Jianyong Zheng

Dr. Jianyong Zheng


Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

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

Representative Publications

View All Publications
9 April 2019 in Applied Sciences
Verified
Journal Article
Open access
PDF file
FMFei Mei
Fei Mei
QWQingliang Wu
Qingliang Wu
STShi Tian
Shi Tian
JLJixiang Lu
Jixiang Lu
YPYi Pan
Yi Pan
JZJianyong Zheng
Jianyong Zheng
Recently, a large number of distributed photovoltaic (PV) power generations have been connected to the power grid, which resulted in an increased fluctuation of the net load. Therefore, load forecasting has become more difficult. Considering the characteristics of the net load, an ultrashort-term forecasting model based on phase space reconstruction and deep neural network (DNN) is proposed, which can be divided into two steps. First, the phase space reconstruction of the net load time series data is performed using the C-C method. Second, the reconstructed data is fitted by the DNN to obtain the predicted value of the net load. The performance of this model is verified using real data. The accuracy is high in forecasting the net load under high PV penetration rate and different weather conditions.

15 November 2019 in Energies
Verified
Journal Article
Open access
PDF file
STShi Tian
Shi Tian
FMFei Mei
Fei Mei
JLJixiang Lu
Jixiang Lu
JLJinjun Lu
Jinjun Lu
YPYi Pan
Yi Pan
JZJianyong Zheng
Jianyong Zheng
With the refinement and intelligence of power system optimal dispatching, the widespread adoption of advanced grid applications that consider the safety and economy of power systems, and the massive access of distributed energy resources, the requirement for bus load prediction accuracy is continuously increasing. Aiming at the volatility brought about by the large-scale access of new energy sources, the adaptability to different forecasting horizons and the time series characteristics of the load, this paper proposes a phase space reconstruction (PSR) and deep belief network (DBN)-based very short-term bus load prediction model. Cross-validation is also employed to optimize the structure of the DBN. The proposed PSR-DBN very short-term bus load forecasting model is verified by applying the real measured load data of a substation. The results prove that, when compared to other alternative models, the PSR-DBN model has higher prediction accuracy and better adaptability for different forecasting horizons in the case of high distributed power penetration and large fluctuation of bus load.

15 March 2018 in Sustainability
Verified
Journal Article
Open access
PDF file
FMFei Mei
Fei Mei
YPYi Pan
Yi Pan
KZKedong Zhu
Kedong Zhu
JZJianyong Zheng
Jianyong Zheng
A hybrid photovoltaic (PV) forecasting model is proposed for the ultrashort-term prediction of PV output. The model contains two parts: offline modeling and online forecasting. The offline module uses historical monitoring data to establish a weather type classification model and PV output regression submodels. The online module uses real-time monitoring data for weather type identification on target days and the forecasting of irradiation intensity and temperature time series. The appropriate regression submodel can be selected based on the subsequent results, and the ultrashort-term real-time forecasting of PV output can be performed over a short time scale. The model incorporates power generation and historical meteorological data from the PV station and is suitable for practical engineering applications. In addition to the irradiation intensity and temperature, other factors related to photovoltaic output are evaluated; however, they are excluded from the model for simplicity and efficiency. The performance of the model is verified by practical modeling analysis.

23 March 2019 in Energies
Verified
Journal Article
Open access
PDF file
HSHaoyuan Sha
Haoyuan Sha
FMFei Mei
Fei Mei
CZChenyu Zhang
Chenyu Zhang
YPYi Pan
Yi Pan
JZJianyong Zheng
Jianyong Zheng
Voltage sag is one of the most serious problems in power quality. The occurrence of voltage sag will lead to a huge loss in the social economy and have a serious effect on people’s daily life. The identification of sag types is the basis for solving the problem and ensuring the safe grid operation. Therefore, with the measured data uploaded by the sag monitoring system, this paper proposes a sag type identification algorithm based on K-means-Singular Value Decomposition (K-SVD) and Least Squares Support Vector Machine (LS-SVM). Firstly; each phase of the sag sample RMS data is sparsely coded by the K-SVD algorithm and the sparse coding information of each phase data is used as the feature matrix of the sag sample. Then the LS-SVM classifier is used to identify the sag type. This method not only works without any dependence on the sag data feature extraction by artificial ways, but can also judge the short-circuit fault phase, providing more effective information for the repair of grid faults. Finally, based on a comparison with existing methods, the accuracy advantages of the proposed algorithm with be presented.

24 December 2018 in Energies
Verified
Journal Article
Open access
PDF file
FMFei Mei
Fei Mei
YRYong Ren
Yong Ren
QWQingliang Wu
Qingliang Wu
CZChenyu Zhang
Chenyu Zhang
YPYi Pan
Yi Pan
JZJianyong Zheng
Jianyong Zheng
Voltage sag is a serious power quality phenomenon that threatens industrial manufacturing and residential electricity. A large-scale monitoring system has been established and continually improved to detect and record voltage sag events. However, the inefficient process of data sampling cannot provide valuable information early enough for governance of the system. Therefore, a novel online recognition method for voltage sags is proposed. The main contributions of this paper include: 1) The causes and waveform characters of voltage sags were analyzed; 2) according to the characters of different sag waveforms, 10 voltage sag characteristic parameters were proposed and proven to be effective; 3) a deep belief network (DBN) model was built using these parameters to complete automatic recognition of the sag event types. Experiments were conducted using voltage sag data from one month recorded by the 10 kV monitoring points in Suqian, Jiangsu Province, China. The results showed good performance of the proposed method: Recognition accuracy was 96.92%. The test results from the proposed method were compared to the results from support vector machine (SVM) recognition methods. The proposed method was shown to outperform SVM.

19 March 2019 in Energies
Verified
Journal Article
Open access
PDF file
DLDanqi Li
Danqi Li
FMFei Mei
Fei Mei
CZChenyu Zhang
Chenyu Zhang
HSHAOYUAN SHA
HAOYUAN SHA
JZJianyong Zheng
Jianyong Zheng
A self-supervised voltage sag source identification method based on a convolution neural network is proposed in this study. In addition, a self-supervised CNN (Convolutional Neural Networks) voltage sag source identification model is constructed on the basis of the convolution neural network and AutoEncoder. The convolution layer and pool layer in CNN are used to extract the voltage sag characteristics, and the self-supervised network training process is realized based on the principle of AE. In the constructed mode, features which reflect the data characteristics are used rather than artificial features, thus improving the accuracy of practical application. It is unnecessary to input a lot of correct labels before the self-supervised training process. The model can meet the requirements of sag source identification on timeliness, practicability, diversity, and versatility in the context of modern big data. In this study, three-phase asymmetric sag sources in sag sources are classified into more detailed categories according to different fault phases. Therefore, the proposed method can not only identify the voltage sag source, but also accurately determine the specific fault phase. Finally, the optimal parameters of the model are recognized through a case study, and a self-supervised CNN model is established based on the data type of voltage sag. This model extracts features and identifies sag sources through the measured sag data. The superiority of the proposed method is verified by a comparison.