最新刊期
卷 55 ， 期 2 ， 2021
- 摘要：A precise control method for brain-computer cooperation with deep reinforcement learning is proposed to solve the problem that the lack of bidirectional information interaction between human and computer and the change of mental state in precise control seriously affect the precision and safety of limb control. First of all, combining the advantages of human in global planning and machine in fine control, a ‘double-loop’ information interaction mechanism composed of active control loop and passive control loop is established. Secondly, the idea of deep reinforcement learning is introduced, and a mathematical model of brain-computer cooperation is derived based on the Monte Carlo sampling principle, with the electroencephalogram(EEG)representing mental state feature as the input of the model and the robot speed instruction as the output. Thirdly, a mental state perception network with three fully connected layers is established, and the EEG of the last 1 000 ms in the real-time monitoring computer memory of brain-machine interface system is extracted as input signal, then a precise brain-computer cooperation algorithm is designed and developed. Finally, a virtual environment and task scene for trajectory tracking is created, and the precise brain-computer cooperation method is experimentally verified. Results and a comparison with a traditional method show that the proposed method improves both the accuracy and completion time of trajectory tracking control task by 36.55% and 22.81%, respectively.关键词：brain-computer cooperation;deep reinforcement learning;brain-machine interface;trajectory tracking;precise control248|4|3更新时间：2025-07-09
- 摘要：The ocean contains abundant resources, and the world has a major demand for the development of ocean resources. Underwater robots are an important tool to realize the development of marine resources. To solve the disadvantages of the traditional underwater robots, such as high power consumption, high noise, and easy detection of electromechanical characters, inspired by the body shape changes during the stress defense of the puffer, a flexible underwater hovering robot based on liquid-gas phase change driving(hovering robot for short)is designed and manufactured. The base is made of flexible silicone rubber material, and one internal cavity is preset, in which encloses a low-boiling-point driving liquid(3MTM NovecTM 7000, boiling point 35 ℃)is enclosed. Heating the driving liquid causes the liquid-gas phase change, and the saturated vapor pressure acting on the inner cavity wall surface causes volume change of the hovering robot to change its buoyancy. By controlling the driving temperature, the hovering robot dives up and down or stays in fixed depth hovering. Experiments show that the driving temperature is ranged within 50-100 ℃ as the hovering robot contains 15 mL of self-encapsulating driving liquid, and a maximum buoyancy of about 1.95 times than its own weight can be achieved; when the driving temperature is 53 ℃, it can hover at fixed depth; within the driving temperature range, the buoyancy measurement error is less than 5%. The flexible underwater hovering robot driven by liquid-gas phase change avoids the traditional electromechanical drive transmission. It has its own working medium, small size and light weight.关键词：underwater monitoring;hovering robot;liquid-gas phase change drive;fixed-depth hovering154|3|2更新时间：2025-07-09
- 摘要：Oxidation behaviors of four candidate materials for the advanced ultra-supercritical water power plant are studied in supercritical water at 700 ℃/25 MPa for 320 h. These four materials are Inconel 617, Inconel 740, HR6W and Sanicro 25. The 2-D and 3-D morphology, cross-sectional structure of the oxides formed on the alloys after exposure are observed, and the oxide phase are analyzed. The results are compared with those in other literature. Results show that the structure of the oxide films formed on HR6W and Sanicro 25 with high iron content forms a double layer structure with a Fe-rich outer layer and a Cr-rich inner layer after exposed in supercritical water at 700 ℃ for 320 h, while only one Cr-rich spinel structure layer is formed on Inconel 740 and Inconel 617 after exposed to supercritical water. The increase of ambient temperature and ambient pressure promotes the oxidation weight gain of the alloys in supercritical water. The reaction of supercritical water with Fe on the alloy surface to produce hydrogen has an inhibitory effect on the formation of continuous chromium oxide layer on the alloy surface. The order of the weight gain data and the oxide layer thickness of the four candidate materials is Sanicro 25>HR6W>Inconel 740>Inconel 617.关键词：nickel-based alloy;iron-based alloy;boiler;supercritical water;high temperature oxidation113|4|1更新时间：2025-07-09
- 摘要：The applicability of a simplified modeling approach to describe the combustion chemistry of macromolecular hydrocarbon fuels with different functional groups is explored by taking n-octane(n-C8H18), iso-octane(i-C8H18), methyl cyclohexane(CH3cyC6)and n-butyl benzene(A1C4H9)as four macromolecular hydrocarbon fuels. One unified kinetic model for above four fuels is constructed, consisting of 124 chemical species and 854 elementary reactions. The model is verified by using the fundamental combustion data such as distribution of main intermediate products in oxidation and pyrolysis reactions, ignition delay time and laminar flame speed. The carbon distributions of the main intermediate products of four fuels and their influences on the global combustion characteristics are examined. Taking n-octane as an example, the number of chemical species and elemental reactions in the above simplified model is further reduced to 56 and 387, respectively, by the directed relation graph with error propagation. These results show that the above simplified modeling method describes the combustion chemistry of linear and branched alkanes well, and can be applied to that of cycloalkanes and alkyl aromatics; the distinct intermediate products generated from the macromolecular hydrocarbon fuels with different functional groups eventually determine their global combustion characteristics; and this modeling method can be effectively coupled with CFD software or code for the simulation of real engine system after further simplification.关键词：hydrocarbon fuels;simplified kinetic modeling;model reduction;high temperature oxidation chemistry146|4|1更新时间：2025-07-09
- 摘要：The high-speed water jet impact model of 17-4PH(0Cr17Ni4Cu4Nb)is established by using the smoothed particle hydrodynamics(SPH)and finite element method(FEM)coupling algorithm(SPH-FEM)based on the nonlinear explicit software ANSYS LS-DYNA. The grid independence is verified and the impact process between high-speed jet and target is simulated. The influences of jet angle, jet velocity and surface roughness on the water erosion characteristics of the target material are systematically studied. The results show that under the same target surface structure, the greater the initial jet angle of the target surface is, the more serious the water erosion damage of material becomes. Surface roughness has a great influence on water erosion. The average mass loss of the 10 μm groove samples is 1.6 times that of the 2 μm groove samples. And the average mass loss of the 2 μm groove sample is approximately 1.4 times that of a smooth surface sample. The improvement of the surface smoothness can promote the free expansion of the pressure wave of the water droplets and significantly reduce the impact pressure, thereby improve water erosion resistance of the material. The mass loss of the material increases with the increase of jet velocity. There is an exponential relationship between the cumulative water erosion mass loss of the target material and the jet velocity. Fitting results show that the velocity exponent of water erosion is about 3.83.关键词：blade;liquid-solid impact;water erosion;surface roughness;numerical simulation138|4|3更新时间：2025-07-09
- 摘要：To achieve rapid and accurate assembly of aero-engine high-pressure rotor parts, we attempt to predict the assembly eccentricity of the rotor parts via intelligent algorithms and then optimize the phase. The first 30 orders of Fourier series are adopted to simulate the shape error and generate error data. Adding the error data to the finite element model to calculate the assembly eccentricity, BP artificial neural network model is established. The amplitude and phase of the Fourier series are extracted as the input of the neural network and the assembly eccentricity as the network output. The attenuation learning rate, regularization, and moving average algorithm participate in the neural network to calculate the assembly eccentricity more accurately and stably. 200 sets of data are used to complete the neural network training and the trained network verifies three sets of test data. The eccentricity of each phase of different assembly parts is calculated with this neural network. Taking the phase as the objective of particle swarm optimization, the optimized assembly phase of the parts is obtained by error transfer calculation. This approach shows that this neural network model fully considers the morphology of the flange and assembly deformation, and significantly improves the calculation efficiency. Then particle swarm optimization algorithm is used to optimally select different phases to meet the requirements of aero-engine rotor assembly and promote service performance.关键词：aero-engine high-pressure rotor;artificial neural network;eccentric prediction;particle swarm optimization algorithm;phase optimization127|5|6更新时间：2025-07-09
- 摘要：To investigate the characteristics of the fluctuation and diffusion of multi-fluctuation-source and multi-process machining, the features of multi-source and nonlinearity of the quality evolution law are paid attention to. The fluctuation diffusion principle of the machining process is analyzed and a network weighting method is proposed following the complex network modeling theory. Then a weighted fluctuation diffusion network model for multi-fluctuation-source and multi-process machining is constructed. The key nodes of the network are identified with the network characteristic analysis method and the weighted semi-local centrality node importance ranking algorithm. A fluctuation diffusion path search scheme based on the breadth first search(BFS)strategy is proposed to realize the identification of the fluctuation source for the key nodes. Finnally, taking the machining process of steam turbine blades as the research object, the process of blade machining quality fluctuation and diffusion is modeled and analyzed. Through the analysis, the rough milling of blade root inlet steam side plane, re-punching of pinhole, fine milling of blade root profile, and rough milling of blade root are key processing procedures, and the re-punching pinhole procedure is taken as an example to solve the fluctuation diffusion path based on the BFS algorithm. The search results show that the state fluctuation levels of the drilling machine and the drill bit equipment, as well as the machining accuracy of the plane positioning benchmark on the steam outlet side of the blade, will have a direct and significant impact on the processing quality of the process, which is a source of quality fluctuations that need to be monitored. At the same time, other important sources of fluctuations that pass through the fluctuation diffusion path also need to be controlled, which will help further improve the processing quality of the key feature - the pinhole feature. The above analysis and identification of important fluctuation sources and diffusion paths are in line with the actual processing and working conditions, indicating the effectiveness of the fluctuation source identification scheme.关键词：multi-fluctuation-source multi-process machining;fluctuation diffusion;complex network;key nodes;fluctuation source identification96|4|1更新时间：2025-07-09
- 摘要：A novel processing principle of friction stir rivet welding(FSRW)is proposed to solve the defects such as cracks, pores, high forming force requirement and poor air tightness, which are usually observed in the traditional aluminum alloy sheets spot joining technology, and to realize high-performance spot joining for aluminum alloy sheets as a consequence. The feasibility is demonstrated for aluminum alloy sheets spot joining with non-uniform thickness by theoretical analysis and experimental research. The material flow behavior and temperature distribution in FSRW process of aluminum alloy sheets are investigated and the effects of rotational speed and dwelling time on the temperature distribution are discussed emphatically. Then, FSRW experiments are carried out following process parameters optimization by controlling the welding heat input. The consistency between theoretical analysis and experimental research is verified and plastic joining mechanism of this novel process is defined by macro/micro analysis of transverse section of the obtained joint. The results indicate that rotational speed and dwelling time are significant factors affecting mechanical properties of the FSRW joints; the reasonable process parameter ranges of rotational speed(1 000-1 400 r·min-1)and dwelling time(3-15 s)are beneficial for aluminum alloy 6061-T6(AA6061-T6)sheets joining of 3 mm and 4 mm thickness. Metallurgical joining can be observed between upper and lower sheets while mechanical joining between plasticized material and threaded rivet. Therefore, the plastic joining mechanism of FSRW process for aluminum alloy sheets accords with the expectation, and this novel process is effective and feasible.关键词：friction stir rivet welding;temperature distribution;material flow behavior;process parameters optimization;welding heat input200|5|3更新时间：2025-07-09
- 摘要：To optimize the working opening degree range of flue gas damper to improve boiler’s thermal efficiency and the ability to adapt to peak load regulation, the flow fields of a flue gas damper with opening degrees of 10°,20°,30°,40°,50°,60°,70°,80° and 90° are simulated under the condition that the flue gas velocity is 18 m/s. The flow characteristics of the flue gas damper is evaluated according to the flow resistance coefficients, flow coefficients, and the distributions of velocity and pressure of different opening degrees of the flue gas damper. The results show that the uniformity of velocity field and pressure field of downstream flue gas gradually becomes better with the increase of damper opening degree. Along with the increase of damper opening degree, the flow resistance coefficient first decreases rapidly and then decreases slowly and the flow coefficient first increases slowly and then increases rapidly. The resistance of the damper is less than 100 Pa and the flow resistance coefficient of the damper is less than 1.08 when the damper is fully opened. If the total pressure of induced draft fan in the waste heat utilization device is 1 000 Pa, the variation characteristics of flue gas flow is well with the damper opening degree when the damper opens in the range of 17°- 80°. Considering the flow resistance coefficient characteristics of the damper, it is suggested that its opening degree range in common operation is from 40° to 80°.关键词：flue gas damper;numerical simulation;flow characteristics81|5|1更新时间：2025-07-09
- 摘要：To explore the relationship between the vibration mode of a gas insulated switchgear(GIS)basin insulator and the surface contamination, this study detects and analyzes the surface contamination of basin insulators based on the modal analysis method. The effects of pollution mass, area, location and material on the modal characteristics of basin insulators are simulated, and the modal test of basin insulators is carried out on the 220 kV GIS model, and related research on the surface contamination detection technology is carried out. The research results show that the pollution mass is the key factor that affects the modal characteristics of the basin insulator, and the pollution area, location and material have little effect on the modal characteristics. When the pollution mass on the surface of the pot insulator increases, the correlation coefficient will decrease significantly, the pollution accumulation value in the 2 000-3 000 Hz vibration signal will increase significantly,and the pollution index in the signal frequency band 2 000-4 000 Hz will increase obviously. This research may provide a method that can be implemented on-site for the live detection of surface contamination of GIS basin insulators.关键词：basin insulator;surface contamination detection;modal analysis65|4|1更新时间：2025-07-09
- 摘要：An experimental platform of droplet convective combustion is built, and the fuel droplet convective combustion experiment is carried out by using the flying drop method. The flame structure image of droplet combustion and the distance between flame and droplet under different flame stages are obtained. At the same time, the phenomenon of steam explosion is explored, and the generation mechanisms of single steam explosion and continuous steam explosion are clarified. The results show that, with the increase of droplet velocity, the flame structure of droplet combustion changes from full envelope flame to boundary layer flame. After the formation of boundary layer flame, there are two types of transition: boundary layer flame becoming unstably extinct and boundary layer flame converting into cone flame; before the formation of boundary layer flame, the transition points of droplet flame structure are basically the same under all working conditions; and the steam explosion phenomenon is caused by the flame contacting the droplet. After the flame contacts the droplet, the droplet temperature rises suddenly, and a large number of droplets evaporate in a short period of time, thus forming steam explosion combustion. The frequency of continuous steam explosion is equivalent to the frequency of Karman vortex street around the droplet, and the generation of continuous steam explosion is mainly affected by the Karman vortex street effect of droplet flow around the droplet.关键词：fuel droplet;convective combustion;flame;steam explosion66|7|0更新时间：2025-07-09
- 摘要：A cruising unmanned aerial vehicle(UAV)-assisted traffic offloading and secrecy transmission scheme is proposed to solve the problem that the uplink quality of cell-edge users is poor and easy to be eavesdropped in traditional cellular networks. Firstly, according to the line-of-sight propagation characteristics of the air-to-ground channel, a channel access mechanism based on the threshold of the received signal strength is established to solve the problem of user transmission mode selection. Then, the coverage region of UAV’s link at any time is determined by considering important factors such as channel access delay, flight altitude and speed of the UAV. Additionally, an analytic formula of average accessible rate is derived and an appropriate method to evaluate the network secrecy rate is given by using the random geometry theory. Finally, the validity of the proposed scheme and theoretical results is verified by simulation experiments, and the optimal settings of key parameters are presented to achieve the optimum network performances. Simulation results and comparisons with traditional cellular networks show that the proposed mobile UAV-based networking scheme increases the uplink rate by about 40% and the security capacity by about 70% on the basis of ensuring the users fairness.关键词：unmanned aerial vehicle;cellular network;traffic offloading;secrecy transmission98|4|2更新时间：2025-07-09
- 摘要：A classification method of epileptic EEG signals under frequency domain attention mechanism(FDAM)based on a deep learning of residual network structure is proposed to improve the classification accuracy of epileptic electroencephalogram(EEG)signals. Firstly, the algorithm of extracting epileptic EEG signal features is analyzed. Then, according to the characteristics of the amplitude that the signal features are mainly distributed in the time-frequency domain, the amplitude features in the time-frequency domain are extracted twice through the residual network. Finally, in order to focus the features extracted from the residual network on the frequency domain which is more relevant to the classification results, a frequency domain attention mechanism is designed to enhance the amplitude characteristics of the frequency domain in the process of deep learning and effectively improve the classification accuracy of epileptic EEG signals. The classification performance of the proposed algorithm is tested by experiments and the experimental data are obtained from the CHB-MIT Scalp EEG Database in the open PhysioNet database. The experimental results show that FDAM algorithm can classify EEG signals in normal state and epileptic state with 98.05% in accuracy, 99.34% in specificity and 96.12% in sensitivity.关键词：frequency domain attention mechanism;residual network;time-frequency characteristics;epilepsy;electroencephalogram168|4|2更新时间：2025-07-09
- 摘要：A robust power control algorithm with priority of energy efficiency based on cognitive radio is proposed to solve the problem that the performance of traditional power control algorithms based on ideal channel perception is degraded due to channel perception errors in satellite networks. The uncertainty of distance perception in a satellite network is firstly considered. A cognitive power control model is then established by analyzing the distance dependent path loss derived from the related channel gain distribution and shadow effect, so that the power control problem is transformed into an optimization problem with constraint conditions. Probability constraints are introduced and the cognitive user optimal power control is realized by using the fractional programming and Lagrange duality theory. Simulation results show that the algorithm has a faster convergence rate and lower complexity. Compared with the power control algorithm without energy efficiency first, the network energy efficiency can be improved by 50% at most in the case of perception error.关键词：satellite communication;cognitive radio;energy efficiency;robust power control;fractional programming84|4|1更新时间：2025-07-09
- 摘要：A channel estimation scheme constrained by a 3-dimensional(3D)channel model for the massive multi-input and multi-output(MIMO)systems is proposed to improve the estimation accuracy of large scale MIMO channels equipped with one-bit quantized analog-to-digital converters(ADCs). Firstly, a channel estimation optimization problem with a spatial channel model constraint is established for massive MIMO by combining a selected 3D space channel model suitable for the planar array antennas at the base station. Moreover, in order to improve the anti-noise ability of the method, a random white noise term is introduced into the error term of the received signal of the optimization objective function. Secondly, an alternating iterative algorithm is adopted to solve the optimization problem, and an updating process for the channel model parameters is derived in detail. Finally, the amplitude of the low-precision received signal is recovered by using the structure of the channel model and the power of noise, so as to achieve more accurate channel parameter extraction. Simulation results and comparisons with the least squares(LS)estimation and the linear minimum mean square error estimation(LMMSE)based on Bussgang decomposition show that the proposed scheme effectively reduces the influence of the quantization noise and the white noise and improves the accuracy of channel estimation. When the signal to noise ratio(SNR)is 10 dB, the proposed method reduces the mean square error estimation of the channel by 75% compared with the linear minimum mean square error estimation algorithm based on Bussgang decomposition.关键词：massive multi-input and multi-output;one-bit analog-to-digital converter;channel estimation;3-dimensional channel model constraints74|4|0更新时间：2025-07-09
- 摘要：The existing software defined satellite network(SDSN)multi-controller deployment algorithms with poor network reliability usually ignore processing delay to result in load imbalance between controllers. A multi-controller reliable deployment algorithm(MCRDA)is hence proposed in this paper. The multi-controller reliable deployment evaluation indexes, such as control delay, control link reliability and node attractability, are defined based on the satellite network node processing capacity, node failure probability, link failure probability and other parameters. Then the nodes with high node attractability are uniformly selected in the network as the controller deployment locations. The controller deployment locations are once determined, the control delay and control link reliability of each controller-switch combination are evaluated, and the control relationships are constructed for the optimal controller-switch combinations to complete the deployment of multiple controllers. The artificial fish swarm algorithm is used to optimize the multi-controller deployment, and the step size updating function is introduced to improve the convergence speed and calculation accuracy of the artificial fish swarm algorithm. The experimental results and comparison with k-means algorithm, NSGA-Ⅱ algorithm and SoftLEO strategy show that the MCRDA reduces the standard deviation of controller load and control delay by 25% and 17%, and heightens the network reliability by more than 30%.关键词：satellite network;software defined;controller deployment;reliability;artificial fish swarm algorithm99|4|2更新时间：2025-07-09
- 摘要：It is difficult to monitor the concentrations of volatile organic compounds(VOCs)quantitively since the response coefficients of VOCs on the infrared sensor are different due to their molecular structures. To solve this issue, an online testing system based on gas chromatography(GC)and infrared sensor is constructed to determine the concentrations of some specific volatile organic compounds in a container. In the online testing system, the correction coefficients of the infrared sensors towards the specific compounds used in the nuclear containment during maintenance, including ethanol, isopropanol, n-butyl alcohol, toluene, xylol and n-undecane, are calculated based on the data obtained on GC and infrared sensor, and the correction coefficients maintain stable in different concentrations. The correction coefficients can be used to correct the data obtained from the infrared sensors. Six main VOCs in the nuclear containment during maintenance are tested with this system, and the correction coefficients of the six VOCs are obtained based on the data from GC and infrared sensor. The concentration of relative chemical can be obtained through multiplying the infrared sensor datum by the correction coefficient. Furthermore, according to the lower explosive limits of VOCs, the VOCs monitoring method in confined space based on infrared sensor is then established.关键词：volatile organic compounds;concentration monitoring;infrared sensor;nuclear containment77|4|0更新时间：2025-07-09
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0