Search

Fault-slip rockbursts are a major type of rockbursts in coal mine roadways, posing great hazards. Currently, there is a lack of systematic research on the entire process of fault-slip rockbursts, and prediction and prevention remain fundamental challenges. To achieve experimental simulation of fault-slip rockbursts, a testing system was developed, along with the development of low-strength high-brittleness similar simulation materials. A large-scale similar model containing faults and roadways was prepared, and a series of verification experiments were conducted, realizing the simulation of roadway shock failure and fault shear slip. The testing system is equipped with a triaxial six-face loading function, with a vertical loading capacity of 20 MPa and a model size of 1.50 m×0.75 m×0.75 m. The main frame innovatively adopts a key plate connection, ensuring high rigidity of the main frame. A honeycomb loading shell structure was invented, achieving oil cylinder group loading, which can meet the requirements of fault-slip rockburst simulation. A similar material suitable for fault-slip rockburst simulation was developed, which uses water glass as the binder, sodium fluosilicate as the curing agent, and talcum powder as the aggregate. The similar material has characteristics of both low strength and high brittleness, with a uniaxial compressive strength ranging from 3.44 to 7.81 MPa, an impact energy index of 9.2, and an elastic energy index of 8.83. Based on the test system and low-strength and high-brittleness materials, the bursting failure simulation of roadway under static load was realized, and the feasibility of simulating bursting failure of roadway using low-strength and high-brittleness materials was verified. The loading conditions of the critical equilibrium state of the roadway were obtained, with the continuous shedding of the roadway's side as a significant precursor characteristic of roadway damage. The shear slip of rough serrated faults was realized, confirming that the shear slip of rough faults would cause significant dynamic load disturbances to the roadway, obtaining a reasonable loading method for fault slip. The method of loading the top of the dynamic zone first and unloading the bottom later can achieve instantaneous slip of the fault and eliminate the influence on the static zone stress state.

ObjectiveTo establish an indicator system for multi-point trigger surveillance and early warning of infectious diseases, providing valuable evidence to improve the overall prevention and control framework. MethodsThe pool of indicators was initially developed through a literature review and focus group discussions, followed by two rounds of Delphi expert consultations with 21 experts nationwide. Based on the experts' ratings of the importance, usability and familiarity of the indicators, the indicators were screened, statistically analyzed and their final weights determined. ResultsThe positive coefficients of experts in the two rounds of consultation were both 100%, the authority coefficients of experts were both ≥ 0.79, and the coordination coefficients of expert opinions were 0.34 – 0.46 (all P < 0.001). According to the occurrence and development process of infectious diseases and the requirements of related multi-source data fusion and collision, the indicator system was divided into three dimensions: single-source data, multi-source data fusion and risk factors. Finally, 82 indicators for multi-point trigger surveillance and early warning of infectious diseases were established. ConclusionThe index system developed in this study for multi-point trigger surveillance and early warning of infectious diseases is based on sound scientific principles and reasoning. It has the potential to expand the channels of infectious disease monitoring and achieve multi-point trigger surveillance and early warning.

Early warning of geo-hazards based on meteorological factors has played an important supportive role in disaster prevention and mitigation in China since its inception in 2003. This paper summarizes the the 20-year development process, technical methods, and disaster reduction effects of the early warning works. (1) The development process of early warning work process is divided into three stages: initiation and promotion (2003-2009), deepening cooperation (2010-2017), and reform and enhancement (2018-2022). (2) With 24-hour early warning work as the main content, a progressive warning model and a relatively complete warning service system have been gradually formed. (3) Three sets of early warning model technology and method systems have been gradually developed, including critical precipitation threshold model, the threshold model based on geo-hazards risk, and the dynamic early warning models, with the publication of industry standards for warning. (4) The spatial and temporal accuracy of warning products continues to improve, with the national and 26 provincial warning spatial accuracies exceeding 5 km × 5 km. The focus is on 24-hour warnings, with development towards 72-hour and medium- to long-term forecasts. Over 8 provincial-level and some municipal and county-level authorities have implemented 3-hour short-term warnings, gradually forming a work system to support service short impending warning response, medium-term prevention and long term deployment. (5) Where there is warning, there is response. The Ministry of Natural Resources has taken the national early warning as one of the bases for initiating defense responses, and 18 provinces have clarified the working mechanisms of the early warning response linkage. (6) The awareness of multi-party disaster prevention has been continuously enhanced. With strengthened inspections,evacuations, and successful risk aversion after receiving early warning information, the effectiveness of disaster prevention and mitigation is evident. The experience of early warning works in the past 20-year can provide reference for the next step in promoting the early warning of geo-hazards based on meteorological factors, supporting the enhancement of China’s capability and level of geo-hazards prevention and control work.

Early warning of geo-hazards based on meteorological factors in China began in 2003, and early warning informations are issued to the public every year during the flood season to alert the public to guard against geo-hazards such as landslides, debris flows, and mudslides caused by rainfall. To better understand the current situation and future trends of research for early warning of geo-hazards based on meteorological factors, the visualization tool CiteSpace was utilized to review the domestic and foreign literature on geo-hazards meteorological risk warning published from 2003 to 2023 and construct the scientific knowledge map. The CNKI database and SCI-Expanded databases were used as data sources. The result show that: (1) The study of critical rainfall thresholds based on the relationship between historical rainfall and disaster is in the leading position in research on early warning of geo-hazards based on meteorological factors, and the study of rainfall-related thresholds will continue to be a research hotspot in the future. (2) Keyword cluster analysis shows that a large number of studies have conducted research on early warning and prediction models and critical thresholds of geo-hazards under different formation conditions, meteorological conditions, different types, and genetic models, indicating that refinement is an important direction for the development of early warning of geo-hazards based on meteorological factors. (3) In the international research, Chinese scholars account for 39% of the total number of publications, showing an absolute advantage in research on geo-hazards meteorological risk early warning. The research results objectively demonstrate the development context, domestic and foreign research hotspots, and trends of research on geo-hazards meteorological risk warning in China, aiming to contribute to the progress of geo-hazards research and early warning forcast business in China.

Improving the geo-hazard meteorological risk warning capabilities is an important component of the comprehensive prevention system for geological disasters. Since 2003, Sichuan Province has been carrying out geo-hazard meteorological risk warning work. Building on the previous foundation, research has been focused on the analysis and establishment of warning models, system design and implementation, and graded warning operation modes since 2019. A geo-hazard warning model library, an integrated geological disaster meteorological risk warning interconnection system at the provincial, municipal, and county levels, and a graded warning mechanism have been created. We have achieved closed-loop management of interconnection and graded early warning based on integrated platform such as rainfall data transmission, early warning model analysis and calculation, and early warning information release in province, cities, and counties. This has solved the three problems of lack of technology, capacity, and funding in cities and counties, as well as the interference caused by multiple early warning information releases on grassroots levels. In 2023, 86% of the successful risk avoidance cases in Sichuan Province were affected by meteorological risk warning, effectively reducing casualties and property losses caused by geo-hazard.

The rainfall-induced shallow landslides are primarily debris landslides, which features simultaneity with significant hazard, and the hydrological response mechanism of water table and soil moisture content to precipitation of this type of landslide is sophisticated, which makes it difficult to predict the slope stability accurately. To further study the influence of the rainfall-triggered internal hydrological responses on slope-stability, on-site precipitation infiltration monitoring, correlation analysis and mechanical analysis were carried out on the Houshanli landslide in Qingchuan County, Sichuan Province. The relationship between precipitation and water table was proposed based on climate and hydrological monitoring data obtained within three year interval. The response of rainfall infiltration, soil volumetric water content and water table were analyzed. The results indicate that: (1) groundwater exhibits periodic fluctuations throughout the year, characterized by three phases of slow decline, rapid decline, and rapid ascent; a linear negative correlation between precipitation and water table was found, and no significant correlation was observed with the water table increment; (2) through the infinite slope model and the relationship between precipitation and water table, a prediction model for shallow landslide stability was constructed. The precipitation threshold (81.8 mm/d) and water table threshold (0.73 m) were determined which has good agreement with the actual situations. This provides an early warning method for rainfall-induced shallow landslides by monitoring these two factors.

Landslide disasters occur frequently in Fujian Province, and early warning of landslide disasters on a regional scale is an important means of effective disaster prevention and mitigation. Due to the complex mechanism of landslide disasters, the traditional regional landslide early warning methods have such problems as insufficient accuracy. Deep learning mainly refers to the technology of feature extraction, abstraction, representation and learning by constructing the neural network model, which is a kind of machine learning. As a classical deep learning algorithm, convolutional neural network has more powerful classification and representation ability than traditional machine learning. Taking Fujian Province as the research area, this paper introduces the convolution neural network into the field of landslide disaster early warning and constructs a regional landslide early warning model of Fujian Province. The process is as follows: (1) The SMOTE optimization algorithm is used to optimize the sample database of landslide disasters in Fujian Province from 2010 to 2018, enlarging the number of positive samples and expanding the proportion of positive and negative samples from 1∶3.4 to 1∶2, and the total number of samples reaches 18040. (2) Construct a convolution neural network model structure, which includes an input layer, two convolution layers, two maximum pooling layers, a full connection layer and an output layer. (3) Use the convolution neural network to train the optimized samples (80% of the samples from 2010 to 2018 as the training set), and use the Bayesian optimization algorithm to optimize the model parameters to obtain the regional landslide early warning model of Fujian Province. (4) The model is tested with 20% of the samples from 2010 to 2018 as the test set, and the confusion matrix and ROC curve are used to test the model. The results show that the accuracy of the model ranges from 0.96 to 0.97, the AUC value is 0.977, indicating that the model accuracy and generalization ability are good. (5) The actual situation of the landslide disaster in the flood season of 2019 is taken as a positive sample, negative samples are collected through the method of time-space sampling, and the 2019 regional landslide sample verification set (603 samples) is constructed. The model is further verified by using the confusion matrix and ROC curve. The results show that the accuracy of the model ranges from 0.75 to 0.85, and the AUC value is 0.852. Although only the actual landslide samples in the flood season of 2019 is used for verification, good results is also achieved. In this paper, the convolution neural network algorithm is applied to the regional landslide early warning, which provides a new way to establish the regional landslide early warning model. The preliminary verification shows that the model is effective and will be further applied and verified in Fujian Province in the future.

Aiming at the problems of low intelligence of current tunneling equipment, difficulty in describing over-excavation, under-excavation and abnormal collision in tunneling process, and difficulty in adapting traditional automatic cutting and memory cutting technology to complex geological conditions, a digital twin-driven virtual teaching memory cutting control method for cantilever roadheader is proposed. By analyzing the research situation of digital twin technology in the field of intelligent coal mining, the overall scheme of memory cutting control system of cantilever roadheader driven by digital twin is designed, and the key technology of memory cutting of cantilever roadheader under complex working conditions is studied. Firstly, the characteristics of digital twin and virtual reality technology are fully utilized to study the virtual teaching strategy under complex working conditions. Based on the Unity3D platform, the virtual twin model of the working face and equipment with the same size of the corresponding entity, the kinematics model of the cutting unit and the virtual collision detection model are established. The virtual model movement is controlled through the intelligent interactive interface at the virtual end, and the teaching trajectory is designed and optimized according to the worker’s experience, so that it can be used as the target expected trajectory of trajectory tracking to make up for the excessive dependence on the worker’s experience caused by the traditional underground manual teaching due to the harsh working conditions. Secondly, in order to improve the quality of section forming, the control method of teaching trajectory tracking and reproduction in memory automatic cutting stage is studied. The dynamic model of cutting part is established by Lagrange method, and the tracking control accuracy of end effector to teaching trajectory is improved by combining iterative learning with sliding mode control. Finally, the simulation control platform of the memory cutting of the cantilever roadheader is built. Through the real-time data transmission and interaction between the virtual space and the physical space and between the modules, the three-dimensional visual simulation of the memory cutting virtual teaching and trajectory tracking control process is completed in the virtual space, and then the memory automatic cutting trajectory tracking control command is generated and sent to the end effector of the physical entity of the cantilever roadheader to drive it to carry out the section forming cutting according to the teaching trajectory. At the same time, the physical sensor collects the pose data of the cantilever roadheader fuselage and the cutting arm, and reversely drives the virtual model to move synchronously. The closed-loop control of robot virtual model and physical entity is realized. On this basis, the virtual and real synchronization of the system, the motion consistency between the virtual prototype and the physical prototype, and the trajectory tracking and reproduction control accuracy are verified. The experimental results show that the system data transmission delay is low, which can ensure the virtual and real consistency and synchronization, and the trajectory tracking control accuracy meets the actual use requirements. This method provides a new idea for memory cutting and intelligent control of tunneling equipment.

Objective To investigate the effects of different anticoagulants on platelet-rich plasma(PRP) release content of growth factor and injection pain. Methods A total of 15 voluntary blood donors were selected, with each blood donor using four kinds of anticoagulant tubes with EDTA-K2 anticoagulation, EDTA-NA2 anticoagulation, citrate anticoagulation, ACD-A anticoagulation respectively as group A, B, C and D. PRP was isolated and prepared by the rich plasma method, and the contents of PDGF-AA, TGF-β, IGF-1, VEGF, and PF-4 were detected by enzyme-linked immunosorbent assay. Meanwhile, SD rats (20, 4 / group) were injected subcutaneously or intradermally with the supernatant of PRP and PG gel prepared in the 4 groups and normal saline in the control group. The pain status of SD rats during the injection was observed and recorded. The pain status of the 5 groups of experimental animals was evaluated according to the American Laboratory Animal Pain Guide. Results The platelet counts in PRP in group D was the highest [(1 294.53±277.37) × 109/L], which was significantly higher than that in group A [ (789.13±377.13) ×109/L] and group C [ (990.94±493.12) ×109/L] (P0.05). The OD value of VEGF was 0.13±0.04, 0.21±0.14, 0.08±0.02 and 0.13±0.04 respectively, with statistically significant difference between group B and C (P0.05) . The median pain scores of the PG supernatant injection of group A, B, C, and D were 4 (2.25), 3 (2.75), 4 (3), 1 (1.5), and the difference was not statistically significant (P>0.05). There was no significant difference between the PRP injection group and the PG supernatant group (P> 0.05). Conclusion PRP prepared by two-step centrifugation with ACD-A anticoagulant can obtain the higher platelet counts and the maximum release of PDGF-AA, PF-4, IGF-1, and TGF-β. In terms of pain, ACD-A anticoagulant injection has the lowest pain with the animals.

Objective To study prevalence characteristics and change pattern of dominant virus strain of influenza epidemics in Guangzhou city and to provide evidence for formulating region-specific prevention and control measures. Methods Weekly influenza surveillance data of Guangzhou city, Guangdong province from 2015 to 2022 were collected from China Influenza Surveillance Information System. The positive rate series of influenza virus and its pathogen subtype were decomposed into trend/season/remainder component by using the seasonal-trend decomposition procedure based on locally weighted regression. By calculating covariance, the contribution rate of the obtained components to the fluctuation of positive rate was analyzed. Results Of 39 198 surveillance specimens sampled in the city during the 8-year period, 3 632 (9.27%) were positive for influenza virus nucleic acid. Trend series analysis showed that the positive rate fluctuated from 4.57% to 16.07%, with an increasing trend, during 2015 – 2019 and varied with a U-shape trajectory during 2020 – 2022. Winter-spring and spring-summer peaks of the positive rate were observed in season series analysis and the remainder component analysis revealed the highest fluctuation range of 48.96% for the positive rate in 2022. The positive rate of influenza A (H3N2) presented a similar season series to that of influenza B (Victoria), with winter and summer peaks, but the summer peaks of influenza A (H3N2) was later than that of influenza B (Victoria); the peak shape and prevalence duration (single peaks mainly in winter-spring season) of influenza A (H1N1) pdm09 positivity were similar to those of influenza B (Yamagata). In general, the remainder component was the main contributor to the fluctuation of the total positive rate, accounting for 59.16% of overall fluctuation (covariance of remainder component versus that of original series: 80.81/136.60), followed by season and trend component, accounting for 25.19% (34.41/136.60) and 15.68% (21.38/136.60) of overall fluctuation, respectively. Conclusion There were virus subtype-specific differences in seasonal peak and prevalence duration of influenza infection and random fluctuation exerted a greater effect than long-term trend and seasonal variation on influenza epidemic pattern in Guangzhou city.

Aiming at the problems of expensive collection cost and missing data due to the privacy and continuity of time series data set, a multi-discriminator generative adversarial network model based on recurrent neural network was proposed, which could synthesize time series dataset that were approximately distributed with real data of a small scale dataset.Multi-discriminator included four discriminators in time domain, frequency domain, time-frequency domain and autocorrelation.Different discriminators could effectively recognize the features of the time series in different domains.In the experiment, the convergence of loss function, principal component analysis and error analysis were performed to evaluate the performance of the model from qualitative and quantitative perspectives.The experimental results show that the proposed model has better performance than other reference models.

The landslide disaster in China is widespread and serious. Regional landslide risk assessment has always been one of the most important contents of landslide disaster prevention and mitigation. In recent years, with the rapid development of big data and artificial intelligence technology, machine learning technology has gradually been widely used in landslide hazard assessment andachieved good results. Based on a large number of literatures, this paper systematically expounds the research status of landslide risk assessment methods based on machine learning technology. This paper reviews and analyzes the existing research results from three key links: evaluation factor selection and quantization normalization, data cleaning and sample set construction, model selection and training evaluation, and finally puts forward some suggestions on the development trend of machine learning landslide risk evaluation methods.

In Qingchuan County of Sichuan Province, landslide disasters occur in a large number of places and cover a wide range of areas. Early warning of regional landslide disaster is an important means of effective disaster prevention and mitigation, and an early warning model is the core of successful early warning. The traditional regional geological disaster warning model is limited by the lack of big data and analysis methods of the complicated investigation and monitoring mechanism of the landslide in the study areas, and it has some problems, such as limited warning precision and insufficient refinement. In this paper, the training sample set of landslide disaster in Qingchuan County is constructed on the basis of the integrated collation and data cleaning of the results of geological disaster investigation and monitoring and precipitation monitoring. The sample set includes 27 input feature attributes such as geological environment rainfall and 1 output feature attribute, covering the total number of the samples in Qingchuan County in the past 9 years (2010—2018) up to 1826 (613 positive samples, 1213 negative samples). Based on the logistic regression algorithm, the study and training of the sample set is carried out with a 50%-fold cross validation. The Bayesian optimization algorithm is used for model optimization, and the accuracy and model generalization ability of the model are verified by such indicators as accuracy, ROC curve and AUC value. The ROC curve is also known as the “Receiver Operating Characteristic” curve. AUC value represents the area under the ROC curve. The verification results show that the training result model based on logistic regression algorithm is of good accuracy and generalization ability (accuracy 94.3% and AUC 0.980). Finally, it is proposed that in the actual warning of regional landslide, 27 characteristic attributes of each warning unit in the research area are input according to the format of characteristic attributes of training samples, and the pre-learned and trained model is called to output the probability of occurrence of landslide disaster, and the warning level of landslide disaster is segmented according to the output probability. A yellow alert is issued when the output probability P is greater than or equal to 40% and P is less than 60%. An orange alert is issued when the output probability P is greater than or equal to 60% and P is less than 80%. A red alert is issued when the output probability P is greater than or equal to 80%. In the next step, the accuracy of the model will be further verified in the landslide disaster early warning business in Qingchuan county.