A5021516433- Metaheuristic Optimization Algorithms Research
- Artificial Immune Systems Applications
- Smart Materials for Construction
- Advanced Sensor and Control Systems
- Thermal Analysis in Power Transmission
- Immune Cell Function and Interaction
- Immunotherapy and Immune Responses
- Icing and De-icing Technologies
- Water Quality Monitoring Technologies
- Air Quality Monitoring and Forecasting
- Advanced Algorithms and Applications
- Water Quality Monitoring and Analysis
China University of Geosciences
2021
Zhengzhou University of Light Industry
2016-2018
Xiamen University of Technology
2018
Institute of Software
2018
Artificial immune system is one of the most recently introduced intelligence methods which was inspired by biological system. Most algorithms are based on clonal selection principle, known as (CSAs). When coping with complex optimization problems characteristics multimodality, high dimension, rotation, and composition, traditional CSAs often suffer from premature convergence unsatisfied accuracy. To address these concerning issues, a recombination operator combinatorial proposed at first....
The traditional BP neural network has a slow convergence speed and requires long training time; it is easy to fall into local optimum; the always of high redundancy; learning memory are unstable. This paper uses genetic algorithm modify weights thresholds in order obtain global optimal value. simulation results show that this method characteristics strong detection ability pollution concentration efficiency, which can help improve accuracy SO2 urban air.
The safety of the transmission lines maintains stable and efficient operation smart grid. Therefore, it is very important highly desirable to diagnose health status by developing an prediction model in grid sensor network. However, traditional methods have limitations caused characteristics high dimensions, multimodality, nonlinearity, heterogeneity data collected sensors. In this paper, a novel called LPR‐MLP proposed predict power consists two parts: (1) local binary pattern (LBP),...
In view of the fact that when BP neural network algorithm is trapped into local extremum and converges to minimum point, convergence rate becomes slow, structures are different, there a contradiction between application examples scale, this paper uses particle swarm optimization optimize initial weights thresholds network. This method effectively enhances ability handle nonlinear problems, improves algorithm's speed search global optimal values at same time. A project case was selected for...