|
|
Enhancing retinal fundus images based on MPA optimization algorithm
Số 92 - Số Tiếng Anh (12/2024)
>
trang 02-08
|
Tải về (1212.42 KB)
Phu-Hung Dinh
|
Tóm tắt
The retinal fundus image serves as a crucial tool in diagnosing numerous prevalent diseases, given that many significant conditions manifest on the retina. Nevertheless, images obtained in real clinical environments often lack quality for accurate diagnosis due to issues such as uneven lighting, blurriness, and low contrast. In this article, we present a simple yet effective approach to enhance retinal fundus images. Initially, the input images are transformed from the RGB color space to the YUV color space, yielding three components: Y, U, and V. The Y component is enhanced using the Contrast-limited Adaptive Histogram Equalization (CLAHE) method. Subsequently, it is decomposed into three layers: structure layer (L_S), texture layer (L_T), and noise layer (L_N) using the Relative Total Variation (RTV) method and Anisotropic Diffusion Filter (ADF). Next, the structure component is enhanced by the BLLI (Brighten low-light image) method, resulting in the L_B component. The MPA optimization algorithm is employed to find optimal parameters for the L_B, L_S, and L_C components. The components are then multiplied by the optimal parameters and summed to obtain the enhanced image. Experimental results demonstrate the effectiveness of the proposed method in improving the quality of retinal fundus images.
Keywords: Relative total variation (RTV) method, anisotropic diffusion filter (ADF), marine predators algorithm (MPA) |
|
|
Some issues regarding surface water quality in the Ha Nam area, Quang Yen town, Quang Ninh province
Số 92 - Số Tiếng Anh (12/2024)
>
trang 09-15
|
Tải về (3148.79 KB)
Nguyen Thi The Nguyen*, Nguyen The Toan, Nguyen Xuan Thanh
|
Tóm tắt
Ha Nam is a crucial area in the socio-economic development of Quang Yen town in particular and Quang Ninh province in general. This study was conducted to identify the current status of surface water pollution in this area to provide a basis for proposing environmental management solutions linked to the implementation of socio-economic development plans. Statistical analysis of water quality monitoring results at 19 points during the dry and rainy seasons of 2023 reveals that the study area is currently contaminated with organic matter, nutrients, microorganisms, and suspended solids. The overall Water Quality Index (WQI) for the entire area is 39.0 ± 16.4 in the dry season and 47.4 ± 18.9 in the rainy season. The difference in water quality between the two seasons is statistically significant with a sig value of 0.02. Pearson correlation analysis shows correlations between several physical and chemical parameters. Cluster analysis indicates similarities in water quality among 5 monitoring points in both monitoring periods. The research results also demonstrate that statistical analysis helps to gain a deeper understanding of the status of water quality and aid managers in implementing better water quality management and monitoring programs, reducing the resources required for monitoring.
Keywords: Ha Nam area, water quality, statistical analysis |
|
|
Estimation for hydrodynamic coefficient of an autonomous underwater vehicle by using dynamic mesh
Số 92 - Số Tiếng Anh (12/2024)
>
trang 16-23
|
Tải về (2796.41 KB)
Vu Duc Thanh, Vu Lam Dong*, Dang Ngoc Anh, Nguyen Huu Hai
|
Tóm tắt
The hydrodynamic coefficient of the device strongly affects the working ability as well as the accuracy and stability of the automatic control algorithms of an autonomous underwater vehicle (AUV). Expensive experimental methods are not even possible in practice. The determination of these coefficients by empirical formulas will lead to large deviations due to the difference in the geometrical structure of the real device with the reference model. In this paper, we use numerical methods to determine these parameters. Complex movements of the AUV will be simulated by ANSYS software on the basis of dynamic mesh method. The obtained numerical results will be used to approximate the hydrodynamic coefficients for the AUV based on the optimization algorithms.
Keywords: Hydrodynamic coefficient, dynamic mesh, UDF, CFD |
|
|
Applying the k-nearest neihbour algorithm to study response for irrigation water fee of water users
Số 92 - Số Tiếng Anh (12/2024)
>
trang 24-29
|
Tải về (695.14 KB)
Bui Thi Thu Hoa*, Nguyen Thi Thu Ha
|
Tóm tắt
Studying water use behavior is one of the solutions from the demand side to help control water resources, which is increasingly depleted. Machine learning approach using the k – nearest neighbour (k-NN) algorithm is used in this paper to study water users' reactions through their ability with irrigation water fees. K-NN is considered as one of the modern methods to classify consumers. The consumer behavior for the possibility of increasing irrigation water fee compared to current levels is also found out. The study was conducted based on investigated data of 222 households in 3 provinces of Thai Nguyen, Nam Dinh and Phu Tho.
Keywords: Machine learning, k-nearest neigbour, water user behavior |
|
|
|
|
Green synthetic of iron nanoparticles from plants
Số 92 - Số Tiếng Anh (12/2024)
>
trang 36-44
|
Tải về (3522.57 KB)
Dinh Thi Lan Phuong
|
Tóm tắt
In this study, iron nanoparticles (Fe NPs) were synthesized from Syzygium nervosum leaves, green tea leaves, and guava leaves. The aim of this study is to utilize polyphenols from leaf extract to stabilize Fe(0), replacing toxic chemicals such as NaBH₄, thereby reducing environmental pollution. The polyphenols in the leaf extracts acted as both reducing agents and stabilizers for the Fe NPs. Modern techniques, including SEM, TEM, XRD, FTIR, and EDX, were employed. Results showed that S. nervosum leaves synthesized Fe NPs with sizes ranging from 40 to 60 nm, while green tea and guava leaves produced particles larger than 100 nm. The appearance of the characteristic X-ray diffraction peak of zero-valent iron at 47.21° and 61.75° indicates the significant potential of S. nervosum leaf extract for the complete reduction of ferric ions. The XRD and FTIR spectra indicated that the polyphenols (peaks at 1614.04 cm⁻¹ (C=C), 1572.89 cm⁻¹ and 1415.28 cm⁻¹ (C-C), 701.50 cm⁻¹ and 908.02 cm⁻¹ (aromatic C-H), and 3289.21 cm⁻¹ (O-H)) in S. nervosum leaf extracts act as protective agents, enhancing the stability of Fe NPs [603.69 cm⁻¹ (Fe-O)] suggests that the Fe NPs are coated with polyphenols).
Keywords: Green materials, green tea leaves, guava leaves, iron nanoparticles, Syzygium nervosum leaves |
|
|
Improving water use efficiency for the Nhon Hai - Thanh Hai irrigation system with ODA in Ninh Thuan
Số 92 - Số Tiếng Anh (12/2024)
>
trang 45-53
|
Tải về (891.33 KB)
Le Minh Thoa
|
Tóm tắt
Ninh Thuan province has a hot climate, with frequent droughts and local water shortages in Ninh Son, Bac Ai, Ninh Hai, Thuan Bac, and Thuan Nam districts. Droughts are likely to cause localized saltwater intrusion deep into the river from (0.5 ÷ 1.5) km. To solve the problem of water resources effectively, with many resources, the province has promoted investment in the construction of strategic irrigation works, contributing greatly to the added value of agricultural production and construction. Building new rural areas and sustainably adapting to climate change with capital ADB has sponsored Ninh Thuan province to implement the subproject "Irrigation development to serve high-tech agriculture in Nhon Hai - Thanh Hai, Ninh Thuan province”. This article improves water use efficiency for the Nhon Hai - Thanh Hai irrigation system using ODA in Ninh Thuan to solve the above problems as well as apply to current exploitation and use units.
Keywords: Water use efficiency, irrigation system, ODA capital, Ninh Thuan |
|
|
Caculation of the hydraulic jump length in horizontal rectangular channels using artificial neural network models
Số 92 - Số Tiếng Anh (12/2024)
>
trang 54-61
|
Tải về (1608.18 KB)
Hung Viet Ho
|
Tóm tắt
This study aims to develop artificial neural network (ANN) models that are effective and simple in forecasting the length of a hydraulic jump in a horizontal rectangular canal. This study also recommends applying the Buckingham Pi theorem to ascertain four inputs and one output of the ANN models, corresponding to five dimensionless Pi terms. The effect of the channel roughness, fluid viscosity, and gravity has been considered to improve the model's performance. The ten models, each containing an input layer with four neurons, an output layer with one neuron, and one or two hidden layers, were meticulously developed and evaluated using the four statistical indicators MAE, RMSE, NSE, and MAPE. According to the test results and comparison with empirical equations, all models perform better than empirical equations in forecasting the hydraulic jump length. The study's findings show that the number of neurons per hidden layer should be between 16 and 32. The best model comprises two hidden layers, each containing sixteen neurons. This model's NSE value reaches 0.9905 during testing. Two-hidden-layer models require fewer epochs but are more efficient than one-hidden-layer models. Consequently, when the inflow Froude number ranges from 1.73 to 19.67, the proposed ANN models might be utilized instead of empirical formulae for determining the hydraulic jump length.
Keywords: ANN, hydraulic jump, machine learning, model |
|
|
Advanced vibration analysis in a half-car model
Số 92 - Số Tiếng Anh (12/2024)
>
trang 62-68
|
Tải về (3324.38 KB)
Nguyen Dang Quy, Vu Ngoc Minh*, Bui Tien Tung, Nguyen Truong Sinh
|
Tóm tắt
A half-car vibrating model will be investigated in terms of time and frequency responses. The equations of motion are simulated numerically to determine the displacement of the vehicle body, two wheels, and roll motion. A novel system of equations is introduced for three cases related to the tire-road separation phenomenon. The dynamic responses are studied over a wide range of input frequencies of road excitations. Furthermore, a comparison of vibration dynamics between the separation and no-separation assumptions is presented using phase graphs. The body bounce is also studied under the separation condition using a detailed mesh plot. The simulation outcomes demonstrate that the developed model provides greater precision compared to previous studies that did not consider separation. The research provides useful background knowledge for designing future vehicles under real-world conditions.
Keywords: Half-Car Model, advanced vibration, tire-road separation, ride comfort |
|
|
Design and implementation of a 2.500m offshore seawater pumping system for high-tech aquaculture in Ba Ria - Vung Tau, Vietnam
Số 92 - Số Tiếng Anh (12/2024)
>
trang 69-76
|
Tải về (3715.52 KB)
Nguyen Minh Tuan*, Nguyen Thi Nho
|
Tóm tắt
This study focuses on the design and implementation of an offshore seawater pumping system to support high-tech aquaculture in Dat Do District, Ba Ria - Vung Tau Province. This is the first application in Vietnam of a seawater pumping system that draws clean water directly from a 2.500m offshore distance without an intermediate reservoir, providing stable salinity levels and minimizing pollution risks. The system uses a dual-suction centrifugal pump made of SUS316 stainless steel, paired with a large-diameter (1.200mm) HDPE pipeline of significant length, meeting requirements for pressure loss, pressure, and corrosion resistance in seawater environments. Experimental results show that the system reliably supplies 110.000 m³ of seawater per day to a 300-hectare farming area. The results confirm the feasibility and high efficiency of the pumping system, meeting water supply requirements with suitable head and flow rate under real-world conditions.
Keywords: Pumping system, aquaculture, high-tech farming, seawater pumping |
|
|
The necessity of 3D CFD method on simulating fluid flow over conveyance structures
Số 92 - Số Tiếng Anh (12/2024)
>
trang 77-85
|
Tải về (701.56 KB)
Le Thi Thu Hien*, Nguyen Van Chien
|
Tóm tắt
This study investigated the application of 3D CFD models in alignment with the project's objectives and the spatial extent of the target region. It included a comparative analysis of the advantages and disadvantages of 1D and 3D CFD models, considering factors such as flow representation, geometric complexity, and computational resource requirements. A case study of the Giang Ma spillway, characterized by its intricate geometry, was conducted to simulate flow depth, velocity, and rapid flow fields using both 1D and 3D models. Significant differences observed in numerical results at specific cross-sections highlighted critical considerations for design engineers.
Keywords: Conveyance structures, 3D CFD model, 1D CFD model |
|
|
Assessing the capacity of the water source to receive wastewater for issuing discharge of wastewater permits
Số 92 - Số Tiếng Anh (12/2024)
>
trang 86-93
|
Tải về (8290.78 KB)
Doan Thuy Kim Phuong, Vo Ngoc Duong*, Nguyen Trung Quan, Nguyen Cong Phong
|
Tóm tắt
The environmental carrying capacity refers to the maximum load that an environment can sustain while still meeting quality requirements for specified uses within a designated research area. In this study, we applied the MIKE 21 model with two modules: Hydrodynamic (HD) and Ecological (Ecolab). The purpose was to assess the water’s capacity to receive wastewater and predict trends in water quality changes resulting from development activities in the Vu Gia-Thu Bon River estuary region.
After calibrating and validating the MIKE 21 model in 2005 and 2009, the results indicated that the organic matter decay coefficient kd = 0,11 and the longitudinal dispersion coefficient Ex ranges from 50 to 150 m²/s for each river segment, facilitating the simulation of water quality in the Vu Gia-Thu Bon estuary. The simulation results of the numerical model indicate a current increase in pollutant concentration in the Vu Gia-Thu Bon estuary area. However, this increase is mild and does not significantly impact environmental pollution, as most measured parameters remain within permissible limits. Nonetheless, it is crucial to implement appropriate wastewater management and treatment plans, including contingency measures, to prevent the direct discharge of pollutants into the river basin.
Keywords: MIKE 21 model, Vu Gia-Thu Bon river, wastewater, Da Nang |
|
|
Analysis and evaluation of the combined wind power and pumped hydro storage system: Efficiency and challenges
Số 92 - Số Tiếng Anh (12/2024)
>
trang 94-101
|
Tải về (698.65 KB)
Nguyen Thi Nho, Nguyen Van Nghia*
|
Tóm tắt
In this paper, we use the Levelized Cost of Energy (LCOE) metric to assess the feasibility of a combined model of pumped hydro storage and wind power. Additionally, the study proposes a method for regulating and storing electricity to maximize the use of excess wind energy by pumping water into reservoirs, which can then be used to generate electricity during peak demand periods, ensuring optimal electricity supply in the region. The results show that the system can meet 81.3% of the electricity demand, although there is still 13% surplus energy and an 18.7% shortfall. The combined system's LCOE is calculated at 0.14 USD/kWh, which is higher than the global average. The methods in this study provide a scientific and practical foundation for implementation of renewable energy storage systems in Vietnam.
Keywords: Levelized cost of energy (LCOE), pumped hydro storage, wind power, hybrid energy system. |
|
|
Research on the optimal operation calculation model of large-scale hydropower reservoir systems using a hybrid algorithm
Số 92 - Số Tiếng Anh (12/2024)
>
trang 102-109
|
Tải về (1400.45 KB)
Ho Sy Mao
|
Tóm tắt
Researchers have applied many optimization techniques to different reservoir systems, the most popular of which are dynamic programming (DP) techniques and genetic algorithms (GA). Still, these techniques are not effective for multi-reservoir systems. The discrete difference dynamic programming algorithm (DDDP), a variation of dynamic programming, is the solution of choice for reservoir system problems. Still, its drawback is that it must find the optimal initial trajectory. If recursive equations are used to find the optimal trajectory, this solution also encounters the problem of dimensionality of the dynamic programming problem. Meanwhile, GA can find the optimal solution from random solution sets. Still, its disadvantage is that finding the global optimal solution is tough, while calculation time is also a limitation of GA for significant system problems. Combining GA and DDDP for multi-reservoir problems is an effective method for these two algorithms. The case study is the system of five cascade hydropower reservoirs in the Da river basin in Northern Vietnam.
Keywords: Reservoir system, genetic algorithm, discrete differential dynamic programming, cascade hydropower |
|
|
|