Professional Documents
Culture Documents
https://doi.org/10.1093/iti/liad020
Research Article
*Corresponding author: Department of Geotechnical Engineering, China Institute of Water Resources and Hydropower Research, Beijing 100038, China.
Email: zhaoyf@iwhr.com
Abstract
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. This paper provides a comprehensive overview of the key
technologies employed in the intelligent construction process of earth–rock dams. Starting with the multidimensional perception tech-
nologies and equipment utilized in earth–rock dam filling construction, it covers various aspects, such as the fine-grained management
of unit construction based on building information modeling, rapid detection and qualification testing of dam materials’ physical
characteristics, fast perception technology for material compaction, and unmanned driving technology for dam compaction equipment.
Additionally, this paper highlights an overview of the existing technical challenges and future prospects in intelligent construction for
earth–rock dams. These research results provide important references and lessons for the construction and development of high earth–
rock dam engineering.
Keywords: artificial intelligence, rolled earth–rock dam, intelligent detection of gradation, BIM model, rapid assessment on compaction
quality.
Received: June 29, 2023. Revised: October 16, 2023. Accepted: October 16, 2023
© The Author(s) 2023. Published by Oxford University Press and Southwest Jiaotong University.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
2 | Liu et al.
the digital dam intelligent compaction monitoring system in the focused on several key areas, including fine management of dam
Nuozhadu Hydropower Station also signifies a new stage in the construction using building information modeling (BIM)-based
development of digital dam intelligent compaction monitoring unit engineering, rapid detection and qualification testing of dam
systems. Chen et al. (2019) have developed unmanned driving material’s physical characteristics, quick sensing of compaction
technology for compaction in dam construction that does not degree for dam materials, and unmanned driving technology for
require changes to the construction machinery’s hydraulic circuit, dam compaction construction equipment. These advancements
electrical control system, and mechanical structure. Combined have greatly enhanced the efficiency and effectiveness of dam
Figure 3. Display of the actual construction of the downstream gravel and rock fill layer.
framework for object detection and U-Net network for semantic practical effectiveness. However, the advantages and disadvan-
segmentation. Attention mechanisms were applied in both the tages of applying these methods to solve particle size distribution
detection and segmentation modules to focus on task-related in actual engineering scenarios for soil and rock materials in dam
features, providing a new approach for visual recognition and construction have not been thoroughly discussed and require
statistical analysis of rock fragment size distribution during tun- further analysis and exploration.
nel boring machine excavation. Regarding size distribution trans-
formation, Wang et al. (2007) achieved fast detection of coarse
aggregate 3D digital image grading by introducing the particle 2.3 Fast perception technology for material
thickness-to-width ratio and conversion coefficients between cir- compaction
cular and square sieves. The proposed method demonstrated high In response to the limitations of direct methods for measuring
accuracy and universality for different types of rock materials and compaction density, many researchers have proposed the use of
gradations. Lyu et al. (2019) utilized the independently developed compaction sensing technology, specifically the use of on-board
SMAS digital image analysis system to perform microstructure compaction monitoring devices, as an indirect method to assess
analysis on collected images of sand particles. They used the the compaction characteristics of dam materials in real-time.
particle equivalent diameter (equivalent area circle diameter) This approach aims to improve the efficiency of quality control in
and particle area to equivalently represent the particle size and earth–rockfill construction projects. The research on compaction
mass of actual soil particles, achieving particle size distribution sensing technology initially started in the context of quality
analysis through transformation. Zhang et al. (2013) found that control for road subbase materials, and the corresponding detec-
using the short axis of an equivalent ellipse and the width of the tion equipment and techniques have been applied in highway
best-fit matrix are the most suitable parameters for representing construction for several decades. Researchers install monitoring
the particle size of coarse particles, with an accuracy of over devices on compaction machinery to collect signals related to
85 per cent. Existing research reveals that many scholars have the vibration characteristics of the compaction roller, such as
established relationships between particle planar morphologi- harmonic frequencies and accelerations. These signals are then
cal features and 3D volumetric features to achieve particle size analyzed and processed from different perspectives to obtain
distribution transformation and quantification. Various methods monitoring indicators that assess the real-time compaction
for quantitative solution exist, all of which have shown certain state of the materials, such as subgrade soil, granular soil, and
Key technologies for intelligent and fine-grained construction | 5
asphalt (Thompson and White 2008; Vennapusa et al. 2010; continuous compaction control acoustic detection system. The
Vennapusa and White 2016). With the advancement and abovementioned researches started mainly with the study of sub-
integration of modern sensor technology, computer technology, grade filling material, where the technology has been relatively
and GPS technology, it has become possible to collect, transmit, mature, and are subsequently applied to evaluate the compaction
and visualize various compaction indicators in real-time. This assessment of earth–rock dam materials. The accuracy of the
has led to the development of Roller-Integrated Compaction compaction quality evaluation is high for fine-grained materials.
Monitoring (RICM) technology. RICM technology serves as a tool In the presence of large-sized and hard dam materials in the
real-time processing of sensor information, and multisensor for aggregate grading characteristics is limited. Additionally, dig-
information fusion technology), auto-driving-based intelligent ital images only ref lect the exposed surface of the earth–rock
compaction for dam construction has gradually been developed materials after spreading. The sizes of the earth–rock materials
and applied in engineering projects (Zhang et al. 2019). obtained through planar image recognition, as well as the derived
By utilizing a combined Browser/Server (B/S) and Client/Server aggregate grading curves, may deviate from the actual results
(C/S) technological architecture, the reliability of autonomous obtained through excavation and sieve analysis. Therefore, how
driving technology in dam compaction construction has been to utilize the particle size composition characteristics obtained
on earth–rock dam materials with a larger particle size distri- of relevant intelligent standards and specifications, the level of
bution range (0–400 mm). Additionally, the relationship between smart dam construction will reach new heights.
compaction quality and various factors is complex and ambigu-
ous. Existing evaluation models inadequately consider the factors
inf luencing compaction quality, leading to poor universality of the Conflict of Interest statement
models. Furthermore, when evaluating the compaction quality The authors declare that they have no known competing financial
of earth–rock dam materials, limitations of the evaluation model interests or personal relationships that could have appeared to
Compaction Monitoring Technology’, Automation in Construction, Measurements’, Journal of Geotechnical and Geoenvironmental Engi-
44: 234–46. neering, 136: 813–22.
Lyu, C. et al. (2019) ‘Gradation Analysis of Sand Particles Based on Wang, H. N. et al. (2007) ‘Investigation into Grading Characteristic of
Digital Image Processing Technology’, Geological Journal of China Coarse Aggregate Via Digital Image Processing Technique’, Journal
Universities, 25: 431–6 (in Chinese). of South China University of Technology(Natural Science Edition), 35:
Mgangira, M. B., Anochie-Boateng, J., and Komba, J. J. (2013)‘ 54–58+62.
Quantification of aggregate grain shape characteristics using White, D. J. et al. (2005) ‘Real-Time Compaction Monitoring in Cohe-