The spatial and temporal distributions of water bodies are of great significance for monitoring the use of water resources. The most common remote sensing applications are classifying the surface cover types and analyzing changes in the spatial distributions of these water resources. In recent years, satellite remote sensing data has been used to extract key features such as the location, area, morphology and river width especially for large water bodies or for those in inaccessible regions in high mountains. This not only saves manpower and promotes safety, but also improves work efficiency. This study analyzes methods for using satellite remote sensing data for water body information extraction with reviews of the reflection characteristics of water bodies in various spectral regions of the electromagnetic spectrum; water extraction methods based on radar and optical remote sensing data at home and abroad since 1980; the working principles, advantages and disadvantages of various water information extraction methods; and some challenges of water information extraction applying remote sensing and the key issues to solve these problems. Finally, this paper forecasts future applications of remote sensing for water body information extraction.

The green energy revolution is leading to power systems with high penetrations of renewable energy sources and high penetrations of inverter-interfaced devices. The dynamic characteristics of these power systems are very different from conventional systems, which has led to many new stability problems. This paper focuses on large-disturbance stability issues and the large-disturbance instability phenomena that appear in these “double-high” power systems. This paper also discusses the shortcomings of existing methods and upcoming challenges. Finally, this paper presents our perspectives on future theoretical models and methods to provide large-disturbance stability in “double-high” power systems.

Intelligent construction is one of the most popular research directions in the field of infrastructure engineering for various fields including hydropower and civil engineering projects. Intelligent construction has been used in China in recent years and is essential for the future development of engineering systems and project management. This study defines intelligent construction as the integration of sensors, communication systems, data systems, construction methods and project management to perceive, analyze and control the safety, quality, environmental effects, schedule and construction costs. This article then describes the main characteristics of intelligent construction systems. Current experience shows that the general closed-loop control theory of intelligent construction including perception, analysis, control and continuous optimization leads to current construction activities forming new construction activities from quantitative descriptions and learning from existing activities. This intelligent behavior provides better control of the construction process. Finally, the integration of intelligent construction methods with management decisions leads to new value creation with new intelligent construction methods needed to further develop construction techniques.

The wide-band oscillation problem caused by the high penetration of renewable energy sources and the high proportion of power electronic equipment, “double-high” systems, seriously affects normal power equipment operation and power system stability. Such systems require on-line monitoring and analyses of the multi-mode and time-varying wide-band oscillation frequency. Traditional wide-area measurement systems (WAMS) can monitor low-frequency oscillations in power grids in real-time. However, they cannot monitor wide-band electromagnetic oscillations. This paper presents a wide-area measurement and early warning system (WAMWS) for monitoring wide-band oscillations in “double-high” power systems. This system has all the functions in the existing WAMS while monitoring wide-band oscillations in “double-high” power systems. The warning system provides wide-band state estimates, oscillation source identification, and security and stability evaluations of the wide-band oscillations. The effectiveness of this system for monitoring wide-band oscillations is verified in simulations. Finally, this paper considers applications of WAMWS.

Four large cascade hydropower stations have been constructed or are under construction along the downstream section of the Jinsha River in southwest China. The complex geological conditions and intensive tectonic activity in this region have created a number of challenging rock mechanics problems for these dams. This study used the experience accumulated during the construction of these four hydropower stations, the general layouts of the hydro-power projects, and analyses of the dam foundations to review key problems related to the rock mechanics, the excavation and prevention of large underground caverns, the steep slope stability and precision blasting methods used in these projects. This study also summarizes the problems and key methods used to solve those problems. The successes of these four hydropower stations benefited from a guiding ideology of understanding, utilization, protective monitoring and feedback on the surrounding rock masses during construction period, and an engineering procedure of the excavation, analysis, inspection, and prediction of each rock layer. These successful experiences can guide similar large rock engineering projects.

Safety resilience is an important field in urban safety research. This paper defines the level of urban safety resilience based on a resilience curve and presents an urban safety resilience model that includes an urban structure model, urban safety resilience model, emergency model and urban recovery model. The model is used to analyze a virtual city including the architecture, traffic, power, communication and water supply sub-systems. The urban seismic safety resilience is then analyzed quantitatively using the Monte Carlo method. The model framework and method have good scalability and can provide support for constructing and assessing the safety resilient of a city.

Surveillance videos are important for maintaining social welfare. This paper classifies and summarizes the traditional and advanced video anomaly detection algorithms. First, the algorithms are classified into different classes according to their development stages, model categories and detection criteria and then they are summarized by class. Then, the advantages and the disadvantages of the different algorithms are identified by comparing the algorithms belonging to different classes. This paper specifically analyses the characteristics of the cluster criterion and the reconstruction criterion in different development stages. Finally, this paper identifies the commonly used model assumptions and the domain knowledge and summarizes the accuracies of the various algorithms. Future research directions are also discussed.

Data cleaning is the process of detecting and repairing dirty data which is often needed in data analysis and management. This paper classifies and summarizes the traditional and advanced data cleaning techniques and identifies potential directions for further work. This study first formally defines the cleaning problem for structured data and then describes error detection methods for missing data, redundant data, conflicting data and erroneous data. The data cleaning methods are then summarized based on their error elimination method, including constraint-based data cleaning, rule-based data cleaning, statistical data cleaning and human-in-the-loop data cleaning. Some important datasets and noise injection tools are introduced as well. Open research problems and future research directions are also discussed.

Microblog sentiment analysis is used to get a user's point of view. Most sentiment analysis methods based on deep learning models do not use emotion symbols. This study uses a double attention model for microblog sentiment analysis that first constructs a microblog emotion symbol knowledge base based on existing emotional semantic resources including emotion words, degree adverbs, negative words, microblog emoticons and common Internet slang. Then, bidirectional long short-term memory and a full connection network are used to encode the microblog text and the emotion symbols in the text. After that, an attention model is used to construct the semantic representations of the microblog text and emotion symbols which are combined to construct the final semantic expression of the microblog text. Finally, the emotion classification model is trained on these semantic representations. The combined attention model and emotion symbols enhance the ability to capture the emotions and improve the microblog sentiment classification. This model gives the best accuracy for many sentiment classification tasks on the Natural Language Processing and Chinese Computing (NLPCC) microblog sentiment analysis task datasets. Tests on the 2013 and 2014 NLPCC datasets give F₁-scores for the macro and micro averages that are 1.39% and 1.26% higher than the known best model for the 2013 dataset and 2.02% and 2.21% higher for the 2014 dataset.

[Significance] Landslide hazards are widely distributed in China and are severely harmful. The registered landslide hazards have achieved remarkable benefits in disaster reduction through a comprehensive prevention and control system. However, approximately 80% of all geo-disasters in China still occur outside the scope of identified hazards yearly. Therefore, monitoring and early warning are important means to actively prevent landslide disasters and achieve great success in disaster mitigation owing to promptness, effectiveness, and relatively low-cost advantages. Deformation is the most significant monitoring parameter for landslides and has become a focus and general trend. Landslide deformation monitoring engineering has strict requirements for controlled cost and high reliability to achieve widespread application and accurate early warning. Therefore, the commonly used monitoring instruments focus on surface deformation and rainfall to meet the requirements for easy equipment installation and low implementation cost. However, surface deformation and rainfall are not sufficient conditions to determine the occurrence of landslides. Various challenges exist in the existing monitoring technologies and early warning methods regarding engineering feasibility and performance improvement. Thus, it is important and urgent to summarize the existing research to rationally guide future development.[Progress] The deformation monitoring methods are divided into surface and subsurface monitoring. Most surface deformation monitoring technologies are vulnerable to the interference of terrain, environment, and other factors; therefore, their timeliness and reliability are not easily guaranteed. Additionally, slope subsurface deformation monitoring technologies can directly obtain the development and damage information of the sliding surface; thus, they can recognize the disaster precursor. Subsurface monitoring has advanced early warning ability; however, the existing instruments have problems, such as high cost, small measuring range, or difficult operation. Acoustic emission technology has the advantages of low cost, high sensitivity, and continuous real-time monitoring of large deformation, and has gradually developed into an optional method for landslide subsurface deformation monitoring. Thus, efficient landslide monitoring should comprehensively use multiple technologies to overcome the limitations of a single technology, and an integrated monitoring system becomes the state-of-the-art trend. The purpose of landslide monitoring is to provide a basis for decision-making of disaster early warning, thus, avoiding casualties and property losses through effective early warning efforts. In the field of early warning, regional meteorological and individual landslide early warning methods are gradually developed and improved. Deformation monitoring data are the main basis for landslide early warning, and experts analyze the deformation trend and sudden change characteristics. Different early warning levels could be triggered by the threshold values of velocity, acceleration, or other criteria. However, a landslide has complex dynamic mechanisms and individual differences; thus, the generic early warning model needs further exploration. The intelligent early warning model integrates machine learning technology with geological engineering analysis to improve the accuracy and automation level of landslide early warning.[Conclusions and Prospects] Deformation monitoring is essential in landslide prevention, and deformation data are the main basis for landslide early warning. Moreover, surface monitoring technologies have been widely used in the perception and decision-making process of landslides. Subsurface monitoring technologies can detect early precursors of landslide evolution to continuously improve early warning accuracy. Analyses show that early warning methods can be improved in the future by integrating machine learning models and geotechnical engineering.

After the corona virus disease 2019 (COVID-19) outbreak in late 2019, civil engineering will play important roles in the urban comprehensive, resilient disaster and pandemic prevention construction. This paper combines the demand for urban pandemic prevention and control in China with the advantages of urban underground spaces for disaster prevention and pandemic responses in a comprehensive development plan. Emergency responses and pandemic control in urban underground spaces can provide safety, stability, adaptability to emergency medical conditions, low energy consumption underground emergency material storage, unified dispatching of emergency resources, and adaptability to urban development. An evaluation framework is given for urban underground space disaster and pandemic prevention to evaluate the resilience of such measures. The advantages of using urban underground spaces are illustrated by comparisons with other systems. The four topics, "stock", "increment", "variable" and "unified dispatching" in the disaster and pandemic prevention, are discussed to develop a resilience construction framework for urban underground spaces. Then, a comprehensive, resilient disaster and pandemic prevention construction framework for urban underground spaces is proposed to provide reference for the construction of resilient cities and urban underground spaces.

The legal field is using more artificial intelligence methods such as legal judgment prediction (LJP) based on case description texts using natural language processing. Charge prediction and law article recommendations are two important LJP sub-tasks that are closely related and interact with each other. However, previous studies have usually analyzed them as two independent tasks that are analyzed separately. Furthermore, charge prediction and law article recommendations both face the problem of confusing charges. To this end, this paper presents a multi-task learning model for joint modeling of charge prediction and law article recommendations. Confusing charges are handled by using a set of charge keywords extracted from case description texts using statistical techniques for integration into the multi-task learning model. This method was evaluated using the CAIL2018 legal dataset. The results show that incorporating the charge keywords into the multi-task learning model effectively resolves the confusing charge problem and significantly improves both the charge prediction and the law article recommendation results.

High dam construction is continuing to develop with new requirements for intelligent dam construction. New information technology capabilities are providing paths for improved intelligent dam construction. The key to achieving safe, quality, efficient, economic, green construction projects is to integrate these new information technology capabilities into intelligent construction methods. New systems enable intelligent construction of dams and the construction of intelligent dams. This article summarizes these two paths for intelligent construction, identifies three stages in the development of intelligent construction systems for dams, and analyzes the technical characteristics, goals, theory, methods, and management models with engineering examples for each stage of the intelligent construction process. The analysis shows the relationship between intelligent dam construction and intelligent dams, the three stages of intelligent dam construction, the changes in manager thinking for solving key problems in the intelligent era, and future developments in intelligent dam construction.