Current Issue : July-September Volume : 2022 Issue Number : 3 Articles : 5 Articles
Most electronic and automotive parts are affixed by bolts. To prevent such bolts from loosening through shock and vibration, anti-loosening coating is applied to their threads. However, during the coating process, various defects can occur. Consequently, as the quality of the anti-loosening coating is critical for the fastening force, bolts are inspected optically and manually. It is difficult, however, to accurately screen coating defects owing to their various shapes and sizes. In this study, we applied deep learning to assess the coating quality of bolts with anti-loosening coating. From the various convolutional neural network (CNN) methods, the VGG16 structure was employed. Furthermore, the gradient-weighted class activation mapping visualization method was used to evaluate the training model; this is because a CNN cannot determine the classification criteria or the defect location, owing to its structure. The results confirmed that external factors influence the classification. We, therefore, applied the region of interest method to classify the bolt thread only, and subsequently, retrained the algorithm. Moreover, to reduce the learning time and improve the model performance, transfer learning and fine tuning were employed. The proposed method for screening coating defects was applied to a screening device equipped with an actual conveyor belt, and the Modbus TCP protocol was used to transmit signals between a programmable logic controller and a personal computer. Using the proposed method, we were able to automatically detect coating defects that were missed by optical sorters....
In this study, an online cell screening algorithm is proposed to estimate the maximum peak current considering the cell inconsistencies in battery packs for electric vehicles. Based on the equivalent circuit model, the maximum peak current is mathematically defined, and the inconsistency parameters affecting the maximum peak current are analyzed. The proposed algorithm compares the inconsistency parameters of each cell and subsequently selects a cell or a group of cells whose voltage can exceed the allowable voltage range. The maximum peak current is determined based on the selected cells, while ensuring that all the cells are charged and discharged within the allowable voltage range. The feasibility and superiority of the proposed algorithm are verified through an experiment conducted on a commercially manufactured battery pack for electric vehicles....
With the continuous advancement in bearing speed, bearing assembly clearance (pocket, radial, and guide clearances) has a particularly significant impact on dynamic characteristics of bearing. In order to improve its performance, it is necessary to generate an optimized design for assembly clearance based on dynamic analysis. In this paper, a dynamic model of cylindrical roller bearing has been put forth based on the variable step fourth-order Runge–Kutta method, while the simulation results have been verified by a high-speed bearing cage motion testing machine. Considering pocket, radial, and guide clearances as independent variables and maximum impact force, whirl deviation ratio, and minimum power loss as the objectives, a multiobjective optimized model of cylindrical roller bearing has been developed using central composite experimental design (CCD) and response surface method. After optimization, the maximum impact force was reduced by 9.26%, the whirl deviation ratio was reduced by 2.87%, and power loss was reduced by 1.45%....
In order to help select high-quality electric buses, we established a performance index system for pure electric buses based on an extensible cloud model. With the rapid development of electric buses, choosing a suitable pure electric bus considering its applicability is challenging. Based on the analysis of the characteristics of the passenger car industry, a preliminary evaluation index system for pure electric passenger cars was constructed. The preliminary indicator system was formed based on the optimization of the main points of current laws and regulations, and divided into four aspects: safety assistance system, comfort, convenience, and economy. Then, the index system was determined from multiple perspectives, and the analytic hierarchy process and the entropy weight method were applied to determine the comprehensive weight. Meanwhile, the evaluation level of the index system of pure electric buses was calculated by the extensible cloud model. At last, six electric buses were selected from Chinese electric bus companies as examples to determine the relevant level. The results show that the method has satisfactory feasibility and applicability in the comprehensive evaluation and that it provides a reference for pure electric bus selection based on application performance....
The wheel-side direct-driven transmission system (WDTS) is a new intelligent transmission technology, which has significant advantages in high-efficiency and few malfunctions for the electric bus. Based on the Lagrange–Maxwell equation, the WDTS electromechanical coupling dynamic model, whose effectiveness is verified by the PMSM speed, is constructed for analyzing the system torsional vibration destabilization characteristics. Then, by determining the resonance curve equation for the torsional vibration response amplitude of the WDTS with the direct multi-scale method, the influences of the torque ripple amplitude of the PMSM and the transmission clearance on the system torsional vibration stability are analyzed. The results indicate that the WDTS torsional vibration response shows complex nonlinear characteristics especially under the effect of the system transmission clearance, which has an important impact on the system stable operation. The research results can lay a theoretical foundation for the design of the WDTS of the electric bus....
Loading....