Dam Inspection Research Articles

A Comprehensive Survey on Visual Perception Methods for Intelligent Inspection of High Dam Hubs.

There are many high dam hubs in the world, and the regular inspection of high dams is a critical task for ensuring their safe operation. Traditional manual inspection methods pose challenges related to the complexity of the on-site environment, the heavy inspection workload, and the difficulty in manually observing inspection points, which often result in low efficiency and errors related to the influence of subjective factors. Therefore, the introduction of intelligent inspection technology in this context is urgently necessary. With the development of UAVs, computer vision, artificial intelligence, and other technologies, the intelligent inspection of high dams based on visual perception has become possible, and related research has received extensive attention. This article summarizes the contents of high dam safety inspections and reviews recent studies on visual perception techniques in the context of intelligent inspections. First, this article categorizes image enhancement methods into those based on histogram equalization, Retinex, and deep learning. Representative methods and their characteristics are elaborated for each category, and the associated development trends are analyzed. Second, this article systematically enumerates the principal achievements of defect and obstacle perception methods, focusing on those based on traditional image processing and machine learning approaches, and outlines the main techniques and characteristics. Additionally, this article analyzes the principal methods for damage quantification based on visual perception. Finally, the major issues related to applying visual perception techniques for the intelligent safety inspection of high dams are summarized and future research directions are proposed.

High-detail and low-cost underwater inspection of large-scale hydropower dams

Abstract. The article presents a practical method that combines low-cost camera systems with remotely operated vehicles (ROVs) to accomplish a comprehensive but economically feasible underwater survey of large hydropower infrastructures. Typically, inspecting reservoirs entails draining them off to allow for visual inspections, which are time-intensive, pose risks to operators' safety and are associated with generation losses. In this regard, ROVs are a much safer and more efficient alternative to traditional methods. The study was conducted at the Pack reservoir in Austria, where a reference framework was set up using terrestrial laser scanning and checkerboard markings for the above-water components. A ROV equipped with a GoPro camera and lighting system for the underwater recordings has been employed. Via a close-range photogrammetric approach, it was possible to generate 3D point clouds of the submerged infrastructure with a survey-grade accuracy level. Various strategies were explored to perform bundle block adjustment (BBA), among these were strategies where ground control points (GCPs) were used, strategies without the use of GCPs but pre-calibrated initial camera parameters and strategies with a combination of using both GCPs and pre-calibrated camera parameters in the BBA. The deployment of an inspection technique using low-cost sensors that can generate highly detailed three-dimensional models of submerged infrastructure areas is presented and discussed, allowing easy detection and localization for maintenance inspection, all while being cost-effective. The paper strengthens the suggestion of best practices that optimize camera settings, considering the effect of electronic image stabilization, suggesting its avoidance, and using advanced calibration methods.

CNN-Transformer hybrid network for concrete dam crack patrol inspection

Regular patrol inspection of concrete dams can detect cracks at an early stage. However, conventional crack segmentation models based on deep learning (DL) are difficult to be deployed in resource-constrained mobile devices due to the large number of parameters. This paper describes a lightweight semantic segmentation model, termed as CrackTrNet, for images of concrete dam cracks. CrackTrNet is a hybrid U-shaped model based on convolutional neural network (CNN) and Vision Transformer. The CNN is adopted to extract low-level visual features and the Transformer focuses on learning the global contextual information. The results demonstrate that its segmentation accuracy can reach 97.60%, while the model size is only 34.86 MB, which is 66.12%–87.85% lower than that of current mainstream DL-based models. To make the model more practical, a crack inspection mobile application (APP) is developed using Android Studio. The integration of lightweight CrackTrNet and APP can effectively assist the intelligent inspection of dam cracks to ensure structural safety.

Evaluation of Probable Maximum Precipitation (PMP) for West Virginia to Predict Current and Future PMP

Probable maximum precipitation (PMP) is essential for dam design criteria. West Virginia Department of Environmental Protection Dam Safety utilizes the 6-hr, 10-mi2 (26-km2) PMP established by Hydrometeorological Report 51 based on the analysis of extreme rainfall. These PMP are outdated because there have been unaccounted for extreme rainfall events that have occurred since 1978. In addition, a zone of uncertainty around the Appalachian Mountains exists, including West Virginia (WV). The objective of this research was to determine the PMP throughout WV by studying the area of uncertainty and consider rising dew points to predict future PMP. This study created an updated PMP map of WV, focusing on the maximization of the 1942 Smethport, PA storm. A watershed-specific analysis was conducted at all dams in the state with an approximate watershed area of 10-mi2 (26-km2). Current PMP estimations and projections for 2100 were determined. Iso-lines were determined to develop the new updated PMP map for WV that displays current and future PMP estimations. Results suggest that the greatest values of PMP occur in the western region of the state; lowest values of PMP occur in the south-eastern part of the state. The Smethport, PA storm (1942) was considered the governing storm for the entire state; however, it may not apply to the eastern panhandle and Potomac highlands. These regions should be considered for additional analysis. The updated PMP map will be a resource for those involved in dam inspection, repair, design, and modification.

Improved ant colony optimization for safe path planning of AUV

In order to address the autonomous underwater vehicle navigation challenge for dam inspections, with the goal of enabling safe inspections and reliable obstacle avoidance, an improved smooth Ant Colony Optimization algorithm is proposed for path planning. This improved algorithm would optimize the smoothness of the path besides the robustness, avoidance of local optima, and fast computation speed. To achieve the goal of reducing turning time and improving the directional effect of path selection, a corner-turning heuristic function is introduced. Experimental simulation results show that the improved algorithm performs best than other algorithms in terms of path smoothness and iteration stability in path planning.

A New Method for the Compaction Quality Inspection of High Rockfill Dams Based on 3D Laser Scanning Technology

The compaction quality is directly related to the deformation and stability of the rockfill dam. Measuring the test pit volume efficiently and accurately is the most critical step during the compaction quality inspection. A new method for calculating the test pit volume based on point cloud data is proposed. An auxiliary device that can change the scanning distance and angle of the handheld 3D laser scanner is developed to collect the initial point cloud. The segmentation method of the initial point cloud data including the test pit and the compaction surface outside the pit is to divide the data into two parts according to the order number of the segmentation point, after slicing and sorting point clouds, which is the key to ensuring the computational precision. The segmentation points are the adjacent two points with the greatest order number difference in these point clouds whose distance from the line connecting the end points of the slicing point clouds is less than dz. The compaction surface point clouds are used to construct a plane by the least-squares algorithm so that the closed three-dimensional model is formed by registering it with the test pit point clouds. After converting the test pit surface to the horizontal plane by the Rodrigues formula, the test pit point clouds are divided into n2 parts with equal projection areas on the horizontal plane, and n2 prisms are constructed using them and their projection areas. The test pit volume is the sum of the intersection space volumes of all prisms and the test pit model, and the intersection space is determined by comparing the Z-values of point clouds. The new method was programmed in MATLAB and applied to the Shuangjiangkou rockfill dam with a height of 315 m. The relative error of volume results between the new method and the old water-filling method is 0.14–2.31%. The cause of the error is analyzed, and it is proved that the results of the new method are closer to the real volume of the test pit in theory. This method reduces the inspection cost but greatly improves the level of precision, efficiency, and intelligence for compaction quality inspection.

Study on the Estimation of Concrete Defects Volume on Dam Body Surface

To maintain the safety and functionality of dams over the long term, it is necessary to make inspections more labor-saving and efficient using the latest technology and to improve the sophistication of inspections based on data. Although dam inspections cover a wide range of items, this study focuses on the continuous monitoring of popouts, a phenomenon of concrete deterioration occurring on the surface of a dam body. It is difficult to predict whether a popout will occur from information on the body surface of the dam, owing to the generation mechanism of the popout. The number of popouts was monitored over time; however, no examples of shape changes were monitored over time. Advancements in various digital technologies are required to accurately evaluate changes in the dam body's surface over time; therefore, in this study, three-dimensional (3D) point-cloud data is created by the Structure from Motion (SfM) from images captured by a Unmanned Aerial Vehicle (UAV) of the concrete defect area due to the popout in an arch dam in the Tohoku region of Japan. The volume of concrete defects of a popout in each of two different periods was calculated by estimating the plane shape of the surface of the dam body. In addition, the shapes of two popouts were compared to confirm the possibility of predictive signs of change.

The role of artificial intelligence and digital technologies in dam engineering: Narrative review and outlook

This narrative review paper explores the diverse applications of artificial intelligence (AI) in the field of dam engineering. Authored by research engineers specializing in civil engineering and data science, and reviewed by experienced dam engineers and AI experts, the paper aims to provide a comprehensive overview of the subject. The paper is structured into three parts. Part 1 offers a concise introduction to AI, covering its historical background, major categories, and current state of development. In Part 2, the focus shifts to the specific applications of AI in hydropower and dam engineering. This section delves into the utilization of AI for predictive modeling, real-time monitoring, optimization, and planning and design. Notable case studies and examples of machine learning techniques applied to predictive models and numerical simulations in the context of dams and levees are also discussed. Part 3 explores the current and emerging technologies being implemented in the industry, including automated decision-making systems and the use of AI-powered drones for efficient dam inspection. Additionally, the section sheds light on the challenges that must be addressed to fully integrate AI into dam engineering practices. This paper targets dam engineers, AI experts, and individuals interested in the intersection of these fields. Its primary contribution lies in providing a comprehensive and up-to-date review of the applications, challenges, and potential future developments of AI in dam engineering. Over 3300 articles on the intersection of dam engineering and AI were identified, and approximately 550 articles (about 17%) were further processed for inclusion in this study. Moreover, it emphasizes the need for robust data management, algorithmic transparency, and ethical considerations in the implementation of AI systems.

Semi-universal geo-crack detection by machine learning

Introduction: Cracks are a key feature that determines the structural integrity of rocks, and their angular distribution can be used to determine the local or regional stress patterns. The temporal growth of cracks can be monitored in order to predict impending failures of materials or structures such as a weakened dam. Thus, cracks and their spatial-temporal distributions should be automatically monitored for assessing their structural integrity, the associated stress patterns and their potential for failure.Method: We show that the U-Net convolutional neural network, semantic segmentation and transfer learning can be used to accurately detect cracks in drone photos of sedimentary massifs. In this case, the crack distributions are used to assess the safest areas for tunnel excavation. Compared to the coarse performance of ridge detection, the U-Net accuracy in identifying cracks in images can be as high as 98% when evaluated against human identification, which is sufficient for assessing the general crack properties of the rock faces for the engineering project.Result: Based on approximately 100 h of manual cracks labeling in 127 drone photos and 20 h of network training, the U-Net was able to successfully detect cracks in 23,845 high-resolution photographs in less than 22 h using two Nvidia V100 GPUs. Meanwhile, the network was able to detect more than 80% of the observable cracks of a volcanic outcrop in Idaho without additional training. With a modest amount of extra labeling on photos of the volcanic outcrop and transfer training, we found that the accuracy significantly improved. The surprising outcome of this research is that the U-Net crack detector laboriously trained on photos of sedimentary rocks can also be effectively applied to photos of volcanic rock faces. This can be important for real-time assessment of geological hazards and lithology information for dam inspection and planetary exploration by autonomous vehicles. For another application, we accurately detected fractures and faults with a scale of tens of kilometers from Martian photographs.Conclusions: In summary, our methodology of using CNN with transfer training suggests that it can be used as a semi-universal detector of cracks in across a range of diverse geological settings.

Structural health monitoring and inspection of dams based on UAV photogrammetry with image 3D reconstruction

This study presents a three-dimensional (3D) reconstruction model based on unmanned aerial vehicle (UAV) images for dam emergency monitoring and inspection. The structure from motion method is adopted to generate a high-precision 3D dam model with scene geometry by matching the features in multiple overlapping images. Then, the generated model is used to calculate the 3D point coordinates and camera pose through the bundle adjustment system. Effect of ground control points on the model accuracy is analyzed by different position sets for adaptation to various situations. The proposed emergency monitoring model is verified on two rock-fill dams, inspection model is utilized on a concrete dam, and damage detection method is achieved on a small-scale dam model with loadable panels. Results show that the proposed 3D dam reconstruction model based on UAV images can obtain satisfactory accuracy and result in a significant improvement in the dam monitoring and inspection efficiency.

Development of Dam Safety Remote Monitoring and Evaluation System

More than 4,000 dams are constructed in Thailand for several purposes, including water supply, flood control, irrigation, and hydropower generation. Among these dams, 14 large dams are operated by the Electricity Generating Authority of Thailand (EGAT). As a dam operator, EGAT is committed to ensuring dam safety by regularly conducting dam inspections and maintenance. This paper presents the development and practical applications of the Dam Safety Remote Monitoring System (DS-RMS). The objective of DS-RMS is to enhance the EGAT’s implementation of its dam safety program in terms of dam monitoring by instrumentation to satisfy international recommendations. DS-RMS consists of five subsystems: Dam Behavior, Reservoir Operation, Earthquake Monitoring, Expert System and Public Communication. DS-RMS has been deployed at 14 large EGAT-operated dams across the country since 2016. Results show that the novel features of DS-RMS enable faster and more reliable dam safety monitoring and evaluation processes.

SUAS Remote Sensing for Closed-canopy Tree Inventory on Earthen Dams

More than half of the U.S. dams are privately owned and experienced the overgrowth of trees. There is a need to improve dam inspection and maintenance in a timely manner. Small Unmanned Aircraft Systems (sUAS) have been increasingly utilized for near-surface landscape mapping and reconnaissance. This study tests an sUAS protocol of closed-canopy tree survey on earthen dams. A DJI Matrice 100 flight was launched on September 22, 2020. The orthoimage and 3D point cloud are extracted, and the canopy height model is built. Treetops and crowns are delineated using an integrated watershed segmentation and image segmentation procedure. The results include a tree survey inventory that contains the locations, tree heights and crowns of 284 trees growing on the downslope of the dam. Given the flight flexibility and fine 3D details acquired from inexpensive drones, sUAS has a high potential for assessing tree overgrowth toward remediation solutions of earthen dams.

A Robotic Cognitive Architecture for Slope and Dam Inspections.

Big construction enterprises, such as electrical power generation dams and mining slopes, demand continuous visual inspections. The sizes of these structures and the necessary level of detail in each mission requires a conflicting set of multi-objective goals, such as performance, quality, and safety. It is challenging for human operators, or simple autonomous path-following drones, to process all this information, and thus, it is common that a mission must be repeated several times until it succeeds. This paper deals with this problem by developing a new cognitive architecture based on a collaborative environment between the unmanned aerial vehicles (UAVs) and other agents focusing on optimizing the data gathering, information processing, and decision-making. The proposed architecture breaks the problem into independent units ranging from sensors and actuators up to high-level intelligence processes. It organizes the structures into data and information; each agent may request an individual behavior from the system. To deal with conflicting behaviors, a supervisory agent analyzes all requests and defines the final planning. This architecture enables real-time decision-making with intelligent social behavior among the agents. Thus, it is possible to process and make decisions about the best way to accomplish the mission. To present the methodology, slope inspection scenarios are shown.

THE KA LOKO TRAGEDY

The North Shore of Kauai is known for its beautiful beaches, scenery, and tropical paradise. However, the community has faced adversity and tragedy due to the lack of necessary engineering inspections, failure to apply for building permits, negligence, and complacency on behalf of property owner James H. Pflueger, the State of Hawaii, and County of Kauai. The Ka Loko dam broke on March 14, 2006 and the rushing waters wreaked havoc on the community causing mass destruction and ultimately claimed the lives of seven individuals. Multiple civil lawsuits and a criminal case followed where the owner pled no contest to seven counts of manslaughter and one count of reckless endangerment. The purpose of this paper is to conduct a brief analysis of the events surrounding the tragedy and extract the lessons learned and impacts it may have to prevent future disasters from occurring. Among the issues, it was discovered that the emergency spillway had been filled with earth, and that dam inspections were irregularly conducted.

Innovative methods of concrete dams’ inspection

A methodology for the operation of remote controlled uninhabited vehicles during the inspection of concrete dams is proposed. The main elements of concrete dams and the damages arising in them are indicated.

Development of dam inspection robot with negative pressure effect plate

AbstractThis paper describes the development of an underwater robot that performs visual inspection while making mechanical contact with a dam surface by a pulling force generated from thrusters with negative pressure effect plates. In general, small and lightweight remotely operated vehicles (ROVs) have low‐power thrusters. Their positioning performance is inferior with respect to external disturbances. Moreover, it is difficult for untrained operators to control the ROV position to check and inspect dams visually. A negative pressure effect plate attached to a thruster produces negative pressure that maintains mechanical contact between the robot and a dam wall surface and ensures stable robot motion on a dam surface to acquire clear continuous images. As described herein, we theoretically investigate the negative pressure effect plate characteristics and experimentally measure the force generated by the pressure difference. Results show that the force of a thruster with the effect plate is five times greater than the nominal thrust force provided by the thruster alone. Based on those results, we designed and developed the underwater robot with the negative pressure effect plates for dam inspection. Moreover, we conducted an experiment in a water tank and a field test at Shorenji Dam, Mie‐prefecture, Japan. The experimentally obtained results indicated that the negative pressure effective plate is effective at generating sufficient force and at realizing stable robot motion on the dam surface. Results demonstrate that the developed robot acquired clear images of the dam surface continuously with no sophisticated operator or controller.

Dynamic BIM-augmented UAV safety inspection for water diversion project

As indispensable infrastructure for supplying industrial and residential water, structural failure to water diversion projects will result in unmeasurable loss. Safety inspection is critical for the operation of these projects. Traditional manual inspection is inefficient, has limited coverage, and cannot collaborate the off-site decision-makers with field workers to evaluate the detected abnormal phenomena. In addition, building information modeling (BIM), which is an intuitive and data-driven information platform, has not been integrated with the existing safety inspection procedure for better understanding the detected risk issues. To tackle these drawbacks, a safety inspection method that integrates unmanned aerial vehicle (UAV) and dynamic BIM is proposed in this paper. An overall workflow for dynamic BIM-augmented UAV safety inspection is developed, which consists of data collection, dynamic BIM construction, and UAV-BIM coupling. A dynamic BIM model is created by aggregating timely-updated safety information with a BIM model in the Web environment. The synchronous navigation of UAV video and dynamic BIM is realized by matching the virtual camera parameters with the real ones. The field experiment implemented in Tianjin, China demonstrates the efficacy of the proposed method, which can not only improve the safety inspection efficiency, but also enable the off-site managers to view the inspection video, and make timely and comprehensive safety evaluation with the support of dynamic BIM. Although the proposed method is introduced from the perspective of water diversion projects, it can also be applied to the inspection of other large-scale infrastructures, e.g. pavements, dams, and railways.

Insights into the Oroville Dam 2017 Spillway Incident

In February 2017, a failure occurring in Oroville Dam’s main spillway risked causing severe damages downstream. A unique aspect of this incident was the fact that it happened during a flood scenario well within its design and operational procedures, prompting research into its causes and determining methods to prevent similar events from reoccurring. In this study, a hydroclimatic analysis of Oroville Dam’s catchment is conducted, along with a review of related design and operational manuals. The data available allows for the comparison of older flood-frequency analyses to new alternative methods proposed in this paper and relevant literature. Based on summary characteristics of the 2017 floods, possible causes of the incident are outlined, in order to understand which factors contributed more significantly. It turns out that the event was most likely the result of a structural problem in the dam’s main spillway and detrimental geological conditions, but analysis of surface level data also reveals operational issues that were not present during previous larger floods, promoting a discussion about flood control design methods, specifications, and dam inspection procedures, and how these can be improved to prevent a similar event from occurring in the future.

Development of dam inspection underwater robot

Development of dam inspection underwater robot

Development of Dam Inspection Underwater Robot

Development of Dam Inspection Underwater Robot