| Motor and motor system are the power source and core components of high-end equipment such as industrial drive,transportation,high-end CNC machine tools,aerospace and national defense industry,which is an important embodiment of national core competitiveness.As a core component of the motor,the motor winding is prone to surface cracks,insulation damage,insulation thinning and other defects in the manufacturing,assembling and service stages under the combined action of production process,motor overload and various environmental factors,which lead to inter turn short circuit or broken rotor bars.The existence of defects will greatly reduce the efficiency,reliability and life of the motor,and even affect the stability of the whole power supply system.Therefore,it is important to carry out defect detection,failure analysis and fault diagnosis of motor windings.The traditional detection methods have the disadvantages of intrusion,low resolution,invisible,and slow.It is urgent to develop a non-invasive and accurate visual detection technology for motor winding defects.Eddy current thermography is a new nondestructive testing technology,which has the characteristics of non-contact,high resolution,high sensitivity,large area,rapid detection,and superior in detecting micro cracks.Based on the eddy current pulse thermography technology(ECPT),this paper studies the nondestructive testing and quantitative evaluation methods of electromagnetic and thermal imaging of multi-physical fields of various motor winding defects,aiming to provide theoretical basis and technical guidance for the high efficiency factory inspection and health status evaluation of motor winding.The main contents and innovation points are as follows:(1)Due to the lack of consideration of influence on inspection effect caused by defect size change of motor winding with traditional methods and limitations of contact inspection,the coupling mechanism and numerical simulation method of multiphysical fields are studied.The model of electric-magnetic-thermal coupling and defect analysis are established to simulate the heat generation process of defects under eddy current excitation.The distribution of electromagnetic and temperature fields of different types of defects is given,and the influence mechanism of defect size change(depth,width,insulation film thickness)on temperature distribution and detection results is revealed.Through the analysis of three-dimensional dynamic scanning simulation model,the influence mechanism of defect width change on thermal response of defect and non-defect is discussed.The influence of scanning position and heating time on defect temperature distribution is explored.It provides a theoretical foundation for the construction of ECPT detection system and results analysis.(2)To fill the technical gap of non-contact and visual detection of motor winding defects,a new detection method based on ECPT is proposed.A digital electromagnetic induction thermography defect detection system is established.Visual inspections of copper bar surface cracks defects,insulation breakage,complex defect of insulation thinning and copper surface crack are realized by the quantitative inspection.The thermal image sequence of motor winding is obtained and the temperature field information of defects is collected.The thermal response characteristics is extracted and analyzed by transient temperature response method.Fast Fourier transform(FFT),principal component analysis(PCA),independent component analysis(ICA)and nonnegative matrix factorization(NMF),are applied to enhance the contrast and the visual detection of composite defects in motor windings is realized.It provides theoretical support and technical guidance for delivery test and state evaluation of industrial motor windings.(3)To solve the problems of low efficiency and manual dependence of industrial motor winding testing,a joint scanning electromagnetic thermographic automatic inspection method is proposed.The theoretical derivation is carried out and the scanning speed model is obtained.A joint scanning electromagnetic thermographic detection system is designed,which simulates the factory production and inspection line.A new 3D data reconstruction algorithm is proposed to achieve accurate speed estimation and sub-pixel spatial-temporal alignment of the image sequence.Various feature enhancement algorithms are applied to enhance the thermal contrast and eliminate the background noise.Otsu segmentation and image morphology are used to extract the response intensity of different algorithms.The optimal experimental parameters are obtained by evaluating the maximum detection number.A quantitative evaluation algorithm is proposed to assess the detection capability of different types,sizes,and positions under the optimal parameters.(4)To solve the problem of possible damage caused by long time and high-density excitation and further improve the detection resolution,an excitation optimization design and defect quantitative evaluation method based on coded excitation and pulse compression technology is proposed.By establishing a 3D simulation model based on eddy current pulsed compression(ECPu CT)technology,the thermal wave signal of the motor winding surface is extracted,and the frequency domain cross-correlation algorithm is used for pulse compression processing.The values of correlation delay,correlation phase,correlation peak,cross-correlation peak and their corresponding delay time are extracted,and the information extraction ability of each characteristic value is compared and analyzed,and the quantitative relationship between features and defect size is established.The defect detection system of ECPu CT excited by Barker code is built.By analyzing the characteristics of crossing-point,skewness and kurtosis of pulse compression response in defect area and non-defect area,the influence of defect size change on the characteristic quantity of pulse response is discussed.By comparing with ECPT technology,it is proved that the proposed excitation source optimization method based on ECPu CT can improve the signal-to-noise ratio and increase the detection depth of the system while using low-power heat sources. |
PDF Full Text Request