Table of Content

28 September 2023, Volume 33 Issue 9

Previous Issue    Next Issue

Safety science theory and safety system science Safety social science and safety management Safety engineering technology Public safety

Safety science theory and safety system science

Piezoelectric safety & security management and its model

WANG Bing, WANG Yuanjie, ZHANG Zhihong

2023, 33(9):  1-9.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0879

Asbtract ( 278 )   HTML ( 2)   PDF (3040KB) ( 10 )  

Figures and Tables | References | Related Articles | Metrics

In order to innovate the safety & security management paradigm and further enrich the safety & security management theories, the research work of piezoelectric safety & security management was carried out based on the piezoelectric theory and safety & security management theory. Firstly, the inspiration of piezoelectric theory for safety & security management was analyzed, the connotation of piezoelectric safety & security management was explored, and the piezoelectric safety & security management conceptual model was proposed. On this basis, the model for piezoelectric safety & security management was constructed and explained by combining the whole life cycle theory of safety & security management. Finally, the characteristics of piezoelectric safety & security management were described. The results show that, piezoelectric safety & security management is a new paradigm, which can take targeted measures to meet the system's safety & security needs according to the changes of system's safety & security status in a certain time and space. Moreover, piezoelectric safety & security management includes two safety & security management modes (namely, normal safety & security management and non-normal safety & security management), and covers three safety & security management behaviors (namely, safety & security risk prevention and control, emergency response, as well as adjustment and optimization).

Safety social science and safety management

Construction safety risk analysis of bridge girder-erecting machine based on complex network integrating N-K model

WANG Haiying, CHU Lin, XU Jian

2023, 33(9):  10-17.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1560

Asbtract ( 198 )   HTML ( 2)   PDF (2060KB) ( 8 )  

Figures and Tables | References | Related Articles | Metrics

To identify the critical causes and paths of bridge girder-erecting machine accidents in construction engineering, a risk factor analysis method based on complex network combined with N-K model was proposed. Firstly, the cases database of bridge erecting machine accidents between 2000-2022 was established, and the accident causes of bridge girder-erecting machine were classified into four first-level risk factors and twenty-four second-level risk factors by analyzing the cause of the accidents. Then, a complex network model with major accident types and the second-level risk factors as nodes and their correlations as edges was established, and the closeness centrality and betweenness centrality of risk factor nodes were calculated. Finally, the N-K model was developed for the first-level risk factor coupling analysis, and the centrality of the risk factor nodes was optimized by using the coupled interactive information. Through the accessibility analysis of risk factors and global main path analysis, 7 key causes and 7 key links of bridge machine accidents were obtained respectively. The results show that the weak management of construction unit and the insufficient safety production inspection have the high value of closeness centrality and betweenness centrality. The biggest risk is from the path of duty dereliction of supervision unit→the weak management of construction unit→the insufficient safety production critical. So, all these key causes and paths can be taken as the focus of safety control during the construction of bridge girder-erecting machine. The key cause of bridge girder-erecting machine construction safety obtained by complex network integrating with N-K model is consistent with the conclusion of accident investigation, and the risk identification of girder-erecting machine construction safety accident is accurately realized.

Construction and application of occupational safety and health management model in Ashele copper industry

WU Song, SUN Pengyi, CHAI Wenhao, LIU Xianpeng, TONG Ruipeng

2023, 33(9):  18-24.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0763

Asbtract ( 112 )   HTML ( 1)   PDF (5951KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

In order to improve the level of occupational safety and health management in mining enterprises, based on the analysis of the elements and theories of occupational safety and health management, a safety and health management model with safety culture, organizational guarantee, behavior control, and technological support as the core elements were proposed, and its operational mechanism was clarified. Based on its theoretical framework, combined with the specific reality of occupational safety and health management in Ashele Copper, the "double ten and triple three" occupational safety and health management model was constructed, and taking the application practice of the enterprise as an example to verify the model's effectiveness. The results indicate that the "double ten and triple three" occupational safety and health management model is based on "Ten Aphorisms", supported by the "Ten Guidelines", with the "Three Systems", "Three Constructions" and "Three Complete Management" as the core contents, has formed a unified and standardized discourse system, which has a high degree of generality, scientificity, feasibility, and scalability, and can contribute to the development of occupational safety and health management in mining enterprises.

Safety engineering technology

Rockburst prediction model based on improved Smote-GBDT algorithm

SONG Yinghua, JIANG Chen, LI Moxiao, QI Shi

2023, 33(9):  25-32.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0850

Asbtract ( 127 )   HTML ( 0)   PDF (4162KB) ( 8 )  

Figures and Tables | References | Related Articles | Metrics

This paper aims to accurately predict rockburst levels and ensure the safety of construction personnel and equipment. First, from the perspective of rock burst mechanism, eight indicators of burial depth (D), uniaxial compressive strength (UCS), uniaxial tensile strength (UTS), rock brittleness index (B₁、B₂), maximum tangential stress (MTS), stress concentration coefficient (SCF) and elastic deformation energy index (W_et) were analyzed, and a rock burst prediction index system was established. Secondly, to address the problem of data imbalance in rockburst samples, the Tomek Link of under sampling method was introduced to improve the (Smote) for mixed oversampling of rockburst training samples. Finally, the SmoteTomek-GBDT rockburst prediction model was constructed, and the validity of the model was verified with 38 sets of data and compared with other models. The results show that the accuracy of SmoteTomek-GBDT is 92.1%, and it is a 5.3% improvement over unsampled and 10.5% improvement over Smote sampled, which is better than the random oversampling model, and avoids cross-grade rockburst misclassification, which is of some significance for accurate rockburst prediction.

Improved system layout design of mountain tunnel construction site safety layout

WANG Jingchun, LI Yonghao, LIU Kailin, ZHAO Fei

2023, 33(9):  33-40.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1313

Asbtract ( 107 )   HTML ( 0)   PDF (11470KB) ( 5 )  

Figures and Tables | References | Related Articles | Metrics

In order to improve the level of safety layout of the construction site of the mountain tunnel, taking the construction site of Niuluan village tunnel as an example, an improved SLP method was proposed to lay out the site for 13 units such as concrete mixing station, machinery storehouse, cement storehouse, etc. The method was combined with a Bayesian network model, and the weight of important factors of the site layout of the mountain tunnel was determined by comparing the factors influencing the layout of the building construction site and the comprehensive interrelationship between the laying units, forming 2 site layout optimization schemes, and determining the relative optimal site layout scheme by establishing a mathematical model with safety level intensity, material flow intensity and transportation time as the objective function, and comparing the optimization scheme with the original scheme by combining the factor weighting method. The results show that the optimal site layout scheme derived from the improved SLP method is smaller than the original scheme in terms of the altered area, and improves 12.30% in terms of the safety level intensity, material flow intensity and transportation time, which helps to improve the efficiency of mountain tunnel construction and enhance the level of construction site safety.

Evaluation model of rockburst intensity of diversion tunnel based on WOA-SVM

JIN Chunling, JI Zhaotai, GONG Li, AN Xiang, ZHOU Yi

2023, 33(9):  41-48.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1143

Asbtract ( 86 )   HTML ( 0)   PDF (3806KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

This study aims to reduce the occurrence of rockburst accidents in the construction of diversion tunnels. Before construction, rockburst intensity assessment and prevention measures were put forward. Four evaluation indexes were selected as rock burst evaluation indexes in the study, namely, rock uniaxial saturated compressive strength R_c, rock uniaxial tensile strength R_t, maximum tangential stress σ_θ of surrounding rock and rock elastic energy index W_et. According to the research results, 120 groups of rockburst instances were selected as the machine learning sample data, and the WOA-SVM assessment model was proposed. Taking a section of a deep-buried and long diversion tunnel in Xianglushan of the Central Yunnan Water Diversion Project as an example, the intensity assessment results of rockburst intensity were verified. The results show that machine learning can better avoid human factors, and it is completely data-driven with an assessment accuracy of 97.22% for WOA-SVM. By comparison, its assessment accuracy and generalization are better than those of PSO-SVM, GA-SVM and WOA-BP neural network models. The results show that the WOA-SVM model can better capture the link between rockburst levels and indicators for rockburst problems.

Remaining useful life prediction method of rolling bearing based on EWM and SVR

GU Yingkui, WANG Yuanjin, SHI Changwu

2023, 33(9):  49-55.  doi:10.16265/j.cnki.issn1003-3033.2023.09.2009

Asbtract ( 84 )   HTML ( 0)   PDF (3313KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

In order to solve the problem that the RUL prediction accuracy of rolling bearings was not high due to the distortion of degradation feature distribution under the condition of limited full-life monitoring data of rolling bearings, a prediction method of the RUL of rolling bearings based on EWM and SVR was proposed. Firstly, the time-domain and frequency-domain features of the vibration signal were extracted, and the logarithmic transformation was performed on the features. Then, the index weights were determined by EWM to realize the feature selection. Finally, SSA was used to optimize the SVR model, and the low-dimensional features after dimensionality reduction by principal component analysis(PCA) were used as the input of the optimized SVR model, and the RUL percentage was used as the output, so as to realize the prediction of the RUL of the bearing. The results show that under the condition of limited monitoring data, compared with other methods, the proposed method not only has a more stable prediction performance, but also has an average reduction of 19.51% in absolute error and 17.73% in mean square error(MSE).

Mine airflow speed sensor data cleaning model for intelligent ventilation

ZHAO Dan, SHEN Zhiyuan, SONG Zihao, XIE Li'na, LIU Baichen

2023, 33(9):  56-62.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1032

Asbtract ( 95 )   HTML ( 0)   PDF (1996KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

At present, there are some problems such as incomplete data and information loss in monitoring data cleaning of mine airflow speed sensor. Therefore, a data cleaning method for mine airflow speed sensor based on SDAE was proposed. The airflow speed data samples of the ventilation system under normal conditions were trained by the SDAE algorithm to obtain the upper limit of reconstruction error and tolerance time by kernel density estimation (KDE). By comparing reconstruction error and error duration of test samples with the upper limit of reconstruction error and tolerance time, the "dirty" data was resolved. Using the monitoring data of wind speed sensor in Dongshan coal mine, the data cleaning test of faulty samples and non-faulty samples was carried out. The results show that this method can automatically identify the noise points and missing values. The "dirty" data is repaired directly by reconstruction. In the case of ventilation system fault, this method can effectively retain the state information while filtering out the interference data. Compared with other data cleaning models such as denoising autoencoder (DAE), long short-term memory (LSTM) neural network, and Kalman filter (KF), the average mean absolute error (MAE) and root mean squared error (RMSE) values of this model are reduced by 75.42% and 74.98% respectively.

Recognition algorithm on safe states of tower crane pins based on optimized Swin Transformer

ZHOU Qinghui, LIU Haoshi

2023, 33(9):  63-68.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0173

Asbtract ( 82 )   HTML ( 0)   PDF (6188KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

In order to reduce the hidden danger during tower crane operation, and improve the accuracy of machine vision when identifying states of pins, a recognition algorithm was proposed based on an optimized Swin Transformer. Firstly, A data set was created by collecting the pinned images of the tower crane on the construction site. Secondly, the safe state of pins was classified and encoded using the One-hot-coding method in the data set. Then, a recognition model for the safety status of pins was established based on an optimized Swin Transformer in which the loss function was adjusted. Through updating the gradient by the AdamW optimizer, a final model was obtained after 1000 training iterations. Finally, an experimental verification was conducted on the pinned image dataset. The results show that the proposed method can improve the identification ability of the safe state of tower crane pins, and its accuracy can reach 99.4%. The average accuracy, the average recall rate and the average specificity can reach 99.4%, 99.4%, and 99.6%, respectively. Its accuracy is higher than typical algorithms, such as DenseNet, ShuffleNet and EfficientNet. When opposed to the original Swin Transformer, the accuracy is also added by 3.6%.

Computer vision based safety inspection of high abutting edges

LI Hua, WU Lizhou, XUE Xicheng, ZHONG Xingrun

2023, 33(9):  69-75.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0202

Asbtract ( 156 )   HTML ( 1)   PDF (3992KB) ( 11 )  

Figures and Tables | References | Related Articles | Metrics

To address the issues of high danger, low efficiency, and complex recognition scenarios in the safety inspection work of construction site managers, a new detection method for high-altitude safety protection equipment was proposed using convolutional neural network (CNN) technology. This method combines computer vision to detect the wearing status of personal protective equipment such as safety helmets and harnesses, as well as any damage to protective nets. Additionally, based on the YOLOv5 algorithm, the attention model was modified and a lightweight detection software was developed. The results indicate that after lightweighting, the model size decreased to 1.9 MB, a reduction of 86.8% compared to before the modification. Under graphics processing unit(GPU) operating conditions, the single-frame image detection time was optimized to 40-50 ms, representing a reduction of 65%-80% compared to before, greatly improving the detection speed.

Study on transient energy transformation and influence factor during perforation explosion

DING Liangliang, LEI Qisong, CHEN Wenkang, XUE Yongzhi, ZHANG Qiang

2023, 33(9):  76-85.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0223

Asbtract ( 86 )   HTML ( 0)   PDF (8069KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

In order to effectively control the risk of safety accidents such as string and packer failure under the perforating well-test combination, a jet penetration model of the perforating shaped charges was established based on the fluid-structure coupling theory. A study for the jet kinetic energy and shell melting energy of the perforating shaped charges was presented to evaluate the transformation law of the jet kinetic energy and shell melting energy and the influence factors of the remnant energy of the perforating shaped charges by taking into account the shell structure, the mass and type of the explosive, the conical liner structure, the perforating gun, and the casing. The obtained results indicate that the remnant energy of the perforating shaped charges is significantly affected by the cone angle of the conical liner, the thickness of the shell, and the mass and type of the explosive. The remnant energy ratio is decreased with an increase in the cone angle of the conical liner between 40° and 70° and an increase in the thickness of the shell between 1 mm and 4 mm. The remnant energy ratio is increased with an increase in the mass of the explosive. With the same explosive mass, the Octogen(HMX) explosive has the smallest remnant energy ratio among Trinitrotoluene (TNT)、Hexanitrostilbene (HNS)、Hexogen (RDX), and HMX explosives.

Longitudinal collision risk control in land route based on PID

LU Fei, CHEN Haonan

2023, 33(9):  86-93.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1126

Asbtract ( 73 )   HTML ( 0)   PDF (4351KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

In order to improve the safety of land route operation and reduce the hidden danger of interval conflict, the longitudinal collision risk of land route was simulated. Firstly, based on the position error theory, the mathematical model of the longitudinal collision risk of land route was established. Considering the comprehensive influence of human factors, the front and back boundaries of the safe area for aircraft operation were determined by the target safety level. Then, using the System Identification toolbox of Matlab to fit and identify the quick access recorder (QAR) data, the transfer function that reflects the corresponding relationship between the angle of the thrust handle and the speed of the aircraft was obtained. Based on this, a PID control system to control the longitudinal spacing of the aircraft in a safe area was designed. Finally, Simulink was used to verify the function of the control system. The results show that under the given air route flow (50 aircrafts/h), the control system can calculate the longitudinal safety interval between two aircrafts (3 326-16 674 m) in real time, and adjust the interval between the front and rear aircrafts to the standard value within 70 seconds when there is an interval conflict. PID control system can always keep the collision risk above the target safety level, which improves the safety of aircraft operation.

SOH monitoring of new energy vehicle lithium batteries based on IPSO-LSTM

LIU Dan, WANG Ruihu, LYU Wei, QIN Ling, LIN Shuichun

2023, 33(9):  94-102.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1250

Asbtract ( 94 )   HTML ( 0)   PDF (3152KB) ( 4 )  

Figures and Tables | References | Related Articles | Metrics

In order to monitor the SOH of lithium batteries of new energy vehicles and prevent the risk of safety accidents caused by battery failures, a model combining IPSO and LSTM was proposed. First, Spearman correlation analysis method was used to extract the SOH of lithium batteries. Secondly, the linear inertia weight and asymmetric learning factor were used to improve the traditional particle swarm optimization(PSO) algorithm, and the IPSO algorithm was used to optimize the key parameters of the hidden layer neuron number, neuron inactivation rate and batch processing value of the LSTM model, and then the LSTM model was further optimized, and the IPSO-LSTM lithium battery SOH monitoring model was established. Finally, the model was verified with the 18650 lithium battery data set which was the mainstream of new energy vehicles, and compared with BP, LSTM and PSO-LSTM models. The results show that the mean absolute error (MAE) of IPSO-LSTM model is less than 0.02, the root mean square error (RMSE) is less than 0.03, and the monitoring error is less than 15%. Compared with BP, LSTM and PSO-LSTM model, the error index value of IPSO-LSTM model is the smallest, and the model has higher accuracy and stability.

Optimal design of pre-inflatable airbag under multiple child sitting postures

HONG Liang, LIU Peng, LIU Gang, GE Ruhai

2023, 33(9):  103-112.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1937

Asbtract ( 79 )   HTML ( 0)   PDF (11701KB) ( 4 )  

Figures and Tables | References | Related Articles | Metrics

In order to improve the protection ability of pre-inflatable airbags for 6-year-old child occupants in the frontal collision of school buses, parameter analysis and design optimization were carried out. Based on the sled test, the simulation model of a school bus was built and verified. The coupling model including a pre-inflatable airbag was established. For the normal sitting posture (NP), forward-leaning sitting posture (OOP₁) and right-leaning sitting posture (OOP₂), the effects of the design parameters on the head injury criterion(HIC), thorax 3 ms resultant acceleration a_3ms were investigated. The results show that when the upper strap length increases, the volume of the airbag upper part increases and HIC decreases significantly. The middle strap length determines the outer contour of the middle and lower parts, and the installation height affects the contact position and contact moment between the child and airbag. For NP and OOP₂, when the middle strap length decreases from 0.29 m to 0.26 m, a_3ms decreases gradually. When the installation height increases, HIC also increases. Decreasing and increasing the opening pressure and degree of the deflation valve can protect the head and thorax. Taking the weighted injury criteria as the optimization target, the improved non-dominated sorting genetic algorithm II is used to optimize the main design parameters. The pre-inflatable airbag with the optimal configuration can reduce child injuries.

Residual life prediction of gas generator set based on deep learning

LIN Qing, YAO Junming, LIANG Wei, YANG Fang, LIAO Chunyan, WANG Youchun

2023, 33(9):  113-121.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0231

Asbtract ( 78 )   HTML ( 1)   PDF (7921KB) ( 5 )  

Figures and Tables | References | Related Articles | Metrics

To ensure the safe and stable operation of critical equipment in oil and gas stations, reduce safety accidents and improve maintenance efficiency while reducing operating costs, this paper proposed a novel residual life prediction model for gas engine generator sets in oil and gas stations based on deep learning models and genetic optimization algorithms. Firstly, the operational data of gas engine generator sets collected from the oil and gas stations were preprocessed, and the Principal Component Analysis (PCA) algorithm was applied to extract the main components of the multi-dimensional sensor features as input features. A deep learning residual life prediction model was constructed using 1-Dimensional Transposed Convolution (TransConv1D) and Long Short-Term Memory (LSTM) for training. To solve the problem of feature stacking redundancy during model training, the QGA was employed to optimize two hyperparameters of the model. Through case verification and comparative analysis with traditional methods, the proposed model achieved at least a 2.46% reduction in MAE and a 7.85% reduction in RMSE in the prediction results. The research demonstrates that the proposed model outperforms the traditional models with smaller prediction errors and can predict the residual life of gas engine generator sets more accurately.

Study on instability criterion of rib pillars in open-pit during highwall mining

LI Guanghe, HU Zihuan, WANG Dong, WANG Laigui, ZHOU Zhiwei, DING Chunjian

2023, 33(9):  122-128.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0454

Asbtract ( 66 )   HTML ( 0)   PDF (10211KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

To investigate the instability criteria of the rib pillar, determine the reasonable setting of the rib pillar during highwall mining in an open-pit coal mine and realize the optimal design of parameters for both the open-pit highwall and the rib pillar, a mechanical model of roof-rib pillar was constructed using Winkler elastic foundation theory. Theoretical formulas of vertical displacement of the roof and the vertical stress of the rib pillar during highwall mining were derived and validated through numerical simulation. Considering safety requirements for rid pillars, combined with Wilson's ultimate strength theory of the coal pillar, the instability criteria of rid pillars in open-pit mining was established. The study explored the impact of slope and pillar shape parameters on the safety of the rib pillar theoretically. The results show that the safety factor of the rib pillar exhibits a negative exponential correlation with mining depth, end-slope angle and end-slope height. Conversely, it shows a linear positive correlation with the total width of the coal pillar, and the theoretical curves of the vertical displacement of the roof and the vertical stress of the rib pillar exhibit a saddle-shaped distribution. Theoretical calculations are in good agreement with numerical simulation results, thereby verifying the rationality and accuracy of the theoretical calculations.

Test method for fire resistance of cables under temperature rise condition of building fire

HU Linming, HE Yan, GAO Fengli, FENG Jun, ZHANG Xiang, WANG Jinyuan

2023, 33(9):  129-135.  doi:10.16265/j.cnki.issn1003-3033.2023.09.2227

Asbtract ( 93 )   HTML ( 0)   PDF (5968KB) ( 4 )  

Figures and Tables | References | Related Articles | Metrics

In order to test the fire resistance of cables at the fully developed stage of fire, two test methods (method A and B) for fire resistance of cables under the temperature rise conditions of building fire were put forward from the aspects of fire conditions, judgment method, fire resistance time, and sample installation. Moreover, a test platform of cable fire resistance under the temperature rise conditions of building fire was set up. Finally, the traditional fire resistance test method and two new test methods were used to test the fire resistance of 21 cables. The results show that compared with the traditional fire resistance test method, the new test method can effectively distinguish fire resistance time of cables. The new test method can test the fire resistance of 1 to 12 cables at the same time, which improves the test efficiency. The fire resistance time of 9 cables in two fire resistance tests are approximately the same, indicating that the new test method has good reproducibility in testing the fire resistance of cables.

Regression analysis model of coal spontaneous combustion temperature in goaf based on SSA-RF

WANG Wei, CUI Xinchao, QI Yun, LIANG Ran, JIA Baoshan, XUE Kailong

2023, 33(9):  136-141.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0846

Asbtract ( 79 )   HTML ( 0)   PDF (3034KB) ( 5 )  

Figures and Tables | References | Related Articles | Metrics

In order to accurately and quickly analyze the spontaneous combustion temperature of coal in goaf and avoid spontaneous combustion fire, the SSA-RF regression analysis model combining SSA and RF algorithm was proposed. Firstly, based on the data obtained from the spontaneous combustion characteristics test in Dongtan coal mine, the regression results of SSA-RF model and RF, back propagation neural network (BPNN), particle swarm optimization algorithm (PSO)-BPNN and SSA-BPNN model were compared and analyzed. Then, the reliability of the SSA-RF model was verified by taking the test data of 1204 coal face in Zhengjia coal mining as an example. Finally, the model was applied to Donggucheng coal mine. The results show that the mean absolute errors (MAE) of SSA-RF, RF, BPNN, PSO-BPNN and SSA-BPNN are 11.203 1, 14.342 0, 19.599 1, 15.530 6 and 14.352 8, respectively. The mean absolute percentage error (MAPE) is 14.89%, 16.91%, 18.55%, 18.43% and 18.11%, respectively. The root mean square errors (RMSE) are 13.761 0, 16.525 0, 20.786 6, 18.022 7 and 17.735 5, respectively. The coefficients of determination (R²) are 0.927 4, 0.882 7, 0.815 3, 0.843 6 and 0.868 8, respectively. All indexes of SSA-RF model are the best, which indicates that it is universal and stable, and it is more suitable for regression analysis of coal spontaneous combustion temperature.

Detection method and experiment of small hole leakage based on electrostatic sensing

YIN Yibing, ZHANG Yongliang, LIU Zunmin, ZHANG Qiang

2023, 33(9):  142-149.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1787

Asbtract ( 65 )   HTML ( 0)   PDF (3295KB) ( 4 )  

Figures and Tables | References | Related Articles | Metrics

According to the phenomenon that high-pressure gas accumulate a large amount of charge near the hole area when they are spewed from small hole, the principle of small-hole leakage detection based on electrostatic sensing signal and the signal processing method are proposed, and exploratory electrostatic sensing experiment are carried out. The mechanism of abnormal electrification in the nearby areas caused by gas ejection from small hole was analyzed first, and a theoretical method of leakage detection based on electrostatic sensing was introduced. To extract information from electrostatic signals, a signal-based method for noise reduction and feature extraction using sparse decomposition was proposed. The gas leakage spray electrostatic perception test was conducted on the pressure pipeline gas leakage test platform. Finally, the application scenarios of small hole leakage electrostatic detection were discussed. The results indicate that the electrostatic signal of the jet gas exhibits high activity during leakage incidents, and the average peak value of the signal shows a quadratic positive relationship with the jet pressure. This suggests that electrostatic sensing technology is promising as a novel approach for monitoring leakage.

Research on sealing failure risk assessment of station flange system based on EWM-AHP-cloud model

LI Feng, ZHANG Laibin, DONG Shaohua, CHEN Lin, ZHANG Hang

2023, 33(9):  150-156.  doi:10.16265/j.cnki.issn1003-3033.2023.09.2040

Asbtract ( 82 )   HTML ( 0)   PDF (3222KB) ( 5 )  

Figures and Tables | References | Related Articles | Metrics

To improve the accuracy of safety assessment of station flange connection system, a seal failure risk assessment method based on EWM-AHP-cloud model was proposed. EWM and AHP were used to combine the weights to reduce the subjectivity of the single AHP method in calculating the weights. Cloud model was used to solve the problems of fuzziness, randomness and difficulty in quantifying the sealing failure risk factors of flange connection system in station. An example of a gas transmission station in North China was used for verification. The sealing failure risk assessment index system of flange system was established based on the historical failure data of station flanges, and the EWM-AHP method was used to determine the combined weight of each index. The standard cloud was established according to the hierarchical quantization standard, and the cloud digital eigenvalue of each risk index was calculated by Matlab, and the cloud map was drawn. The risk level of each risk index was determined based on the EW-type index closeness. The results indicate that the comprehensive risk level of sealing failure of the flange system being evaluated is relatively high, and the evaluation results are basically consistent with the actual situation. The EWM-AHP cloud model can effectively assess the risk of sealing failure in flange systems, and this method has a certain degree of scientificity and effectiveness.

Study on trip chain clustering of hazardous materials transportation vehicle based on GPS data

CHEN Ranran, XU Jiali, LI Jian

2023, 33(9):  157-163.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1676

Asbtract ( 85 )   HTML ( 1)   PDF (1977KB) ( 4 )  

Figures and Tables | References | Related Articles | Metrics

In order to assist the government in regulating the transportation of hazardous materials, a cluster analysis framework based on GPS data was proposed. We set the spatiotemporal threshold, extracted the vehicle base station and halfway effective stop point from the GPS data, and generated trip chains of the hazardous material vehicle. Based on calculating the number of trip chains, the average number of stop points, the average stop time of trip chains and the average distance of trip chains, the average risk of trip and trip point were proposed to measure the risk of hazardous materials transportation vehicles. The characteristics of vehicle trip chain were taken as clustering indicators, silhouette coefficient and the sum of squares due to error were taken as evaluation indicators, and by comparing the results of K-Means, K-Means++, PAM and FCM algorithms, the trip chains of hazardous materials transportation vehicles were clustered. The feasibility of the framework was verified by case analysis. The results show that the K-Means++ algorithm has the best effect, and the hazardous material transportation vehicles are divided into 5 categories: chemical park connection (44.19%), intercity transportation (31.42%), city distribution (13.23%), freight terminal connection (9.76%) and non-working state (1.40%).

Effects of fracture dip and aperture of submarine metal mine surrounding rock on seepage characteristics

FU Cuicui, ZHANG Yongliang, QU Min, YIN Shili, LIU Yunfei, WANG Mingbin

2023, 33(9):  164-172.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1311

Asbtract ( 48 )   HTML ( 0)   PDF (7047KB) ( 5 )  

Figures and Tables | References | Related Articles | Metrics

In order to study the influence of fracture inclination and opening degree of high pressure surrounding rock of undersea metal mine on seepage characteristics and reveal the influence law of fracture inclination and opening degree of high pressure surrounding rock of undersea metal mine on seepage velocity and outlet flow, a multi-fracture seepage model of surrounding rock was established based on the coupling theory of fracture state and seepage velocity in Xinli mining area of Sanshandao, an undersea metal mine. The numerical simulation software COMSOL Multiphysics was applied to obtain the change of seepage velocity and outlet flow of groundwater in the crack of surrounding rock of a deep well by changing the crack dip Angle and crack opening. The results show that the Angle of fracture inclination has an important influence on the seepage velocity of groundwater, and the intersection of multiple fractures causes sudden changes in the seepage velocity of the groundwater. In the range of simulated inclination Angle, the larger the Angle of fracture inclination, the smaller the seepage velocity of groundwater outlet, the smaller the outlet discharge, and the more obvious the seepage velocity change at the intersection of cracks. When the crack opening is between 0.5 and 2 mm, the seepage velocity at the intersection of the crack changes obviously, and the crack opening has a significant effect on the seepage flow.

Public safety

Prediction of rear-end collision risk of freeway trucks based on machine learning

WEN Huiying, HUANG Kunhuo, ZHAO Sheng

2023, 33(9):  173-180.  doi:10.16265/j.cnki.issn1003-3033.2023.09.2142

Asbtract ( 111 )   HTML ( 0)   PDF (5158KB) ( 6 )  

Figures and Tables | References | Related Articles | Metrics

In view of the frequent rear-end accidents of trucks on highways, the risk of rear-end collision of trucks on highways was assessed, and the influence of traffic flow characteristics was analysed to reduce the occurrence of rear-end accidents. According to the German HighD open-source dataset, the TTC thresholds of different conflict risk levels were used as the classification standard, and the vehicle trajectories and traffic parameters of trucks were extracted. The risk models of rear-end collision of trucks were established based on Random Forest (RF) model, Support Vector machine (SVM) model, and Artificial Neural Network (ANN) model respectively. The overall forecasting ability of each model was compared with the evaluation indexes such as confusion matrix, area under the receiver operating characteristic curve (AUC) and Kolmogorov-Smirnov (KS) test. The model with the best prediction accuracy was selected to analyse the influence of each characteristic parameter on the rear-end risk. The results show that the prediction accuracy of the RF model is 75%, which is 8% and 10% higher than that of the SVM model and the ANN model, respectively. The prediction accuracy, recall, AUC and KS values of the RF model are better than those of the SVM and ANN models. The three parameters of minimum headway, standard deviation of vehicle speed and standard deviation of acceleration have the highest influence on the risk of rear-end collision of trucks.

Vehicle scheduling for hazardous materials transportation based on dissimilar paths

CHAI Huo, HE Ruichun, JIA Xiaoyan, DAI Cunjie

2023, 33(9):  181-188.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1503

Asbtract ( 72 )   HTML ( 0)   PDF (1932KB) ( 5 )  

Figures and Tables | References | Related Articles | Metrics

In order to effectively mitigate potential risks during the simultaneous transportation of hazardous materials by multiple vehicles, this study addressed the problem of simultaneous multi-vehicle transportation on distinct paths between 2 nodes within a hazardous materials transportation network. A method for calculating safe departure time intervals of vehicles was proposed to select paths and allocate vehicles. A mathematical model was established for simultaneously scheduling multiple vehicles to travel on different paths through path selection and vehicle allocation. A multiobjective optimization algorithm based on NSGA-II was designed for solving the proposed model. Using an example of transporting 10 hazardous material vehicles in an 8-node and 13-edge hazardous materials transportation network, the proposed method yielded 13 transportation schemes satisfying single-trip risk constraints and 15 transportation schemes satisfying total risk constraints. A comparative analysis was performed on transportation risk, distance, and time under different risk constraints. The results indicate that compared with the total risk constraints, the single-trip risk constraints are more stringent, which is suitable for road authorities to select safe transportation paths, while total risk constraints are suitable for carrier enterprises to make vehicle scheduling decisions.

A pyramid model of community resilience construction in a risk society

XU Wenwen, LIU Yiliang

2023, 33(9):  189-195.  doi:10.16265/j.cnki.issn1003-3033.2023.09.0362

Asbtract ( 90 )   HTML ( 0)   PDF (1643KB) ( 3 )  

Figures and Tables | References | Related Articles | Metrics

In order to effectively improve the dynamic adaptability of grassroots communities in China to risks, the concept of community resilience was introduced, and the construction mode and practical path of community resilience was studied. Firstly, through the analysis of policy text and field investigation, the three-dimensional dilemma of community resilience construction in China was discussed. Then, the pyramid model of community resilience was constructed. Finally, a practical path to enhance community resilience was put forward. The research shows that there are three major dilemmas in the construction of community resilience in China: insufficient policy guidance, vague construction goals and lack of construction consciousness. The community resilience construction in China can be practiced in six dimensions: spatial resilience, facility resilience, technical resilience, institutional resilience, organizational resilience and cultural resilience. The practical ways to improve community resilience include improving policy system and strengthening resource input, making clear goals to enhance the construction consciousness, and designing evaluation system to promote construction.

Construction and stability analysis of an evacuation model for contrary movement personnel

DENG Qing, ZHOU Zheng, DENG Li, JIANG Huiling, ZHOU Liang, ZHOU Zhengqing

2023, 33(9):  196-203.  doi:10.16265/j.cnki.issn1003-3033.2023.09.2178

Asbtract ( 89 )   HTML ( 0)   PDF (2375KB) ( 4 )  

Figures and Tables | References | Related Articles | Metrics

To accurately depict the evolution process of individual behavior conflicts during evacuations and improve evacuation efficiency, the evolutionary game theory was applied to analyze the evolution of evacuation behaviors in two typical scenarios involving contrary movement personnel. The impact of population density, level of emergency severity, and individual psychological gains on the choice of cooperative or competitive strategies by individuals was discussed. The dynamic evolution process of game players and the influence of parameter adjustment on the strategy selection of both sides could be demonstrated with the help of Matlab. The simulation results showed that under certain constraints, there were five equilibrium points which were (1,0), (0,1), (1,1), (0,0), and a saddle point. It is found that different types of retrograde pedestrians benefit from different proportions, but the influencing parameters of their behavior evolution have the same influence trend. When the psychological benefits of regular evacuees, the risk level of the accident, and the additional time loss of counterflow individuals increase, evacuees tend to adopt a yielding strategy. Conversely, when the cost of competition for regular evacuees and the rewards of successful competition for counterflow individuals decrease, the counterflow individuals tend to adopt a yielding strategy.

Copula-BN based risk assessment methodology of marine ship collisions

LI Xinhong, FU Yaqian, LIU Yazhou, HAN Ziyue, ZHANG Renren

2023, 33(9):  204-213.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1469

Asbtract ( 92 )   HTML ( 0)   PDF (12055KB) ( 4 )  

Figures and Tables | References | Related Articles | Metrics

In order to effectively prevent marine ship collisions, a Copula-BN-based dynamic risk assessment method for marine ship collisions was proposed. Gaussian Copula method was used to identify the edge distribution of each node, so that the complex relationships between risk factor and collision risk was more accurate and the Bayesian model inference results were more reasonable. Firstly, the data collected on ship collisions were combined with expert knowledge to identify and analyze the specific risk factors that led to ship collisions from the perspective of the human-ship-loop system. Gaussian Copula method was used to identify the edge distribution of each node and determine the optimal marginal distribution function of each node. Then BN structure was determined through expert experience and data structure analysis, thus completing the establishment of the Copula-BN model. Based on the correlation analysis, forward probability prediction analysis and backward fault diagnosis analysis, the key risk factors leading to the occurrence of ship collision were introduced based on the dependency between each risk factor and the target factor, and the whole risk system was dynamically analyzed through the state update of the nodes. The study shows that this method can identify the key risk factors leading to ship collision accidents and realize the dynamic assessment of ship collision risk.

Research overview of roadside accident risk assessment and roadside safety design

CHENG Rui, PAN Ye, DAI Junji, WANG Tao, XIE Zhongcao

2023, 33(9):  214-226.  doi:10.16265/j.cnki.issn1003-3033.2023.09.1864

Asbtract ( 89 )   HTML ( 0)   PDF (2078KB) ( 7 )  

Figures and Tables | References | Related Articles | Metrics

In order to comprehensively understand the research progress of roadside safety and formulate prevention and control programs, the development directions of roadside safety were shown from three aspects: the proportion of publication, the source of publication and the country of publication. Then, the research status, existing issues and future development trends of roadside safety were analyzed from two aspects: the risk assessment of roadside accidents and the roadside safety design. Finally, the different evaluation models and prediction methods for the risk assessment of roadside accidents were summarized. The results show that the first ten significant cause factors that lead to high-risk roadside accidents are, in decreasing order of importance, driver age ≤ 25 or ≥ 65, alcohol, small radius curve, speeding, failure to use seat belts, heavy trucks, heavy traffic, objects adjacent to the lane, no lighting/night, and the narrow shoulder. In terms of roadside accident risk assessment, future research should integrate mobile and static multi-source data, put forward a method for assessing the risk situation of roadside accidents, and further explore the interaction mechanism of accident causation factors. In the aspect of roadside safety design, a set of reasonable and effective protection methods should be suggested from the aspects of accurately quantifying the width of the clear zone, formulating the setting conditions of clear zone, reasonably matching the grade of guardrail and reducing the design cost of energy dissipation of traffic facilities. In addition, under the current research focus on intelligent network connection, it is also the future research direction to improve the vehicle running stability by constructing a human-vehicle-road collaborative comprehensive perception system and a road network comprehensive operation monitoring and warning system.

DBN model of external corrosion accident in gas network system

LI Cong, XU Zixuan, ZHUANG Yufeng, YANG Rui, XU Yabo, CHEN Chen

2023, 33(9):  227-236.  doi:10.16265/j.cnki.issn1003-3033.2023.09.2314

Asbtract ( 91 )   HTML ( 2)   PDF (5192KB) ( 7 )  

Figures and Tables | References | Related Articles | Metrics

In order to understand the occurrence regularity and characteristics of external corrosion accidents in the middle and later period of pipeline operation and prevent external corrosion leakage accidents of gas pipeline network, an aggregation network of external corrosion accidents was organized and constructed by combining CN and DBN. The key factors of the accident were screened by degree analysis. On this basis, the parameter learning and Leaky Noisy-or gate correction model were combined to construct the DBN of external corrosion accidents of gas pipeline network. Finally, according to different failure scenarios, the influence characteristics of different dynamic elements on external corrosion accidents and the influence of maintenance measures on pipeline failure probability were obtained. The results show that the impacts of different disaster-causing elements on pipeline network failure are different and staged. Among them, aging, accident accumulation and chemical corrosion have significant impacts on the corrosion perforation and pipeline rupture in the middle and late service stages, while the pressure cycling dominates the pipeline rupture accident in the middle service stage. Through CN topology analysis, the importance of accident elements can be identified from the overall point of view. Through the construction analysis of DBN, the dynamic influence of accident factors on accident consequences can be obtained.