[Objective] To reveal the characteristics of hydrological situation evolution in Pailugou watershed of Qilian Mountains, and to provide a basis and reference for future water resource management and optimal allocation in the watershed. [Methods] Based on the measured runoff and hydrological data of Qilian Mountain Field Observatory from 2000 to 2019, the effects of precipitation and temperature on runoff were investigated by using the linear trend method, Pettitt′s test, and wavelet analysis, et al., and a BO-LSTM runoff prediction model for the Pailudou Basin was established.[Results] 1)From 2000 to 2019, precipitation, air temperature and runoff in Pailugou Watershed showed a two-stage upward trend, and the cutoff point was in 2010, precipitation and runoff, the first stage of the upward trend are higher than the second stage, the slope is 10.74, 3.16 in turn; air temperature is the opposite, the second stage is higher than the first stage, the slope is 0.11. And precipitation, air temperature and runoff of the MK mutation test z-value are greater than 0.(2)Precipitation in the May-October months on the runoff changes of the contribution rate is larger; and air temperature in the December-April months on the runoff changes of the contribution rate is large.(3)The air temperature in Pailougou Basin mainly has two main cycles, 3 a and 14 a, of which the first main cycle is 14 a; runoff exists in three main cycles, 19 a, 9 a and 3 a, of which the first main cycle is 19 a; precipitation mainly exists in two main cycles, 4 a and 11 a, of which the first main cycle is 11 a.(4)The BO-LSTM runoff prediction model for Pailougou, with an accuracy of R² of 0.63 and a root-mean-square error of 14 047 m³, and the prediction accuracy of the model is greater in months with smaller runoff than in months with larger runoff. [Conclusion] Precipitation, air temperature and runoff in Pailugou Basin have been on an upward trend in the past 20 years. Runoff, precipitation and air temperature in Pailugou Basin have obvious cyclicity. Air temperature and precipitation are important factors affecting the runoff in Pailugou Basin. The runoff prediction model can be applied to Pailugou Basin. The above results provide scientific support for the study of water resource effects in the Qilian Mountains and the prediction of water resources in inland river basins.

[Objective] The frequent occurrence of global extreme climate events and the rapid pace of urbanization have resulted in increasingly severe urban waterlogging issues, leading to a significant rise in drainage pressure within urban pipe networks. It is of utmost importance to promptly assess the operational status of these pipe networks. [Methods] The flood process analysis mechanism model of the Wenchong drainage system in Huangpu District, Guangzhou, was established based on the SWMM model to determine the hydraulic capacity of pipeline sections under different rainfall scenarios. Subsequently, a data model incorporating both single-well and multi-well computing models was developed to predict the hydraulic capacity of urban pipe sections, with careful selection of the most suitable data model. Finally, an evaluation of operational risk for the urban pipe network was conducted using an improved analytic hierarchy process and comprehensive risk index method. [Results] The results show that:(1) The decision tree model based on the multi-well calculation mode has the highest accuracy, with an RMSE of 0.077 and a MAE of 0.030, and the single-well calculation time is 600,000 times faster than the SWMM model.(2) In the area where sensitive points are concentrated, social factors exert the most significant influence on the operational risk of the pipe network, with a weightage of 0.72. In contrast, in the area where sensitive points are not concentrated, economic factors play a dominant role in determining the operational risk of the pipe network, carrying a weightage of 0.68. [Conclusion] The decision tree model based on the multi-well calculation mode can meet the precision and timeliness requirements for urban flood emergency response. The risk of pipe network operation varies across different areas of the city, necessitating the proposal of improvement measures tailored to local conditions.

[Objective] Affected by the uncertainty of water forecast, power generation scheduling, ecological scheduling test, and bank stability requirements, the centralized water level scheduling mode of the Three Gorges Reservoir has some problems, such as difficulty in precise water level regulation and room for optimization of water resource utilization efficiency. In order to effectively control the flood control risk and improve the operation flexibility and comprehensive benefit, the optimal operation mode of extending the pre-flood period of the Three Gorges Reservoir was proposed. [Methods] Based on the analysis of the changes of hydrological regime in the Three Gorges Reservoir, the flood control effects and benefits of different schemes were compared and analyzed through the coupling of the extension of pre-flood centralized water level, the flood control calculation and the flood evolution model of the downstream river. Under the principle of not affecting the safety of flood control and effectively improving the comprehensive benefit, the pre-flood water level centralized subsiding method was optimized. [Results] The result show that the Three Gorges Reservoir can pre-discharge the water level to 145 m when it is transferred to flood control operation in 1954, 1998 and other major floods in the Yangtze River basin, and the pre-discharge process does not increase the number of days beyond the warning water level of major hydrological stations such as Shashi and Chenglingji downstream, and do not raise the water level during flood peak. [Conclusion] On the premise that the flood control risk can be controlled, when the downstream river water level is low in mid to late June, the pre-flood water level of the Three Gorges Reservoir can be appropriately extended to 145 m, which can better coordinate the pre-flood flood control and water supply needs, and effectively connect the dynamic control of the reservoir′s operating water level during the flood season, which is conducive to improving the flexibility of the centralized flood control and water resource utilization efficiency, and provide better conditions for the actual operation of the Three Gorges reservoir..

[Objective] The impoundment operation of the Wudongde Hydropower Station has caused significant landslide deformation. In order to find out the development characteristics of landslide and the water storage response law of wading landslide deformation, [Methods] SBAS-InSAR, optical remote sensing interpretation, field surveys, mathematical statistics and theoretical analysis were used to study the number, development characteristics, deformation patterns and trend of landslides induced by the impoundment of the Wudongde Hydropower Station. [Results] The result show that there are 62 landslides along the coast of Wudongde Reservoir, including 39 water-related landslides. These landslides are concentrated in Jurassic, Cretaceous red beds and Proterozoic shallow metamorphic rock areas. The slope of the landslide is 10°~40°, the height difference between the front and rear edges is 200~800 m, the slope types are mostly convex forward-facing slopes, with linear or fold line sliding surfaces being predominant. A total of 114 obvious deformations have occurred in 39 water-related landslides since the initial impoundment, and the maximum deformation rate is 63.72 mm/a. Currently, the number of landslide deformation and deformation rate are increasing year by year, which is in the active stage. The deformation trend of landslide is categorized into accelerated deformation, uniform deformation, and deceleration stabilization deformation. Among them, the accelerated deformation landslide accounts for 61.54%, which is dominated by the landslide with a straight line and a high degree of wading. Its water storage response is significant and the deformation is large. The influence of reservoir water level decline on landslide is greater than that of reservoir water level rise. [Conclusion] The deformation law of Wudongde reservoir landslide is revealed, which provides a scientific basis for the prediction and prevention of reservoir landslide disaster.

[Objective] Precipitation and runoff are the key elements of the hydrological cycle. Studying the trend variations of precipitation and runoff, especially further exploring changes in their relationship, is of great significance for understanding climate change, water resources management, and ecological protection. [Methods] Based on a long series of precipitation and runoff data from 1956 to 2019, the innovative trend analysis method was used to analyze the variation trends of precipitation, runoff, and runoff coefficient at different spatiotemporal scales in the Pearl River Basin. [Results] The result showed that:(1) the overall annual precipitation in the Pearl River Basin showed a downward trend, with changes in low-value categories declining by more than 5%, indicating an increased likelihood of drought in the Pearl River Basin, and the change trend of annual runoff was consistent with that of precipitation. Across different sub-basins, the West River and East River showed significant downward trends, while the North River showed an upward trend.(2) From a seasonal perspective, spring and autumn precipitation in the Pearl River Basin showed a significant downward trend, while summer and winter precipitation showed an upward trend. The spring high precipitation values and all types of autumn precipitation showed a downward trend of more than 5%, and the summer high precipitation values and winter low-to-median precipitation showed an upward trend of more than 5%. The variation trends of annual runoff in spring, autumn, and winter were consistent with precipitation, but were opposite in summer.(3) The runoff coefficient showed a significant downward trend, especially in high values. Preliminary analysis suggested that the decrease in runoff coefficient was related to the increase of evapotranspiration, changes in vegetation coverage, and the regulation and storage of water engineering.(4) The consistency rate between the innovative trend analysis method and the traditional Mann-Kendall method was 70% for trend type and 22% for significance. [Conclusion] Spatiotemporal variation characteristics of precipitation, runoff, and runoff coefficient in the Pearl River Basin: From a temporal perspective, the annual precipitation and runoff in the Pearl River Basin show an overall declining trend, with a significant decrease in the runoff coefficient. From a spatial perspective, the precipitation and runoff in the West River and East River exhibit a noticeable downward trend, while the North River shows an upward trend. The innovative trend analysis method, compared to the Mann-Kendall method, demonstrates significant differences in trend significance testing.

[Objective] Aiming at revealing the synergistic impacts of climate change and anthropogenic activities on vegetation ecology, the spatiotemporal dynamics of net primary productivity(NPP) in the Weihe River Basin since the implementation of the ecological restoration policy in 2000 are explored, which provide a theoretical basis for the ecological conservation and sustainable development of the region. [Methods] Based on the Google Earth Engine(GEE) cloud platform, MOD17A3HGF6.1 datum are utilized to investigate the spatiotemporal distribution patterns of NPP, CPP, and VPG. By constructing six vegetation change scenarios, it examines the influence mechanisms of climatic factors and human interventions on vegetation changes. The contributions of each factor were quantitatively evaluated using least squares regression, Pearson correlation analysis, and contribution indices. [Results] (1) From 2000 to 2023, NPP in the Weihe River Basin exhibited unimodal seasonal fluctuations, with an average annual increase of 9.91 gC·m^-2·a^-1, and demonstrated a spatial gradient of “higher in the south, lower in the north”.(2) During the vegetation restoration phase, temperature and precipitation were positively correlated with vegetation growth, with correlation coefficients of 0.26 and 0.43, respectively. Solar radiation displayed a marginal positive correlation in vegetation restoration driven by human activity, with a correlation coefficient of 0.12. Population density and road density accounted for approximately 34.77% of the restored vegetative area.(3) In the vegetation degradation phase, precipitation was responsible for 30.43% of the vegetation decline, with correlation coefficients for population density and road density reaching 0.71 and 0.64, respectively. [Conclusion] Throughout the study period, NPP in the Weihe River Basin exhibited a general upward trajectory. Climate variables such as temperature and precipitation, population and road density, exert substantial influence on NPP, while solar radiation has a comparatively limited effect. Climate change emerges as the predominant driver of vegetation restoration, while human interventions such as ecological restoration efforts significantly enhance regional vegetation productivity.

[Objective] To achieve accurate prediction of tunnel squeezing, [Methods] an eXtreme Gradient Boosting(XGBoost) model tuned by Grey Wolf Optimization(GWO) was constructed for tunnel squeezing prediction. Training and testing of the GWO-XGBoost model were conducted on an imbalanced dataset with missing data that had undergone imputation and oversampling techniques. The input features of the GWO-XGBoost model included tunnel burial depth(H), rock tunnelling quality index(Q), diameter(D), strength stress ratio(SSR), and support stiffness(K). The performance of the GWO-XGBoost model was rigorously evaluated using a suite of metrics, including accuracy(ACC), the F1 score, the Kappa coefficient, and the Matthews correlation coefficient(MCC). [Results] The result indicated that the presented GWO-XGBoost model achieved an impressive prediction accuracy of 98.94% on both the training set and the test set. Moreover, on the test set, the cumulative value of the evaluation metrics soared to 5.913 1, underscoring the model's exceptional predictive capabilities. The average Shapley Additive exPlanation(SHAP) values for SSR, D, K, Q, and H were 3.06, 1.07, 0.82, 0.73, and 0.51, respectively, indicating that SSR was the most influential feature affecting the model's output result. [Conclusion] The application of the GWO-XGBoost model to the Huzhubeishan Tunnel and Muzhailing Tunnel has yielded squeezing predictions that closely align with the actual conditions observed, proving the high applicability and predictive accuracy of the presented model in tunnel engineering.

[Objective] The simulation method of the construction process of roller compacted concrete dam(RCCD) provides a scientific and effective technical means for analyzing the complex dynamic system of dam construction. The traditional discrete event simulation(DES) method considers the main factors affecting the construction schedule(such as concrete production, concrete transportation, and construction rainfalls) in the simulation process. However, the influence of construction machinery failure on the construction process in the warehouse surface operation is not effectively considered, and the accuracy of the construction progress simulation result and the three-dimensional visualization display effect need to be further improved. To address the aforementioned challenges, a coupled SD-DES visual simulation method for RCCD construction considering mechanical failures is developed through the integration of system dynamics(SD) with conventional construction simulation approaches. [Methods] First of all, the SD model of mechanical failure in silo construction was established based on system dynamics, so as to realize the causality analysis of the behavior of the roller operation, failure, and maintenance in the storehouse construction. Secondly, based on the traditional DES simulation model, the coupled SD-DES simulation model of RCCD construction is established by coupling the SD model of mechanical failure and taking the time of roller compacted construction as the interface variable of the coupled model, so as to realize the data communication and interaction of the model, which effectively improves the accuracy of the simulation result. Then, the system simulation technology is combined with the visualization technology to construct a three-dimensional dynamic scene of the dam construction based on Unity 3D to realize the three-dimensional dynamic visualization display of the simulation result. Finally, taking a high roller compacted concrete dam project in the southwestern region of China as an example, construction simulation analysis and result display were conducted. [Results] Compared with the traditional DES model simulation result(deviation of 2.49% from the actual), the simulation result based on the coupled SD-DES model(deviation of 0.28% from the actual) are more in line with the actual situation, and the level of three-dimensional dynamic visualization display has been further enhanced. [Conclusion] The study demonstrates that the coupled SD-DES visualization simulation method effectively accounts for the impact of construction machinery failures on construction schedules, enhances simulation accuracy and visualization capabilities of the construction process, and contributes to the rational formulation of dam construction plans and informed management decisions for on-site construction.

[Objective] Monitoring the deformation of foundation pits is crucial to ensuring the safe construction of foundation pits. To enhance the application value of monitoring data and ensure the safety of foundation pit construction, this study relies on a 330kV substation foundation pit project in Xi'an, Shaanxi Province, and is based on actual deformation monitoring result. [Methods] Using the mean square error(MSE) as the fitness function of the genetic algorithm(GA), the trend, seasonality, and holiday(sporadic event) parameters of the Prophet model were optimized, with particular attention to the trend parameters consistent with the deformation pattern of the foundation pit. The GA-Prophet foundation pit deformation prediction model was constructed, and its feasibility and effectiveness were verified using evaluation metrics such as MAE, RSS, RMSE, and Theil Inequality Coefficient values. Additionally, this model was employed for the early prediction of horizontal and vertical deformation of the foundation pit to evaluate the safety status of the foundation pit structure. [Results] The result indicate that the prediction curve of the GA-Prophet model closely aligns with the measured data curve, attributed to the adoption of a saturation model that conforms to the actual engineering displacement change pattern in the prediction model. Taking the horizontal displacement prediction result of the JC8 measurement point as an example, the MAE, RSS, RMSE, and Theil Inequality Coefficient values of the prediction result were 0.480, 1.310, 0.512, and 0.052, respectively, all superior to the prediction result of the Prophet, LSTM, ARIMA, and BP models. Moreover, the early prediction result of foundation pit deformation by this model show that the maximum predicted values of horizontal and vertical deformation at each measurement point did not exceed the deformation alarm values specified by the standards, indicating that the foundation pit structure is in a safe state. [Conclusion] The model demonstrates good applicability for predicting foundation pit deformation and improves the accuracy of the prediction result. It can be used for the safety prediction of foundation pit deformation.

[Objective] Due to the complex conditions in deep canyon areas, the pullout capacity of micro-pile foundations configured with small-diameter towers in power grid projects is insufficient. The use of tapered micro-piles, which combine traditional micro-piles with varying diameters, may be an effective solution to enhance pullout resistance. This approach aims to improve the pullout performance while offering greater cost-effectiveness. [Methods] Five groups of indoor model tests were conduceted to comparatively analyze the load-bearing characteristics of traditional micro-piles and tapered micro-piles in sandy silt. It also examined the relationship between tapered arrangements and the load-bearing mechanisms of tapered micro-piles. Furthermore, numerical simulations were employed to reveal how various structural parameters of the tapered design and soil properties affect the vertical load capacity of tapered micro-piles. [Results] The result indicated that the load capacity and material efficiency of tapered micro-piles exceed those of conventional micro-piles by over two times, with displacements under ultimate loads significantly greater than those of traditional micro-piles. Additionally, the group effect of tapered micro-pile groups was enhanced, and under uplift loads, different parts of the pile experienced varying load distributions, with the role of enlarged heads becoming increasingly prominent, carrying more than 50% of the load. [Conclusion] The influence of different tapered structural parameters on the vertical load capacity of the pile foundation varied with changes in the enlarged head and side friction resistance. For practical applications, it is essential to optimize the design parameters rationally, as tapered micro-piles demonstrate greater advantages in the treatment of high-strength foundation soils.

[Objective] The channel straightening project of the Pinglu Canal has fragmented the river course, compromising the integrity of original river course and causing ecosystem patchiness. Understanding the current status of fish resources and the characteristics of their diversity is crucial for the ecological management of the Pinglu Canal. [Methods] During the spring and autumn in 2021 and 2022, a survey of fish resources and species diversity in the Pinglu Canal was conducted using multi-mesh gill nets. A total of 125 fish species were collected, belonging to 10 orders, 34 families, and 89 genera. [Results] The result showed that the Pinglu Canal contained three nationally protected Class II species, two endemic species of the Qinjiang River, three anadromous/migratory species, and eight invasive species, accounting for 2. 4%, 1. 6%, 2. 4%, and 6. 4% of the total species, respectively. The fish community primarily consisted of mid-and bottom-dwelling, adhesive-egg-laying, and omnivorous species. The Shannon-Wiener, Simpson, Margalef, and Pielou indices of the fish community in the Pinglu Canal ranged from 2. 347 to 2. 757, 0. 081 to 0. 151, 3. 493 to 4. 382, and 0. 812 to 0. 892, respectively. These indices showed relatively uniform distribution across different river reaches. [Conclusion] The result indicate that the fish community structure in the Pinglu Canal is relatively uniform. The reach from the Yujiang River to the Shaping River shows higher stability, while other river reaches experience moderate or severe disturbances. This study provides supplementary baseline data on the fish community structure in the Pinglu Canal and explores the potential impact of inter-basin connectivity on fish resources, aiming to provide a scientific basis for habitat restoration assessments after the channel straightening project.

[Objective] Clarifying the spatiotemporal evolution characteristics and temperature lag effect of high dams and large reservoirs can provide a scientific basis for stacked beam gate scheduling, which is fundamental for alleviating discharge of low-temperature water and protecting the survival and reproduction of fish in the lower reaches. [Methods] Through numerical simulation and based on the measured hydrological data, water temperature data, meteorological boundary data from 2020 to 2022, and design operation data in an average water year, a three-dimensional numerical simulation model of water temperature for the Wudongde Reservoir area was developed. The spatiotemporal evolution characteristics of water temperature structure and lag effect of discharge water temperature after reservoir operation were analyzed. [Results] The result showed that:(1) In spring and summer, a distinct vertical stratification of water temperature was observed in the Wudongde Reservoir, with a temperature difference of 11~12 ℃. After the flood season, as the thermal exchange within the water body increased, the vertical water temperature gradually became uniform. No significant water temperature stratification was observed in autumn and winter.(2) There was a clear issue of low-temperature water discharge in spring and summer, and high-temperature water discharge in winter from the reservoir. Monitoring data from 2020 to 2022 indicated that the discharge water temperature in May was, on average, 2.68℃ lower than the natural water temperature at the dam site, and the discharge water temperature in December was, on average, 4.78℃ higher. Simulation result during the stable operation period indicated that from March to June, the monthly average discharge water temperature was 0.91~4.09 ℃ lower than the natural water temperature at the dam site. From December to February of the following year, the monthly average discharge water temperature was 1.34~4.34 ℃ higher.(3) After the impounding of reservoir, the time for the water temperature to reach 14℃ and 18℃ each year was delayed to varying degrees compared to before the dam construction. The occurrence time of suitable spawning water temperature for fish in the lower reaches was postponed by 30~50 days, with the delay ranked as follows: Rhinogobio ventralis > Jinshaia sinensis, Four major Chinese carps, Common carp, Crucian carp > Lepturichthys fimbriata. [Conclusion] The result show that the stratification of water temperature in the Wudongde Reservoir, along with the low-temperature discharge water it causes, has varying degrees of negative effects on fish reproduction in the lower reaches of the river. Therefore, scientific measures must be implemented to address these issues.

[Objective] Understanding the migration and transformation patterns of contaminants in the groundwater level fluctuation zone is crucial for uncovering the root causes of groundwater pollution. [Methods] Based on the theory of groundwater solute transport, soil column simulation experiments were conducted to investigate the migration and transformation patterns of arsenic and mercury under different water level fluctuation amplitudes(ΔH=0 cm, 10 cm, 20 cm) in the water level fluctuation zone. [Results] The results showed that the variation amplitude of total arsenic and As(V) concentrations in soil solution at sampling points followed the same trend as the water level fluctuation amplitudes, with an overall performance of column 3(ΔH=20 cm)> column 2(ΔH=10 cm). However, the concentration of mercury in the soil solution was not affected by the water level fluctuation and showed an overall increasing trend. [Conclusion] The result indicate that the periodic fluctuation of the water level mainly affects the characteristics of key environmental factors in the medium of the fluctuation zone, which in turn impacts the migration and transformation patterns of arsenic in this zone. The amplitude of water level fluctuation is positively correlated with the degree of this effect. Therefore, in groundwater arsenic pollution risk assessments, the impact of groundwater level fluctuation on arsenic migration should not be overlooked.

[Objective] This study proposes a network traffic anomaly detection method that addresses the issues of data imbalance, high feature dimensionality, and low detection efficiency in water conservancy industrial control networks. The method integrates an improved Conditional Generative Adversarial Network(ICGAN), Deep Residual Shrinking Network(DRSN), and Long Short-Term Memory Network(LSTM). [Methods] ICGAN was used to construct a balanced network traffic dataset, and a DRSN-LSTM hybrid deep learning model was employed for anomaly detection in network traffic. DRSN was responsible for extracting spatial features, with residual connections addressing network degradation and overfitting issues. The compression and excitation network automatically assigned weight coefficients to each feature map to improve detection performance. Lastly, LSTM extracted temporal features from the data. [Results] The method was tested in the application scenario of the Qinhuai River Wudingmen Sluice Station. The result showed that models trained on the ICGAN-optimized dataset achieved higher traffic classification accuracy than those trained on the original dataset. Overall, DRSN-LSTM achieved an accuracy of 98.76% in detecting network traffic anomalies. P, R, and F1 values for normal data classification were 99.22%, 99.69%, and 99.46%, respectively, which outperformed the comparison models in terms of these evaluation indicators. [Conclusion] By integrating the advantages of ICGAN, DRSN, and LSTM algorithms, the anomaly detection method for water conservancy industrial network traffic effectively alleviates the type imbalance in the original dataset, improves the detection ability of abnormal industrial control network traffic, and ensures the safe and stable operation of water conservancy projects.

[Objective] Tension rupture represents the predominant rupture form in the process of rock collapse, with the potential to cause significant disruption. Consequently, there is a clear need to study the acoustic emission signal characteristics and strain evolution law in the process of rock tension damage. This will facilitate the identification of precursor information of rock damage, which can then be used for the monitoring and early warning of rock collapse. [Methods] A three-point bending test(1.0 m×0.5 m×0.15 m) was conducted on a large-scale tuff to monitor the rupture process in real time using acoustic emission technology and digital image correlation technology. The acoustic emission signals and deformation evolution characteristics of large-scale tuff in the process of tensile damage were then analyzed by combining principal component analysis and hierarchical clustering algorithms. [Results] The application of acoustic emission detection technology allows for the accurate identification of the damage state of the rock mass, as well as the provision of effective damage precursor information. [Conclusion] The result show that:(1) According to the evolution of acoustic signal parameters, the tensile failure of rock can be divided into four stages: microcrack initiation stage, small-scale and stable cracking stage, unstable cracking stage and failure stage;(2) The acoustic signals can be divided into six categories using the acoustic emission parameters using clustering and principal component analysis algorithm. The cumulative changes and proportions of the two types of characteristic signals, the gradual low-amplitude type signal and the sudden high-amplitude type signal, are capable to characterize the cracking process;(3) The findings indicate that the acoustic emission characteristic parameters and their evolution characteristics can effectively reflect the rock rupture process, providing precursor information for rock damage up to 141 seconds in advance compared with the simultaneous deformation monitoring. The result of this study offer insights that can inform the development of effective method and techniques for monitoring and early warning of rock failure. This research could provide support for rock collapse monitoring and early warning.

[Objective] Concrete, as the cornerstone of national economic construction, necessitates the accurate prediction of its compressive strength for the design and safety of engineering structures. This study aims to predict concrete compressive strength using Deep Neural Network(DNN) models and proposes the RF-NSGA-II algorithm to optimize concrete mix proportions, achieving dual optimization of compressive strength and cost. [Methods] Fifteen DNN model architectures with different hidden layers and neuron numbers were constructed and evaluated for performance, selecting the best model. Hyperparameter optimization strategies and Bayesian optimization were employed to enhance the predictive performance of the DNN model. The performance of the DNN model was compared with Support Vector Regression(SVR) and Random Forest(RF) models. The RF-NSGA-II algorithm was used to optimize concrete mix proportions to meet strength requirements and cost control. [Results] The result showed that the optimal model had 3 hidden layers and 64 neurons(3L-64u). After optimization, the DNN model′s MAE and MSE decreased by 18% and 27%, respectively. Compared to the SVR and RF models, the optimized DNN model reduced MAE and MSE by 4% and 12%, and 11% and 15%, respectively. [Conclusion] Case validation demonstrated that the DNN3-L64u-BOP model′s predictions aligned well with experimental values, and the RF-NSGA-II algorithm effectively reduced costs while meeting engineering strength requirements. The Bayesian-optimized DNN model successfully predicted concrete compressive strength, and the RF-NSGA-II algorithm exhibited excellent performance in multi-objective optimization of concrete mix proportions, showing significant practical value in engineering applications.

[Objective] To improve the high-value utilization of reservoir sediment, reservoir sediment foamed concrete was prepared by taking the water binder ratio, reservoir sediment mixing amount and foam mixing amount as variables. [Methods] The response surface method in the Design-Expert 13.0 software was used to optimize the mix ratio design, analyze the effects of variables on the compressive strength and thermal conductivity of the reservoir sediment foamed concrete. SEM and XRD were used to explore the microstructure of the reservoir sediment foamed concrete. [Results] The results show that the influence of each factor on the 28 d compressive strength of reservoir sediment foamed concrete is in the following order: water binder ratio>foam content >reservoir sediment content, and the thermal conductivity is in the following order: water binder ratio> reservoir sediment content >foam content. With the increase of water binder ratio, the compressive strength increases at first and then decreases, and the thermal conductivity gradually decreases. Increasing foam and reservoir sediment mixing result in a decrease in both compressive strength and thermal conductivity. Porosity showed a negative correlation with thermal conductivity, consistent with an exponential model. [Conclusion] The optimal mix ratio of reservoir sediment foamed concrete is 0.4, 30% of reservoir sediment and 4.2% of foam content, and the 28 d compressive strength of the foamed concrete is 18.19 MPa, and the thermal conductivity is 0.121 4 W/(m·K), apparent density is 701.2 kg/m³. The analysis of microscopic result showed that the internal material of the specimen was tightly bound and the pores were uniformly distributed under the condition that the amount of reservoir sediment was 30%. Reservoir sediment foamed concrete provide a new way for the use of reservoir sediment in the field of construction materials.

[Objective] Reasonably and accurately describing the mechanical behavior and failure mechanism of Carbon Fiber Reinforced Polymer(CFRP) when used to reinforce corroded steel-bar concrete bending members is crucial for in-depth research and improvement of CFRP reinforcement technology in enhancing the mechanical performance and durability of corroded steel-bar concrete structures. [Methods] Acoustic emission monitoring technology was employed to conduct bending performance tests on steel-bar concrete beams with different degrees of corrosion which reinforced by CFRP. By comprehensively analyzing key indicators such as deflection, rebar strain, carbon fiber strain, and characteristics of acoustic emission signals, the mechanical characteristics of CFRP-reinforced corroded steel-bar concrete beams were revealed. [Results] The result showed that the effect of CFRP reinforcement was closely related to the degree of corrosion of the steel-bar. Under a determined reinforcement method, as the corrosion rate of the steel-bar increased, the “tension arch” effect was intensified due to the degradation of the bond performance between the steel-bar and concrete, leading to a gradual decrease in the reinforcement effect of CFRP. [Conclusion] The spalling and detachment of the concrete cover made the “tension arch” effect between CFRP and concrete more significant than that of the steel-bar, which had a greater impact on the overall mechanical performance of the test beams. In addition, combining the characteristics of acoustic emission monitoring signals with mechanical test indicators could more comprehensively reflect the mechanical response of the test beams at various loading stages, providing strong technical support for the evaluation and design of CFRP-reinforced corroded steel-bar concrete structures.

[Objective] Photovoltaic output in water-light complementary optimal scheduling is characterized by volatility, randomness, and intermittency. Its solution space is typically high-dimensional, complex, and continuous. A variety of continuous control decision-making problems are involved in water-light complementary optimal scheduling. [Methods] The Deep Deterministic Policy Gradient(DDPG) algorithm in the deep reinforcement learning algorithm was suitable for solving continuous and complex problems in the solution space. The water-light complementary problem was modeled using reinforcement learning. Based on the water-light complementary mechanism, the concepts of “demand for adjustment” and “capacity for adjustment” were considered to set up the environment, actions, reward function, and penalty function. The DDPG algorithm was then used for optimization. The applicability and effectiveness of the model were assessed by comparing and analyzing the optimization result using only the initial DDPG algorithm and those using the genetic algorithm. Taking the large-scale water-light complementary base in the upper reaches of the Lancang River as an example, three cascade hydropower station configuration schemes and three representative hydrological years were set up for a case study. [Results] The analysis indicated that:(1) the DDPG algorithm performed faster. After considering the mechanisms of “demand for adjustment” and “capacity for adjustment”, the photovoltaic power consumption reached 12.993 billion kWh, which was the highest among the three models.(2) The lower the water inflow, the stronger the photovoltaic consumption capacity was. When the installed capacity of cascade hydropower stations was 8.62 million kW, the photovoltaic consumption capacity only increased by 1% from normal year to dry year. At this time, the complementary capacity of hydropower system could be maximized.(3) The photovoltaic consumption capacity was relatively low during the wet season, and the photovoltaic consumption rates of the three schemes were 77.43%, 79.85%, and 89.39%, respectively. [Conclusion] The deep reinforcement learning algorithm demonstrates the advantage of rapid convergence in the water-light complementary optimal scheduling. Integrating mechanisms of “demand for adjustment” and “capacity for adjustment” into reinforcement learning modeling can significantly enhance the photovoltaic consumption efficiency, achieve better resource utilization, and effectively improve the operation efficiency of the water-light complementary system and the photovoltaic power consumption capacity. This method shows promising result in the capacity configuration and operation scheduling of clean energy base, providing a theoretical and practical foundation for the future expansion and application of clean energy system.

[Objective] In order to explore the optimal soil and water conservation measures for hillside citrus orchards in karst rocky desertification area, [Methods] a runoff observation plot method was developed to monitor the hillside citrus orchards in the karst area of northern Guangxi. The runoff, sediment, and nutrient loss under three soil and water conservation measures—vegetation planting, brick-and-stone enclosure, and PVC board enclosure—were analyzed at different rainfall levels(heavy rain, rainstorm, and heavy rainstorm). Additionally, the entropy-weight fuzzy comprehensive evaluation method was used to perform a quantitative analysis of the comprehensive benefits. [Results] The result showed that:(1) whether rainfall levels were considered or not, vegetation planting demonstrated the best nitrogen and phosphorus reduction benefits, and PVC board enclosure had the best sediment reduction benefits. In general, the reduction benefits of the brick-and-stone enclosure were lower than those of the other two measures. Vegetation planting provided the best runoff reduction benefits under heavy rain and rainstorm, while PVC board enclosure had higher runoff reduction benefits than vegetation planting under heavy rainstorm.(2) The water, soil, and nutrient loss, as well as the reduction benefit indicators of soil and water conservation measures, did not all change monotonically with the increase in rainfall level. Specifically, runoff yield and nutrient loss increased monotonically with the increase in rainfall level, while most other indicators did not show a monotonic change. The change trends in indicators such as runoff reduction, sediment reduction, nitrogen reduction, and phosphorus reduction of the same measure varied with different rainfall levels.(3) When rainfall levels were not considered, the comprehensive benefits of implementing PVC board enclosure in hillside citrus orchards in karst rocky desertification area were the best, followed by vegetation planting. When rainfall levels were considered, the comprehensive benefits of implementing vegetation planting were the best under rainfall levels of heavy rain and rainstorm, and the comprehensive benefits of implementing PVC board enclosure were the best under heavy rainstorm. [Conclusion] The implementation of vegetation planting, PVC board enclosure, and brick-and-stone enclosure in hillside citrus orchards in karst rocky desertification area all provide certain soil and water conservation benefits.