Search

• Fertilizer distribution depends on the speed at the outlet of the mixing cavity. • Uneven dispersion of fertilizer granules due to directly falling from blade gaps. • Fertilizer mixed by double-blade has lower outlet speed than that of single-blade. • Mixing effect of multi-fertilizers by air blowing & double-blade stirring is best. In proportional variable-rate fertilization, improving the uniformity of multi-fertilizer mixing helps to improve the overall utilization efficiency of fertilizers. In this paper, a multi-fertilizer mixing by air blowing and blade stirring was presented. Using fertilizer granules mixing uniformity as an indicator, the method optimized the structures and working parameters of single-blade group stirring, double-blade group stirring and air blowing & double-blade stirring. EDEM and Fluent software were used for the simulation and the results were verified by experiments. The simulation results showed that the appropriate speed range of single-blade group stirring was 600 ∼ 900 r⋅min−1 and the highest mixing uniformity was achieved with the blade mounting angle α of 20°. The best structure of double-blade group stirring was achieved with an upper blade mounting angle β of 20°, a lower blade mounting angle θ of −20° and a spacing h of 89 mm. For the air blowing & double-blade stirring method, the best results were achieved when the input air velocity v a was 8.4 m⋅s−1 and the blade speed of agitator n b was 740 r⋅min−1. The mixing uniformity of nitrogen (N), phosphate (P) and potash (K) fertilizer granules reached 98.41%, 98.12% and 98.31%, respectively. The bench test results showed that the mixing uniformity by air blowing & double-blade stirring was better than that by double-blade group, and the mixing uniformity by double-blade group was better than that by single-bladed group, which were consistent with the simulation results. Meanwhile the differences between experimental values and simulation values were within 2%, which showed that the simulation models were reliable. Results from this study can be used as reference for the design of mixing devices of variable-rate fertilization. [ABSTRACT FROM AUTHOR]

• Hindquarter of cow's back in depth image was located. • 3D structure feature map of point clouds was constructed. • Two-step EfficientNet model was built for BCS. • 91.21% images were scored with the error less than 0.25. Body condition score (BCS) is an important indicator to evaluate the nutritional and health status of perinatal dairy cows. The insufficient accuracy of the automatic body condition score is the core bottleneck restricting the popularization and application of this technology. To further improve the accuracy of automatic scoring of cow body condition, a 3D structure feature map based on convex hull distance of point clouds was constructed and used as the input of a two-level model based on EfficientNet network, which realized the improvement of recognition accuracy within the error of 0.25 in the automatic scoring of cow body condition. First, the depth image of the cow's back was preprocessed to obtain the point cloud from the hook bone to the needle bone region containing the main body condition information. Second, the point cloud was voxelized and convex hulled, and the distance between each peripheral voxel and the nearest convex hull surface was calculated and projected to the X-Y plane to obtain the structural feature map. Then, a two-level classification model based on the EfficientNet network was constructed, and the whale optimization algorithm (WOA) was used to optimize the network scaling coefficient in the model. Finally, 5119 depth images of 77 cows were used to train and test the model, and the dataset ratio was 1:1. The results showed that when the BCS value was in the range of 2.25 to 3.75 and the two-class classification limit was 2.75, the proportions of images with recognition errors less than 0.25 and 0.50 were 91.21% and 97.60%, respectively, and the average recognition speed was 3.446 frames/s. The recognition effect was better than that of the MobileNet-V2, XceptionNet and LeNet-5 network models. This method could realize the noncontact assessment of individual body conditions of dairy cows in large-scale farms and had the characteristics of high accuracy, strong applicability and low cost. [ABSTRACT FROM AUTHOR]