Weizhen Dong

Weizhen Dong

Auto-generated profile, based on publicly available information

Weizhen Dong

Department of Electrical and Computer Engineering, The University of Alabama, Tuscaloosa, AL, USA

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
29 January 2021 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
WDWeizhen Dong
Weizhen Dong
Shuhui LiShuhui Li
Shuhui Li
XFXingang Fu
Xingang Fu
ZLZhongwen Li
Zhongwen Li
MFMichael Fairbank
Michael Fairbank
YGYixiang Gao
Yixiang Gao
This paper proposes a novel artificial neural network (ANN) based control method for a dc/dc buck converter. The ANN is trained to implement optimal control based on approximate dynamic programming (ADP). Special characteristics of the proposed ANN control include: 1) The inputs to the ANN contain error signals and integrals of the error signals, enabling the ANN to have PI control ability; 2) The ANN receives voltage feedback signals from the dc/dc converter, making the combined system equivalent to a recurrent neural network; 3) The ANN is trained to minimize a cost function over a long time horizon, making the ANN have a stronger predictive control ability than a conventional predictive controller; 4) The ANN is trained offline, preventing the instability of the network caused by weight adjustments of an on-line training algorithm. The ANN performance is evaluated through simulation and hardware experiments and compared with conventional control methods, which shows that the ANN controller has a strong ability to track rapidly changing reference commands, maintain stable output voltage for a variable load, and manage maximum duty-ratio and current constraints properly.

2 February 2021 in IEEE Journal of Emerging and Selected Topics in Power Electronics
Verified
Journal Article
Closed access
ZLZhongwen Li
Zhongwen Li
ZCZhi ping Cheng
Zhi ping Cheng
Shuhui LiShuhui Li
Shuhui Li
JSJiKai Si
JiKai Si
JGJinfeng Gao
Jinfeng Gao
HDHimadry Shekhar Das
Himadry Shekhar Das
The voltage-source grid-supporting inverters (GSIs) are important to establish and support the frequency and voltage of a microgrid (MG) in both islanded and grid-connected modes. Traditionally, a cascaded control strategy that includes an external droop-based control loop and an inner proportional-integral (PI)-based voltage and current control loop is adopted to control a GSI, which barely shows inertia and limited control performance. In this article, a novel cascaded control strategy is proposed for a GSI. In the external loop, a VSG-based control strategy is adopted to increase the frequency stability of the MG, and a novel synchronization control method that does not use a complex PLL unit is proposed to satisfy the plug-and-play requirements of a GSI. In the inner voltage and current control loop, this article proposes an SMC-based vector control strategy which uses the fixed switching frequency and inherits the robustness and fast response speed of the SMC control. Both simulation and experiments demonstrate the effectiveness and improved performance of the proposed control strategy.

19 May 2022 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
YGYixiang Gao
Yixiang Gao
Shuhui LiShuhui Li
Shuhui Li
Yang XiaoYang Xiao
Yang Xiao
WDWeizhen Dong
Weizhen Dong
MFMichael Fairbank
Michael Fairbank
BLBing Lu
Bing Lu
Buildings account for nearly 40% of primary energy and 36% of greenhouse emissions, which is one of the main factors driving climate change. Reducing energy consumption in buildings toward zero-energy buildings is a vital pillar to ensure that future climate and energy targets are reached. However, due to the high uncertainty of building loads and customer comfort demands, and extremely nonlinear building thermal characteristics, developing an effective zero-energy building energy management (BEM) technology is facing great challenges. This paper proposes a novel learning-based and iterative IoT system to address these challenges to achieve the zero-energy objective in BEM of connected buildings. Firstly, all buildings in the IoT-based BEM system share their operation data with an aggregator. Secondly, the aggregator uses these historical data to train a deep reinforcement learning model based on the Deep Deterministic Policy Gradient method. The learning model generates pre-cooling or pre-heating control actions to achieve zero-energy BEM for building heating ventilation and air conditioning (HVAC) systems. Thirdly, for solving the coupling problem between HVAC systems and building internal heat gain loads, an iterative optimization algorithm is developed to integrate physics-based and learning-based models to minimize the deviation between the on-site solar photovoltaic generated energy and the actual building energy consumption by properly scheduling building loads, electric vehicle charging cycles and the energy-storage system. Lastly, the optimal load operation scheduling is generated by considering customers’ comfort requirements. All connected buildings then operate their loads based on the load operation schedule issued by the aggregator. The proposed learning-based and iterative IoT system is validated via simulation with real-world building data from the Pecan Street project.

10 May 2023 in IEEE Transactions on Transportation Electrification
Verified
Journal Article
Closed access
WDWeizhen Dong
Weizhen Dong
Shuhui LiShuhui Li
Shuhui Li
YGYixiang Gao
Yixiang Gao
BBBharat Balasubramanian
Bharat Balasubramanian
YHYang-Ki Hong
Yang-Ki Hong
BCBing Cheng
Bing Cheng
An interior permanent magnet (IPM) motor is a prime electric motor used in electric vehicles (EVs), robots, and electric drones. In these applications, maximum torque per ampere (MTPA), flux-weakening, and maximum torque per volt (MTPV) techniques play a critical role in the efficient and reliable torque control of an IPM motor. Although several approaches have been proposed and developed for this purpose, each has its specific limitations. The objective of this paper is to develop a neural network (NN) method to determine MTPA, flux-weakening, and MTPV operating points for the most efficient torque control of the motor over its full speed range. The NN is trained offline by using the Levenberg-Marquardt backpropagation algorithm, which avoids the disadvantages associated with online NN training. A cloud computing system is proposed for routine offline NN training, which enables the lifetime adaptivity and learning capabilities of the offline-trained NN and overcomes the computational challenges related to online NN training. In addition, for the proposed NN mechanism, training data are collected and stored in a highly random manner, which makes it much more feasible and efficient to implement lifetime adaptivity than any other methods. The proposed method is evaluated via both simulation and hardware experiments, which shows the great performance of the NN-based MTPA, MTPV, and flux-weakening control for an IPM motor over its full speed range. Overall, the proposed method can achieve a fast and accurate current reference generation with a simple NN structure, for optimal torque control of an IPM motor.

26 October 2020 in IEEE Transactions on Transportation Electrification
Verified
Journal Article
Closed access
YSYang Sun
Yang Sun
Shuhui LiShuhui Li
Shuhui Li
XFXingang Fu
Xingang Fu
WDWeizhen Dong
Weizhen Dong
MRMalek Ramezani
Malek Ramezani
BBBharat Balasubramanian
Bharat Balasubramanian
To improve the efficiency of a permanent magnet (PM) motor, it is usually required to extend the motor operation from linear to over or even six-step modulation regions based upon the space vector pulse width modulation technique. Traditionally, a PM motor is controlled by using the standard vector control technique. However, a recent study shows the decoupling inaccuracy of designing PM motor control using the conventional standard vector control approach. This issue has caused a challenge to control a PM motor, particularly when the motor operates from linear to over-modulation regions. In this paper, a novel approximate dynamic programming (ADP) vector controller is developed to overcome the challenge. The ADP controller is developed using the full dynamic equation of a PM motor and implemented using an artificial neural network (ANN). A feedforward control strategy is integrated with the ANN-based ADP controller to enhance the performance of the controller in both linear and over-modulation regions. The ANN-based ADP control is compared with the conventional standard vector control and a model predictive control. Simulation and hardware experiments demonstrate that the proposed controller can track reference changes better with high efficiency and reliability for PM motor operation in linear and over-modulation regions.

16 August 2024 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
YGYixiang Gao
Yixiang Gao
Shuhui LiShuhui Li
Shuhui Li
WDWeizhen Dong
Weizhen Dong
Yang XiaoYang Xiao
Yang Xiao
BLBing Lu
Bing Lu
This paper proposes an intelligent building energy management (BEM) strategy based on two different time-scale neural networks (NNs) to minimize building electricity expense and maximize building user comfort considering load demand uncertainties and variation in dayahead and real-time electricity prices. First, in terms of the dayahead BEM, an intelligent and optimization problem on an hourly time scale is formulated to generate a day-ahead baseline BEM solution. Then, in terms of the real-time BEM, another intelligent and optimization problem in a 5-minute time scale is proposed to balance the real-time deviations from the energy demand scheduled one day ahead based on the real-time electricity price. In this paper, the NN models are developed to handle the energy management challenge of highly nonlinear building heating, ventilation, and air conditioning (HVAC) systems while physics-based models are developed for building non-HVAC loads. A joint particle swarm optimization and mixed-integer linear programming technique is developed to generate optimal scheduling for building HVAC and non-HVAC loads, in which the thermal coupling issues between HVAC and non-HVAC loads are solved through an iterative approach. Finally, the proposed strategy is evaluated for several different scenarios by considering uncertainties of customers’ comfort requirements and weather prediction. The results demonstrate that the proposed intelligent BEM is economical, efficient, and robust to address the challenge of demand uncertainty and day-ahead and real-time electricity price inconsistency.