Piezoresistive acceleration sensors are widely used in various fields of the industrial Internet of Things because of their lightweight, fast response, and small size. The structural sensitivity of the sensor affects the accuracy of the measurement. And the sensitivity that the traditional method designs are only a feasible solution, not an optimal solution. Due to the differences in factory processes, the optimization of structural sensitivity is an NP-hard problem. To solve the design problem of structural sensitivity, we adopt the swarm intelligence algorithm in this paper, and we design a model for the structural sensitivity of the piezoresistive acceleration sensor. In addition, an improved grasshopper optimization algorithm (CC-GOA) that combines chaos strategy and Cauchy mutation is proposed to optimize the structural sensitivity of the piezoresistive acceleration sensor, and the structure of the sensor is composed of four beams and mass block. The experiments are compared with six well-known algorithms on 16 benchmark functions to verify the algorithm performance of CC-GOA, and then, the structural sensitivity of the piezoresistive acceleration sensor is optimized by CC-GOA. The results indicate that the piezoresistive acceleration sensor is designed with high sensitivity and superiority.
Aiming to solve the problems of poor performance and low stability in the automatic clip-feeding of a grafting machine, an automatic clip-feeding mechanism with a precise single-clip discharge mechanism was designed, and a clip-feeding performance test was carried out. Taking the grafting clip of the 2TJGQ-800 type of vegetable-grafting robot as the research object, the clamping-force analysis model of the grafting clip was constructed by ABUQUS finite-element analysis software, and the variation law of clamping force, steel wire diameter, and opening deformation, as well as the calculation equation of clamping force, were obtained. The grafting clip model was verified by mechanical test, and test results showed that the grafting clip with a steel wire diameter of 0.7 mm proved safe and reliable for grafted cucumber and watermelon seedlings; the grafting clip with steel wire diameter of 0.8 mm had a risk of producing injury to grafted cucumber and watermelon seedlings when clamping. The method of single-clip discharge in the inclined discharging slideway was put forward, and the components for clip discharge and clip pushing were designed. The critical thrust for sending out the grafting clip in the clip-feeding slideway was 0.603 N after analyzing the force status of the grafting clip in the clip-feeding slideway. Test results showed that the success rate of automatic clip-feeding reached 98.67% when inclination angle of row-discharging slideway was 50° and the thrust of clip-pushing cylinder (input air pressure of 0.4 MPa) was 8.04 N, which met the technical requirements of mechanical grafting. The inclination of the grafting clip and the damaged clip in the feeding slideway is the main reason for the failure of clip-feeding. The research results can provide theoretical and design references for the innovative research of the automatic clip-feeding mechanism of grafting robots.
In recent years, with the increase of data scale, multi-label learning with large scale class labels has turned out to be the research hotspots. Due to the huge solution space, the problem becomes more complex. Therefore, we propose a multi-label algorithm based on kernel learning machine in this paper. Besides, the Cholesky matrix decomposition inverse method is adopted to calculate the network output weight of the kernel extreme learning machine. In particular, in terms of large matrix inverse problem, the large matrix is divided into small matrices for parallel computation through using matrix block method. Compared with several state-of -the-art algorithms on several benchmark data sets, results of the experiments show that the proposed algorithm makes a better performance with large scale class labels.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.