Xuan Li

Mr. Xuan Li

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
11 January 2022 in IET Generation, Transmission & Distribution
Verified
Journal Article
Open access
PDF file
FMFei Mei
Fei Mei
XLXuan Li
Xuan Li
JZJianyong Zheng
Jianyong Zheng
HSHaoyuan Sha
Haoyuan Sha
DLDanqi Li
Danqi Li
The utilization of data‐driven artificial intelligence technology for ultra‐high voltage converter valve state assessment can improve efficiency and accuracy, but there are still problems such as imbalance of raw data and poor model interpretability. This paper aims to propose a new type of ultra‐high voltage converter valve state assessment method to solve these problems. An oversampling method that considers characteristic boundary information is proposed, which can retain and enhance the boundary information of minority samples. The balanced data are imported into categorical boosting to realize offline training and online evaluation of converter valve state assessment. An interpretability analysis framework of the converter valve state assessment model based on the Shapley additive explanations is proposed, which can explain the operation logic of the model and find the key factors that affect the assessment results. The calculation example shows that the evaluation accuracy of this method can reach up to 98.5%, and the running speed is faster than the conventional method. The results can provide guidance for the maintenance and repair of UHV converter valves.

22 October 2020 in Energies
Verified
Journal Article
Open access
PDF file
XLXuan Li
Xuan Li
JZJianyong Zheng
Jianyong Zheng
FMFei Mei
Fei Mei
HSHaoyuan Sha
Haoyuan Sha
DLDanqi Li
Danqi Li
With the continuous improvement of the voltage level of the power system, the electromagnetic interference problem of the converter station has become more and more serious. The thyristor control unit (TCU) is the core equipment of the converter valve, and its normal operation is related to the safe and stable operation of the entire converter valve. This paper starts with the actual electromagnetic environment in the converter valve hall, analyzes the failure principle of the TCU under electromagnetic disturbance, and observes the electromagnetic field distribution and sensitive components on the circuit board. Then, a TCU failure early warning method based on pattern matching and support vector regression (SVR) is proposed. The failure trend is deduced by constructing an abnormal information vector, and then the failure predictor is constructed using support vector regression optimized by grid search (GS), genetic algorithm (GA), and particle swarm optimization (PSO). Considering the failure type and warning time comprehensively, an early warning is issued when the failure mode probability increases to the threshold. When new failure modes appear, the failure mode library will continue to expand. The calculation example shows that this method can effectively warn the TCU failure in the electromagnetic environment, and its prediction accuracy can reach 89.2%, which is better than the traditional failure prediction method.

1 December 2020
Verified
Conference Paper
Closed access
XHXiao Han
Xiao Han
SSShan Song
Shan Song
YWYang Wang
Yang Wang
MWMengfan Wang
Mengfan Wang
XLXuan Li
Xuan Li
JZJianyong Zheng
Jianyong Zheng
With the development of wind-solar-battery power system and the increasingly strict requirements of consumers for power quality, it is of great significance to evaluate the power quality of wind-solar-battery power system both accurately and comprehensively. This paper constructs the evaluation index system of power quality applicable to the wind-solar-battery power system. On this basis, it establishes a comprehensive evaluation model based on the fuzzy analytic hierarchy process (FAHP) and improved grey relational analysis (GRA). The model takes the existing experience into account during the decision-making process, integrating the index value coefficient into the objective correlation degree. While introducing the adjustment strategy of dynamic resolution coefficient, this model makes up for the shortcomings of the traditional grey relational analysis, and realizes the good combination of subjective expert experience and objective node data. Finally, a case study was provided to verify the feasibility and effectiveness of the proposed method.