Search

In traditional rockburst hazard assessment methods, there are problems of large computational complexity and low precision in dividing hazardous areas. In order to meet the development needs of intelligent and visual prevention and control of rockburst, an intelligent assessment method for rockburst hazard areas based on image recognition technology is proposed. Using a semi quantitative estimation method, the method quantitatively characterizes the main controlling factors of dynamic and static loads for 11 types of rockburst hazards. Based on OpenCV machine vision library and deep learning model, the method achieves image recognition for a single main control factor. By constructing a mapping matrix between the grayscale of the image and the stress concentration coefficient, linear and nonlinear superposition of a single influencing factor is achieved to obtain the stress concentration coefficient matrix of the assessment area. Using the min max standardization method to construct a 4-level discrimination standard of 'no, weak, moderate, and strong' for the hazard area of rockburst, the method achieves graded and division assessment. A software for intelligent assessment of rockburst hazards is developed based on Python language, and the actual application effect of the software is tested. The results show that the software improves the traditional one-dimensional linear hazard area division method for roadways to a two-dimensional plane division method for the entire mining space. It significantly improvies the assessment efficiency and precision of hazard area division and reduces labor costs. The assessment results are highly consistent with the microseismic energy density cloud map and the on-site measured mining pressure pattern, which can provide effective guidance for the prevention and control of on-site rockburst.

Multivariate time series have more complex and high‐dimensional characteristics, which makes it difficult to analyze and predict the data accurately. In this paper, a new multivariate time series prediction method is proposed. This method is a generative adversarial networks (GAN) method based on Fourier transform and bi‐directional gated recurrent unit (Bi‐GRU). First, the Fourier transform is utilized to extend the data features, which helps the GAN to better learn the distributional features of the original data. Second, in order to guide the model to fully learn the distribution of the original time series data, Bi‐GRU is introduced as the generator of GAN. To solve the problems of mode collapse and gradient vanishing that exist in GAN, Wasserstein distance is used as the loss function of GAN. Finally, the proposed method is used for the prediction of air quality, stock price and RMB exchange rate. The experimental results show that the model can effectively predict the trend of the time series compared with the other nine baseline models. It significantly improves the accuracy and flexibility of multivariate time series forecasting and provides new ideas and methods for accurate time series forecasting in industrial, financial and environmental fields. [ABSTRACT FROM AUTHOR]

In view of the obvious characteristics of mining pressure in Wangjialing Coal Mine under the loess gully landform, the Get Data software was used to extract the elevation data within the mining range, and Surfer was used to process the data to construct the loess gully three-dimensional landform; based on numerical simulation technology, the influence of loess gully landform on the stability of overburden structure in fully mechanized top coal caving mining was studied. The results show that the loess gully landform has a significant effect on the original rock stress in the mining area, and the effect gradually decreases with the increase of mining depth; it shows that the shallow original rock stress is mainly affected by the loess gully landform, and the deep original rock stress is mainly related to the buried depth of rock strata. Under the influence of loess gully landform, with the advance of mining face, the overlying strata are subjected to more and more force, the key strata begin to lose stability and fracture, the arch area with the most serious subsidence of goaf continues to lag behind the geometric middle of goaf, and the concentration degree of advance abutment pressure of working face is obviously affected by the thickness of overlying loess layer.

In this paper, we derive and analyze a new one-parameter family of modified Cauchy method free from second derivative for obtaining simple roots of nonlinear equations by using Padé approximant. The convergence analysis of the family is also considered, and the methods have convergence order three. Based on the family of third-order method, in order to increase the order of the convergence, a new optimal fourth-order family of modified Cauchy methods is obtained by using weight function. We also perform some numerical tests and the comparison with existing optimal fourth-order methods to show the high computational efficiency of the proposed scheme, which confirm our theoretical results. The basins of attraction of this optimal fourth-order family and existing fourth-order methods are presented and compared to illustrate some elements of the proposed family have equal or better stable behavior in many aspects. Furthermore, from the fractal graphics, with the increase of the value m of the series in iterative methods, the chaotic behaviors of the methods become more and more complex, which also reflected in some existing fourth-order methods.

Aims: Inconclusive findings of lipid‐lowering medications (LLMs) on cancer survival benefit require more evidence. We tested the hypothesis that adherence to this drug is associated with reduced cancer‐specific mortality in a homogeneous population who had used this drug before cancer diagnosis. Methods: The Australian Cancer Database was linked to the Pharmaceutical Benefits Scheme database, and to the National Death Index (up to 2015). Medication adherence was calculated by proportion of days covered. Cox regression models with time‐varying covariates were used to derive multivariable‐adjusted cause‐specific hazard ratio (HR) and 95% confidence interval (CI) for the associations between adherence to LLMs, statins, lipophilic, and hydrophilic statins and cancer‐specific mortality. Results: From 2003 to 2013, 3 separate cohorts of 20 046, 11 719 and 6430 female patients with newly diagnosed breast, colorectal cancer, and melanoma respectively were identified. The 1‐year adherence was similar at 1‐year prediagnosis in the 3 cohorts, on average 82%. Each 10% increase in 1‐year adherence to LLMs was inversely associated with cancer‐specific mortality among women with breast cancer (fully adjusted HR = 0.92, 95% CI 0.91–0.93), colorectal cancer (fully adjusted HR = 0.92, 95% CI 0.91–0.93), or melanoma (fully adjusted HR = 0.97, 95% CI 0.94–1.00). The reductions in cancer‐specific mortality were more pronounced for women who adhered to lipophilic than hydrophilic statins in all 3 cancers albeit not statistically significant for melanoma. Conclusion: Among LLM users, adherence to this drug is associated with a decrease in cancer‐specific mortality. If confirmed, LLMs could be considered as an adjuvant cancer therapy to improve prognosis in cancer survivors. [ABSTRACT FROM AUTHOR]

• Deep forest algorithm is used to detect epileptic EEG signal for the first time. • A hybrid model framework of decomposition, feature selection and ensemble learning is constructed, and an improved variational mode decomposition, weighted maximum relevance and minimum redundancy, and deep forest model is established to identify epileptic EEG signals. • Based on the limitations of ridge regression, The elastic net is introduced to improve VMD and improve the accuracy of solving the variational problem. The study of EEG signals is of great significance for the diagnosis and prevention of brain disease. Most of the previous studies are based on the binary classification of nonictal and ictal EEG signals, and there are few studies on the detailed division of EEG signals. In this paper, in addition to the binary classification of EEG signals, the multiclassification of EEG signals is also studied. An EEG signals recognition framework based on improved variational mode decomposition (VMD) and deep forest is proposed. Firstly, the L 1 penalty term is introduced into the variational problem of VMD to improve the Tikhonov regularization term. The improved VMD algorithm is used to decompose the original signal. Second, a weighted minimum redundancy maximum relevance criterion is constructed for feature selection. Finally, a deep forest model is built to classify EEG signals. The feasibility of the proposed method is verified by EEG data from Bonn University and the Centre for Neurology and Sleep, Hauz Khas, New Delhi. The experimental results are compared with the traditional machine learning methods and the existing methods. Experimental results show that this method can recognize epileptic EEG signals effectively. [ABSTRACT FROM AUTHOR]

Bio-inspired optimization algorithms are competitive solutions for engineering design problems. Chicken swarm optimization (CSO) combines the advantages of differential evolution and particle swarm optimization, drawing inspiration from the foraging behavior of chickens. However, the CSO algorithm may perform poorly in the face of complex optimization problems because it has a high risk of falling into a local optimum. To address these challenges, a new CSO called chicken swarm optimization combining Pad$ \acute{e} $ approximate, random learning and population reduction techniques (PRPCSO) was proposed in this work. First, a Pad$ \acute{e} $ approximate strategy was combined to help agents converge to the approximate real solution area quickly. Pad$ \acute{e} $ approximate was grounded in a rational function aligning with the power series expansion of the approximated function within a defined number of terms. The fitting function used in this strategy employs the above rational function and the extreme points are calculated mathematically, which can significantly improve the accuracy of the solution. Second, the random learning mechanism encouraged agents to learn from other good agents, resulting in better local exploitation capability compared to traditional CSO. This mechanism has a special idea that when it comes to selecting random individuals, it selects from the same type of high-performing agents, rather than selecting them completely at random. Third, a new intelligent population size shrinking strategy was designed to dynamically adjust the population size to prevent premature convergence. It considers fitness function calls and variations in recent optimal solutions creatively. To validate the algorithm's efficacy, PRPCSO was rigorously tested across 23 standard test functions and six kinds of practical engineering problems. We then compared PRPCSO with several mainstream algorithms, and the results unequivocally established PRPCSO's superior performance in most instances, highlighting its substantial practical utility in real engineering applications.