Tunnel lining detection is an important element of quality management in tunnel construction and maintenance. Due to the variety of internal lining defects and unclear boundaries makes it challenging to identify these problems and train models effectively. Relying on manual detection or existing models, it is not possible to achieve fast and accurate defect detection. To address the above problems, A dataset consisted of 1 922 liner radar samples collected from Yunnan Tunnel B-scan was developed for training the model. A tunnel lining defect detection model YOLO-Tunnel based on YOLOv5 was proposed, which improved the model feature extraction ability, increased the receptive field, and improved the model localization ability by upgraded the Backbone and Neck. And further improved the model detection ability by selected the appropriate model size and balanced weight based on the dataset's scale and target size proportions. The results show that YOLO-Tunnel has better defect detection accuracy compared to YOLOv5s and also meets the real-time detection requirements, in which the precision, recall, and mAP are increased by 2.5, 9.0, and 8.1 percentage points, respectively, with the inference time increases by 2.7 ms to 21.8 ms. The research results provide a reference for further improving the performance of the detection of tunnel lining detection and the direction of optimization of the model reference.

To address limitations in the engineering application of neural network based maximum power point tracking(MPPT) algorithms, an improved lightweight neural network MPPT algorithm was proposed. The complexity and memory usage of the neural network were reduced through a knowledge distillation compression algorithm, and a lightweight model was obtained. The inherent theoretical error of model predictions was corrected using an optimized variable step-size perturb and observe method. In the initial stage, the neural network predicted the voltage range of the maximum power point. In the later stage, disturbance observation progressively refined this range until it converged at the maximum power point. A simulation model was developed in MATLAB/Simulink, and a physical model was constructed for comparative experiments. Results indicate that the proposed algorithm achieves higher tracking efficiency, improved ripple voltage suppression, and lower resource consumption rate in embedded devices.

The advantages of renewable wind energy lead to a rapid growth in the scale of wind power, while lightning strike accidents on wind farm delivery systems have a significant impact on the new power system. The traditional lightning strike warning method requires high data types and sample sizes, and lacks consideration of relative location as well as the distribution of lightning density. A lightning strike warning method for wind farm delivery systems based on the stepped lightning strike probability calculation method was proposed. Firstly, the data of lightning points around a wind farm in Hainan, China in 2020 were analyzed, and the Monte Carlo method was used to find the center of mass of the clusters as well as the density of lightning points to fit the trajectory of the thunderclouds. Then, based on the relative position of the movement trajectory and transmission line, the stepped lightning strike probability calculation method was combined to calculate the value of the lightning strike probability in a short period of time. Finally, the simulation was combined with the operation monitoring data of a wind farm in Hainan from 2020 to 2022. The results show that the relative error of the proposed method is within 15%, and the impact of the difference in the density of lightning points on the warning accuracy is effectively reduced, which ensures the safety of the wind farm delivery system.

To explore the factors affecting customers' evaluation of fresh logistics service quality, a logistics service quality evaluation model was proposed and established based on sentiment analysis of online reviews and latent Dirichlet allocation (LDA). A convolutional neural network (CNN) model integrating a multi-head self-attention mechanism and bidirectional long short-term memory network (BiLSTM) was constructed for sentiment analysis of online comments. Additionally, LDA topic model was carried out for positive and negative comments after classification. The key factors affecting the evaluation of fresh product logistics service quality were obtained by exploring the focus of customers' demand for fresh product logistics service. The sentiment analysis based on CNN-BiLSTM-Attention was implemented through Python programming, and the results of sentiment analysis on online comments were compared with those of support vector machine (SVM), CNN, BiLSTM, and CNN-BiLSTM. The comparison results show that, compared with the classification results of other models, the CNN-BiLSTM-Attention model is superior in accuracy, precision, recall rate, F₁, and other indexes, effectively improving the accuracy of text emotion classification. The research results demonstrate that researching the factors affecting the logistics service quality of fresh e-commerce based on online review data can help e-commerce enterprises better improve logistics efficiency and service quality from the perspective of consumer demand.

In response to the current situation of relying on manual alignment of the optical path in existing velocity interferometer system for any reflector(VISAR) devices, and to meet the future demand for remote automated control, a new method for automatic alignment of the optical path was proposed. The complementary metal oxide semiconductor(CMOS) of this method was measured indirectly, and the pixel deviation of the light spot was used as a system input. Coefficient matrix transformation and discrete fuzzy feedback control methods were used to quickly eliminate the errors. Based on the modules such as vision and motion in the Windows control and automation technology(TwinCAT), each of which was run in a different real-time kernel, the communication link between the vision and motion control modules was eliminated, and fast real-time closed-loop control was realized. After the experimental verification of shock wave velocity measurement, the remote “one-button” automatic alignment was realized. The system can shorten the alignment time to 2 s and improve the alignment accuracy to 4.5 μm. The problem of inefficient manual adjustment of the existing device was solved, and the accuracy and stability of the system were improved.

Gas risk assessment of tunnel construction is one of the key issues to ensure the safe construction of tunnels passing through gas sections. Aiming at the three stages of investigation, design and construction in the process of tunnel construction, the interpretative structural model was used to reveal the hierarchical key of gas risk influencing factors in the above three stages. The hierarchical model of risk assessment was established, and the weight calculation method of risk assessment index was defined. Combined with data collection, literature collation and engineering investigation, the assignment standard of risk assessment index was proposed based on membership degree theory, and the gas risk assessment method in tunnel construction process was constructed. Combined with engineering examples, the rationality of the proposed method was verified. The results show that the pregnancy risk environment in the survey stage is the precondition, and the survey disturbance is the inducing factor. The pregnancy risk environment in the design stage is the precondition, and the design factor is the inducing factor. In the construction stage, the pregnancy risk environment is the precondition, the construction disturbance is the inducing factor, and the site management is the root cause. The risk is revealed in the survey stage, the risk is reduced in the design stage, and the safety is ensured in the construction stage. The proposed method is consistent with the on-site disclosure, which verifies the rationality of the proposed method. Through the above research, it can provide a theoretical basis for the risk determination of the tunnel crossing the gas area, and provide a reference for the selection of subsequent engineering measures.

In the petroleum industry, the CO₂ flooding technology plays an important role in many EOR methods. In recent years, CO₂ flooding technology has attracted more attention because of its positive contribution to carbon storage, but CO₂ injection will greatly increase the risk of corrosion failure of oil casing, and the development of corrosion inhibitors and the research on inhibition mechanism have achieved certain results. The research progress of organic corrosion inhibitors was systematically summarized from the perspective of the inhibition mechanism of organic corrosion inhibitors on CO₂ corrosion in the petroleum industry. The adsorption, reinforcement, bridging and hydrophobic film formation mechanisms of organic corrosion inhibitors were mainly introduced. The inhibition effects of organic amines, imidazolines, surfactants, polymers and carbon dots on CO₂ corrosion were compared and analyzed from the mechanism of functional groups and metal surfaces. The research on the inhibition mechanism of organic corrosion inhibitors in CO₂ environment and the development trend and focus of corrosion inhibitors were prospected.

Cardiac arrest and global cerebral ischemia-reperfusion injury after cardiopulmonary resuscitation (CPR) are common pathological conditions in critically ill patients, with high mortality and disability rates. Currently, the conventional asphyxia method is commonly used to establish a brain injury model after CPR. However, it has a low resuscitation success rate and postoperative survival rate. Therefore, a modified asphyxia model of cardiac arrest and global cerebral ischemia-reperfusion injury after CPR was established and compared it with the conventional asphyxia method to evaluate its advantages. Sprague Dawley(SD) rats were randomly divided into a conventional group and a modified group. The conventional asphyxia method and the modified asphyxia method were used to establish the models, respectively. The resuscitation success rate and 24-hour survival rate were observed. Neurological deficits were assessed using neurological deficit scores. Hematoxylin-eosin staining was used to observe pathological changes in brain tissue. Transmission electron microscopy was used to examine neuronal ultrastructure. Western Blot analysis was performed to detect inflammatory, oxidative stress, and apoptotic markers. The results show that compared with the conventional asphyxia method, the modified asphyxia method has a higher resuscitation success rate and 24-hour survival rate. No significant differences are observed in brain injury, inflammation, oxidative stress, and apoptosis between the two methods. The modified asphyxia method meets the requirements for studying global cerebral ischemia-reperfusion injury.

The reservoir of Chang 7 in Heshuinan area of Ordos Basin has high salinity of formation water and strong heterogeneity of mud content, which leads to unclear understanding of oil-water distribution and main controlling factors. It seriously restricts the efficient development of the oilfield. In order to improve the efficiency and accuracy of the identification of oil-water layer and clarify the distribution law of oil and water, the intelligent random forest method was used to interpret the fluid properties on the basis of logging, oil test and production test data, and defined the distribution law of oil and water in Chang 7 in Heshuinan area. In addition, the controlling factors of oil and water distribution were analyzed from various angles by combining the experiments of constant rate mercury injection, CT (computerized tomography) and nuclear magnetic resonance. The results show that the accuracy of the fluid properties interpretation can reach 78.9% by using random forest method. In the vertical direction, the oil layer developed in Chang 7₁ is better than that in Chang 7₂, and the oil-water layer and water layer are mainly developed in Chang 7₂. In the plane direction, the eastern oil layer is thicker and distributed more continuously, and the southwest oil layer near the B60 well area has higher water content. The fluid properties inside the reservoir are controlled by the shale content, the pore structure and the viscosity of crude oil. The pore size and the connectivity of pore throat of the reservoir are affected by the content of mud. The pore structure of the reservoir and the viscosity of crude oil together control the fluid mobility, thus affecting the spatial distribution of oil and water in Heshuinan area.

The spatial variation characteristics of soil fertility in potato growing area were clarified to provide theoretical basis for soil precise fertilization and fertilizer management in the study area. Taking Keshan Farm in Heilongjiang Province as the study area, 100 sample points were selected in the potato growing area, and soil pH, organic matter, total nitrogen, total phosphorus and total potassium were selected as the indicators to evaluate soil fertility. Geostatistics and geographic information system(GIS) were combined to analyze the spatial variation characteristics of soil nutrients, and soil comprehensive evaluation method was used to evaluate soil fertility in the study area. Results show that the soil is weakly acidic and the pH variation coefficient is small. The contents of organic matter, total nitrogen, total phosphorus and total potassium are at medium and high levels, belonging to moderate intensity variation. Soil pH is a moderate spatial autocorrelation, and the spatial autocorrelation of organic matter, total nitrogen, total phosphorus and total potassium is weak. The spatial accumulation of organic matter and total nitrogen is significant. The spatial variation of soil nutrients in the study area is obvious, showing an east-west direction, and the content of soil nutrients in the middle of the study area is relatively low. The soil fertility in the study area is above the medium level, and the area with good fertility accounts for 72% of the total area. The soil fertility of Keshan farm is good, which can meet the needs of potato growth. Human factors are the main factors affecting soil nutrient content.