A5100425865- Advanced Chemical Sensor Technologies
- Gas Sensing Nanomaterials and Sensors
- Insect Pheromone Research and Control
- Spectroscopy and Laser Applications
- Analytical Chemistry and Sensors
- Innovative Microfluidic and Catalytic Techniques Innovation
- Nanoparticle-Based Drug Delivery
- Water Quality Monitoring and Analysis
- Nanoplatforms for cancer theranostics
- Pickering emulsions and particle stabilization
- Peptidase Inhibition and Analysis
- Fluid Dynamics and Thin Films
Yantai University
2023
Nanjing University of Aeronautics and Astronautics
2020-2023
Jiaxing University
2023
Harbin University of Science and Technology
2022
United States Naval Research Laboratory
2012
Although gas-borne ultrasound catalysis has been developed as a new method to discriminate gas species and measure the concentration, applications of machine learning methods in analyses with single metal oxide (MOX) sensor catalyzed by are still scarce. In this work, an ultrasonically MOX sensor, we explored effectiveness K-nearest neighbors (KNN), support vector (SVM), single-hidden-layer BP-ANN (SHBP) discrimination application SHBP concentration measurement. The target gases work...
In this article, a system design method and Support Vector Machine (SVM) gas identification algorithm for the ultrasonically catalyzed single-sensor E-nose is proposed investigated first time. aspect of hardware, consists an ultrasonic module, sensing data processing personal computer, power supply module. software, microcontroller in module coded by C programming language acquisition operational control, user interface made on PC C# display sending commands. Performance integrated tested...
We develop a tricomponent (ternary) hydrodynamic model for multiphase flows of biomass and solvent mixtures, which we employ to simulate biofilm. In this model, the three predominant effective components in biofilms, are extracellular polymeric substance (EPS) network, bacteria, (consisting nutrient, etc.), modeled explicitly. The fluid mixture is assumed incompressible as whole, while intercomponent mixing, dissipation, conversion allowed among components. Bacterial growth EPS production...
Electronic noses are one of the predominant technological choices for gas mixture detection, but their application in real-world atmospheric environments still leaves several issues to be resolved. The key bottleneck is effect turbulence caused by diffusion gases atmosphere on quantitative and qualitative analytical performance electronic nose. In light this, this paper presents a analysis strategy monitoring. This adopts baseline manipulation raw sensor data reduce drift interference, then...
In the mining process of coal, oil and natural gas, there are often a large number toxic, hazardous explosive gases, these gases usually exist in form mixtures. When flammable reach certain concentration atmosphere, they will explode when encounter an open flame. Therefore, accurate detection mixed is great significance to prevention safety accidents. order improve accuracy gas component recognition, this paper proposes composition identification method. This method uses kernel principal...
This paper has proposed and explored two drift calibration methods for the ultrasonically catalyzed single-sensor electronic nose, which retrain machine learning (ML) model gas identification concentration regression by combining newly collected data with old data. The are termed update method incremental method, respectively. In former, randomly selected samples at certain concentrations replaced new similar same sample number, to ML models. latter, experimental merged create a set, results...
Ultrasonic radiation can catalyze the redox reaction at sensing surface of a gas sensor, which provides new method to significantly raise sensors' sensitivity and discriminate species. In this work, convolutional neural network (CNN) based model for species discrimination concentration regression is proposed developed an ultrasonically assisted thermoelectric in order achieve accurate results. The experiment evaluates sensor's ability detect quantify four gases clean air: acetone, methanol,...
The ultrasonic radiation method provides a new solution to the single sensor based gas analysis. But it has been unknown that whether artificial neural network (ANN) can be effectively applied in analysis with an ultrasonically radiated and how apply. In this work, BP-ANN model which implement identification concentration measurement catalytic combustion is explored, prominent performance found. BP-ANN, named GWO-MRMR-DHBP, two hidden layers, its feature set computed by minimal redundancy...