self-help after register in the system.The system management platform can save historical fault information of elevators and send real-time running data to the cloud.Lastly,a cloud platform system was designed and developed.The cloud platform is developed based on Alibaba Cloud which realizes the functions of fault diagnosis and message sending.The system management platform sends the fault alarming information to the cloud platform,and then the cloud platform sends the fault alarming information to the relevant people by We-chat or Message.An elevator fault prediction model is also studied in this paper,which can predict faults of elevators such as sticking layer,colliding to top,collapsing to bottom,over speed and so on.The fault prediction model is deployed on the cloud platform,and then the cloud platform receives the running data of elevator which is sent by the system management platform and predicts the running status of elevator through fault prediction model.The cloud platform will send a warning massage to the relevant personnel promptly when it predicts the elevator may fail.