A5101877666- Electrowetting and Microfluidic Technologies
- Marine and coastal ecosystems
- Neural Networks and Applications
- Water Quality Monitoring and Analysis
- Nanomaterials and Printing Technologies
- Electrohydrodynamics and Fluid Dynamics
- Advanced Algorithms and Applications
- Fault Detection and Control Systems
- Advanced Control Systems Optimization
- Remote Sensing in Agriculture
- Iterative Learning Control Systems
- Adaptive Control of Nonlinear Systems
- Land Use and Ecosystem Services
- Intravenous Infusion Technology and Safety
- Building Energy and Comfort Optimization
- Water Quality Monitoring Technologies
- Coral and Marine Ecosystems Studies
- Water Quality and Pollution Assessment
- Plant Water Relations and Carbon Dynamics
- Control Systems and Identification
- Machine Learning and ELM
- Modular Robots and Swarm Intelligence
- Advanced Sensor and Control Systems
- Soil Geostatistics and Mapping
- Neural Networks and Reservoir Computing
PowerChina (China)
2025
Inner Mongolia Normal University
2012-2024
Nanjing Institute of Vegetable Science
2024
University of Jinan
2019-2024
Central South University
2020-2024
Shanghai Electric (China)
2024
China Aerodynamics Research and Development Center
2022
Tianjin University of Science and Technology
2018
Nanyang Technological University
2006-2017
Beijing University of Posts and Telecommunications
2009-2014
This paper proposes a robust support vector machine for pattern classification, which aims at solving the over-fitting problem when outliers exist in training data set. During phase, distance between each point and center of class is used to calculate adaptive margin. The incorporation average techniques standard (SVM) makes decision function less detoured by outliers, controls amount regularization automatically. Experiments bullet hole classification show that number vectors reduced,...
Global lakes have suffered from algae blooms and loss or expansion of aquatic vegetations in recent decades. Remote sensing is considered as an effective approach to monitor blooms. However, individual spectral index unable separate them due the similarity features. In this paper, characteristics analyses were conducted find suitable indices for distinguishing emergent vegetation, submerged vegetation floating yellow a complex environment, Ulansuhai Lake, China. It was found that near...
In the centralized heating, ventilating and air-conditioning (HVAC) system, air handling units (AHUs) are traditionally controlled by single-loop proportional-integral-derivative (PID) controllers. The control structure is simple, but performance usually not satisfactory. this paper, we propose a cascade strategy for temperature of AHU. Instead fixed PID controller in classical scheme, neural network (NN) used outer loop. This approach only overcomes tedious tuning procedure inner loop...
The estimation of water clarity in coastal and inland waters from satellite images is limited by significant uncertainties attributable to the optical complexity these waters. To address this issue, we propose an improved semi-analytical model for retrieving Secchi disk depth (ZSD) based on mechanistic (Lee model) developed Lee et al. (2015). In model, a quasi-analytical algorithm (QAA) was employed estimate total absorption coefficient (a(λ)) backscattering (bb(λ)) retrieve diffuse...
Lakes at a global level have increasingly experienced algal blooms in recent decades, and it has become key challenge facing the aquatic ecological environment. Remote sensing technology is considered an effective means of bloom detection. This study proposed novel detection index (ABDI) based on Sentinel-2 Multispectral Instrument (MSI) data. The ABDI was evaluated by application to Hulun Lake, China. Areas detected were consistent with those identified from visual interpretation maps [the...
Grassland ecosystem dominates in the Mongolian Plateau, mostly located its arid and semi-arid regions. Although is an important source for agriculture, it also a fragile system ecologically. This one of most sensitive areas to global climate change. Precipitation (PPT) soil moisture (SM) are water sources grassland ecosystem, their changes would greatly affect vegetation growth. paper generates precipitation use efficiency (PUE) (SWUE) Plateau based on multi-source remote sensing data...
An important component in improving the quality of forests is to study interference intensity forest fires, order describe fire and vegetation recovery, improve monitoring ability dynamic change forest. Using a event Bilahe, Inner Monglia 2017 as case study, this extracted burned area based on BAIS2 index Sentinel-2 data for 2016–2018. The leaf (LAI) fractional cover (FVC), which are more suitable changes area, were calculated by comparing biophysical spectral indices. results showed that...
This paper considers the design of robust neural network tracking controllers for nonlinear systems. The is used in closed-loop system to estimate function. We introduce conic sector theory establish a control system, with guaranteed boundedness both input/output (I/O) signals and weights network. trained by simultaneous perturbation stochastic approximation (SPSA) method instead standard backpropagation (BP) algorithm. proposed guarantees stability estimation good performance. performance...
The Sentinel-1 imaging radar mission provides a short revisit-time, continuous all-weather, and day-and-night imagery at the C-band, which brings opportunities for dynamic extraction of lake water areas. For wetland-type lakes, it is difficult to distinguish between water, submerged plants, mudflats edge lake, leads difficulty in recognizing affects accuracy area extraction. In this paper, level prediction algorithm based on dual-polarization data decomposition proposed solve problem. We can...
During the tunneling process of a double-shield TBM, vibrations generated by rock–machine interaction can affect its safe, efficient, and stable operation. This study was based on Eping Water Diversion TBM Project. By deploying vibration monitoring system, original signals were acquired. A denoising method combining Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) Multi-scale Permutation Entropy (MPE) applied for reconstruct. The time-domain...
Stability is a key issue during spiking neural network training using SpikeProp. The inherent nonlinearity of Spiking Neuron means that the learning manifold changes abruptly; therefore, we need to carefully choose steps at every instance. Other sources instability are external disturbances come along with sample as well internal arise due modeling imperfection. unstable scenario can be indirectly observed in form surges, which sudden increases cost and common occurrence SpikeProp training....
Proximal sensing offers a novel means for determination of the heavy metal concentration in soil, facilitating low cost and rapid analysis over large areas. In this respect, spectral data model variables play an important role. Thus far, no attempts have been made to estimate soil content using continuum-removal (CR), different preprocessing statistical methods, modeling variables. Considering adsorption retention metals spectrally active constituents study proposes method determining...
Hulunbuir, Otindag, Horqin and Mu Us are the four greater sandy lands in China. They typical ecologically fragile areas with limited ability to resist interference susceptible changes external environment. With climate change recent decades, these have experienced severe land desertification decreased groundwater levels. The ecological environment is at risk of further deterioration, continuous vegetation restoration management should be adapted local maintain water balance natural...
Bottle gourd [Lagenaria siceraria (Mol.) Stand] is a widely cultivated succulent crop species. In December 2022, serious bottle disease occurred in the protected vegetable planting base of Xingguo County, Ganzhou City, Jiangxi Province, China, with 85% 2,100 plants having gray mold disease-like symptoms, including spots on infected fruit. They quickly expanded at suitable temperature and humidity, forming layer inward depressions, which spread to fruit stem causing watery rot, flesh turned...
We propose a robust recurrent kernel online learning (RRKOL) algorithm based on the celebrated real-time approach that exploits trick in training manner. The novel RRKOL guarantees weight convergence with regularized risk management through use of adaptive hyperparameters for superior generalization performance. Based new concept structure update error variable parameter length, we are first one to detailed error, such and stability proof can be integrated sparsification scheme solid...
For training algorithms of recurrent neural networks (RNN), convergent speed and error are always two contradictory performances. In this letter, we propose a normalized adaptive learning (NARL) to obtain tradeoff between transient steady-state response. An augmented term is added gradient exactly model the derivative cost function with respect hidden layer weight. The influence induced gain activation on stability also taken into consideration. Moreover, rate employed improve robustness...
Water clarity ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Z<sub>SD</sub></i> ) is a critical water quality parameter that requires remote sensing mapping. Although great progress has been made in retrieval over clear waters during past decades, challenges remain turbid waters. To address this issue, new model was proposed to retrieve oceanic extremely coastal waters, by improving the Firstly, were optically classified into three classes...
Currently, hyperspectral remote sensing technology used for vegetation monitoring mainly uses empirical and semi-empirical statistical methods to calculate heavy metal content. Combining physical models machine learning algorithms is an effective method estimating biochemical parameters without much ground measurement data. However, a deep extreme (DELM) has faster training speed better generalization ability. By introducing the Aquila Optimizer (AO) algorithm, process of DELM can be...
Surges during training process are a major obstacle in Spiking Neural Network (SNN) using Spike-Prop algorithm and its derivatives [1]. In this paper, we perform weight convergence analysis to understand the proper step size SpikeProp learning hence avoid surges process. Using results of analysis, propose an optimum adaptive rate each iteration which will yield suitable within bounds condition. The performance is compared with existing methods via several simulations. It observed that use...
Abstract A new drop analysis method for liquid property study, called image (IDA), with CCD, has been developed and its abilities to monitor the growth measure volume by acquisition digital processing technology are introduced. The remarkable advantage of IDA is that there no need determine dielectric constant tested it can be used measuring unknown liquids. Image provides not only volume, but also shape information during growth. By merging fiber (FDA) into a fiber‐image analyzer (FIDA),...