A5051320045- Energy Load and Power Forecasting
- Metaheuristic Optimization Algorithms Research
- Traffic Prediction and Management Techniques
- Machine Learning and Data Classification
- Rough Sets and Fuzzy Logic
- Advanced Optical Sensing Technologies
Jianghan University
2021-2023
Deep learning algorithms mostly have network parameters that can affect their training results, and the combination of neural architectures also has a significant impact on algorithm performance. The performance deep is usually proportional to overall number parameters, leading excessive resource consumption for exploring with large hyper-parameters. To solve this problem, vector representation proposed which architectures, multi-objective optimization model established based genetic in...
The geomagnetic sensor system needs to meet the low power consumption and high real-time requirements of single-chip microcomputer, which may cause vehicle flow detection algorithm be insensitive high-frequency signals with small mean value changes. Slow vehicles, end-to-end vehicles large-scale are also prone false detections other issues. This paper studies method based on waveform, is perform filtering collected signal dynamically adjust threshold, calculate magnetic field strength its...
In actual prediction scenarios, attribute features include time data, weather and energy consumption data. The relationship between is very complex. Exploring the feature attributes decision sets with fast reduction can reduce amount of model training Since seasonal temporal variations greatly influence moving horizon method used to update improve accuracy prediction.From above, an based on weighted neighborhood rough set long short-term memory neural network(LSTM) proposed in this paper....