The world today is being hit by COVID-19. As opposed to fingerprints and ID cards, facial recognition technology can effectively prevent the spread of viruses in public places because it does not require contact with specific sensors. However, people also need to wear masks when entering public places, and masks will greatly affect the accuracy of facial recognition. Accurately performing facial recognition while people wear masks is a great challenge. In order to solve the problem of low facial recognition accuracy with mask wearers during the COVID-19 epidemic, we propose a masked-face recognition algorithm based on large margin cosine loss (MFCosface). Due to insufficient masked-face data for training, we designed a masked-face image generation algorithm based on the detection of the detection of key facial features. The face is detected and aligned through a multi-task cascaded convolutional network; and then we detect the key features of the face and select the mask template for coverage according to the positional information of the key features. Finally, we generate the corresponding masked-face image. Through analysis of the masked-face images, we found that triplet loss is not applicable to our datasets, because the results of online triplet selection contain fewer mask changes, making it difficult for the model to learn the relationship between mask occlusion and feature mapping. We use a large margin cosine loss as the loss function for training, which can map all the feature samples in a feature space with a smaller intra-class distance and a larger inter-class distance. In order to make the model pay more attention to the area that is not covered by the mask, we designed an Att-inception module that combines the Inception-Resnet module and the convolutional block attention module, which increases the weight of any unoccluded area in the feature map, thereby enlarging the unoccluded area’s contribution to the identification process. Experiments on several masked-face datasets have proved that our algorithm greatly improves the accuracy of masked-face recognition, and can accurately perform facial recognition with masked subjects.

Perforated completion is a main method of horizontal well completion. Based on the mass conservation equation, the momentum conservation equation and the variable mass flow in the horizontal wellbore, pressure drop calculation models of the wellbore fluid are established in perforated completion of horizontal wells. The results of calculation and analysis show that the frictional pressure drop, acceleration pressure drop and mixing pressure drop have different effects on total pressure drop of the wellbore fluid, and the frictional pressure drop plays a major role while the acceleration pressure drop and mixing pressure drop have little influence. The liquid viscosity, production and horizontal wellbore length also have different effects on various kinds of pressure drop. When liquid viscosity is smaller and the length of the horizontal wellbore is shorter, the effects of the acceleration pressure drop and mixing pressure drop cannot be neglected. The theoretical basis and the calculation model of the variable mass flow pressure drop of horizontal wellbore are provided.

In the early evaluation stage of offshore gas reservoirs, the DST test is mainly utilized to evaluate gas well deliverability. However, due to the specificity of offshore operations, the quality of DST test is generally poor, resulting in the anomaly in deliverability calculation, particularly in the high temperature and high pressure gas reservoirs. Since conventional binomial or exponential deliverability equation cannot be applied; and the open flow capacity can only be calculated by one-point method experience formula of other blocks. Therefore, the accuracy is hard to be guaranteed. On the basis of the binomial deliverability equation, this study proposed an advanced deliverability equation named stable point pseudo pressure deliverability equation, which means that as long as there is stable well testing data and pressure build up data gathering from one choke, the absolute open flow can be accurately calculated, and that it can achieve the same effect with the normal deliverability test. Practices have shown that this method can successfully solve the abnormality in deliverability calculation of offshore gas reservoirs with high temperature and high pressure.

Construction industries have poor cost performance in terms of finishing projects within a budget. A fuzzy model for evaluating the critical factors of cost overrun for construction projects in China is developed by identifying, classifying and ranking cost overrun factors of the construction industries. Sixty-five cost overrun factors are identified and classified into four clusters (project macro, project management, project environment, and core stakeholders) through a detailed literature review process and a discussion with experts from the Chinese construction industry. A questionnaire survey was conducted for data collection to calculate an index of the project-influenced factors and clusters in the construction industry in China. With the help of the proposed model, it is possible to guide project managers and decision makers to make better informative decisions such as project macro, project management, project environment, and core stakeholders.

No abstract