Article
Chemistry, Multidisciplinary
Huan Wang, Lixin Zhang, Jiawei Zhao, Xue Hu, Xiao Ma
Summary: This paper proposes a method for soil moisture and organic matter content detection based on hyperspectral technology. By using CNN and LSTM modules to extract features, optimizing network parameters with genetic algorithm, and combining grey correlation analysis, the prediction accuracy and computational efficiency are improved. The method outperforms other models in soil moisture and organic matter prediction, and has significant application value.
APPLIED SCIENCES-BASEL
(2022)
Article
Environmental Sciences
Guangcai Yin, Xingling Chen, Hanghai Zhu, Zhiliang Chen, Chuanghong Su, Zechen He, Jinrong Qiu, Tieyu Wang
Summary: A novel interpolation method based on genetic algorithm and neural network model (GANN model) was proposed to improve the prediction accuracy of soil heavy metals (HMs) by spatial interpolation. The results showed that the GANN model had a good prediction performance with relatively lower root mean square error values compared to other traditional interpolation methods.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Green & Sustainable Science & Technology
Zhuolong Jia, Changgen Yan, Bo Li, Han Bao, Hengxing Lan, Zherui Liang, Yuling Shi, Jing Ren
Summary: During the rainy season on the China Loess Plateau, Guar gum (GG) is used to stabilize loess slopes and control erosion. Lab experiments showed that GG can effectively improve the mechanical and hydraulic properties of loess, with the best results obtained at a GG content of 1.0%.
JOURNAL OF CLEANER PRODUCTION
(2023)
Article
Agriculture, Multidisciplinary
Yan Li, Songhua Yan, Jianya Gong
Summary: This study proposes a new Bayesian neural network framework that quantifies the uncertainty in retrieving soil moisture (SM) and ultimately reduces the uncertainty and improves accuracy using techniques such as Monte-Carlo dropout and Deep Ensembles.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Environmental Sciences
Ran Wang, Jianhui Zhao, Huijin Yang, Ning Li
Summary: Soil moisture is essential in meteorology, hydrology, and agricultural sciences. This study proposes a SSA-CNN model based on Sentinel-1 microwave remote sensing data and Sentinel-2 optical remote sensing data for inverting farmland soil moisture. The model achieved better accuracy compared to other machine learning approaches with an average coefficient of determination of 0.80, an average root mean square error of 2.17 vol.%, and an average mean absolute error of 1.68 vol.%. The inversion results using the SSA-CNN model demonstrated good performance in local situations.
Article
Environmental Sciences
Yang Sun, Xiaowei Gu, Xiaochuan Xu
Summary: The study found that vetiver is more advantageous for reinforcing the slope of the tailings reservoir, and the optimal root content of the root-soil composites is within the range of 0.3-0.4%. These results can provide important theoretical support for the comprehensive control and ecological restoration of a tailings reservoir slope.
ENVIRONMENTAL EARTH SCIENCES
(2021)
Article
Multidisciplinary Sciences
Yanqing Liu, Cuiqing Jiang, Cuiping Lu, Zhao Wang, Wanliu Che
Summary: This study proposed an optimized BP neural network model using an improved genetic algorithm (IGA) to predict soil nutrient time series with high accuracy. Empirical evaluation using annual soil nutrient data from China showed that the IGA-BP method accurately predicted soil nutrient content for future time series.
Article
Environmental Sciences
Nathan C. Dadap, Alexander R. Cobb, Alison M. Hoyt, Charles F. Harvey, Andrew F. Feldman, Eun-Soon Im, Alexandra G. Konings
Summary: When organic peat soils in Southeast Asian peatlands become dry, they become flammable and can lead to catastrophic fire events. This study used neural networks to model soil moisture and found that future climate change, including reduced precipitation and increased evaporative demand, will decrease soil moisture. This may accelerate peat carbon emissions and suggest that degraded areas with less tree cover are more vulnerable to climate change, highlighting the need for urgent peatland restoration.
ENVIRONMENTAL RESEARCH LETTERS
(2022)
Article
Environmental Sciences
Aifeng Lv, Zhilin Zhang, Hongchun Zhu
Summary: The study proposed a method to downscale soil moisture spatial resolution using a neural network and additional data, showing improved retrieval accuracy. The method demonstrated an increase in spatial and temporal correlations, as well as a decrease in root mean square error and mean absolute error. The analysis concluded that the neural network-based approach is promising for obtaining high-spatial-resolution soil-moisture data.
Article
Environmental Sciences
Soo-Jin Lee, Chuluong Choi, Jinsoo Kim, Minha Choi, Jaeil Cho, Yangwon Lee
Summary: Soil moisture is crucial for the global cycle of energy, carbon, and water, as well as plant growth and crop yield. Deep learning techniques have enabled accurate estimation of soil moisture, which can be used for drought and flood prevention, as well as assessing crop growth potential based on soil moisture conditions.
Article
Chemistry, Multidisciplinary
Yiwen Wang, Dongna Liu, Haiyu Dong, Junwei Lin, Qi Zhang, Xiaohui Zhang
Summary: Through stability evaluation, a landslide geological disaster can be identified and project safety and risk control can be ensured. An improved sparrow search algorithm is proposed to optimize the slope safety factor prediction model of a BP neural network, improving the traditional model and providing better accuracy. The improved model effectively predicts the safety factor of slopes under different conditions, providing a new technology for slope disaster warning and control.
APPLIED SCIENCES-BASEL
(2023)
Article
Green & Sustainable Science & Technology
Yuyan Liu, Fei Shi, Xuan Liu, Zihui Zhao, Yongtao Jin, Yulin Zhan, Xia Zhu, Wei Luo, Wenhao Zhang, Yuefang Sun, Xuqing Li, Yancang Wang
Summary: Soil moisture is a critical element of the Earth system, essential for studying the terrestrial water cycle, ecological processes, climate change, and disaster warnings. This study compared different training samples and used a BP neural network model to simulate the monthly surface soil moisture in China for 2019 and 2020, finding that evapotranspiration and precipitation have the greatest influence on soil moisture.
Article
Multidisciplinary Sciences
Dingming Yang, Zeyu Yu, Hongqiang Yuan, Yanrong Cui
Summary: A novel improved genetic algorithm is proposed in this paper, which has shown superior performance in terms of searchability, convergence efficiency, and precision compared to other mainstream swarm intelligence optimization algorithms. In addition, the algorithm also has great application potential in adversarial attacks.
Article
Multidisciplinary Sciences
Dingming Yang, Zeyu Yu, Hongqiang Yuan, Yanrong Cui
Summary: The choice of crossover and mutation strategies is crucial in genetic algorithms. A novel improved genetic algorithm was proposed in this paper, showing superior performance in global search ability, convergence efficiency, and precision in comparison to other mainstream swarm intelligence optimization algorithms. The algorithm performed significantly better in most tested functions and was successfully applied to neural network adversarial attacks.
Article
Computer Science, Artificial Intelligence
Ahmet Kara, Engin Pekel, Erdener Ozcetin, Gazi Bilal Yildiz
Summary: This research introduces an integrated deep learning-based framework for soil moisture prediction, which utilizes LSTM layers and an attention mechanism to extract dependencies and discriminative information from time series soil moisture data. The hyperparameters of the proposed network are determined using a genetic algorithm to enhance the prediction accuracy.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Mechanics
Dunwen Liu, Yu Tang, Min Cao, Jianjun Zhang, Qian Xu, Caiwu Cai
Summary: Field and rock-like materials model tests were conducted to analyze the cumulative damage of tunnel surrounding rock and watery fractured rock under multiple cycle blasting. The influence range of blasting on tunnel surrounding rock was found to be within 1.2 m, and nonlinear variations of cumulative damage were observed in watery fractured rock mass.
ENGINEERING FRACTURE MECHANICS
(2021)
Article
Chemistry, Analytical
Dunwen Liu, Wanmao Zhang, Yu Tang, Yinghua Jian, Chun Gong, Fengkai Qiu
Summary: A method for detecting and evaluating the uniformity of concrete surface coatings using infrared imaging and cluster analysis was proposed in this study. The method involves processing and analyzing infrared images to extract temperature distribution data of pixel points on the concrete surface. By combining cluster analysis and hierarchical analysis, an evaluation method for concrete surface coating uniformity was established.
Article
Chemistry, Analytical
Dunwen Liu, Haofei Chen, Yu Tang, Chao Liu, Min Cao, Chun Gong, Shulin Jiang
Summary: The rapid development of highway engineering has made slope stability an important issue in infrastructure construction. Long-term monitoring of high-slope micrometeorology has important practical significance for green vegetation growth, ecological recovery, landscape beautification, and the economy.
Article
Chemistry, Analytical
Dunwen Liu, Chun Gong, Yu Tang, Yinghua Jian, Kunpeng Cao, Haofei Chen
Summary: This paper presents an evaluation method of corrosion damage for sulfate-attacked concrete using CT and ultrasonic velocity testing. The coarse aggregate information was extracted using CT, and the proportion of coarse aggregate in the ultrasonic test line was calculated. The degree of sulfate corrosion in the concrete structure was evaluated using the analytic hierarchy process (AHP). The results show that this evaluation method can accurately assess the corrosion damage in sulfate-attacked concrete structures.
Article
Chemistry, Physical
Dunwen Liu, Wanmao Zhang, Yu Tang, Yinghua Jian, Yongchao Lai
Summary: Machine-made sand is a necessary choice for sustainable development in the concrete industry. This study investigates the performance of machine-made tuff sand concrete through orthogonal experiment and grey correlation analysis. The results show that the sand rate, mineral admixture, and water-cement ratio significantly affect the properties of the concrete. Comparing different types of fine aggregate concrete, the study finds that stone powder content has minimal effect on compressive strength.
Article
Chemistry, Physical
Yu Tang, Weichao Qiu, Dunwen Liu, Wanmao Zhang, Ruiping Zhang
Summary: This article examines the properties and replacement ratio of machine-made sand in tunnel slag machine-made concrete, and presents key findings through experiments and analysis, providing guidance for the quality control of machine-made sand concrete.
Article
Chemistry, Multidisciplinary
Dunwen Liu, Shulin Jiang, Yu Tang
Summary: The acoustic emission and wave velocity characteristics of metasandstone under graded cyclic loading and unloading were studied, revealing that the acoustic emission signals are mainly generated during loading and the wave velocity rapidly decreases when the stress exceeds a threshold. These findings can be used to predict the instability and damage of rock masses in engineering.
APPLIED SCIENCES-BASEL
(2022)
Article
Construction & Building Technology
Zhi Li, Xianqing Meng, Dunwen Liu, Yu Tang, Tan Chen
Summary: A comprehensive evaluation system based on the cloud model and analytic hierarchy process (AHP) theory was established to assess the geological disaster risk of super long tunnels. The evaluation was carried out on a super long highway tunnel in China. The results show that the evaluation system can accurately determine the disaster risk level of the tunnel under different conditions, and the cloud model theory addresses the issue of subjective evaluation indicators. This evaluation system can be used as a new approach for tunnel risk assessment in similar projects.
ADVANCES IN CIVIL ENGINEERING
(2022)
Article
Chemistry, Analytical
Dunwen Liu, Yinghua Jian, Yu Tang, Kunpeng Cao, Wanmao Zhang, Haofei Chen, Chun Gong
Summary: This study investigates the protective performance of silane coating on in-service concrete structures in a sulfate environment. Concrete samples are collected and subjected to accelerated erosion with wetting-drying cycles to simulate the erosion process. The samples are divided into protected, exposed, and control groups, representing corrosive environments with silane protection, corrosive environments without protection, and general environments. Various methods, including ultrasonic velocimetry, CT scan imaging, NMR pore structure analysis, and strength testing, are used to analyze the characteristics of the specimens during sulfate erosion. The research quantitatively analyzes the protective effect of silane coating on concrete structures under sulfate attack and proposes an index for judging the damage rate of specimens. The results show that silane coating can reduce the damage of concrete structures in a sulfate-attack environment by more than 50%.
Article
Chemistry, Analytical
Qiangqiang Zhang, Gonglian Dai, Yu Tang
Summary: This paper presents an analytical method to determine and decompose the temperature distribution of concrete slab track, and validates the method through experimental tests. The results show that solar radiation plays a significant role in non-linear temperature distribution, while atmospheric temperature has little effect.
Article
Geochemistry & Geophysics
Shan Wu, Weichao Qiu, Dunwen Liu, Yu Tang
Summary: This paper introduces an integrated geological forecasting system incorporating an industrial endoscope to detect watery karst areas in a tunnel. Numerical simulations were used to analyze the effects of different grouting thicknesses on settlement, bulge, and water surge in the tunnel. The practical application of grouting engineering demonstrated good grouting effect.
Article
Chemistry, Multidisciplinary
Dunwen Liu, Chong Wang, Yu Tang, Haofei Chen
Summary: In this study, theoretical analysis, field tests, and vibration monitoring were conducted to determine suitable rock mass cracking parameters for excavation of hard rock tunnels near historic sites using high-pressure gas expansion rock-cracking technology. The results showed that the desired rock mass cracking effect could be achieved using the cutting mode of central vertical empty hole + double wedge cutting hole and the auxiliary hole network parameter of 0.8 m x 0.7 m. The measured vibration velocity at a monitoring point 60 m away from the tunnel face in the field test was less than 0.1 cm/s, meeting the vibration control requirements of the historic sites. The research results demonstrate that the high-pressure gas expansion rock-cracking technology has the advantages of low vibration, noise, and flying rocks when there is high-quality hole plugging and no punching, providing a technical reference for excavation of hard rock tunnels near ancient buildings and historic sites.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Physical
Yinghua Jian, Dunwen Liu, Kunpeng Cao, Yu Tang
Summary: In this study, an accelerated erosion experiment was conducted using field sampling to investigate the deterioration characteristics of concrete under sulfate attack and the protective effect of silane coatings. Ultrasonic velocity measurement and CT scanning were used to compare the samples protected by silane coatings with those that were not protected. A method for evaluating sulfate damage to concrete based on CT images and ultrasonic velocity analysis was proposed.
Article
Chemistry, Multidisciplinary
Chengtao Yang, Ruiping Zhang, Dunwen Liu, Yu Tang, Rendong Huang, Weichao Qiu
Summary: The demand for tunnel construction is increasing rapidly, leading to more attention on tunnel mechanization construction to improve excavation ergonomics. In order to enhance the evaluation of tunnel drilling and blasting method excavation ergonomics, a set of evaluation methods based on the game theory G2-EW-TOPSIS model is proposed. This evaluation method includes a comprehensive index system and calculation of subjective and objective weights using different methods. The proposed model is applied to analyze the excavation ergonomics of the Shangtianling Tunnel, showing its effectiveness in identifying main factors and providing reference value.
APPLIED SCIENCES-BASEL
(2023)
