No. 4 (2022)

Article

• Investigating and applying business process management to build the leave application process for teachers

  Lâm Quế Thanh, Nguyễn Duy Phúc (Author)

  Abstract | PDF (Vietnamese) | Views: 6 Downloads: 8

  Currently, in our country's education, the issue concerning teachers' leave application has not been unanimously agreed upon. At each educational unit, there will be specific regulations on the leave application process for teachers. However, the management process is still limited and fragmented between the steps in the leave application process. The article presents the results obtained after applying business process management to the leave application process for teachers. The objective of the study is to build a process of applying for teacher leave using Bonitasoft software [2] by building a diagram showing the steps to apply for teacher leave. Therefore, it is possible to closely manage the operation of the leave application process for teachers and reduce a lot of management time and operating costs for administrators. 

   

• Research on the powertrain system on Tesla electric cars

  Cao Hùng Phi, Võ Hồng Hải (Author)

  Abstract | PDF (Vietnamese) | Views: 7 Downloads: 7

  This paper presents the results of research on the powertrain system on Tesla electric cars. In this study, the torque transmitted from the electric motor to the active wheels was investigated by Matlab Simulink software. The simulation process was performed through the cases of 50%, 60%, 70% and 100 % of engine power. The simulation results show that the torque at the active wheels increased sharply compared to the engine's torque, which proves that the powertrain responded well to the vehicle's needs. It also shows the acceleration of the car.

• A study of stable settlement in soil cement pilereinforced foundation

  Lê Bá Vinh, Phan Hữu Hoàng (Author)

  Abstract | PDF (Vietnamese) | Views: 7 Downloads: 6

  In the calculation of soil cement piles, one of the most important factors is the stable settlement of the ground after being reinforced with soil cement piles according to different calculation methods. 

   

• The problem of homogeneous backward heat conduction on a bounded interval

  Phạm Minh Trí, Nguyễn Thành Nhân (Author)

  Abstract | PDF (Vietnamese) | Views: 8 Downloads: 7

  The problem of homogeneous backward heat conduction on a bounded interval is an ill-posed problem. The problem is concerned with the equation Ax = y. For the given data y, we find x = Aly. The existence and uniqueness of the solution to the problem may not be obtained. In case of existence, the solution to the problem is also unstable because the operator A1 is not continuous. In this paper, we study the ill-posedness of the problem and a regularization is given. In the following parts, the Fourier method is used to obtain the weak form of solution. Moreover, the method of quasi-boundary value is used to construct the regularized solution. 

   

• Using multivariate LSTM model in forecastingtemperature and rainfall

  Dương Thị Hà, Nguyễn Thái Nghe (Author)

  Abstract | PDF (Vietnamese) | Views: 12 Downloads: 11

  Forecasting temperature and rainfall is an important issue in the agricultural sector to assist farmers in planning appropriate planting. Several techniques have previously been proposed based on statistical analysis, learning machine and deep learning techniques. This paper proposes a method to build a model to forecast monthly temperature and rainfall using the Multivariate Long Short-Term Memory (MLSTM) model. The parameters of the model are adjusted to suit the proposed problem. The model is evaluated using RMSE and MAE measurements. Besides, other forecasting models (such as LSTM, MLP, and SVR) are also used to compare the effectiveness of the proposed model. Experimental results on the data set of monthly average temperature and rainfall in Vietnam from 1901 to 2015 show that the MLSTM model is quite effective with the RMSE on the temperature set of 1.311 and the MAE of 1.051, respectively on the rainfall data set of 2.299 and 2.450. 

   

• Research on the rotational dynamics of a 1.25-ton truck

  Nguyễn Đức Hải, Lương Văn Vạn (Author)

  Abstract | PDF (Vietnamese) | Views: 16 Downloads: 11

  This paper presents the research results of the truck's motion trajectory using a single-track planar model. In this study, the truck's motion trajectory was investigated using Matlab software with the steering rule identified in advance for the purpose of determining the influence of the following factors: Steering angle, the grip coefficient between the tire and the road surface, the effect of the truck's motion speed and load on the motion trajectory in case of the truck turning around. The survey results show that an increase in the steering angle led to the reduction in the rotation radius, and an increase in the grip coefficient led to the reduction in the rotation radius and an improvement in the rotational stability. The change in truck speed is proportional to the change in the rotation radius of the truck. 

   

• Analysis of the vertical displacement of a wheel moving on the floating plate by the combination of the moving element method and the boundary element method

  Phan Nhật Tân, Dương Chung Nguyện, Nguyễn Tấn Hưng (Author)

  Abstract | PDF (Vietnamese) | Views: 11 Downloads: 4

  This paper presents a numerical approach with the combination of the moving element method and the boundary element method to analyze the hydroelastic behavior of the three-degree freedom system moving on a floating plate structure. Based on this approach, the vertical displacement of a wheel moving on the floating plate has been determined. Furthermore, the influence of material parameters on the hydroelastic of the wheel has also been identified. The results of the study show that the displacement of the wheel had a positive relationship to the drag coefficient of the bogie, the drag coefficient of the wheel, the stiffness of the bogie and the stiffness of the wheel. Among the parameters observed, the stiffness of the wheel has the most influence on the displacement of the wheel. In the future, the study on the behavior of the three-degree freedom system on the floating plate structure by experimental evaluation should be conducted to examine the accuracy of the calculation method presented in this article.

• Building the system for detecting fake information in the text form by deep learning method

  Nguyễn Trường Vương, Cù Vĩnh Lộc (Author)

  Abstract | PDF (Vietnamese) | Views: 9 Downloads: 13

  In this paper, we present an approach to textual fake information detection using deep learning method combined with a Vietnamese-specific pre-training model (PhoBert) to recognize and detect fake information. Our proposed models are tested and evaluated on Vietnamese dataset [1]. From input data through input data processing steps, we need to apply word embedding to convert text data into vector model. Then this vector will go through the PhoBert transform model to get the feature vectors as output. This output vector is the input of an MLP (Multi-layer Perceptron) feedforward network to be used for the classification process. The softmax() function for the classification process is used. This function returns the probability on two sets of fake or real labels. In this paper, three different models such as CNN, LSTM and PhoBert are used to perform model training on the same data set, then compare the obtained results and the accuracy of each model to then select the best and most accurate model to build a fake information detection system. The test results show that the PhoBert model has achieved outstanding results with 89% accuracy in detecting fake information.

   

• The construction of a model to determine the load acting on the universal shaft of a 1.25-ton truck

  Lương Văn Đen, Lương Văn Vạn (Author)

  Abstract | PDF (Vietnamese) | Views: 8 Downloads: 14

  The powertrain system is responsible for transmitting torque from the engine to the active wheels. The load acting on the powertrain system is determined by the loads acting on the truck during the movement as well as the dynamic parameters of the truck. In order to calculate and determine the load acting on the powertrain system of the truck in specific cases, it is necessary to build a diagram describing the operation of the system and establish equations representing the physical processes occurring in the system. This paper presents a method to build a model and determine the load acting on the universal shaft of a 1.25-ton truck. The survey results have determined the variable values of torque acting on the universal shaft with different gears. 

   

• The infuence of drying time and temperature on the isolated yeast cell density during yeast dough production from Saccharomyces cerevisiae

  Nguyễn Thị Thu Hương, Đinh Thị Bích, Lý Thị Ngọc Yến (Author)

  Abstract | PDF (Vietnamese) | Views: 8 Downloads: 7

  This study was carried out to produce high quality yeast powder from isolated Saccharomyces cerevisiae. The essential factors affecting the quality of yeast dough were studied, including time (15 hours, 17 hours and 19hours) and drying temperature (40°C, 42°C, 44°C and 46°C). The results show that the drying time of yeast powder for 17 hours at 42°C achieved the highest isolated yeast cell density (6.4x10 cfu/ml) and the possibility of fermenting wine with papaya with high alcohol level (8.48% v/v at 20°C) was higher than the other experiments. 

   

• A multimedia information system in the cloud

  Nguyễn Quốc Thoại Anh, Nguyễn Quốc Anh Thư (Author)

  Abstract | PDF (Vietnamese) | Views: 7 Downloads: 7

  Vietnam is one of the first countries in the world to issue a national digital transformation program, making Vietnam become a country being aware of digital transformation in parallel with advanced countries in the world. In particular, in the years 2020 and 2021, despite the impact of the COVID-19 epidemic, the national digital transformation took place vigorously and drastically in all sectors and fields. Most notably, the Government has built the national database system on the population and the system of producing, issuing and managing citizens'identification. As a result, Vinh Long Radio and Television Station has gradually digitized their program contents. Previously, the content management was through magnetic tapes, but now the management has been digitized into files. Therefore, the management of huge volumes of data is an urgent task today. In order to well manage the existing data and update new data on a daily basis, it is necessary to build a multimedia information system in the cloud. 

   

• Detecting and segmenting lung injury caused by COVID-19 with deep learning technique

  Phan Anh Cang, Trần Thị Kim Ngân (Author)

  Abstract | PDF (Vietnamese) | Views: 6 Downloads: 7

  The world is going through a global health crisis known as the Covid-19 pandemic. To check if the patient is infected with Covid-19, the Realtime RT-PCR test method is selected. However, this method depends on many factors such as the time of specimen collection, preservation process, or specialized machinery system. In this report, a deep learning algorithm based on the above Resnet50, ResNet101 networks has been proposed to help detect Covid-19 cases using X-ray images. Through testing, the results were obtainted with accuracy and mAP of approximately 99%. This method gives doctors more bases to identify patients infected with Covid-19 and have timely treatment. 

   

• The modification of land compensation principles in case of land acquisition by the State

  Nguyễn Ngọc Anh (Author)

  Abstract | PDF (Vietnamese) | Views: 8 Downloads: 11

  This article provides an in-depth analysis of the land compensation principles in case of land acquisition by the State for national defense and security purposes or socio-economic development for national and public interests on the basis of legal provisions and actual application. Therefore, the legal limitations and inadequacies leading to difficulties in actual application have been identified and recommendations for modification have been given.

• Classification and localization of 14 types of lung lesions using neural networks

  Trần Quốc Thịnh (Author)

  Abstract | PDF (Vietnamese) | Views: 4 Downloads: 7

  The risk of lung disease is immense for many people, especially those in developing countries. Doctors have difficulty interpreting chest X-rays, which can lead to misdiagnosis. Deep learning methods help locate lung abnormalities. A dataset of 112,000 images is labeled by experienced physicians. A neural network is trained based on Faster-RCNN and Retina framework towards big data approach to reduce the training time. Furthermore, Grad-CAM is used to highlight the lesion region in making predictions. The results demonstrate that the method using deep learning techniques can achieve the accuracy of up to 95.5%. 

   

• Cardiovascular disease detection on thoracic X-ray images using VGG16 transfer learning method

  Nguyễn Văn Bình, Nguyễn Thái Nghe (Author)

  Abstract | PDF (Vietnamese) | Views: 6 Downloads: 7

  Cardiovascular disease is one of the most dangerous and common diseases in Vietnam and in the world today. More worrying is that this disease is increasingly common among young people. Especially in the context of the complicated developments of the COVID-19 pandemic, people with cardiovascular disease are at high risk of being infected with Corona virus. Therefore, the identification and early diagnosis of cardiovascular disease is one of the important and necessary research directions today. In this article, we propose to use a deep learning model with transfer learning to detect and identify two common types of cardiovascular disease, which are enlarged heart disease and aortic aneurysm, through X-ray images. - Chest radiograph. Specifically, this study used the transfer learning method with the VGG16 network model - a form of deep learning model - combined with data preprocessing to identify cardiovascular diseases. Experiments on available data of VinBigData have been labeled by experts in the field of cardiology. Experimental results from three scenarios show that this approach is very satisfactory with accuracy of 0.95, 0.96, and 0.70, respectively. 

   

• Investigating, measuring and designing the harvesting blade in the combination of shallot harvesting and transporting machines

  Cao Hùng Phi, Nguyễn Thuận Hải Đăng, Lê Trung Hậu, Huỳnh Chí Linh (Author)

  Abstract | PDF (Vietnamese) | Views: 10 Downloads: 10

  The study on shallot harvesting technology is an integral part of the combination of shallot harvesting and transporting machines, which is an important mechanism that determines the digging and separation of shallot from the soil, thus affecting the yield. The harvesting blade works in an environment with many external forces due to direct contact with the soil and is responsible for removing shallot from the soil. The study on the measurement and design of harvesting blades in the combination of shallot harvesting and transporting machines was conducted on Autodesk Inventor Professional software. The study involves the design, the manufacturing orientation, and the first-time application test in the specialized cultivation area of shallot in Vinh Chau district, Soc Trang province. 

   

• University autonomy to meet the requirements of the industrial revolution 4.0

  Hồ Minh Trung, Phạm Minh Trí (Author)

  Abstract | PDF (Vietnamese) | Views: 4 Downloads: 10

  The article mentions the concepts of “industrial revolution 4.0", "education 4.0”, "university autonomy" as well as the current status of higher education in Vietnam. Some orientations are suggested to implement the university autonomy, in which the new point is to build "a smart university" by applying the achievements of the fourth industrial revolution.