In order to improve the prediction accuracy of lithology affected by imbalanced geological data, an ECA-MSCB ResNet model was proposed. The model integrates ECA (efficient channel attention) and MSCB (multi-scale convolutional block) into the traditional ResNet architecture to achieve efficient extraction and representation of lithological data features. For the issue of imbalanced lithology categories, prior probability-balanced logit bias was introduced during model training, and the focal loss function was modified to enhance the recognition of minority lithology classes. Experimental results show that the model based on ECA-MSCB ResNet performs well on the imbalanced geological lithology dataset, achieving an average prediction accuracy improvement of approximately 7.45% compared to the original ResNet model and 27.33% compared to the random forest method. Notably, the recognition of minority lithology classes improves by an average of 17.9%. Furthermore, the model demonstrates strong lithology classification ability on public datasets, achieving an F₁-score of 75.77%. In addition, the recognition accuracy of the proposed model outperformed both traditional and mainstream methods. The ECA-MSCB ResNet method holds significant application value in the field of imbalanced geological lithology recognition.

Fire flooding is a primary method used to enhance heavy oil recovery, often replacing steam stimulation. However, it encounters challenges such as low sweep efficiency and delayed effective times in heavy oil development. A new fire flooding PGI (pulse gas injection) technology was used to solve these problems. The combustion sweep effect of fire flooding can be improved by adjusting the working system of the gas injector. Based on the geological characteristics of the Hongqian 1 heavy oil reservoir in Xinjiang, the feasibility and mechanism of PGI were elucidated through numerical simulations. The influence of geological and engineering factors was studied on the development effect of PGI, and the consequences are applied to the fire flooding field. The combustion front can be controlled by the peak-valley value stage of PGI, accelerating the uniform movement of the combustion front and improving the oil displacement efficiency. Specifically, PGI can reduce the effective time by approximately 600 days and increase combustion sweep by more than 30%. This technique is particularly suitable for medium heterogeneous reservoirs with a vertical permeability contrast of less than 15 and crude oil viscosities ranging from 2 000 to 10 000 mPa·s. The optimal pulse amplitude ranges between 1.5 and 2, with a recommended step length of 30 days. When applied to the Hongqian 1 fire flooding industrial area in Xinjiang, daily oil production increased by 0.5 to 2.8 t for the well group, and the air-oil ratio decreased by 34%. PGI can achieve a better production increase effect for field production.

Oil transfer station plays a crucial role in the oil and gas gathering and transportation system of an oilfield, ensuring stable production and continuous supply of oil and gas. However, given the complexity of its process system and the ambiguous uncertainty surrounding fault modes and relationships, a systematic reliability assessment method integrating T-S fuzzy fault trees with BNs(Bayesian networks) was proposed. Firstly, a T-S fuzzy fault tree was established based on T-S gates and their descriptive rules, which is subsequently converted into a Bayesian network model. Secondly, leveraging limited fault samples and general data sources, Bayesian updating estimation was employed to determine the failure rates of basic events, addressing the uncertainty inherent in fault sample data. Lastly, the T-S fault tree and BN model were synergistically utilized for forward reasoning to predict the reliability of the process system and the contribution of basic events, while reverse diagnosis is conducted to pinpoint the key factors causing different fault states of the system. Research conducted on typical oil transfer station process systems has demonstrated that the proposed method can effectively predict system failure rates and diagnose weak links even under conditions of uncertainty in basic data and event relationships. This provides crucial decision support for the optimal design and reliability maintenance of complex oil and gas process systems.

In order to advance the analysis and mining of TCM(traditional Chinese medicine) text data and achieve intelligent extraction and processing of knowledge, the BIO(begin, inside, outside) sequence labeling method, the BiLSTM-CRF model, and manually defined rules were adopted to complete the knowledge extraction task. Utilizing the Py2neo library in Python 3.6 and the Neo4j database, a spleen and stomach disease knowledge graph was constructed based on Neo4j, and a TCM spleen and stomach disease named entity recognition system was developed using the Flask framework. The results show that the BiLSTM-CRF model achieves high performance and good generalization ability on the test set, with accuracy, precision, recall, and F₁ scores of 96.19%, 86.64%, 88.82%, and 87.71%, respectively. The constructed knowledge graph includes eight types of node labels, such as prescriptions or patent medicines, Chinese medicines, and clinical manifestations, as well as ten types of relationships. It supports the querying and discovery of nodes and relationships among Western medical diagnosis, TCM syndromes, and TCM treatment principles for spleen and stomach diseases. It is concluded that the BiLSTM-CRF model demonstrates excellent generalizability in named entity recognition of TCM spleen and stomach disease. It exhibits outstanding performance in handling complex text structures and domain-specific terminology, providing strong support for the research on knowledge extraction and knowledge graph construction in Traditional Chinese Medicine for spleen and stomach diseases.

The existing CFRP(carbon fiber reinforced plastic) plate clip-type anchorage exhibits arching deformation during the tensioning process due to compressive forces on the inside wall of the anchor cup. This causes “voids” at the contact interface between the anchor cup and the clip, leading to an uneven distribution of lateral forces within the CFRP plate anchorage section. The sides of the CFRP plate are prone to cracking failure due to stress concentration. An optimized design for arching clip-type CFRP plate anchors was proposed to address this issue. Finite element numerical simulations and static loading tension tests were conducted on clip-type anchors with varying arch heights. The findings show that the primary failure mode of conventional CFRP plate clip-type anchors is initial cracking followed by fragmentation, with an anchoring efficiency of only 68.75%. When the arch height is low, the voids in the anchor cup are not adequately filled, resulting in lower compressive stress and an anchoring efficiency of 56.67%. When the arch height is too high(0.5 mm), stress concentration occurs in the middle section of the CFRP plate anchorage, which increases the anchoring efficiency to 81.25%, but this is still suboptimal. Notably, when the arch height is set to 0.25 mm, anchoring efficiency increases to 90.83%, and the failure mode shifts to explosive failure, indicating that the CFRP material has been fully utilized. The rational adjustment of the clip’s arch height effectively prevents cracking failures in the CFRP plate anchorage due to voids, demonstrating the significant engineering application value of this research.

The external human-machine interface is used to enhance communication between autonomous vehicles and road users like pedestrians and cyclists, thus traffic safety and user experience are improved. The recognizability of the external interface is considered the foundation for ensuring effective and understandable signal functions, and it is explored to ensure pedestrians crossing safety. The form of information expression, interface location, and the speed of autonomous vehicles were considered as independent variables, and eye-tracking technology was used to collect eye movement and behavioral data. The identifiability of the interface was evaluated through repeated measures analysis of variance and logistic regression. The results show that the form of interface information, location, and vehicle speed significantly affect identifiability. The light band has the best identifiability; higher recognition efficiency is observed when the vehicle is traveling at a low speed; and the highest recognition efficiency is found when the interface is at P3, while the lowest is noted at P1 and P4.From the perspective of enhancing pedestrian traffic safety, this study provides a reference for the design of the external interface of autonomous vehicle while driving, and helps to improve pedestrian attention and recognition accuracy of the external interface.

Located in the south of the Central Asian orogenic belt, the southern Beishan belt in Gansu Province is a key area for studying the tectonic evolution of the Central Asian orogenic belt. Its late Paleozoic tectonic setting has been controversial for a long time. In order to further explore the late Paleozoic tectonic evolution of the southern Beishan belt,the geochronology and geochemical characteristics of the Changshan monzogranite body were analyzed. The analysis results show that the LA-ICP-MS zircon U-Ph weighted average age of the Changshan monzogranite is (291.1±1.5) Ma, and the emplacement of the plutons occurred in the early Permian. Geochemical datas show that the plutons are high potassium calc-alkaline and peraluminous series rocks. The results show that SiO₂ ranges from 72.07% to 72.94%, K₂O ranges from 4.93% to 5.10%, and the contents of K₂O>Na₂O, A1₂O₃ ranges from 13.52% to 13.97%. The curves of chondrite-normalized REE are obviously right inclined, and the LREE are relatively enriched (LREE/HREE are 10.96~14.98), δEu are 0.78~0.92, with weak negative Eu anomaly. Trace elements are relatively enriched in LILE (large ion lithophile elements), depleted HFSE (high field strength elements), and significantly depleted in high field strength Elements Nb, Sm, Y. According to the regional tectonic setting, petrological and geochemical characteristics, the Changshan adamellite plutons are considered to be the product by post-collisional magmatic activity, reflecting the completion of the collision collage on the southern margin of the Central Asian orogenic belt in the early Permian.

In the exploration and exploitation of oil and gas, artificial intelligence models are extensively employed in the prediction of formation pore pressure. Among them, single models tend to encounter problems such as overfitting or unstable prediction outcomes, leaving room for improvement in aspects like prediction accuracy and generalization ability. To enhance the prediction accuracy of formation pore pressure, a CNN-Attn neural network-based formation pore pressure prediction model was established by virtue of deep learning technology. In this research, five types of logging and while-drilling data were optimally selected, and the linear correlation between the data and formation pore pressure was verified using the Pearson correlation coefficient method. Through the optimization of the structure of the one-dimensional CNN, the model can effectively capture the local characteristics of the data and, when combined with the self-attention mechanism, strengthen the model’s ability to capture global dependencies, thereby elevating the model’s expressiveness and comprehension. To validate the prediction accuracy of this model, two wells in the Bayan block were subjected to prediction. The average absolute errors of the prediction results were both less than 1 MPa, the root mean square errors were both less than 1 MPa, the average relative errors were both less than 1.3%, and the determination coefficients were both greater than 0.9, with higher accuracy compared to the BP, CNN, and LSTM models. This model has improved the prediction accuracy of formation pore pressure and provided data support for drilling safety.

In the ASIST system, data on 86 917 abnormal events from 2017 to 2023 are collected as research objects, and an indicator system for abnormal events was established. To ensure the safety of aviation operations, accurate and reliable risk assessment models were developed to analyze abnormal events in depth, thereby achieving effective risk management. Firstly, the principle of catastrophe theory was introduced into the fuzzy inference system, which enables it to better handle complex issues and enhance the accuracy of evaluations. Then, a risk assessment model based on catastrophe theory and fuzzy inference system was developed to assess the risks of abnormal aviation events. Additionally, 56 cases with detailed background information records were selected for instance analysis, and compared with the cloud model, to verify the feasibility and accuracy of the model. Finally, relevant indicators were controlled using fuzzy methods, providing guidance for the safety management work of aviation operations.

The two important characteristics of microseismic are: tiny and shear rupture. The resulting monitoring characteristics are significantly different from those used to monitor natural earthquakes and artificial seismic exploration sources. Microseismic and its monitoring characteristics are the cornerstone of the development, application, and judgment of microseismic monitoring methods. First, different monitoring methods were investigated, suggestions were puts forward for their development and scope of application, and the reasons why some methods have not improved much were exploved. Among them, the most important ones are: when the number of microseismic, positive and negative initial motion, and signal-to-noise ratio are not easy known, it is necessary to conduct large-scale trial calculations and statistically investigate the combination of focal mechanisms with a high probability, so as to complete reasonable migration stacking. Mathematical statistics in the denoising should be used throughout all steps of detection, and so on. From the perspective of probability and mathematical statistics, following the characteristics of microseismic and its monitoring, the results show that microseismic monitoring has to be based on the fact of low signal-to-noise ratio, summarizes and improves the principle and denoising of VS(vector scanning). In the process of VS processing and interpretation automation, a large number of mathematical statistics are implemented to confirm the noise coherence parameters and analyze the microseismic activity. It makes up also for the defect that the vertical accuracy of ground monitoring is poor and cannot confirm the vertical height of the stimulation rock volume. VS has formed a relatively complete ground monitoring system after more than 20 years of research and development. Probability and mathematical statistics are important concepts and tools to ensure the success of the development and application of microseismic monitoring methods.