The traditional geological method relies too much on the resolution of seismic reflection and the quality of well data for the determination of geological profiles. In view of the fact that the number of well data that can be used for calibration in the early stage of development is very small, and the traditional geological modeling method generates geological profiles. The efficiency is low and it is difficult to support model establishment and frequent updating. A geological profile generation method based on improved Pix2Pix network was proposed. Firstly, the initial three-dimensional data was sliced. Based on the comprehensive analysis of deep learning network, a Pix2Pix network model based on residual and multi-scale discriminator was constructed. The residual mechanism was introduced in the generator part to improve the learning ability of the network to geological features, and a multi-scale discriminator was set for the model to enhance the discriminant performance of the network. The real seismic reflection data and geological profile data of the oilfield were used to train the model. The experimental results show that the performance of the network model is significantly improved after the introduction of residual mechanism and multi-scale discriminator. The SSIM (structural similarity) score of the generated results and the real geological profile can reach 91.89 %, and the geological features in the generated results are highly fitted with the actual situation.

In order to enhance the understanding of the characteristics and potential of geothermal resources in the Dunhuang Basin to support exploration and development initiatives, through the application of geothermal exploration techniques and hydrogeochemical analysis, the methods including hydrogen and oxygen isotope testing, Piper diagram analysis, the K-Mg geothermometer, and thermal storage estimation was used to investigate the chemical properties of geothermal water, the recharge source and age, as well as to calculate the thermal storage temperature and geothermal water circulation depth. Furthermore, the potential of geothermal resources in the basin was comprehensively evaluated. The results indicate that the thermal reservoir is primarily composed of argillaceous sandstone, pebbly fine sandstone, and glutenite located in the lower member of the Neogene Shulehe Formation, with the caprock consisting of Quaternary loose rock as well as mudstone and sandy mudstone in the upper member of the same formation. The primary heat source is heat conduction from the upper mantle and deep crust. The geothermal waters are dominated by Cl·SO₄-Na hydrochemical components, with temperatures ranging between 28.7 ℃ and 38.0 ℃ and total dissolved solids (TDS) values varying from 1 146 mg/L to 3 250 mg/L. Isotopic analysis, including δD, δ¹⁸O,³H, and¹⁴C, reveals that the geothermal water represents a mixture of deep groundwater and modern precipitation. The estimated reservoir temperatures range from 39.49 ℃ to 42.75 ℃, and the calculated circulation depths are between 1 020.65 m and 1 268.34 m. These findings suggest that the geothermal resources in the Dunhuang Basin are derived from deep groundwater circulation, replenished by atmospheric precipitation from the southern mountainous regions. The calculated thermal potential modulus of the geothermal fluid is 1.78 × 10⁹ kJ/(km²·a), indicating significant resource potential for future utilization.

In order to solve the problems of wind power fluctuations and intermittency during grid integration, which affect the stable operation of the power grid, a capacity optimization configuration scheme for a flywheel-lithium battery hybrid energy storage system was proposed. This scheme combined empirical mode decomposition (EMD) and variational mode decomposition (VMD). Firstly, typical daily data was obtained using the K-means algorithm, and EMD was applied to decompose the output power signal of these typical wind power daily data into grid-connected power that meets fluctuation limits and power that needs to be smoothed by the hybrid energy storage system. Then, the sparrow search algorithm was used to optimize the number of decomposition modes K and the quadratic penalty factor α in the VMD algorithm. By decomposing the power that needs smoothing using VMD, a reasonable allocation between lithium batteries and flywheel energy storage was achieved. Finally, considering the constraints of energy storage charging and discharging power and state of charge, an economic model was constructed with energy storage cost as the objective function. The actual power generation data of Qiejidunqu wind farm in Gonghe County, Hainan Prefecture, Qinghai Province were simulated and calculated by MATLAB platform. The results show that the proposed strategy not only effectively mitigates wind power fluctuations but also improves the overall economy of the system.

The slewing and luffing coupling motion of tower cranes can easily induce structural vibrations in the crane mast and swing angles of the payload, potentially leading to operational faults. To investigate these vibration patterns, an elastic crane model was developed under combined slewing and luffing dynamic motions, incorporating Lagrange dynamics, air resistance, and beam deflection. The model was analyzed across phases from acceleration to constant speed and then to deceleration. The effectiveness of this nonlinear coupled motion model was validated using a designed experimental platform. The study examined the effects of varying accelerations and initial swing angles. The results indicate that luffing acceleration influences structural vibration, while slewing acceleration has a significant impact on it. Additionally, initial angles greater than or equal to 0.2 rad greatly affect structural vibration. When the slewing acceleration exceeds 0.04 m/s², the frequency of mast vibration increases. Understanding the structural vibration law during coupled motion is crucial for enhancing the design of dynamic systems.

In order to explore the problem of fracturing fluid filtration in the process of open hole fracturing, based on the classical filtration model, combined with the two-dimensional model of hydraulic fracturing, ABAQUS finite element software was used for simulation calculation. It is found that in the initial stage of fracture propagation, the pore pressure increases rapidly and increased linearly, and the filtration loss of fracturing fluid also increase rapidly, which is the initial stage of fracture propagation. With the continuous injection of fracturing fluid, the increasing trend of pore pressure becomes slower, and the filtration loss of fracturing fluid also increases slowly. As the crack width gradually widens, it is the crack propagation stage. In the later stage of fracture propagation, when the fracture length reaches a certain length, the change of pore pressure is gradually stable, and the filtration loss of fracturing fluid is slightly reduced compared with the pressure holding state and the fracture propagation stage. In the whole process, the calculation of filtration loss takes into account the dynamic expansion of cracks, which is of certain significance for the actual engineering filtration situation.

To study the surface deformation law of insufficient mining goaf with thick loose layers after grouting, and evaluate the effectiveness of grouting and filling method in treating strip goaf areas, taking a thick loose layer strip goaf of Daizhuang Coal Mine in Jining as an example, theoretical analysis, numerical calculation, surface deformation observation, and deep rock optical fiber monitoring methods were used to compare and analyze the surface deformation laws of the goaf before and after grouting treatment. The results show these as follows. Under the influence of superposition of strip type insufficient mining in the goaf of thick loose layer, the subsidence of goaf decreases during the active period of surface movement, while the residual deformation increases in the later period, the maximum total subsidence of goaf is 611.8 mm and the residual subsidence is 157.8~288.1 mm by using probability integral method and numerical calculation, the surface residual deformation of goaf has great influence on the proposed high-rise buildings. The cumulative subsidence of the surface in the goaf treated by grouting filling method during observation period after grouting is -5.6~-1.5 mm. The comprehensive analysis of deep distributed optical fiber monitoring and surface leveling observation shows that the subsidence of the overlying rock in the goaf after grouting is less than 1 mm/a, and the analysis shows that the surface deformation after grouting is mainly caused by the settlement of the upper thick loose layer. The adoption of grouting filling method can significantly shorten the movement time of overlying strata in the goaf and effectively reduce the deformation of deep strata in the goaf, calculate and analyze the surface deformation value after grouting tends to be stable, the goaf site is in a stable state, and the grouting treatment effect is good, meeting the needs of high-strength engineering construction. The research results can provide guidance for surface deformation prediction, site stability evaluation and treatment effect detection after grouting in goaf under similar conditions.

Aiming at the nonlinearity and multi-disturbance problems of the electric regulating valve control system in the actual production process, a control method based on the improved ant colony algorithm to optimize the single neuron PID (proportional integral derivative) was proposed and applied to the valve opening control. The self-learning and self-adaptive ability of the single-neuron network was used to achieve the online tuning of PID control parameters. The improved ant colony optimization algorithm was adopted to optimize the learning rate and neuron ratio coefficients in the single-neuron PID, which effectively overcomed the shortcomings of the single-neuron PID where the learning rate and neuron ratio coefficients could not achieve the expected control effect due to the empirical setting. The simulation comparison results show that, compared with the traditional PID, single neuron PID, and single neuron PID based on ant colony optimization algorithm optimization of the three control methods, the control method proposed overshoots the amount of reduction of 10.2%, 6.1%, and 1.8%, respectively. At the same time, the regulation time is correspondingly shortened by 0.22s, 0.07s, and 0.03s. It shows a stronger adaptive and anti-interference ability, which can make the valve opening control more stable and reliable.

In order to explore the deformation and failure mechanisms, as well as degradation characteristics of silt slopes under rainfall conditions, a large-scale slope model test apparatus was designed with silt slopes as the research subject. Multi-sensor internal monitoring and 3D laser scanning technology were applied. During slope instability under a rainfall intensity of 30 mm/h, data on moisture content, pore water pressure, soil pressure, and deformation and failure characteristics at various slope locations were collected. Results indicate that sensors at the slope toe have the fastest response, showing the highest rate of change. Rainwater accumulates at the slope toe, causing horizontal seepage, which leads to rapid increases in moisture content at the toe and lower middle sections of the slope, along with a reduction in soil shear strength. Under rainfall intensity conditions, the deformation and failure of silty slopes initiate at the slope toe, where small-scale collapses first occur. These progressively develop into transverse through-cracks, accompanied by minor-scale failures that extend upslope. Ultimately, these processes lead to overall slope failure. The findings offer theoretical insights to support engineering construction and protection in silty slope regions.

Network intrusion detection systems (NIDS) are critical for maintaining cybersecurity. However, due to the complexity of network traffic data and the issue of class imbalance, existing detection models often exhibit high false alarm rates and insufficient detection accuracy for different types of attacks. To address these challenges, an imbalanced learning method for network intrusion detection, based on topological data analysis (TDA) and named TopoSMOTE, was proposed. This method aims to balance the training dataset by generating new minority class samples. The core of TopoSMOTE lied in constructing topological graphs to synthesize new samples. Firstly, the method used TDA to map the spatial relationships and connection patterns in network traffic data, forming a topological graph. Then, based on the topological graph, a minority class sample selection strategy was designed, which synthesized new data by selecting the nearest neighbor samples with topological relationships in a low-dimensional mapped space. Experiments were conducted on two imbalanced datasets. The experimental results show that the TopoSMOTE method achieves higher detection accuracy and lower false alarm rates compared to advanced oversampling methods and intrusion detection models.

A large number of towns in the western mountainous areas are within the hazard range of disaster chains such as landslides and debris flows, making the safety and disaster risk of mountain towns a key focus. The landslide-debris flow disaster chain in Lijie Gully, Lijie Town, Zhouqu County, Gansu Province was taken as a case study. Based on field investigations and remote sensing analysis, surface deformation was analyzed using interferometric synthetic aperture radar (InSAR) technology to determine the source of the disaster chain. The RAMMS dynamic model was then employed for the process and risk analysis of debris flows, followed by vulnerability and risk assessment. The research results indicate these as follows. The main subsidence areas within the basin are located at the edge and top of the ancient landslide area above Lijie Gully on the North Mountain, with the main cause of deformation being the seasonal freezing and thawing of soil. Based on the deformation results and remote sensing image analysis, the RAMMS dynamic model is used to conduct risk analysis of the Lijie Gully debris flow under three rainfall frequencies of 1%, 2%, and 5%. The comprehensive evaluation results show that the high-risk area accounts for 70.51%, largely distributed in the loose accumulation surface of the North Mountain landslide, the gully, both banks of the gully, and the accumulation fan at the gully mouth. Based on the risk and vulnerability assessment results, the high-risk area of the Lijie Gully debris flow accounts for 13.10%, mainly distributed in the gully mouth and along both sides of the gully where buildings are located. The large volume of loose accumulation above Lijie Gully is still in a continuous process of creeping-deformation-sliding, providing a large amount of material source for debris flows. The risk zoning results under different rainfall frequencies provide a reference for the disaster reduction of urban debris flows.