Yufei Sun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yufei Sun has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Bo Yin
Ocean University of China
1 shared publication Kai Su
1 shared publication- Kexin Li
School of Information Science and Engineering, Ocean University of China, Qingdao, China
1 shared publication Liangli Ma
1 shared publication- Xiaoming Guo
Southeast University
1 shared publication - Zehua Du
Qingdao University of Science and Technology, Qingdao, China
1 shared publication
Latest Publications
- Open Access
- 24 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 1376-1387
A Nonintrusive Load Identification Model Based on Time-Frequency Features Fusion
Nonintrusive load monitoring (NILM) plays a key role in the real-time electricity consumption monitoring of household appliances. However, it is difficult to realize high precision load identification by using a single waveform feature. Therefore, this article proposes a two-stream convolutional neural network based on current time-frequency feature fusion for nonintrusive load identification. First, a time series image coding method for current time-frequency multi-feature fusion is proposed. The method can extract the time domain and frequency domain features of the current timing signal effectively. Then, we present a two-stream neural network combining the gated recurrent unit (GRU) and a two-dimensional convolutional neural network (2D-CNN) to improve the load identification performance. Finally, the experimental results on the PLAID and IDOUC datasets show that the proposed model outperforms the state-of-the-art methods.
- 24 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 1376-1387
Open AccessA Nonintrusive Load Identification Model Based on Time-Frequency Features Fusion
Nonintrusive load monitoring (NILM) plays a key role in the real-time electricity consumption monitoring of household appliances. However, it is difficult to realize high precision load identification by using a single waveform feature. Therefore, this article proposes a two-stream convolutional neural network based on current time-frequency feature fusion for nonintrusive load identification. First, a time series image coding method for current time-frequency multi-feature fusion is proposed. The method can extract the time domain and frequency domain features of the current timing signal effectively. Then, we present a two-stream neural network combining the gated recurrent unit (GRU) and a two-dimensional convolutional neural network (2D-CNN) to improve the load identification performance. Finally, the experimental results on the PLAID and IDOUC datasets show that the proposed model outperforms the state-of-the-art methods.
- 1 March 2014
- Published by International Academy Publishing (IAP) in Journal of Software
- Vol. 9 (3) , 709-715-715
An Efficient Discrete Invasive Weed Optimization Algorithm for Web Services Selection
Efficient algorithm is required to select component services with end-to-end QoS constraints in dynamic composition of Web Services while optimizing the QoS of the composite service. In this paper, the services selection probelm is modeled as a nonlinear optimization with constraints, then a novel discrete invasive weed optimization web services selection algorithm is proposed. The proposed solution consists of two steps:first, a set of randomly generated feasible solutions are transformed into decimal code. Second, We utilize Gaussian diffusion to guide the population to spread in the solution space. The mutation probability and mutation step size of individual is dynamically adjusted by changing the stantard deviation of Gaussian distribution. Accordingly, the population diversity is ensured in the early stage to expand the search space, while the local search nearby excellent individuals is focused in the latter stage to ensure the global convergence. Theoretical analysis and experiment results indicate the efficiency, robustness and feasibility of our approach.
- 1 March 2014
- Published by International Academy Publishing (IAP) in Journal of Software
- Vol. 9 (3) , 709-715-715
An Efficient Discrete Invasive Weed Optimization Algorithm for Web Services Selection
Efficient algorithm is required to select component services with end-to-end QoS constraints in dynamic composition of Web Services while optimizing the QoS of the composite service. In this paper, the services selection probelm is modeled as a nonlinear optimization with constraints, then a novel discrete invasive weed optimization web services selection algorithm is proposed. The proposed solution consists of two steps:first, a set of randomly generated feasible solutions are transformed into decimal code. Second, We utilize Gaussian diffusion to guide the population to spread in the solution space. The mutation probability and mutation step size of individual is dynamically adjusted by changing the stantard deviation of Gaussian distribution. Accordingly, the population diversity is ensured in the early stage to expand the search space, while the local search nearby excellent individuals is focused in the latter stage to ensure the global convergence. Theoretical analysis and experiment results indicate the efficiency, robustness and feasibility of our approach.