Vol. 4, No. 2 (2020): Regular Issue

In this paper, the global stability and bifurcations of equilibriums in the (2m+1)th -degree polynomial system are discussed for a better understanding of the complexity of bifurcations and stability of equilibriums in such a (2m+1)th-degree polynomial system. The appearing and switching bifurcations for simple and higher-order equilibriums are presented. The broom-teethcomb-appearing, broom-spraying-appearing, and broom-sprinkler-spraying-appearing bifurcations for simple and higher-order equilibriums are presented. The antenna-switching bifurcations for simple and higher-order equilibriums are discussed, and the parallel straw-bundle-switching and ower-bundle-switching bifurcations for simple and higher-order equilibriums are also presented.

In the modern oil & gas industry ,specifically, in hydrocarbon transportation by pipelines, maintenance decision making for the functional availability of equipment has a major in uence on the profitability of the facilities. In this perspective, this work aims to increase the capacity to monitor the behavior of a TORNADO type gas turbine, installed at an oil pumping station in Hassi R'Mel situated in southern Algeria, with modeling the degradation actions of this machine to make the effective decisions of their maintenance. The maintenance decision support process proposed in this work is based on the concept of fuzzy logic, helps the maintenance operator to choose the most appropriate maintenance actions depending on availability, productivity and quality of the transported oil. This minimizes costs and environmental impact, by ensuring an optimal level of oil transportation, for the monitoring of examined turbine component failures with a sustainable and cost-effective maintenance strategy.

Aiming at the problems that traditional machine learning algorithms such as SVM rely too much on emotional dictionary, grammatical mechanism and low recognition rate in processing text sentiment by natural language, a different ratio CNN-ATTBiLSTM algorithm based on the algorithm is proposed in this paper. First, the CNN algorithm is used to process the local features, and the CNN feature fusion method is used to integrate ATTBiLSTM (two-way long-term and short-term memory network based on attention mechanism) at the output, Then the two algorithms are combined in different ratios to find the highest F1 value as the combination ratio. Finally, the algorithm proposed in this paper is compared with the traditional machine learning algorithms and the proposed algorithms in the experimental data set. The results show that the new algorithm is 0.85% better than the equal weight ratio CNN-ATTBiLSTM algorithm. So it is the best algorithm compared with other algorithms.

This work is motivated by finding dynamic quantities by which some insects are able to sense the direction of vibration sources. Many insects communicate by substrate vibrations alone. We introduce a two-degree-of-freedom dynamic model with time delay in the inputs to the system. Through frequency response functions we obtain surprisingly simple results with regard to the vibrational responses to waves from two opposite directions. The two-degree-of-freedom systems are divided into two types. For those whose rotational resonance frequency is lower than the translational resonance frequency, type I, the vibration on the side toward the wave source is larger than the back side for all wave frequencies. This difference in vibration amplitudes reverses for those systems whose rotational resonance frequency is above the translational resonance frequency { type II. This directional dependence is further demonstrated by numerical simulations of bandlimited random input. Our results provide the basic principle guiding the reconstruction of "Seismoscope" that was recorded to have been invented by Zhang Heng in ancient China before 132AD.

The method of aircraft positioning in the anti-UAV(unmanned aerial vehicle, UAV) system is detecting objects through an image. Its preprocessing process needs to extract the edge of the target aircraft, however, due to the influence of image acquisition technology and electromagnetic environment, the obtained aircraft image is often interfered by a lot of noise. The traditional edge detection algorithm and the method of using local uniform sparsity to enhance the features of the aircraft can not meet the engineering requirement of detecting the edge of the aircraft which has complex structure. Aiming at this problem, an algorithm of edge extraction for aircraft under strong noise is proposed. Firstly, searching high definition images of other aircrafts similar to the structure of the target aircraft in the database, the algorithm is used to learn the texture features of these images. Then, the missing texture features corrupted by strong noise are repaired, this process can achieve the goals of enhancing the target aircraft's detail features and weakening influence of the background's features. Finally, the edge of the target aircraft is successfully detected. The experimental results show that the proposed algorithm can effectively detect the edge of the aircraft under strong noise.

This paper studies quasiperiodic vibration-based energy harvesting in a delayed Rayleigh-Duffing oscillator coupled to a piezoelectric harvester device. Analytical investigation is performed using the multiple scales method to obtain approximation of periodic and quasiperiodic responses as well as the corresponding output power. The influence of different parameters of the harvester on the amplitude of solutions and powers is examined. Results show that for appropriate values of the delay parameters, quasiperiodic vibration-based energy harvesting is significantly improved far from the resonance while the periodic vibration-based one is almost absent near the resonance.