- Water Quality and Pollution Assessment
- Geotechnical Engineering and Underground Structures
- Solar Thermal and Photovoltaic Systems
- Climate change and permafrost
- Advanced Memory and Neural Computing
- Methane Hydrates and Related Phenomena
- Environmental Changes in China
- Infrared Target Detection Methodologies
- Advanced Image Fusion Techniques
- Landslides and related hazards
- Phase Change Materials Research
- Speech Recognition and Synthesis
- Iterative Learning Control Systems
- Environmental Quality and Pollution
- Integrated Energy Systems Optimization
- Sensorless Control of Electric Motors
- Geothermal Energy Systems and Applications
- Cellular Automata and Applications
- Nanofluid Flow and Heat Transfer
- Energy Load and Power Forecasting
- Thermal Radiation and Cooling Technologies
- Thermography and Photoacoustic Techniques
- Geotechnical Engineering and Analysis
- Structural Integrity and Reliability Analysis
- Neuroscience and Neural Engineering
Harbin Institute of Technology
2020-2025
China Academy Of Machinery Science & Technology (China)
2024
Changchun University of Science and Technology
2023-2024
Beijing Jingshida Electromechanical Equipment Research Institute
2023
Taiwan Forestry Research Institute
2021
Beijing University of Civil Engineering and Architecture
2020
Universidad del Noreste
2020
Geological Exploration Institute of Shandong Zhengyuan
2019
Southwest Petroleum University
2018-2019
Jiangsu Provincial Academy of Forestry Science and Management
2019
In order to achieve high-quality repair of complex curved parts, a remanufacturing process method utilizing laser cladding and reverse engineering technology is proposed be implemented by robots. This study focuses on the oscillating helical surface screw pump rotor. A single-pass test conducted using Response Surface Methodology (RSM) construct predictive model identify optimal parameters. The model’s accuracy validated through analysis variance (ANOVA) index verification, while lap rate...
Abstract. Landslide disasters are one of the main risks involved with operation long-distance oil and gas pipelines. Because previously established disaster risk models too subjective, this paper presents a quantitative model for regional assessment through an analysis patterns historical landslide along Using Guangyuan section Lanzhou–Chengdu–Chongqing (LCC) multiproduct pipeline (82 km) in China as case study, we successively carried out two independent assessments: susceptibility...
Summary Dispersing nanomaterials can effectively improve the thermal energy storage performance of molten salts. However, research on such improvement mechanism is still immature, as understanding levels induced by different nanoparticles has not been consistent. Using silica additives, this work prepares nine salt nanofluids aqueous solution method with use potassium nitrate, sodium and their binary mixture base The key these are evaluated through characterization approaches, differential...
Abstract The detection of road infrared targets is essential for autonomous driving. Different from RGB images, the acquisition images unaffected by visible light. However, signal‐to‐noise ratio still presents significant challenges. This study demonstrates an improved target model detection. An advanced EfficientNet incorporated to replace conventional structure and enhance feature extraction. A Dilated‐Residual U‐Net also introduced reduce noise images. Meanwhile, k ‐means algorithm data...
In order to enhance the synchronization of welding robot arms and improve quality, this study proposes a fuzzy PID-based improved deviation coupling multi-axis synchronous control method. Firstly, in response intricacies inherent compensation mechanism structure substantial volume system computation, integration average speed sub-average is proposed optimize compensator. This aims mitigate errors, minimize adjustment time, elevate overall performance. Moreover, PID algorithm employed design...
More than 30,000 tons of molten salts as efficient heat transfer and thermal energy storage materials have been employed in a single commercial concentrating solar power plant. Nanoparticles could increase the performance salts. However, there is no synthesis method for mass production salt nanofluids by far. In this work, two techniques that are aqueous solution high-temperature melting methods, which be potentially used large scale nanofluids, were comparatively evaluated investigating...
A mathematical model for simulating hydrodynamics and pollutants migration in a tidal river network was constructed, which takes the temporal spatial distribution of rainfall runoff non-point into consideration. Under design hydrologic conditions typical hydrological year, daily concentration change process control section is obtained. Aiming at uncertainty hydrology water quality parameters such as flow direction, rate area, statistical analysis method used to obtain maximum allowable under...
As a typical representative of transformation thermodynamics, which is the counterpart optics, thermal cloak has been explored extensively while most current research focuses on structural design instead adaptability and practicability in dynamic environment. The evaluation energy processes involved under conditions are also lacking, essential to engineering application this functional structure. In paper, based environment sinusoidal form with ambient amplitude, distribution density, phase,...
The spatial distribution of soil depth is the key information watershed planning and design, also one factors that determine classification slope utilization. on hillside requires on-site survey, which labor-intensive. In addition, there are real challenges in updating or supplementing data because surveyed may be altered by subsequent artificial natural forces. Based concept risk assessment, this study calculated external forces (external factors) internal (location topsoil's surface,...
Abstract Cellular automata (CA) are computational systems that exhibit complex global behavior arising from simple local rules, making them a fascinating candidate for various research areas. However, challenges such as limited flexibility and efficiency on conventional hardware platforms still exist. In this study, we propose memristor-based circuit implementing elementary cellular (ECA) by extending the stateful three-memristor logic operations derived material implication (IMP) gates. By...
Abstract. Landslide disaster is one of the main risks involved with operation long-distance oil and gas pipelines. Because previously established risk models are too subjective, this paper presents a quantitative model for regional assessment through an analysis laws historical landslide disasters along Using Guangyuan section Lanzhou–Chengdu–Chongqing (LCC) Long-Distance Products Oil Pipeline (82 km) in China as case study, we successively carried out two independent assessments: hazard...
With the rapid development of optical remote sensing, it is urgent to find a reliable target detection method. Compared with traditional algorithms, convolutional neural network has attracted considerable attention owing its efficiency and high transitivity. However, different from general images, sensing images contain complex background information dense small targets changeable directions that make very challenging. To solve these problems provide stable high-performance method, rotated...
Working memory refers to the brain's ability store and manipulate information for a short period. It is disputably considered rely on two mechanisms: sustained neuronal firing, “activity-silent” working memory. To develop highly biologically plausible neuromorphic computing system, it anticipated physically realize that corresponds both of these mechanisms. In this study, we propose memristor-based neural network firing activity-silent memory, which are reflected as dual functional states...
Flames typically exhibit a progression from red to yellow and even white as temperature increases, accompanied by an increase in image noise. Conventional methods for flame segmentation often yield unsatisfactory results. To address this issue, we propose method based on color features. In the HSV space, brightness layer (V) undergoes adaptive histogram equalization median filtering. We analyze characteristics of common pixels RGB YCbCr spaces derive orthogonal feature formula. The improved...
In order to research the distribution of residual chlorine LNG drainage in ocean, further formulate reasonable wastewater marine disposal scheme and provide scientific support protection environment.According tidal current characteristics receiving waters project design typical hydrological conditions different seasons, a 2D hydrodynamic model was used simulate field cold water discharge Radial sand ridge group shallow sea area Jiangsu maritime space.And as simulation mobility transfer two...
In order to research the distribution of thermal field low temperature discharge from LNG in ocean, further formulate reasonable wastewater marine disposal scheme and provide scientific support protection ocean environment, a two-dimensional hydrodynamic model was used simulate tide current radial sand ridge group shallow sea area Jiangsu maritime space. This based on characteristics near project seasonal design typical hydrological conditions different seasons. as condition that simulated...
Taking a reservoir in South China as an example, we use rainfall-runoff unit hydrograph method to analyze the time changing process of surface runoff inflow, which generated by typical design rainfall. On basis series data flow and water quality control section main rivers Xili Reservoir, establish mathematical response relation between non-point source pollutants flux, such flux COD, NH3-H, catchment area runoff. Then simulate dynamic change progress pollution load generate with initial...
The large-scale wind power incorporation increases the uncertainty of grid. In this paper, we propose a new stochastic optimization method to eliminate it. entropy risk value (EVaR) is used describe safety margin system for limiting security constraint. Finally, use matlab validate validity model and effectiveness by numerical simulation IEEE 30 nodes, compare scheduling results under different confidence level.
In this paper, we propose a small-footprint wake- up-word speech recognition (WUWSR) system with two stages to recognize two-syllable wake-up word. the first stage, convolution neural network (CNN) is trained predict posterior probability of context-dependent state. Thus wake-up-word detected according confidence score obtained by dynamic programming. second cascade bidirectional long short-term memory (LSTM), convolutional modules and deep feed-forward (BLCDNN) successively verify...