Design Of Vibration Monitoring System For Steam Turbine

As a kind of significant equipment in power system,steam turbine must runfor long.Shaft which is the core componet affects the normal functioning of theunit. Shaft vibration whose measurement will not affect the normal functioningof the unit can produce many faults.Therefore,it is a very effective method tojudge the operation condition of unit by analying vibration signal.The traditional vibration monitoring system based on microcontroller.Withthe steam tubine to large-scale and high capacity development,system continuesto increase the requirement of data and computation, microcontroller has beenunable to satisfy real-time analysis.Therefore, the hardware platform needs thecore chip which is fast processing and large storage for data processing andspectrum analysis.In this paper,based on the research of the vibration signal measurement andanalysis method, the program of shaft vibration monitoring system isdetermined.General DSP chip as the main controller is used to design themonitoring system,which is the TMS320LF2407A of TI company.The vibrationsignal is obtained by the sensor and is processing by the signal conditioningcircuit; the key signal is the reference signal of integer-period sampling.Systemwill give an alarm signal and take protective measures,when the parameter valuewhich is calculated from the DSP exceeds the safety threshold.The spectrumanalysis of vibration signal through Fast Fourier transform(FFT) obtain theamplitude and phase of different frequncies.The system design some modula hardware circuits,and the software use themeathod of mixed promramming by CCS,which increase readablity andportability.During the vibration signal sampling,the unsynchronous of sampling period and signal period and the uniform sampling interval can generateerror.According to the problem,the paper adopts the method of combininghardware and software, which not increasing the cost and complexity ofcircuit.The sampling interval is adjusted dynamically in a signalperiod.Simulation results demonstrate that the proposed algorithm can reduce thesynchronous error and improve the sample accurary.