A5085013881- Textile materials and evaluations
- Industrial Vision Systems and Defect Detection
- Imbalanced Data Classification Techniques
- EEG and Brain-Computer Interfaces
- Electricity Theft Detection Techniques
- Neural dynamics and brain function
- Anomaly Detection Techniques and Applications
- Smart Parking Systems Research
- Color Science and Applications
- Dyeing and Modifying Textile Fibers
- Advanced Algorithms and Applications
- Traffic control and management
- Heart Rate Variability and Autonomic Control
- Machine Learning and ELM
- Mechatronics Education and Applications
- Vehicle License Plate Recognition
- Energy Load and Power Forecasting
- Neuroscience and Neural Engineering
- Power Line Inspection Robots
- Fault Detection and Control Systems
- Traumatic Brain Injury Research
- Petroleum Processing and Analysis
- Software Engineering Research
- Advanced Data Processing Techniques
- Spectroscopy and Chemometric Analyses
Nanchang University
2023-2025
Zhengzhou University of Light Industry
2021-2023
Soochow University
2023
The heart rate variability (HRV) of patients with disorders consciousness (DOC) differs from healthy individuals. However, there is rarely research on HRV among DOC following treatment deep brain stimulation (DBS). This study aims to investigate the modulatory effects DBS-on central-autonomic nervous system based variations. We conducted DBS surgery eight DOC. Postoperatively, all underwent short-duration for 3 days, frequencies 25 Hz, 50 and 100 Hz respectively. Each day comprised four...
Emerging brain-computer interface (BCI) technology holds promising potential to enhance the quality of life for individuals with disabilities. Nevertheless, constrained accuracy electroencephalography (EEG) signal classification poses numerous hurdles in real-world applications.
Aiming at solving the problem that existing artificial neural networks (ANNs) still have low accuracy in predicting yarn strength, this study combines traditional expert experience and an ANN to propose a hybrid network, named weighted network. Many studies shown it is reliable predict strength based on technology. However, most training models face with problems of easy trapping into their local minima. The prediction yarns relies experience. Obvious can help model perform preliminary...
Abstract Considering the under-maintenance and over-maintenance of existing equipment maintenance methods, this paper studies a Condition Based Maintenance method for silk dryers. The entropy is used to eliminate influence subjective factors more objectively reflect weight different input parameters; optimizing number nodes in hidden layer network improve prediction accuracy; using GA-BP neural establish state model solve disadvantages BP network, example, unstable prediction, easily falling...
With the continuous development of deep learning, due to complexity neural network structure and limitation training time, some scholars have proposed broad Broad Learning System (BLS). However, BLS currently only verifies that it has excellent effects on data sets, does not necessarily actual sets. In response this, this paper uses effect in predicting unevenness yarn quality set, proposes a BLS-based multi-layer (MNN) for problems, which is called Multilayer Neural Network (BMNN).
Unevenness is one of the important parameters for evaluating yarn quality, but current prediction accuracy unevenness low. One reasons that there are few sample dataset prediction. For this problem, paper applies generalized regression neural network to predict yarn. Then, optimized by using particle swarm optimization, fruit fly optimization algorithm, and gray wolf optimizer, respectively. Finally, models were experimentally validated their effectiveness. The experimental results show...
Traditional hydraulic drive experiments present a number of challenges. During the transmission experiment, equipment is easily damaged and must be frequently updated, which makes it difficult for large students to study at same time; traditional offline, monotonous, boring make increase their interest in learning from what inherent; most undergraduate have home due impact COVID-19. Therefore, need an excellent teaching system that allows them perform improve efficiency. A course was...
<p> The renewable wind power sources are difficult to be predicted in view of the fluctuating factors such as bearing, pressure, speed, and humidity surrounding atmosphere. An attempt is made this paper propose a difference method build neural network long short term memory (LSTM) model for prediction. First, correlation each data analyzed then per-forming processing on original solve problem that cannot by probability distribution. prediction building LSTM feeding difference-processed...