Lei XI

Dr. Lei XI

College of Electrical Engineering & New Energy, China Three Gorges University, Yichang 443002, China
orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
9 December 2019 in IEEE Transactions on Sustainable Energy
Verified
Journal Article
Closed access
Lei XILei XI
Lei XI
LYLu Yu
Lu Yu
YXYanchun Xu
Yanchun Xu
SWShouxiang Wang
Shouxiang Wang
XCXi Chen
Xi Chen
The widely adoption of distributed renewable energy sources (DREs) effectively reduces carbon emission and beat atmospheric haze in developing countries. However, random disturbance issues emerge in power grids with DREs when applying traditional centralized automatic generation control (AGC) strategies. Therefore, a multi-agent distributed control strategy is proposed for AGC in this work, which is mainly based on the concept of deep reinforcement learning, and developed by the strategy of action discovery. Moreover, area control error and the amount of carbon emission are employed in reward functions to obtain optimal solutions in the implementing process of the proposed strategy. Simulations are provided in the work to show the effectiveness of the strategy, while comparisons are also offered, where the simulating results obtained by two other intelligent AGC algorithms are used as references, according to which, the superiority of the proposed strategy is confirmed.

14 July 2020 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
Lei XILei XI
Lei XI
JWJunnan Wu
Junnan Wu
YXYanchun Xu
Yanchun Xu
HSHongbin Sun
Hongbin Sun
As the conventional automatic generation control (AGC) is inadequate to deal with the strong random disturbance issues induced by the ever-increasing penetration of renewable energy to the power grids, this article proposes a deep-reinforcement-learning-based three-network double-delay actor-critic (TDAC) control strategy for AGC to handle the above problem, which is mainly developed by multiple neural networks to fit the system action strategies and evaluate the value. The proposed strategy can increase the exploration efficiency and the quality of AGC and improve the system control performance using the modified actor-critic (AC) method with incentive heuristic mechanism, while a novel iterative way of the value function is also used to reduce the bias of optimization effectively for achieving optimal coordinated control of the power grid. The simulations are provided in the work to show the control performance of the strategy. Compared with other smart methods, the simulation study demonstrates that TDAC has excellent exploratory stability and learning ability. Meanwhile, it also can improve the dynamic performance of the power system and achieve the regional optimal coordinated control.

1 June 2018 in Energy
Verified
Journal Article
Closed access
Lei XILei XI
Lei XI
JCJianfeng Chen
Jianfeng Chen
YHYuehua Huang
Yuehua Huang
YXYanchun Xu
Yanchun Xu
LLLang Liu
Lang Liu
YLYudan Li
Yudan Li

24 December 2020 in IEEE Transactions on Sustainable Energy
Verified
Journal Article
Closed access
Lei XILei XI
Lei XI
ZLZhou LIpeng
Zhou LIpeng
YXYanchun Xu
Yanchun Xu
XCXi Chen
Xi Chen
As the penetration rate of new energies, energy storage devices and electric vehicles increases continuously in power grid, power grid faces strong random disturbances, as well as the issues of downfall in frequency control ability induced by the factors such as system inertia reduction and frequency control under-capacity in traditional power units. Therefore, an algorithm of automatic generation control with DQ(σ, δ) is proposed for the multi-area interconnected power grid. By integrating the eligibility trace, DQ(σ, δ) adopts the principle of double Q learning to unify double Q(δ) and double Expected-Sarsa(δ), thereby the over-estimation of Q in the algorithm and the phenomenon of over-fitting can be avoided. The simulations are provided to demonstrate the effectiveness of the proposed algorithm, where the improved IEEE standard two-area load-frequency control model and the model of the multi-area interconnected power grid based on Central China Power Grid are adopted. The results show that the phenomena of random disturbances and frequency instability can be eliminated in multi-area interconnected power grid. The convergence is faster and the dynamic performance is better in the proposed algorithm compared with the traditional algorithms.

27 September 2022 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
LXLihui Xie
Lihui Xie
JWJunnan Wu
Junnan Wu
YLYanying Li
Yanying Li
Qiuye SunQiuye Sun
Qiuye Sun
Lei XILei XI
Lei XI
The integrated energy system based on ubiquitous power Internet of Things (IoT) has the characteristics of ubiquitous connection of everything, complex energy conversion mode, and unbalanced supply-demand relationship. It brings strong random disturbance to the power grid, which deteriorates the comprehensive control performance of automatic generation control. Therefore, a novel deep reinforcement learning algorithm, namely, collaborative learning actor–critic strategy, is proposed. It is oriented to different exploration horizons, has the advantage on experience sharing mechanism and can continuously coordinate the key behavioral strategies. Simulation tests are performed on the two-area integrated energy system and the four-area integrated energy system based on ubiquitous power IoT. Comparative analyses show that the proposed algorithm can efficiently solve the problem of strong random disturbance, and has better convergence characteristic and generalization performance. Besides, it can realize the optimal cooperative control of multiarea integrated energy system efficiently.

5 August 2022 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
Lei XILei XI
Lei XI
HLHaokai Li
Haokai Li
Jizhong ZhuJizhong Zhu
Jizhong Zhu
YLYanying Li
Yanying Li
SWShouxiang Wang
Shouxiang Wang
In order to solve the problem of frequency instability of power system due to strong random disturbance caused by large-scale electric vehicles and wind power grid connection, an improved reinforcement learning algorithm, namely, optimistic initialized double Q, is proposed in this article from the perspective of automatic generation control. The proposed algorithm uses the optimistic initialization principle to expand the agent action exploration space, so as to prevent Q-learning from falling into local optimum by greedy strategy; meanwhile, it integrates double Q-learning to solve the problem of overestimation of action value in traditional reinforcement learning based on Q-learning. In the algorithm, the hyperparameter $\alpha_{\tau}$ is introduced to improve the learning efficiency, and the reward $b_{\tau}$ based on exploration times is introduced to increase the $Q$ value estimation to drive the exploration of the algorithm, so as to obtain the optimal solution. By simulating the two-area load frequency control model integrated with large-scale electric vehicles and the four-area interconnected power grid model integrated with large-scale wind power generation, it is verified that the proposed algorithm can obtain the global optimal solution, thus effectively solvinng the frequency instability caused by strong random disturbance in the grid-connected mode of large-scale wind power generation, and compared with many reinforcement learning algorithms, the proposed algorithm has better control performance.