In order to investigate the role of rheumatoid arthritis (RA)-related pathways in lung squamous cell carcinoma (LSCC). By obtaining gene expression data for RA and LSCC from the GEO and TCGA database, differentially expressed genes were screened using GEO2R tool and Rstudio software. GO/KEGG functional enrichment analysis identified key genes in the RA signaling pathway. Combining SNP data from the IEUopenGWAS database, Mendelian randomization analysis was used to assess the causal relationship between the RA signaling pathway and LSCC. The constructed gene-drug and ceRNA networks, along with immune cell infiltration analysis, revealed 188 co-expressed differential genes, mainly enriched in the RA signaling pathway. Mendelian randomization analysis showed that increased activity of the RA signaling pathway is associated with a reduced risk of LSCC. This study provides new insights into the pathogenesis and potential research directions for the treatment of LSCC.

At present, complex interstitial lung diseases have the problems of low classification accuracy and lack of auxiliary diagnostic information. To address these problems, an image retrieval framework based on multi-feature fusion and supervised contrastive learning methods was proposed. Interstitial lung disease features were extracted using Res-Net50 and radiomics feature extraction modules. In order to fuse two features of different modalities and scales, a feature fusion module was designed that can jointly represent the spatial calculation feature correlation of two features. The feature discrimination between interstitial lung disease categories was improved through supervised contrastive learning methods, and a typical interstitial lung disease database was retrieved. The highest precision, recall rate and F₁ score were obtained in the retrieval task of interstitial lung disease data, and a silhouette coefficient of 0.482 was obtained in the feature vector discrimination index for image retrieval. The experimental results show that compared with the traditional deep learning single feature modality method, the proposed method can effectively improve the classification retrieval accuracy of interstitial lung disease images and improve the interpretability of interstitial lung disease diagnosis.

Forest fires have the characteristics of strong suddenness, great destructiveness, many uncertain factors and high risk of fighting. In order to study effective strategies of firefighting, firstly, based on the theory of cellular automata, the forest fire system was analyzed, and a forest fire model considering external factors such as wind and flame retardant was established. Then, on this basis, the fire-fighting agent was modeled and correlated with the forest fire model, so as to build the fire-fighting model. Finally, the simulation algorithm based on cellular automata was designed to simulate the effect of fire-fighting strategies under the influence of different environmental factors, and the effect of different fire force allocation strategies. The results show that the method can combine fire-fighting simulation with actual decision making, and provide visualized and quantified strategy scheme for relevant departments to make fire-fighting decision, which is helpful to reduce forest fire loss and rescue cost.

To investigate the mechanisms of floor heave and support methods for large-section inclined shafts in highly expansive and weak rock, a case study was conducted on the main inclined shaft of a mine in the Yaojie Coal and Electricity Group. Through field surveys and laboratory tests, characteristics such as low rock mass strength, poor integrity, high clay content, and softening and swelling upon water exposure were identified. A Flac3D numerical model was established to simulate the deformation patterns of the tunnel under actual engineering conditions, elucidating the mechanisms of floor heave deformation in large-section soft rock inclined shafts. Three optimized prevention strategies were summarized, and an optimized support scheme of “inverse floor arch and anchor cable” was proposed based on the original support design. Field measurement data indicated that the optimized support scheme significantly improved the floor heave control in soft rock tunnels, enhanced the bearing capacity of the surrounding rock, reduced floor heave deformation, and achieved stability control of the tunnel surrounding rock. This provides an important reference for the control of floor heave in large-section shallow soft rock inclined shafts.

The Yingxiongling shale oil reservoirs in the Qaidam Basin are notably characterized by the development of the laminated shale and thin-layered shale. In order to identify the superior sweet spots, an experimental study on the physical and mechanical properties of laminated and thin-layered shales was conducted, and the sweet spots of the Yingxiongling shale oil reservoirs were systematically evaluated in conjunction with the analysis of oilfield production data. The results show that laminated shale exhibits higher total organic carbon (TOC) content, stronger hydrocarbon generation capacity, and higher initial oil saturation compared to thin-layered shale. Although laminated shale has lower porosity, it exhibits stronger anisotropy, higher stress sensitivity coefficient, and more developed initial natural microcracks. Additionally, the laminated shale demonstrates high horizontal permeability, strong fluid absorption capacity, and effective imbibition oil displacement ability. Its lower compressive strength facilitates the formation of complex fracture networks. The experimental research results are consistent with the in site fluid production analysis. The fracture morphology generated by the laminated shale is more complex with strong oil displacement ability and high oil displacement efficiency, indicating that the laminated shales produce fluids first; therefore, it is concluded that the laminated shales are the preferred sweet spots. The findings of this study have important theoretical significance for the exploration and development of shale oil in Yingxiongling.

Micro-nano pores are developed in unconventional oil and gas reservoirs such as shale and tight sandstone, and the study of oil-water two-phase seepage law and relative permeability in micro-nano pores is the theoretical basis for the effective development of such reservoirs. Addressing the two-phase flow of oil and water within nanopores, a mathematical model was established for nanopore-scale two-phase oil-water flow, grounded on the Hagen-Poiseuille (HP) equation and taking into account microscale seepage mechanisms such as oil-water distribution, viscous coupling, multi-layer adsorption, and slippage. Based on this model, a calculation method for relative permeability was developed. The validity of the model was verified by fitting the results from Lattice Boltzmann method (LBM) simulations. Furthermore, through a parametric sensitivity analysis, the characteristics of two-phase oil-water flow and the influence patterns of relative permeability within nanopores were investigated. The results showed that the slip length significantly impacted the velocity distribution of oil and water. An increase in the wetting contact angle led to varying degrees of augmentation in both oil and water relative permeabilities. As the viscosity ratio rose, the oil-phase relative permeability experienced a notable increase, while the water-phase relative permeability remained relatively unchanged. The presence of positive slip could result in relative permeabilities exceeding unity. As the pore radius enlarged, the pore area available for fluid flow expanded, thereby enhancing both oil and water relative permeabilities. This study holds guiding significance for elucidating the fluid flow mechanisms in micro-porous media and facilitating the exploitation of shale oil.

The small pore throat radius, high starting pressure gradient, poor formation physical property and large gas production difference of horizontal wells after fracturing in tight sandstone reservoirs seriously affect the increase of tight gas reserves and production. In view of the above problems, the production dynamics of gas wells under different levels of sweet spots in the eastern Ordos Basin were analyzed, and a new tight gas fracturing method based on sweet spot distribution was proposed. Based on the evaluation of rock mechanics parameters, energy storage coefficient and argillaceous content parameters, the evaluation model of geological sweet spot and engineering sweet spot was established, and the reservoir quality of tight sandstone gas reservoir was divided by fuzzy comprehensive evaluation method. According to different sweet spot types, the production characteristics and sweet spot distribution characteristics of Class I, II and III horizontal wells were summarized, and three fracturing modes based on geological sweet spot and engineering sweet spot were formed. The results show that when the horizontal well is drilled in Class I geological sweet spot and Class I engineering sweet spot, the cumulative gas production is higher than expected (Class I well), and a larger fracturing scale and displacement can be adopted to maximize the economic benefit (Model 1). When the horizontal well is drilled in Class I geological sweet spot and Class II engineering sweet spot combination, the cumulative gas production is lower than Class I well (Class II well), and the fracturing scale should be reduced appropriately to pursue a certain production capacity and economic benefit (Model 2). When horizontal wells are drilled in geological sweet spots below Class II and combinations of Class I/II engineering sweet spots, the production effect is the worst (Class III wells), and the fracturing scale should be controlled by reducing displacement to avoid the fracture from entering the risk layer and obtain a certain production rate (Model 3). The above new fracturing method was applied to a newly drilled horizontal well in a tight gas reservoir, and the open flow and gas production of the horizontal well after fracturing reached the expectation, which has certain guiding significance for the development of tight sandstone gas reservoir.

After fracturing, the solid particles carried by the high speed liquid will cause serious erosion to the oil nozzle, and it is difficult to ensure the stable operation of the oil nozzle. To address the serious erosion problem of the nozzle, numerical simulation was employed to study the erosion wear of the nozzle, and the influence patterns of sand content, sand grain diameter, sand grain density, pump displacement, and liquid viscosity on the erosion wear of the nozzle were analyzed. The research indicates that: when the sand content and liquid viscosity increase, the maximum erosion rate exhibits linear growth; when the sand grain density and pump displacement increase, the maximum erosion rate exhibits exponential growth; and when the sand grain diameter increases, the maximum erosion rate shows exponential decrease. The orthogonal test method is used to judge the significance of each factor. The factors affecting the erosion wear of the nozzle are as follows: sand content ratio > pump displacement > sand density > sand diameter > liquid viscosity.Based on the results of numerical simulation, the machine learning method is used to compare and analyze SVR(support vector regression), CNN(convolutional neural network), BP(back propagation) neural network and RFR(random forest regression) algorithm to predict the erosion wear results of oil nozzle respectively. By preferring the SVR algorithm and adopting the particle swarm optimization algorithm to optimize the prediction model, a better nozzle erosion prediction model is obtained.

To address the issues of incomplete feature extraction, poor stability, and limited generalization in traditional fault diagnosis models, a model based on a multi-scale convolutional neural networks (MCNN), bidirectional gated recurrent units (BiGRU), and multi-head self-attention mechanism (MSA) was proposed. The model was designed to achieve comprehensive feature extraction from both spatial and temporal perspectives. It took raw vibration signals as input, and multi-scale features were extracted through convolution kernels of different sizes. A multi-head self-attention mechanism was used to dynamically adjust output weights, disregarding redundant information and weighting the extracted features for fusion. Then the fused features were input into a BiGRU network, which utilized a bidirectional information fusion mechanism to explore information from both past and future directions, capturing dependencies between different parts of the input sequence. Finally, Softmax was employed for classification. Experimental validation was conducted using three bearing fault datasets, and the results show that the proposed model has excellent performance metrics on different datasets and showcases good generalization and feasibility.

Conventional diagnostic methods that require a large amount of data support in practical engineering are difficult to effectively perform centrifugal pump fault diagnosis under small sample conditions. Therefore, the residual network (ResNet) in deep learning was combined with dilated convolution and extended into a siamese network to construct a dilated residual siamese network (DRSN). The dilated residual network was used as the feature extraction module of the siamese network, which enhanced the feature extraction ability of the model. Positive and negative sample pairs were constructed to extract more information from each sample, and make more effective use of limited data.The two sub-networks share parameters, the number of free parameters and lowering the risk of overfitting was reduced when the sample was insufficient. The proposed network model alleviated the problem of insufficient training samples, improved the efficiency of data utilization, and realized the fault classification of centrifugal pump under the condition of small samples. The research results show that even in the most sample-scarce situation, the accuracy of the model on the centrifugal pump test dataset can still reach 82.20%, which is at least 8.8 percentage points higher than other models.