Short-Circuit Fault Detection In Laminated Long Stators Of High-Speed Maglev Track

In order to meet the urgent requirements of rapid economic development for high-speed transportation in China,the "National Comprehensive Three-dimensional Transportation Network Planning Outline" puts forward the plan to study and promote the layout of high-speed maglev channels and the construction of test lines between mega cities.The construction of long high-speed maglev trunk lines has been put on the agenda of high-speed transportation development,The importance of maglev track detection and maintenance has gradually become prominent.Different from the traditional wheel rail track,the high-speed maglev track is actually a long stator primary of synchronous linear motor.It is not only the "track" of traditional wheel rail transit vehicles,but also an important electrical component of suspension,traction and operation control system.The high-speed maglev track laid in the open air for a long distance is vulnerable to the impact of vehicle rail collision and environmental corrosion,resulting in the rupture and fall off of the epoxy resin layer on the surface of the stator core,causing in the short circuit fault between the core pieces.If it is not repaired in time,it will cause the magnetic field fluctuation of the traction traveling wave,which will directly affect the safety and stability of vehicle operation.The existing research mostly focuses on vehicle technology,and there is relatively little research on track detection and maintenance,and the existing detection equipment only focuses on the geometric parameters of the track,and the detection of electrical characteristics of long stator is still blank.In order to meet the urgent needs of the development of high-speed maglev transportation for the electrical state detection of maglev track,aiming at the characteristics of long-distance open-air elevated laying of maglev track,few abnormal tracks,uncertain distribution,time-consuming and easy to miss manual inspection,a dynamic,fast and efficient detection method is designed to provide a basis for the daily maintenance and repair of high-speed maglev track.The main research contents are as follows:Aim to clarify the specific impact of short-circuit fault on main leakage magnetic field of the traveling wave,the long stator with local short-circuit fault is analyzed by analytical calculation and finite element simulation.The results show that the anomaly of main leakage magnetic field of the traveling wave caused by short-circuit fault is a typical weak signal,the amplitude increases with the increase of short circuit area and thickness.At the same time,aiming at this typical weak magnetic anomaly signal,the corresponding magnetic anomaly detection system and signal processing method are designed.Aim to realize the detection of weak magnetic anomaly signal under strong background noise and various common mode interference,the magnetic gradient tensor measurement method is used to replace the traditional magnetic scalar and magnetic vector measurement method,so as to reduce the background noise and common mode interference,further enhance the performance of magnetic anomaly and facilitate the extraction of magnetic anomaly signal.Considering the characteristics of small installation space and large magnetic field magnitude in the measurement of main leakage magnetic field of the traveling wave of long stator of high-speed maglev track,the Prewitt operator in the field of image edge detection is instantiated,and a small volume and low-cost magnetic gradient tensor measurement system composed of 8 Hall magnetic sensors is designed.Through the spatial redundancy structure,the shortcomings of zero offset and poor consistency of sensitivity parameters of Hall magnetic sensor are overcome.It also provides a new idea for the spatial layout of sensors in the traditional magnetic gradient tensor measurement system.Aim to realize the rapid detection of a few abnormal targets in a large number of detection data,combined with the sensitivity of fractal analysis to the internal law of data and the sensitivity of wavelet analysis to local mutation,a fractal and wavelet composite magnetic anomaly signal processing method is designed.In the first step,the special similarity of the main leakage magnetic field of the traveling wave is considered.Using the characteristics that the abnormal magnetic field caused by the short circuit between iron cores will destroy this similarity,the multi fractal spectrum feature analysis method is used to quickly screen a large number of data and quickly obtain the data of the suspected fault section.In order to increase the magnetic anomaly detection probability of the primary screening link and ensure the primary screening efficiency,by comparing the time complexity and robustness of different algorithms,the moment method with low time complexity and good robustness is comprehensively selected to calculate the multifractal spectrum of the traveling wave main leakage magnetic field;The objective function is constructed by synthesizing the characteristic parameters of multifractal spectrum.A single objective function value is used to describe the state of a certain track.Combined with the constant false alarm rate detection technology,the amount of data sent to the next link is greatly reduced while ensuring the maximum anomaly detection probability.The second step is the re inspection link.The wavelet modulus maxima corresponding to signal and noise are easy to connect and interrupt respectively,so as to further confirm the suspicious data,and combine the two to realize "fast and accurate" on-board detection.In order to improve the magnetic anomaly detection accuracy in the re inspection link,the performance of different wavelet basis functions is compared,the optimal wavelet basis function is selected,the anomaly detection process based on wavelet modulus maximum line is designed,and the accurate detection of inter core short-circuit fault is completed.Aim to verify the effectiveness of the designed Prewitt Hall magnetic gradient tensor measurement system and the fractal and wavelet composite magnetic anomaly signal processing method,a ground equivalent verification device is designed and the corresponding tests are completed.The experimental results show that compared with the traditional cross and square structure,Prewitt structure can effectively reduce the requirements of the measurement system for the consistency of sensor parameters,avoid the screening of sensors,and doesn’t increase the difficulty of circuit design and welding,which is convenient for implementation;The fractal and wavelet composite magnetic anomaly signal processing method can accurately reflect the local anomaly of magnetic field on the multi fractal spectrum characteristics and the large line of wavelet modulus maximum,so as to realize rapid and accurate detection.In the laboratory environment,under different detection distances of 8 ～ 12 mm,this method can detect all the data containing magnetic anomaly signals.The positioning accuracy of magnetic anomaly is within two cogging cycles,which meets the requirements of short circuit fault detection between cores of high-speed maglev track.