Research On Real-time Acquisition System Of Elevator Network Based On Multi-core DSP

With the rapid development of society and the continuous progress of science and technology,the scale of the city is expanding.In order to save the urban land area,the building is also getting higher and higher.Elevator has become an indispensable lifting tool in people’s daily life.The safety of elevators has also attracted great attention of manufacturers and users.In order to ensure the safe operation of elevators and make the development of urban life more safe and reliable,elevator monitoring technology has emerged.At present,the serious elevator faults are as follows:1)the elevator runs and stops suddenly.2)the elevator vibrates in the horizontal plane.3)the elevator stops in the position of the non-elevator door.This technology can monitor the operation of elevator in real time,and detect unsafe factors that may cause malfunction.So it is of great significance.Nowadays,the scale demonstration network system of Internet of Things of elevator operation safety is divided into two subsystems: one is elevator operation safety parameter acquisition system,the other is elevator audio and video image acquisition,elevator running status display and media platform integrated system.The elevator operation safety parameter acquisition system carries on the real-time monitoring to the elevator operation state,connects the elevator operation safety monitoring data to the elevator emergency disposal platform,is able to alarm when encounters the elevator fault,can make the statistical analysis to the elevator failure,and warns to the elevator maintenance unit.In this paper,the safety monitoring platform of elevator is established,and the elevator safety monitoring system is built on the basis of simplified platform.According to the characteristics of elevator safety monitoring,the hardware and software design of elevator information real-time acquisition system based on multi-core DSP,which is the core part of the monitoring system,is completed.The elevator monitoring system takes TMS320C6678 DSP of TI Company as the core,including hardware circuit,parallel computing,data processing,data input and output,etc.The research goal is that the data processing has been carried out when the node elevator operation data enters the Internet of Things of elevator,the real-time online analysis with the combination of the Internet of Things is carried out,and the operation efficiency of the Internet of Things is improved.Among them,the image processing module is used for matching the obtained current frame on-site video image with a relatively stable reference image.The main method is the bottomfeature matching method,the selected feature extraction algorithm is scale invariant feature transformation algorithm,and the feature matching algorithm is K nearest neighbor algorithm.Through the detection results of each adjacent frame image,it is transmitted to the Online Analysis Center of the Internet of Things to judge the open and close of elevator room and whether people are trapped in an elevator.After judging someone in the elevator and detecting that the elevator door does not move beyond the set time threshold,it judges that someone are trapped in the elevator,and sends out an alarm message.