The health monitoring and fault diagnosis of heavy-duty engines are increasingly important for energy storage ecosystem. During operation, vibration characters corresponding to the specific fault need to be extracted ...The health monitoring and fault diagnosis of heavy-duty engines are increasingly important for energy storage ecosystem. During operation, vibration characters corresponding to the specific fault need to be extracted from the overall system vibration. Faulty characteristics emanating from one single cylinder are also mixed with those from other cylinders. Besides, the change of working condition brings strong nonlinearities in surface vibration. To solve these problems, an improved deep residual shrinkage network (IDRSN) is developed for detecting diverse engine faults at various degrees using single channel surface vibration signal. Within IDRSN, a wide convolution kernel is utilized in first convolution layer to capture the long-term fault-related impacts and eliminate the short-time random impact. The residual network module is adopted to enhance the focus the relevant components of vibration signals. Mini-batch training strategy is used to improve the model stability. Meanwhile, Gradient-weighted class activation map is adopted to assess the consistency between the learned knowledge and the fault-related information. The IDRSN is implemented to diagnosing a diesel engine under various faults, faulty degrees and operating speeds. Comparisons with existing models are analyzed in terms of hyper-parameters, training samples, noise resistance, and visualization. Results demonstrate the proposed IDRSN's superior performance on fault diagnosis accuracy, stability, anti-noise performance, and anti-interference performance. An average accuracy rate of 98.38 % was achieved by the proposed IDRSN, in comparison to 96.64 % and 93.56 % achieved by the DRSN and the wide-kernel deep convolutional neural network respectively. These results highlight the proposed IDRSN's superiority in diagnosing multiple faults under various working conditions, offering a low-cost, highly effective, and applicable approach for complex fault diagnosis tasks.展开更多
High clearance sprayers are widely used in field operations because of their high ground clearance and good passing performance,which can solve the problem of spraying high-stalk crops in the middle and late stages.In...High clearance sprayers are widely used in field operations because of their high ground clearance and good passing performance,which can solve the problem of spraying high-stalk crops in the middle and late stages.In this paper,an air curtain system was designed to address the phenomenon of droplet drift in the operation of high clearance sprayers.Based on static pressure recovery theory,the design and optimization of the flow velocity at the outlet of the air curtain were carried out.Using SolidWorks software for modeling,ICEM CFD software to divide meshes,and Fluent software to solve the problem,the air duct model was simulated and drift characteristics of droplets were studied through continuous phase and discrete phase coupling calculation.Using three-factor and three-level orthogonal test,the optimal solution of the model was obtained as follows:a spray pressure of 0.4 MPa,a horizontal wind speed of 2 m/s,a fan frequency of 40 Hz,and a droplet drift rate of 9.38%.According to the degree of influence from large to small,the factors are arranged as follows:horizontal wind speed,fan frequency,and spray pressure.An air curtain system test prototype and a droplet drift rate test platform was built,and flow rate of the air duct outlet and the droplet drift rate were tested under multiple working conditions.Experimental results showed that:when the horizontal wind speed was 2 m/s and 4 m/s,the droplet drift rates were the lowest when frequency was 25 Hz and 35 Hz,respectively,which were 13.65%and 23.88%,respectively.When the horizontal wind speed was 6 m/s and 8 m/s,the droplet drift rates reached the lowest when frequency was 45 Hz,which were 27.02%and 29.78%,respectively.When the horizontal wind speed was 2 m/s,4 m/s,6 m/s,and 8 m/s,the droplet drift rates of the optimal auxiliary airflow were reduced by 17.33%,34.51%,50.62%,and 67.54%,respectively.Experiments show that the optimal auxiliary air velocity changes when the horizontal wind speed is different.展开更多
基金funded by the National Key R&D Program of China(Grant No.2021YFD2000303)Tianjin Research Innovation Project for Postgraduate Students in China(Grant No.2021YJSB182)Weichai Power Co.,Ltd.in China(Grant No.WCDL-GH-2023-0147).
文摘The health monitoring and fault diagnosis of heavy-duty engines are increasingly important for energy storage ecosystem. During operation, vibration characters corresponding to the specific fault need to be extracted from the overall system vibration. Faulty characteristics emanating from one single cylinder are also mixed with those from other cylinders. Besides, the change of working condition brings strong nonlinearities in surface vibration. To solve these problems, an improved deep residual shrinkage network (IDRSN) is developed for detecting diverse engine faults at various degrees using single channel surface vibration signal. Within IDRSN, a wide convolution kernel is utilized in first convolution layer to capture the long-term fault-related impacts and eliminate the short-time random impact. The residual network module is adopted to enhance the focus the relevant components of vibration signals. Mini-batch training strategy is used to improve the model stability. Meanwhile, Gradient-weighted class activation map is adopted to assess the consistency between the learned knowledge and the fault-related information. The IDRSN is implemented to diagnosing a diesel engine under various faults, faulty degrees and operating speeds. Comparisons with existing models are analyzed in terms of hyper-parameters, training samples, noise resistance, and visualization. Results demonstrate the proposed IDRSN's superior performance on fault diagnosis accuracy, stability, anti-noise performance, and anti-interference performance. An average accuracy rate of 98.38 % was achieved by the proposed IDRSN, in comparison to 96.64 % and 93.56 % achieved by the DRSN and the wide-kernel deep convolutional neural network respectively. These results highlight the proposed IDRSN's superiority in diagnosing multiple faults under various working conditions, offering a low-cost, highly effective, and applicable approach for complex fault diagnosis tasks.
文摘High clearance sprayers are widely used in field operations because of their high ground clearance and good passing performance,which can solve the problem of spraying high-stalk crops in the middle and late stages.In this paper,an air curtain system was designed to address the phenomenon of droplet drift in the operation of high clearance sprayers.Based on static pressure recovery theory,the design and optimization of the flow velocity at the outlet of the air curtain were carried out.Using SolidWorks software for modeling,ICEM CFD software to divide meshes,and Fluent software to solve the problem,the air duct model was simulated and drift characteristics of droplets were studied through continuous phase and discrete phase coupling calculation.Using three-factor and three-level orthogonal test,the optimal solution of the model was obtained as follows:a spray pressure of 0.4 MPa,a horizontal wind speed of 2 m/s,a fan frequency of 40 Hz,and a droplet drift rate of 9.38%.According to the degree of influence from large to small,the factors are arranged as follows:horizontal wind speed,fan frequency,and spray pressure.An air curtain system test prototype and a droplet drift rate test platform was built,and flow rate of the air duct outlet and the droplet drift rate were tested under multiple working conditions.Experimental results showed that:when the horizontal wind speed was 2 m/s and 4 m/s,the droplet drift rates were the lowest when frequency was 25 Hz and 35 Hz,respectively,which were 13.65%and 23.88%,respectively.When the horizontal wind speed was 6 m/s and 8 m/s,the droplet drift rates reached the lowest when frequency was 45 Hz,which were 27.02%and 29.78%,respectively.When the horizontal wind speed was 2 m/s,4 m/s,6 m/s,and 8 m/s,the droplet drift rates of the optimal auxiliary airflow were reduced by 17.33%,34.51%,50.62%,and 67.54%,respectively.Experiments show that the optimal auxiliary air velocity changes when the horizontal wind speed is different.
