1、 原文 ： Design of Fault Diagnosis Monitor System for the Locomotive Bearings Based on DSP Technology Abstract The rolling bearing is one of the key parts of the locomotive running components, because it condition is directly related to the performance and safety of locomotive. In this paper, the monit
2、or system for the locomotive bearings based on DSP TMS320LF2407A is designed. This system diagnoses the rolling bearing fault using vibration analysis method. It is based on comprehensive resonance demodulation and fast Fourier transform technique, and it adopts related methods to handle the result
3、of the FFT. It effectively improves the response characteristics, sensitivity, differentiate and measurement accuracy of the bearing failure monitor system, and it can fulfill the monitor and prediction of the transient fault in the course of the locomotive running. Key words: resonance demodulation
4、 technology； digital signal processor； related methods I. Introduction The higher safety is required to the trains because its speed is raised constantly. Bearing fault is one of the major factors causing eventful traffic accidents and affecting rail safety. Currently the railway system usually uses
5、 the bearing temperature detector to monitor the locomotive bearing condition. Theoretical analysis and a lot of practice show that the bearing temperature detector can prevent accidents from occurring to some extent, but most of the bearing fault is not sensitive to temperature. When the temperatur
6、e of the bearing is beyond the range and the system gives an alarm, the worse damage of the bearing has occurred, and even the incident had happened. Therefore, to find the fault more early and accurately, the more advanced monitoring means must be adopted. Most of the bearing fault is very sensitiv
7、e to vibration signal. The fault can cause vibration of the bearing increased. Compared with monitoring the temperature of bearing, the analysis and processing results to the vibration signal has more advantage than the temperature means. II. System composing and work process Based on the need, the
8、monitor of the bearing fault monitoring system to the locomotive bearing sets two detections: itineration detections and fixed detections. The itineration detection is used in the normal conditions, and the fixed detection is used for the continuous monitoring of the fault bearing. The system adopts
9、 special composite sensor to collect the vibration of the bearing and the temperature signal at the same time. After the data processing, the corresponding fault levels and rise in temperature are got. The data acquisition unit is designed in this system. Alarm information will be transmitted to all
10、 carriages through interfaces so that the staff can handled in time, and the same time, the fault data and the related information of the train such as the current location and speed will be transmitted to the dispatch center through GPS, which is convenient to adopt corresponding measures. The syst
11、em block diagram is in Fig. 1. III. The key technology of the design for the monitoring system A. The spectrum analysis means for diagnosing bearing fault Dispatch Center GSM system Industry computer Instrument for data Instrument for fault diagnosis Instrument for fault diagnosis Instrument for fau
12、lt diagnosis Fig.1 System block diagram Under normal circumstances, all parts of the rolling bearing (inner circle, outer circle, roller, holding frame) will retain the stable relative movement state. If the surface of some element (except for holding frame) has crack, and this crack is in the surfa
13、ce of the rolling adjacent component, the instantaneous vibration impulse must be produced. Assumed that the number of the roller in the bearing is Z； the diameter of the roller is d； the average diameter of the bearing inner circle and the bearing outer circle (the diameter of the roller revolution
14、 path) is D； the frequency of the bearing rotation is f0. Assumed that the inner circle is fixed and the outer circle is circumvolved, the vibration frequency brought by the surface defects of different bearing components can be derived. These frequencies can be called the fault characteristics freq
15、uency of the inner circle, outer circle and the roller. c ir c le )( in n e r 21 01 fDdZf c ir c le )( o u te r 21 01 fDdZf ( r o lle r )1 021 fDddDf B. Resonance demodulation technology We can collect vibration signal using the resonance of the bearing components, and detect the envelope of the fau
16、lt signal using envelop detector, which can fulfill the analysis to the fault character. This is called “resonance demodulation Compound sensor Temperature signal Fault vibration signal DSP digit signal processor Envelop detection Envelop signal Frequency band pass filter Temperature signal processing Fig.2 Signal pretreatment
