Hua Deng

Dr. Hua Deng

orcid

Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 April 2023 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
Jun DaiJun Dai
Jun Dai
YZYi Zhang
Yi Zhang
HDHua Deng
Hua Deng
To avoid the cumbersome calculation of inverse kinematics and improve the efficiency of obstacle avoidance, a novel potential guided bidirectional rapidly-exploring random tree star with the direct connection strategy for redundant robot manipulators in the joint space is proposed. First, an expansion strategy based on the artificial potential field is designed in the joint space. Then, this expansion strategy is combined with the goal-biased bidirectional rapidly-exploring random tree star (GB-RRT * ) to improve the ability of obstacle avoidance. Second, a direct connection strategy is designed to improve expansion efficiency. Finally, the effectiveness and superiority of the proposed algorithm are verified by simulations and experiments. The results show that, compared with bidirectional RRT and GB-RRT * , the proposed algorithm can plan a shorter path with a wider clearance between the redundant robot manipulator and obstacles, generate fewer invalid nodes that collide with obstacles, and spend less time. Therefore, it can avoid obstacles more effectively and obtain an optimal path.

15 July 2008 in IEEE Transactions on Neural Networks
Verified
Journal Article
Closed access
HDHua Deng
Hua Deng
HLHan-Xiong Li
Han-Xiong Li
YWYi-Hu Wu
Yi-Hu Wu
A new feedback-linearization-based neural network (NN) adaptive control is proposed for unknown nonaffine nonlinear discrete-time systems. An equivalent model in afflne-like form is first derived for the original nonaffine discrete-time systems as feedback linearization methods cannot be implemented for such systems. Then, feedback linearization adaptive control is implemented based on the affine-like equivalent model identified with neural networks. Pretraining is not required and the weights of the neural networks used in adaptive control are directly updated online based on the input-output measurement. The dead-zone technique is used to remove the requirement of persistence excitation during the adaptation. With the proposed neural network adaptive control, stability and performance of the closed-loop system are rigorously established. Illustrated examples are provided to validate the theoretical findings.

9 November 2017 in IEEE/ASME Transactions on Mechatronics
Verified
Journal Article
Closed access
HZHaibo Zhou
Haibo Zhou
HDHua Deng
Hua Deng
JDJi-an Duan
Ji-an Duan
This paper presents a hybrid fuzzy decoupling control strategy and its implementation for leveling and positioning a maglev wafer carrier in an IC packaging application. The maglev platform possesses micron scale positioning and can carry the wafers with precision motion between workstations. Because the maglev system is open-loop unstable, there are strong nonlinear interactions between electromagnets, and the exact nonlinear system model is unknown, it is hard to achieve a good decoupling control performance by using conventional control strategy. Therefore, in this paper, a hybrid fuzzy decoupling control with good decoupling performance is proposed. The decoupling control uses a master-slave structure, which consists of a linear master decoupling control term and a nonlinear slave intelligent compensation term. The master decoupling term decouples the main interactions of the coupled maglev system. Using the decoupled main system as reference, the slave intelligent term compensates the nonlinear interactions. Furthermore, for this control strategy, establishing the exact linear or nonlinear model is not necessary, only an approximated linear model is needed. The experimental results show that the proposed control strategy decouples most of the interactions between the electromagnetic actuators. Experimental comparisons also highlight that the master-slave decoupling control has shorter settling time and smaller steady-state error than a conventional adaptive fuzzy controller and Proportional-Integral-Differential controller.

23 February 2021 in IEEE Transactions on Transportation Electrification
Verified
Journal Article
Closed access
YZYu Zhou
Yu Zhou
HDHua Deng
Hua Deng
HLHan-Xiong Li
Han-Xiong Li
SXShengli Xie
Shengli Xie
The monitoring of temperature distribution is critical to the safety performance and cycle life of lithium-ion batteries (LIBs). This article introduces a systematic solution for real-time modeling of the battery thermal process under nonhomogeneous boundary conditions. The proposed method integrates the nonhomogeneity separation and the time-space (T/S) separation. First, the nonhomogeneity separation is designed to homogenize the boundary conditions. Then, the spatiotemporal dynamics can be decomposed by the T/S separation. With the T/S separation, a set of spatial basis functions (SBFs) are designed to deal with the spatial complexity, and then, a corresponding temporal model is constructed to capture the temporal nonlinearity. According to the Rademacher complexity, the generalization bound of the developed model is deduced, thereby ensuring the modeling performance. Experimental studies demonstrate the validity of the proposed modeling method.

22 January 2020 in IEEE Access
Verified
Journal Article
Open access
RHRong Hu
Rong Hu
HDHua Deng
Hua Deng
YZYi Zhang
Yi Zhang
A novel dynamic-sliding-mode-manifold-based continuous fractional-order nonsingular terminal sliding mode control is proposed for a class of second-order nonlinear systems. By designing the parameter in the continuous fractional-order nonsingular terminal sliding mode manifold as an exponential function of the tracking error, a dynamic sliding mode manifold can be obtained by adjusting the parameter online. Even if reference signals change, the parameter does not need repetitive offline optimization. By combining the fast-terminal-sliding-mode-type reaching law, the system states are attracted to the manifold quickly, enhancing the controller's robustness. When a large initial error exists, the control system can still accelerate response and reduce overshoot simultaneously owing to the dynamic changing characteristic of the manifold. The stability and finite-time convergence of the closed-loop system are proven by the Lyapunov stability theory. Simulation results on SISO and MIMO nonlinear systems show that for different reference signals, the proposed method has a better tracking performance than the general fractional-order nonsingular terminal sliding mode control.

1 January 2017 in Control Engineering Practice
Verified
Journal Article
Closed access
WDWen-Hua Ding
Wen-Hua Ding
HDHua Deng
Hua Deng
YXYi-Min Xia
Yi-Min Xia
XDXiao-Gang Duan
Xiao-Gang Duan