Bus duct fault monitoring method

Abstract

The invention relates to a bus duct fault monitoring method, which comprises the following steps of: selecting three sections of bus ducts which are not adjacent to each other in a bus system, and existence of N sections of bus ducts between a left bus duct and a middle bus duct as well as between the middle bus duct and a right bus duct; a sound sensor and a middle vibration sensor are installed in the middle of the middle bus duct, a left vibration sensor is installed in the middle of the left bus duct, and a right vibration sensor is installed in the middle of the right bus duct. According to the application, the sound signals on the bus duct site and the vibration signals on the bus duct are collected, the abnormal sound signals are screened out through signal comparison, and the similarity and correlation degree of the abnormal sound signals and the vibration signals are analyzed, so that whether the abnormal sound signals and the vibration signals are generated by the same vibration source or not is judged; whether abnormal vibration parts and abnormal sound parts exist on the bus duct can be judged, so that automatic detection of potential faults such as mechanical damage and connection looseness is realized, time and labor are saved, and the automatic monitoring level of the bus duct is improved.

Claims

1. A bus duct fault monitoring method, characterized by: Three non-adjacent bus duct sections were selected in the busbar system: the left bus duct, the middle bus duct, and the right bus duct. N bus duct sections were located between the left and middle bus ducts, and between the middle and right bus ducts. A sound sensor and a middle vibration sensor were installed in the middle of the middle bus duct, a left vibration sensor was installed in the middle of the left bus duct, and a right vibration sensor was installed in the middle of the right bus duct. During the operation of the busbar system, the sound detection signal output by the sound sensor is collected, and at the same time, the vibration detection signals output by three groups of vibration sensors are collected, namely the middle vibration signal, the left vibration signal, and the right vibration signal; the sound detection signals are subjected to noise reduction processing to filter out environmental noise, and then waveform separation is performed to obtain different sound signal waveforms; each sound signal waveform is compared with several preset typical sound signal waveforms and the similarity D between the two is calculated. The similarity D is compared with a preset threshold value D0. If the former is less than the latter, the current sound signal waveform is considered to be an abnormal sound signal waveform; If it is determined that there is an abnormal sound signal waveform, the middle vibration signal, the left vibration signal, and the right vibration signal within the same time period are obtained, and the middle vibration signal, the left vibration signal, and the right vibration signal are respectively subjected to noise reduction and waveform separation to obtain respective middle vibration signal waveforms, respective left vibration signal waveforms, and respective right vibration signal waveforms; Each middle vibration signal waveform, each left vibration signal waveform, and each right vibration signal waveform are respectively matched with the abnormal sound signal waveform for correlation. When the correlation G between a certain vibration signal waveform and the abnormal sound signal waveform is greater than the preset threshold G0, it is considered that the vibration signal waveform and the abnormal sound signal waveform are generated by the same vibration source, and the vibration signal waveform is defined as an abnormal vibration signal waveform. At this time, the potential fault hidden danger of the bus duct system is judged.

2. The bus duct fault monitoring method according to claim 1, characterized in that: The process of correlation matching is as follows: establish a plane rectangular coordinate system, in which the horizontal coordinate of the coordinate system is the sampling time point, and the vertical coordinate of the coordinate system is the fluctuation amplitude; place the abnormal sound signal waveform and each vibration signal waveform in the plane rectangular coordinate system, randomly select a certain number of sampling time points, i.e., control time points, on the horizontal coordinate, obtain the vertical coordinates of the abnormal sound signal waveform and each vibration signal waveform at each control time point and calculate the vertical coordinate difference between the two, and then calculate the average vertical coordinate difference A of each vibration signal waveform and the abnormal sound signal waveform at each control time point; The coordinate points corresponding to the points are extracted, and the coordinate points corresponding to each group of vibration signal waveforms at each sampling time point are extracted respectively. The coordinate points on the abnormal sound signal waveform are sequentially connected with straight line segments to obtain the reference broken line L1; the coordinate points on each vibration signal waveform are sequentially connected with straight line segments to obtain the sample broken line L2, the area covered under the reference broken line L1 is divided into three parts, and the areas S1, S2, and S3 of the three parts are calculated respectively; the area covered under the sample broken line L2 is divided into three parts, and the areas S4, S5, and S6 of the three parts are calculated respectively, and the convergence coefficient B is defined as [(S5 / S4-S2 / S1)+(S3 / S2-S6 / S5)+(S3 / S1-S6 / S4)] / 3; the preset correlation calculation formula G=k1 / A+k2 / B, where k1 and k2 are preset calculation coefficients, and G is the correlation value.

3. The bus duct fault monitoring method according to claim 2, characterized in that: The interval to the left of the left bus duct is defined as interval I, the interval between the left bus duct and the middle bus duct is defined as interval II, the interval between the middle bus duct and the right bus duct is defined as interval III, and the interval to the right of the right bus duct is defined as interval IV; when there is an abnormal vibration signal waveform in the left vibration signal waveform, and there are abnormal vibration signal waveforms in the middle vibration signal waveform and the right vibration signal waveform, the cumulative value Sz of the sample areas of all abnormal vibration signal waveforms in the left vibration signal waveform, the cumulative value Sm of the sample areas of all abnormal vibration signal waveforms in the middle vibration signal waveform, and the cumulative value Sr of the sample areas of all abnormal vibration signal waveforms in the right vibration signal waveform are calculated respectively; the fault source interval is predicted according to the size and difference of Sz, Sm and Sr.

4. The bus duct fault monitoring method according to claim 3, characterized in that: A busbar fault probability prediction model is established based on the type of busbar, number of busbar sections, connector structure, hoisting structure, typical operating conditions, and the operating time of each operating condition. Based on the current operating time of the busbar system, the probability prediction model is used to predict the failure probability of each busbar section. Confidence parameters are assigned to each busbar section based on the change in the difference between Sz, Sm, and Sr. Finally, the confidence parameters are used to correct the failure probability to obtain a corrected value for the failure probability of each busbar section. The fault probability correction values ​​are sorted by size, and the top three bus ducts are marked as "suspected" faults to remind management personnel to conduct investigations.

5. The busbar fault monitoring method according to claim 4, characterized in that: When Sz > Sm > Sr, and (Sz - Sm) / N > (Sm - Sr) / N, (Sz - Sm) / N > (Sz - Sr) / 2N, confidence parameter f1 is assigned to each busbar in interval I, confidence parameter f2 is assigned to each busbar in interval II, confidence parameter f3 is assigned to each busbar in interval III, and confidence parameter f4 is assigned to each busbar in interval IV, where f1 > 1 > f2 > f3 > f4; When Sz < Sm < Sr, and (Sm - Sz) / N < (Sr - Sm) / N, (Sr - Sm) / N > (Sr - Sz) / 2N, confidence parameter f1 is assigned to each busbar in interval IV, confidence parameter f2 is assigned to each busbar in interval III, confidence parameter f3 is assigned to each busbar in interval II, and confidence parameter f4 is assigned to each busbar in interval I, where f1 > 1 > f2 > f3 > f4; When Sm > Sr, Sz > Sr, and |Sm - Sz| / N < (Sm - Sr) / N, confidence parameter f1 is assigned to each busbar in interval II, confidence parameter f2 is assigned to each busbar in interval I, confidence parameter f3 is assigned to each busbar in interval III, and confidence parameter f4 is assigned to each busbar in interval IV, where f1 > 1 > f2 > f3 > f4; When Sm > Sz, Sr > Sz, and |Sm - Sr| / N < (Sm - Sz) / N, confidence parameter f1 is assigned to each busbar in interval III, confidence parameter f2 is assigned to each busbar in interval IV, confidence parameter f3 is assigned to each busbar in interval II, and confidence parameter f4 is assigned to each busbar in interval I, where f1 > 1 > f2 > f3 > f4.

Images