Likai Liang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Likai Liang has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Yanling Wang
School of Mechanical, Electrical and Information Engineering Shandong University Weihai China
7 shared publications - Yang Mo
Shandong Univ, Weihai, Peoples R China
3 shared publications - Zhiqiang Niu
State Grid Weihai Power Supply Company, Weihai 264200, China
2 shared publications - Chunli Wang
School of Mechanical, Electrical and Information Engineering, Shandong University, Weihai 264209, China
1 shared publication - Dongxing Gao
School of Mechanical, Electrical and Information Engineering Shandong University Weihai China
1 shared publication - Fan Song
School of Mechanical, Electrical and Information Engineering, Shandong University (Weihai), Weihai 264209, China
1 shared publication
Latest Publications
- Open Access
- 10 July 2026
- Published by MDPI AG in Micromachines
- Vol. 17 (7) , 824
DynamiGraph: A Specialized, Runtime-Aware FPGA Overlay for Ultra Low-Latency GNN Inference on Edge Devices
Graph Neural Networks (GNNs) have become essential for analyzing graph-structured data, yet their deployment on resource-constrained edge devices is severely limited by high computational complexity and irregular memory access patterns. Here, we introduce DynamiGraph, a specialized FPGA-based overlay accelerator engineered for ultra-low-latency GNN inference in edge computing scenarios. Unlike general-purpose accelerators that incur high resource overhead to support a broad range of operators, DynamiGraph adopts a streamlined architecture focusing exclusively on essential General Matrix Multiplication (GEMM) and Sparse–Dense Matrix Multiplication (SpDMM) kernels. We implement a hardware-native runtime optimization mechanism that dynamically exploits graph sparsity via an edge-centric execution flow, eliminating redundant computations without requiring complex static preprocessing. Experimental results on an AXU2CGA edge platform demonstrate that DynamiGraph achieves sub-millisecond inference latencies on small-scale benchmarks (e.g., Cora) and a peak throughput of 1467 inferences per second. Furthermore, our runtime sparsity exploitation yields over 2000× reductions in floating-point operations compared to dense equivalents. These findings indicate that trading off model generality for architectural specialization and runtime awareness offers an efficient architectural alternative for enabling real-time graph intelligence in power- and bandwidth-limited edge environments.
- 10 July 2026
- Published by MDPI AG in Micromachines
- Vol. 17 (7) , 824
Open AccessDynamiGraph: A Specialized, Runtime-Aware FPGA Overlay for Ultra Low-Latency GNN Inference on Edge Devices
Graph Neural Networks (GNNs) have become essential for analyzing graph-structured data, yet their deployment on resource-constrained edge devices is severely limited by high computational complexity and irregular memory access patterns. Here, we introduce DynamiGraph, a specialized FPGA-based overlay accelerator engineered for ultra-low-latency GNN inference in edge computing scenarios. Unlike general-purpose accelerators that incur high resource overhead to support a broad range of operators, DynamiGraph adopts a streamlined architecture focusing exclusively on essential General Matrix Multiplication (GEMM) and Sparse–Dense Matrix Multiplication (SpDMM) kernels. We implement a hardware-native runtime optimization mechanism that dynamically exploits graph sparsity via an edge-centric execution flow, eliminating redundant computations without requiring complex static preprocessing. Experimental results on an AXU2CGA edge platform demonstrate that DynamiGraph achieves sub-millisecond inference latencies on small-scale benchmarks (e.g., Cora) and a peak throughput of 1467 inferences per second. Furthermore, our runtime sparsity exploitation yields over 2000× reductions in floating-point operations compared to dense equivalents. These findings indicate that trading off model generality for architectural specialization and runtime awareness offers an efficient architectural alternative for enabling real-time graph intelligence in power- and bandwidth-limited edge environments.
- Open Access
Coplanar Meta-Surface-Based Substrate-Integrated Waveguide Antennas with Broadband and Low Reflections for K-Band Beam Scanning
Four novel substrate-integrated waveguide (SIW) antennas are proposed, in order to obtain K-band beam scanning through the coplanar meta-surfaces of properly devised complementary split-ring resonators. More specifically, coplanar rhombus- and hexagon-shaped meta-surfaces replace the metallized via holes in the traditional SIW structure, achieving low reflection and wide bandwidth, respectively. Another trapezoid-shaped meta-surface is introduced, in order to realize good leaky-wave radiation performance with high-gain beam scanning in both rhombus- and hexagon-shaped SIW components. These designs are further extended to two different mixed types of two-row meta-surfaces, with the rhombus and hexagon structures combined in different orders to enhance the complex SIW transmission lines and antennas, which can simultaneously obtain good reflection and bandwidth with different priority, depending on the arrangement. We explain the performance differences with rhombus and hexagon meta-surfaces through the analysis of relevant equivalent circuit models and extracting the effective medium parameters, and we verify the bandwidths and radiations of four SIW antennas both numerically and experimentally. The maximum gains of the four antennas are 18.1 dBi, 17.0 dBi, 18.8 dBi and 17.1 dBi, where the corresponding relative bandwidths are 10.74%, 19.42%, 14.13% and 18.38%. The maximum simulated radiation efficiency and aperture efficiency of the proposed antennas are 91.20% and 61.12%, respectively. Our approach for generating flexible and selectable tuned electromagnetic fields from SIWs is applicable for the development of mm-Wave antennas or sensors on PCB-integrated platforms for highly directive scanning radiation.
Open AccessCoplanar Meta-Surface-Based Substrate-Integrated Waveguide Antennas with Broadband and Low Reflections for K-Band Beam Scanning
Four novel substrate-integrated waveguide (SIW) antennas are proposed, in order to obtain K-band beam scanning through the coplanar meta-surfaces of properly devised complementary split-ring resonators. More specifically, coplanar rhombus- and hexagon-shaped meta-surfaces replace the metallized via holes in the traditional SIW structure, achieving low reflection and wide bandwidth, respectively. Another trapezoid-shaped meta-surface is introduced, in order to realize good leaky-wave radiation performance with high-gain beam scanning in both rhombus- and hexagon-shaped SIW components. These designs are further extended to two different mixed types of two-row meta-surfaces, with the rhombus and hexagon structures combined in different orders to enhance the complex SIW transmission lines and antennas, which can simultaneously obtain good reflection and bandwidth with different priority, depending on the arrangement. We explain the performance differences with rhombus and hexagon meta-surfaces through the analysis of relevant equivalent circuit models and extracting the effective medium parameters, and we verify the bandwidths and radiations of four SIW antennas both numerically and experimentally. The maximum gains of the four antennas are 18.1 dBi, 17.0 dBi, 18.8 dBi and 17.1 dBi, where the corresponding relative bandwidths are 10.74%, 19.42%, 14.13% and 18.38%. The maximum simulated radiation efficiency and aperture efficiency of the proposed antennas are 91.20% and 61.12%, respectively. Our approach for generating flexible and selectable tuned electromagnetic fields from SIWs is applicable for the development of mm-Wave antennas or sensors on PCB-integrated platforms for highly directive scanning radiation.
- Open Access
- 21 April 2020
- Published by MDPI AG in Applied Sciences
- Vol. 10 (8) , 2879
Mapping the Academic Landscape of the Renewable Energy Field in Electrical and Electronic Disciplines
Research on renewable energy fields of electrical and electronic disciplines is key to promoting the efficient production and utilization of renewable energy, but its branches are numerous and development is uneven. This article examines the topic distribution and future development trajectory of this field, and aims to provide academic researchers with the clearest development context and organizational structure in this field. This study obtained all fields from 3743 articles in the field of renewable energy from the Web of Science (WoS) database, with a time span of 1992–2018. We applied statistical analysis, the latent dirichlet allocation (LDA) topic model, and the autoregressive integrated moving average (ARIMA) model to map topic landscapes in the field of renewable energy. By analyzing these fields, we discovered the digital characteristics of the field and divided the field into 29 different topics, such as “Power conversion technology”, “Micro-grid”, and “Electric vehicles and hybrid electric vehicles”, and analyzed the growth characteristics of topics in two time periods—1992–2005 and 2005–2018. Finally, based on the development trajectory of each topic, we predicted their future development enthusiasm, which was divided into cold, hot, and stable. We compiled statistics on the most popular outlets and citations for each topic, making it easy for researchers and journal editors to appreciate and apply them.
- 21 April 2020
- Published by MDPI AG in Applied Sciences
- Vol. 10 (8) , 2879
Open AccessMapping the Academic Landscape of the Renewable Energy Field in Electrical and Electronic Disciplines
Research on renewable energy fields of electrical and electronic disciplines is key to promoting the efficient production and utilization of renewable energy, but its branches are numerous and development is uneven. This article examines the topic distribution and future development trajectory of this field, and aims to provide academic researchers with the clearest development context and organizational structure in this field. This study obtained all fields from 3743 articles in the field of renewable energy from the Web of Science (WoS) database, with a time span of 1992–2018. We applied statistical analysis, the latent dirichlet allocation (LDA) topic model, and the autoregressive integrated moving average (ARIMA) model to map topic landscapes in the field of renewable energy. By analyzing these fields, we discovered the digital characteristics of the field and divided the field into 29 different topics, such as “Power conversion technology”, “Micro-grid”, and “Electric vehicles and hybrid electric vehicles”, and analyzed the growth characteristics of topics in two time periods—1992–2005 and 2005–2018. Finally, based on the development trajectory of each topic, we predicted their future development enthusiasm, which was divided into cold, hot, and stable. We compiled statistics on the most popular outlets and citations for each topic, making it easy for researchers and journal editors to appreciate and apply them.
- Open Access
Short-Term Wind Power Prediction Using GA-BP Neural Network Based on DBSCAN Algorithm Outlier Identification
Accurately predicting wind power plays a vital part in site selection, large-scale grid connection, and the safe and efficient operation of wind power generation equipment. In the stage of data pre-processing, density-based spatial clustering of applications with noise (DBSCAN) algorithm is used to identify the outliers in the wind power data and the collected wind speed data of a wind power plant in Shandong Province, and the linear regression method is used to correct the outliers to improve the prediction accuracy. Considering the important impact of wind speed on power, the average value, the maximum difference and the average change rate of daily wind speed of each historical day are used as the selection criteria to select similar days by using DBSCAN algorithm and Euclidean distance. The short-term wind power prediction is carried out by using the similar day data pre-processed and unprocessed, respectively, as the input of back propagation neural network optimized by genetic algorithm (GA-BP neural network). Analysis of the results proves the practicability and efficiency of the prediction model and the important role of outlier identification and correction in improving the accuracy of wind power prediction.
Open AccessShort-Term Wind Power Prediction Using GA-BP Neural Network Based on DBSCAN Algorithm Outlier Identification
Accurately predicting wind power plays a vital part in site selection, large-scale grid connection, and the safe and efficient operation of wind power generation equipment. In the stage of data pre-processing, density-based spatial clustering of applications with noise (DBSCAN) algorithm is used to identify the outliers in the wind power data and the collected wind speed data of a wind power plant in Shandong Province, and the linear regression method is used to correct the outliers to improve the prediction accuracy. Considering the important impact of wind speed on power, the average value, the maximum difference and the average change rate of daily wind speed of each historical day are used as the selection criteria to select similar days by using DBSCAN algorithm and Euclidean distance. The short-term wind power prediction is carried out by using the similar day data pre-processed and unprocessed, respectively, as the input of back propagation neural network optimized by genetic algorithm (GA-BP neural network). Analysis of the results proves the practicability and efficiency of the prediction model and the important role of outlier identification and correction in improving the accuracy of wind power prediction.
- Open Access
Power Transmission Congestion Management Based on Quasi-Dynamic Thermal Rating
Transmission congestion not only increases the operation risk, but also reduces the operation efficiency of power systems. Applying a quasi-dynamic thermal rating (QDR) to the transmission congestion alarm system can effectively alleviate transmission congestion. In this paper, according to the heat balance equation under the IEEE standard, a calculation method of QDR is proposed based on the threshold of meteorological parameters under 95% confidence level, which is determined by statistical analysis of seven-year meteorological data in Weihai, China. The QDR of transmission lines is calculated at different time scales. A transmission congestion management model based on QDR is established, and the transmission congestion alarm system including conductor temperature judgment is proposed. The case shows that transmission congestion management based on QDR is feasible, which improves the service life and operation flexibility of the power grid in emergencies and avoids power supply shortages caused by unnecessary trip protection.
Open AccessPower Transmission Congestion Management Based on Quasi-Dynamic Thermal Rating
Transmission congestion not only increases the operation risk, but also reduces the operation efficiency of power systems. Applying a quasi-dynamic thermal rating (QDR) to the transmission congestion alarm system can effectively alleviate transmission congestion. In this paper, according to the heat balance equation under the IEEE standard, a calculation method of QDR is proposed based on the threshold of meteorological parameters under 95% confidence level, which is determined by statistical analysis of seven-year meteorological data in Weihai, China. The QDR of transmission lines is calculated at different time scales. A transmission congestion management model based on QDR is established, and the transmission congestion alarm system including conductor temperature judgment is proposed. The case shows that transmission congestion management based on QDR is feasible, which improves the service life and operation flexibility of the power grid in emergencies and avoids power supply shortages caused by unnecessary trip protection.
- Open Access
- 5 November 2018
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 435 (1) , 012014
The Design of Game Platform Based on Java
In the face of rich game resources, it is necessary to have a good tool to manage games and improve the user's game experience. This is the meaning of the game platform. The game platform described in this paper adopts Java technology, and uses C/S mode and My SQL database to complete the design of the platform. This game platform has the functions of login module, management module and so on. Through the platform, it can realize jigsaw puzzle, Snake game and real-time battle of LAN tank war game. The paper describes in detail the early preparation, overall design, specific implementation and functional testing of the game platform. All functions of the platform can be used normally, the game can also run normally under the rules, which has completed the requirements.
- 5 November 2018
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 435 (1) , 012014
Open AccessThe Design of Game Platform Based on Java
In the face of rich game resources, it is necessary to have a good tool to manage games and improve the user's game experience. This is the meaning of the game platform. The game platform described in this paper adopts Java technology, and uses C/S mode and My SQL database to complete the design of the platform. This game platform has the functions of login module, management module and so on. Through the platform, it can realize jigsaw puzzle, Snake game and real-time battle of LAN tank war game. The paper describes in detail the early preparation, overall design, specific implementation and functional testing of the game platform. All functions of the platform can be used normally, the game can also run normally under the rules, which has completed the requirements.
- Open Access
- 5 November 2018
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 435 (1) , 012017
Design and Implementation of Wind Resources Web Platform
In the context of China's rapid industrialization and rapid growth of energy demand, wind power development has become an inevitable choice for China's sustainable development. Therefore, it is of great practical significance to study the development status of wind power in China. This paper demonstrates a web platform designed by Dreamweaver CS5 software and HTML5. It clearly and intuitively shows the situation of the development and utilization of wind power in China by four main parts: wind power status, wind power distribution, wind power prediction and wind power application, the development and implementation of this platform can provide convenience for wind power information query and research in China.
- 5 November 2018
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 435 (1) , 012017
Open AccessDesign and Implementation of Wind Resources Web Platform
In the context of China's rapid industrialization and rapid growth of energy demand, wind power development has become an inevitable choice for China's sustainable development. Therefore, it is of great practical significance to study the development status of wind power in China. This paper demonstrates a web platform designed by Dreamweaver CS5 software and HTML5. It clearly and intuitively shows the situation of the development and utilization of wind power in China by four main parts: wind power status, wind power distribution, wind power prediction and wind power application, the development and implementation of this platform can provide convenience for wind power information query and research in China.
- 2 April 2018
- Published by Taylor & Francis in IETE Journal of Research
- Vol. 65 (5) , 716-725
Power System Advanced Scheduling Based on Advanced Thermal Rating
The dynamic thermal rating (DTR) can improve the thermal load capacity of transmission lines. The integration of DTR into power system scheduling achieves full utilization of the line load capacity. Thus, this paper proposes the concept of advanced DTR and applies it to the advanced scheduling of power system. Based on advanced thermal rating, combining with conductor temperature, safe running time, and system power flow, the advanced scheduling is realized to achieve the predictive use of line thermal load capacity. In addition, based on the wind power forecast, the advanced scheduling in the wind turbine system is studied. The influence of the randomness of wind power on the advanced scheduling is presented. The results prove that the advanced scheduling based on advanced thermal rating can improve the economy and safety of power grid operation.
- 2 April 2018
- Published by Taylor & Francis in IETE Journal of Research
- Vol. 65 (5) , 716-725
Power System Advanced Scheduling Based on Advanced Thermal Rating
The dynamic thermal rating (DTR) can improve the thermal load capacity of transmission lines. The integration of DTR into power system scheduling achieves full utilization of the line load capacity. Thus, this paper proposes the concept of advanced DTR and applies it to the advanced scheduling of power system. Based on advanced thermal rating, combining with conductor temperature, safe running time, and system power flow, the advanced scheduling is realized to achieve the predictive use of line thermal load capacity. In addition, based on the wind power forecast, the advanced scheduling in the wind turbine system is studied. The influence of the randomness of wind power on the advanced scheduling is presented. The results prove that the advanced scheduling based on advanced thermal rating can improve the economy and safety of power grid operation.
- Open Access
Impact of Conductor Temperature Time–Space Variation on the Power System Operational State
The conductor temperature of an overhead transmission line varies with time and space, which has an important impact on the system operation. In this paper, the conductor temperature is solved iteratively by the CIGRE heat balance equation. The time–space variation of conductor temperature of a 220-kV transmission line is analyzed using real meteorological data from Weihai. Considering the temporal distribution characteristics, the seasonal model of the conductor temperature is given. Considering the spatial distribution, the mean value model, the weight average model, and the segmentation model are established. The system power flow involving the conductor temperature is established based on the relationship between conductor temperature and transmission line parameters. Through the calculation of power flow and the analysis of the maximum power transmission capability, the accuracy of the segmentation model is verified. The results show that the conductor temperature of overhead lines has obvious time–space variation characteristics. It is necessary to consider the time–space variation when analyzing the operation state of power systems.
Open AccessImpact of Conductor Temperature Time–Space Variation on the Power System Operational State
The conductor temperature of an overhead transmission line varies with time and space, which has an important impact on the system operation. In this paper, the conductor temperature is solved iteratively by the CIGRE heat balance equation. The time–space variation of conductor temperature of a 220-kV transmission line is analyzed using real meteorological data from Weihai. Considering the temporal distribution characteristics, the seasonal model of the conductor temperature is given. Considering the spatial distribution, the mean value model, the weight average model, and the segmentation model are established. The system power flow involving the conductor temperature is established based on the relationship between conductor temperature and transmission line parameters. Through the calculation of power flow and the analysis of the maximum power transmission capability, the accuracy of the segmentation model is verified. The results show that the conductor temperature of overhead lines has obvious time–space variation characteristics. It is necessary to consider the time–space variation when analyzing the operation state of power systems.