Aiming at the problem that coal and gas outburst is difficult to early warning, an warning index system for was analyzed established in this paper. On basis, method based on Bi-directional Long Short-term Memory neural network algorithm (BiLSTM) proposed. The local optimal solution optimized obtain maximum or minimum value of fitness function. An model combination Hunter-prey Optimization BiLSTM (HPO-BiLSTM) established. refined parameters were adjusted using dynamic optimization positions...

The extraction of coalbed methane (CBM) from low-permeability coal beds is a key matter for mine safety. microporous system has significant influence on its gas adsorption capacity. To reveal the evolution and characteristics under solvent treatment, CO2 isothermal experiments were carried out. experimental results show that N-Methylpyrrolidone (NMP) significantly modifies micropore structure coal. After NMP maximum reductions in pore volumes lignite, bituminous coal, anthracite are 64.99%,...

Currently, coal mining faces the uncertainty of risk and gas outbursts inaccurate prediction results. Owing to this, an artificial immune algorithm (AIA) was developed for outburst based on Hamming distance (HD) calculation method antibody antigen affinity called intelligence (HDAIA). The correlation matrix indicators constructed using interpolation function in algorithm. HD used obtain between antigen, minimum screened result. collected dynamic data drilling cuttings desorption index K1...

In all the construction safety accidents, high falling accident is most serious of five big hurts which are threatening building workers. Risk assessment important means how to prevent and control from height accident. Based on "human - machine environment Management" complex system, risk index system about 4 major categories 23 sub-categories include quality factors production personnel equipment environmental conditions management was established. The AHP-Fuzzy evaluation model weight sets...

According to the complexity and nonlinear interaction between influencing factors of gas explosion in coal mine, this paper analyzed main mine accident with Grey Relational Analysis, built risk assessment index system based on four "man-machine-environment- management". This multi-criterion multi-level network calculation model for got weight sets by using Analytic Network Process. Multi-level fuzzy comprehensive evaluation determined level. A South, example, its assessment, results showed...

Summary The excess of gas concentration in the top corner coal working face has always been main factor restricting safe productivity mines. Therefore, rapid and accurate prediction is an effective method to prevent disasters. At same time, development Internet things made monitoring data collected by mine safety system exhibit nonlinear big characteristics. In order characteristic related from a high‐dimensional datasets, model that integrates t‐distributed Stochastic Neighbor Embedding...

City fire risk analysis is a very important basis for scientifically and effectively planning deploying urban protection formulating its related regulations. It also can guarantee facilitate the coordinated development of city's economic construction social economy. Combining characteristic China control safety management state, thesis based on engineering systematic theory combine, has set up fault tree city area. Applying method fuzzy degree sorted basic incident did graduation factors...

When processing the current coal and gas outburst risk index data, it is observed that data sequence does not have intuitive regularity, acquired signals are generally mixed with different degrees of noise, which hinders prediction. To reduce influence noise signal on prediction, proposed to use Translation-invariant Wavelet Threshold Denoising (TIWTD) process result shows in original significantly reduced. On this basis, a Compact Negative Selection Algorithm (CNSA) introduced detect...

In order to accurately predicting gas emission in coal mines, the complicated nonlinear characteristics of was analysis, prediction method put forward for based on self-organizing data mining. It used ternary quadratic polynomial local function and original variable each generation, minimum deviation principle criteria model selected. And then, high-order equation established by mining method. The fitness relative error this ±0.03% predictive ±1.45% mine. results show that: can automatically...

Affected by many involved factors, different dimensions, data with large difference, incomplete information and so on, the most optimal selection of regional outburst prevention measures for mine has become a complicated system project. The traditional way measure belongs to qualitative method, which may cause high-cost gas control, huge quantities drilling work, long construction time even secondary disaster. To solve above-mentioned problems, in light occurrence status coal seam No. 21...

Aiming at the identification of coal and gas outburst risk, using advantages clone selection algorithm (CSA), such as self-adaptation robustness, characteristics fast convergence particle swarm optimization (PSO) algorithm, complex decoding problem, mutation process brought by CSA binary coding are used. It is difficult to control problem. Using PSO optimization, problem abnormal detection in monitoring developed studied, a risk anomaly model based on variation established. The uses index...

In order to accurately predict the possible gas emission quantity in coal mines, it is proposed use multi-thread calculation of Immune Genetic Algorithm (IGA) and injection vaccines improve accuracy prediction combine Estimation Distribution (EDA) distribution probability excellent populations. Calculating, selecting populations for iteration, optimize population generation process Algorithm, so that quality continuously optimized improved, optimal solution obtained, thereby establishing a...

The production practices in Shendong mining area show that the CO concentration at return air corner of shallow coal seams often exceeds limit, but temperature residual goaf has not significantly increased, and no other indicative gas spontaneous combustion been detected. In order to provide scientific basis for management fully mechanized face prevention control seam fires, based on "three zones" distribution parameters characteristic coal, mechanism buried area, source influencing factors...

Artificial intelligence is flourishing, and its research achievements are being extensively applied across various industries. In the field of predicting coal gas outbursts, methods such as machine learning deep have been widely explored, resulting in accurate prediction accuracy excellent predictive effects. This has significantly improved safety mine underground operations.

The monitoring data of the gas in coal mines are characterized by streaming big data, because development Internet Things. To accurately and dynamically predict concentration, improve accuracy risk prediction outburst, first, we established a model based on to using machine learning algorithm. This was proposed principal component analysis linear regression method. In addition, prototype system which supports iterative refreshment for live streams real-time Spark Streaming. Moreover,...