Research On Key Technologies Of Multi-channel Electromagnetic Disturbance Real-time Monitoring For Vehicle Equipment

Electric multiple unit(EMU)is a typical comprehensive system which including a large number of cables linked to varies of on-board equipment.The mass cables and bundles are laid intensively in limited space or in the same slot,and the crosstalk between cables is unavoidable.Because of these,the similar disturbance forms and spectrum are often observed on different wires,which make it difficult to distinguish the actual source of noise.In addition,for a synchronized multi-channel measuring or monitoring system,the coupling between sensor cables will also make cross-influence or lead to errors to the test results.In this thesis,the frequency response of the cable coupling network which represent a multi-channel electromagnetic disturbance monitoring system are analyzed,and used as the decoupling function to correct the port’s voltage.Furtherly,the pre-calibrated frequency response parameters of the network can be used for the monitoring system to compensate the errors to the amplitude and phase of the observed signals,which means we can reverse the test result from the receiver’s port to the original sensor output and remove the cables influences.In this thesis,several key technologies of multi-channel real-time electromagnetic disturbance monitoring system are studied.The system is based on a NI PXIe virtual instrument platform and implemented by programming in the LabVIEW environment.A synchronized dual-channel real-time spectrum analyzer is realized,and the decoupling method which mentioned above is applied to either monitoring ports.The functions are validated by test on an equivalent circuit which simulating a pair of mutual coupling transmission lines.Firstly,a combined time-frequency trigger function is proposed to reduce the data transmission pressure on the system bus,which can significantly lessen the data in the multi-channel acquisition and solve the problem of mass data storage,by only capture and process signals of interest and abandon most useless samples.In this technology,the frequency-domain triggering is realized by setting trigger templates and comparing the single-frame spectrum of acquisitions.Secondly,for the crosstalk analysis of cable-coupled network,multi-channel correlation analysis is designed to solve the frequency domain response between ports.The frequency-domain multi-channel decoupling and test port voltage correction are realized by utilizing the frequency response characteristics between mutual-coupling channels.Finally,the functions of each modules of the monitoring system are verified by a series of measuring and test on a self-made equivalent network,which circuit parameters are adopted to simulate a pair of mutual coupling transmission lines.The frequency response of the network is obtained by a pre-calibration which perform direct-path and coupling-path tests.By comparing the frequency responses either derived from the test or from mathematical simulation,it shows an acceptable compliance.The experimental results also shows good decoupling and correction performance,which can depress the crosstalk between the monitoring system’s test channels for different frequency signals.