Search

With the coverage of sensor-rich smart devices (smartphones, iPads, etc.), combined with the need to collect large amounts of data, mobile crowd sensing (MCS) has gradually attracted the attention of academics in recent years. MCS is a new and promising model for mass perception and computational data collection. The main function is to recruit a large group of participants with mobile devices to perform sensing tasks in a given area. Task assignment is an important research topic in MCS systems, which aims to efficiently assign sensing tasks to recruited workers. Previous studies have focused on greedy or heuristic approaches, whereas the MCS task allocation problem is usually an NP-hard optimisation problem due to various resource and quality constraints, and traditional greedy or heuristic approaches usually suffer from performance loss to some extent. In addition, the platform-centric task allocation model usually considers the interests of the platform and ignores the feelings of other participants, to the detriment of the platform’s development. Therefore, in this paper, deep reinforcement learning methods are used to find more efficient task assignment solutions, and a weighted approach is adopted to optimise multiple objectives. Specifically, we use a double deep Q network (D3QN) based on the dueling architecture to solve the task allocation problem. Since the maximum travel distance of the workers, the reward value, and the random arrival and time sensitivity of the sensing tasks are considered, this is a dynamic task allocation problem under multiple constraints. For dynamic problems, traditional heuristics (eg, pso, genetics) are often difficult to solve from a modeling and practical perspective. Reinforcement learning can obtain sub-optimal or optimal solutions in a limited time by means of sequential decision-making. Finally, we compare the proposed D3QN-based solution with the standard baseline solution, and experiments show that it outperforms the baseline solution in terms of platform profit, task completion rate, etc., the utility and attractiveness of the platform are enhanced.

The RapidIO standard is a packet-switching interconnection technology similar to the Internet Protocol (IP) conceptually. It realizes the high-speed transmission of RapidIO packets at the transport layer, but this greatly increases the probability of network blocking. Therefore, it is of great significance to optimize the RapidIO routing strategy. For this problem, this paper proposes a Double-Antibody Group Multi-Objective Artificial Immune Algorithm (DAG-MOAIA), which improves the local search and global search ability of the population by adaptive crossover and adaptive mutation of the double-antibody groups, and uses co-competition of multi-antibody groups to increase the diversity of population. Through DAG-MOAIA, an optimal transmission path from the source node to multiple destination nodes can be selected to solve the Quality Of Service (QoS) problem during data transmission and ensure the QoS of the RapidIO network. Simulation results show that DAG-MOAIA could obtain high-quality solutions to select better routing transmission paths, and exhibit better comprehensive performance in all simulated test networks, which plays a certain role in solving the problem of the RapidIO routing strategy.

Taking a 1000 MW turbine generator as the research object, the short-circuit fault in electrical disturbance is analyzed. Since it is very difficult to carry out fault analysis experiments and research on actual systems, simulation analysis is one of the more effective means of electrical fault diagnosis; the simulation’s results approach the actual behavior of the system and are ideal tools for power system analysis, and can provide an empirical basis for practical applications. The short-circuit fault model of the SIMULINK power system is built to analyze the two types of faults of generator terminals short-circuit and power grid short-circuit. The impact load spectrum, fault current and speed fluctuation between low-voltage rotors were extracted and analyzed. The conclusion is that the impact value of electromagnetic torque at the generator terminal is greater than that on the power grid side. The impact value of a two-phase short-circuit at the generator terminal is the largest, and that of a three-phase short-circuit on the power grid side is the smallest. The transient impulse current of a three-phase short-circuit at any fault point is greater than that of a two-phase short-circuit; the impulse current of the grid side short-circuit is much greater than that of the generator terminal short-circuit; the speed fluctuation and fluctuation difference caused by the three-phase short-circuit in the grid side are the largest. The alternating frequency of the transient electromagnetic force of the four kinds of faults avoids the natural frequency of the torsional vibration of the shaft system, and the torsional resonance of the shaft system in the time domain of the short-circuit fault will not appear. However, after the fault is removed, the residual small fluctuation torque in the system has a potential impact on the rotor system. This research shows an analysis of the structural integrity and safe operation of turbine generator units after a short-circuit fault, which can not only be applied to engineering practice, but also provide a theoretical basis for subsequent research.

This paper conducts a coupled vibration analysis of a two-stage bladed disk rotor system. According to the finite element method, the bladed disk rotor system is established. The substructure modal synthesis super-element method (SMSM) with a fixed interface and free interface is presented to obtain the vibration behaviors of the rotor system. Then, the free vibration results are compared with the ones calculated by the cyclic symmetry analysis method to validate the analysis in this paper. The results show that the modes of the two-stage bladed disk not only include the modes of the first- and second-stage bladed disk, but also the coupled modes of the two-stage bladed disk.

This paper presents a study on wave propagation in rotating functionally graded (FG) microbeams reinforced by graphene nanoplatelets (GPLs). The graphene nanoplatelets (GPLs) are considered to distribute in the diameter direction of the micro-beam in a gradient pattern, which leads to the functionally graded structure. By using the Halpin-Tsai micromechanics model and the rule of mixture, the effective material properties of the microbeam are determined. According to the Euler-Bernoulli beam theory and nonlocal elasticity theory, the rotating microbeams are modeled. A comprehensive parametric study is conducted to examine the effects of rotating speed, GPL distribution pattern, GPL length-to-thickness ratio, GPL length-to-width ratio, and nonlocal scale on the wavenumber, phase speed and group speed of the microbeam. The research findings can play an important role on the design of rotating graphene nanoplatelet (GPL) reinforced microbeams for better structural performance.

The risk of knee injuries in freestyle skiing athletes that perform aerials is high. The internal stresses in the knee joints of these athletes cannot easily be directly measured. In order to ascertain the mechanical response of knee joints during the landing phase, and to explore the mechanism of damage to the cartilage and ligaments, a finite element model of the knee joint was established. Three successful landing conditions (neutral, backward, or forward landing) from a triple kicker were analyzed. The results demonstrate that the risk of cruciate ligament damage during a neutral landing was lowest. A forward landing carried medium risk, while backward landing was of highest risk. Backward and forward landing carried risk of injury to the anterior cruciate ligament (ACL) and posterior cruciate ligament (PCL), respectively. The magnitude of stress on the meniscus and cartilage varied for all three landing scenarios. Stress was largest during neutral landing and least in backward landing, while forward landing resulted in a medium level of stress. The results also provide the basis for training that is scientifically robust so as to reduce the risk of injury and assist in the development of a professional knee joint protector.

Using consumer depth cameras at close range yields a higher surface resolution of the object, but this makes more serious noises. This form of noise tends to be located at or on the edge of the realistic surface over a large area, which is an obstacle for real-time applications that do not rely on point cloud post-processing. In order to fill this gap, by analyzing the noise region based on position and shape, we proposed a composite filtering system for using consumer depth cameras at close range. The system consists of three main modules that are used to eliminate different types of noise areas. Taking the human hand depth image as an example, the proposed filtering system can eliminate most of the noise areas. All algorithms in the system are not based on window smoothing and are accelerated by the GPU. By using Kinect v2 and SR300, a large number of contrast experiments show that the system can get good results and has extremely high real-time performance, which can be used as a pre-step for real-time human-computer interaction, real-time 3D reconstruction, and further filtering.