A5100615341- Advanced Algorithms and Applications
- Advanced Sensor and Control Systems
- Hydraulic and Pneumatic Systems
- Industrial Technology and Control Systems
- Advanced Neural Network Applications
- Industrial Vision Systems and Defect Detection
- Energy Efficient Wireless Sensor Networks
- Advanced Control Systems Optimization
- Neural Networks and Applications
- Smart Grid and Power Systems
- Neural Networks Stability and Synchronization
- Adaptive Control of Nonlinear Systems
- Surface Roughness and Optical Measurements
- Fault Detection and Control Systems
- Robotic Mechanisms and Dynamics
- Chromium effects and bioremediation
- Chaos control and synchronization
- Surface Treatment and Coatings
- Power Systems and Technologies
- Stability and Control of Uncertain Systems
- Manufacturing Process and Optimization
- Advanced MIMO Systems Optimization
- Geochemistry and Elemental Analysis
- High-Voltage Power Transmission Systems
- Optical Systems and Laser Technology
Taiyuan University of Science and Technology
2011-2024
Beijing Institute of Technology
2007-2023
State Grid Corporation of China (China)
2018
Taiyuan University of Technology
2007-2016
Photoelectrochemical water splitting has long been considered an ideal approach to producing green hydrogen by utilizing solar energy. However, the limited photocurrents and large overpotentials of anodes seriously impede large-scale application this technology. Here, we use interfacial engineering strategy construct a nanostructural photoelectrochemical catalyst incorporating semiconductor CdS/CdSe-MoS2 NiFe layered double hydroxide for oxygen evolution reaction. Impressively, as-prepared...
Due to the fact that intelligent algorithms such as Particle Swarm Optimization (PSO) and Differential Evolution (DE) are susceptible local optima efficiency of solving an optimal solution is low when trajectory, this paper uses Sequential Quadratic Programming (SQP) algorithm for trajectory planning a hydraulic robotic excavator. To achieve high stationarity during operation excavator, trade-off between time jerk considered. Cubic splines were used interpolate in joint space, time-jerk was...
Abstract Reasonable operation of each excavator joint under path constraints directly affects efficiency and energy consumption. Efficiency consumption are considerably important performance indicators, particularly for large equipment, such as excavators. Therefore, we propose a quadratic trajectory optimization method based on time–energy With the pseudo‐path velocity, acceleration, equivalent convex jerk constraints, interior‐point was used to obtain optimal an excavator. The time values...
摘要： 利用分段的多项式插值方法对挖掘机自主作业时的运动轨迹进行规划，存在优化轨迹不平滑或各关节存在不必要的往复运动等问题，导致挖掘机各关节受到较大冲击，影响挖掘机的使用寿命。针对挖掘轨迹规划存在的问题，提出了基于4-3-3-3-4多项式的关节角度分段插值策略以实现挖掘机最优时间轨迹规划。该策略利用4-3-3-3-4多项式，以角速度和角加速度为约束条件，采用差分进化算法求取插值时间的最优值，得到了各关节最优角度曲线，实现了挖掘机最优时间轨迹规划。对基于4-3-3-3-4多项式分段插值策略进行仿真试验，并在相同条件下与其他多项式插值策略进行了对比。试验结果表明，该多项式插值策略得到的挖掘轨迹更加平滑，各关节能够以较小的角速度和角加速度的变化到达目标点，避免了各关节在运行过程中产生较大冲击，同时节省了工作时间，使得挖掘机自主作业更加平稳高效。
Common non‐destructive material testing technology has some well‐known problems such as slow detection, low detection accuracy, and level of information obtained. To solve these problems, this study applied recent advances in convolution neural networks to propose an effective deep learning network using casting datasets. The approach achieves with automatic, intelligent technology. For most existing networks, image is eventually transformed into a multidimensional visual feature vector for...
We deal with the problem of pinning sampled-data synchronization for a complex network probabilistic time-varying coupling delay. The sampling period considered here is assumed to be less than given bound. Without using Kronecker product, new error system constructed by property random variable and input delay approach. Based on Lyapunov theory, delay-dependent criterion derived in terms linear matrix inequalities (LMIs) that can solved effectively MATLAB LMI toolbox. Numerical examples are...
Summary In industrial production, detecting polarizer defects online and in real time is necessary. Existing methods of based on deep learning can ensure the accuracy classification; however, there are several issues associated with these methods. These include models having low detection speed, consuming large amounts memory, being difficult to be transplanted into systems. To solve aforementioned problems, a lightweight efficient network (LWEN) structure was designed, which improves...
For the traditional fuzzy control to stabilize a MIMO system, number of rules will increase exponentially with increasing input variables. The multistage is introduced deal problem explosion for system. and complexity controller design are induced largely based on structure. And it easy capture human expertise. A designed utilized double inverted pendulum as typical satisfied performance obtained. result real time shows that effective robust
The proper operation of On-Load Tap-Changer (OLTC) in converter transformer is essential for the safety and stability HVDC system. Since a well-defined series vibration signals are produced as its signature, it's extremely desirable to indicate inner connections differences among these each condition. mechanical theories OLTC analyzed first. Based on an improved EMD method aimed at reducing aliasing effect analysis nonstationary nonlinear signals, frequency would be effectively extracted....
Object detection involves solving two main problems: identifying the object and its location. This transforms problem into a classification localization problem. Currently, research shows that convolutional neural network (CNN) can be used to solve problem, module incorporating region proposal method locate object. Although CNN based on proposals achieve high recall, which improves accuracy, performance cannot meet actual requirements for small-size precise localization. is mainly due...
A linear interpolation fuzzy controller with NN compensator is proposed for an electro-hydraulic position servo system. First, to overcome the conventional controller's disadvantage of lower precision caused by discretization, bilinear 2-D table lookup method used replace exact method. Then, minimize steady state error force load, a neural network feed forward proposed. Finally, imitate practical system, co-simulation technique between AMESim and MATLAB/Simulink adopted. The model valve...
In this paper, non-linear predictors based on different artificial neural network (ANN) are compared. General radical basis function (GERBF) network, a modified RBF is introduced. From the comparison with BP, RNN and RBF, it obvious that GERBF has priority in prediction of speech signal. The experiment results show: coding systems ANN have better synthesized than ITU's G721 system higher mean segmental SNR but less computation other systems.
Feedback control systems wherein the loops are closed through a real-time network called networked systems. This type of system has many advantages flexibility with respect to traditional The insertion communication in feedback loop makes analysis and design complex. tools methods developed conventional theory not enough for this analysis. In paper, generalized predictive is used deals time delays loop. stability introduced.
The use of a communication network in control loop has gained increasing attention recent years, because it can offer many advantages. One major challenges this so-called networked systems (NCS) is the network-induced delay effect loop. In paper, strategy based on generalized predictive (GPC) presented, which compensate for delays. A stability analysis method proposed. simulation results illustrate effectiveness method.
Feedback control systems wherein the loops are closed through a real-time network called networked (NCS). The insertion of communication in feedback loop makes design controller NCS complex. In this paper, fuzzy method for system is introduced. order to evaluate feasibility, double inverted pendulum implemented on campus network. experimental result shows that an effective due its robustness against parameter uncertainty.