Research On Cooling Oil Channel Detection Technology For Iron Of High-Power Motor Of New Energy Vehicle

Based on the background of the two-carbon goal,the development of energy-saving and environmentally friendly new energy vehicles has become an urgent need for the safety of the energy structure and environmental protection.Good heat dissipation performance of new energy vehicle drive motor is an important guarantee for its efficient operation.However,the cooling oil channel of the iron core of a new type of high-power drive motor may be blocked under the existing production technology,which affects the heat dissipation performance of the motor and has the problem of difficult manual visual inspection and low efficiency.Aiming at this problem,this paper proposes a horizontal non-contact automatic detection method and its system design scheme to improve detection efficiency and reduce labor intensity.In this paper,the cooling oil channel of the motor core is taken as the research and detection object,a set of cooling oil channel penetration detection system based on machine vision is designed,and related image processing algorithms are studied to realize the quantitative judgment of cooling oil channel penetration.The main work of this paper is as follows:(1)According to the production process of iron core and the formation process of cooling oil passage theory,a variety of permeability detection technologies were tested and analyzed to study the permeability detection index of cooling oil channel.According to the test results,combined with the technical requirements of the detection system,the overall structure design and function overview of the detection system.(2)Propose a horizontal non-contact automatic detection method and its system design scheme,and complete the three-dimensional modeling of key components.In view of the characteristics of multiple cooling oil channels and dense distribution,the hardware of the image acquisition system was selected and analyzed to determine the lighting mode.Highquality images containing multiple densely distributed oil channels were collected by largeformat backlight projection in local areas at one time,and then the whole-area oil channel diagram of the whole product was obtained by dynamic rotation control of the iron core.Design and explain the key parts of the detection system,and complete the selection and design of motion control.(3)The cooling oil channel imaging principle and image preprocessing method are studied to improve image edge clarity and contrast.Based on the imaging principle of the backlight projection and the energy gradient function,the evaluation index model of the best imaging parameters of the detection system was explored to obtain the best illumination intensity,light source distance and exposure time of the system,so as to ensure the rapid and high quality image acquisition of the densely distributed cooling oil channel.The results of mean filtering,Gaussian filtering,median filtering and bilateral filtering were analyzed,and the median filtering operator with the best denoising and edge preserving effect was evaluated by means of mean square error(MSE)and peak signal-to-noise ratio(PSNR).In order to solve the problem that the cooling oil channel in one period cannot be collected at the same time due to the uncertain position of the iron core during feeding,the Angle of deviation correction is realized by using the notch feature of the outer edge of the iron core and the shape template matching method,which successfully completes the global image acquisition.The maximum inter-class variance method is used to extract the target region in threshold segmentation.(4)A quantitative judgment model of cooling oil channel blockage was proposed,and the image processing process is simplified by introducing unqualified detection exclusion zone,so as to achieve accurate judgment of cooling oil channel connectivity.The geometric and shape characteristics of the cooling oil channel were analyzed,and the optimal parameters were quantified to determine the distribution law of the cooling oil channel.By comparing the definitions of three models for quantifying the permeability of cooling oil channel,experiments show that a simplified standard model for quantifying the permeability of cooling oil channel proposed by combining morphological processing and image difference method is feasible and reliable,and can realize the evaluation of the permeability of cooling oil channel.(5)Development of online detection human-computer interaction interface based on Visual Studio 2019 and Halcon.Relying on an enterprise,the horizontal non-contact automatic detection equipment proposed in this paper is built,and the image processing algorithm matching the simplified congestion quantization standard model is applied to verify the performance of the detection system scheme and image processing algorithm in practical application.The actual operation results show that the identification accuracy of the detection system can reach 99.7%,and it has the ability of high efficiency detection of the core cooling oil channel.Finally,it is successfully applied to the automatic production line of the electric motor iron core of a new energy vehicle.