Mechanism Of Water-inrush And Early Warning Technology In North-china-type Coalfield

With the increase of mining depth and intensity and large-scale exploration of lower group coal, mining conditions in North-China-Type coalfield become more complex, water pressure, stress and groundwater temperature gradually growing, then mine water inrush hazards also is more prominent with frequency and strength of water inrush increasing frequently, the problem is very urgent to develop monitoring and early warning technology for real-time water inrush monitoring and early warning, to ensure safe coal mine. Although there are some research achievements about water inrush monitoring and early warning in domestic and abroad, theory and technology are not mature, especially the criterion of water inrush and early warning is indeed necessary to further study systematically in the stage of researching.In this paper, Zhaozhuang coal mine and monitoring and early warning system of water inrush as main study object, in view of mechanism and mechanical criteria of main water hazard types, mine water inrush monitoring and early warning mode, monitoring and early warning index system, early warning criteria and key techniques, research work is finished. Then, water inrush monitoring and early warning system is designed and developed, and applied in 5303 working face in Zhaozhuang coal mine.1. Hydrogeological condition is analyzed and main type of water inrush is determined in Zhaozhuang coal mine. Mining coal seams of 3# and 15 # are threatened by several sources of Fengfeng group lime water, Taiyuan group lime water, Shanxi group sandstone fissure water and lower Shihezi group(K8) sandstone fissure water. There are five types of inrush channels of faults, karst collapse columns, poor sealed boreholes, top and floor fractured zone, and which correspondingly assembled into five class of poor sealed boreholes water inrush, fault water inrush, karst collapse columns water inrush, goaf water inrush and seam floor water inrush.2. For five types of water inrush, water inrush mechanism is researched and corresponding water inrush criterion is put forward. Using the theory of thick wall cylinder, mechanical model of water inrush from side wall of poor sealing borehole is proposed, main factors of the inrush are involved in mining damage range, aperture, water pressure in borehole, mechanical properties of surrounding rock, so water inrush criterion is established for this kind of water inrush. For fault, activation mechanism under mining influence is revealed, fault activation factors are summed up of water filling degree, fault dip, fault belt characteristic, fault location etc, and the criterion of fault activation and water inrush is established. After goaf water inrush model and mechanism is researched systematically, mechanical models of shear failure of top effective confined layer, effective tensile failure of top effective confined layer, adjacent goaf water bursting are established separately, integrated criterion of goaf water inrush is performed. Summarizing previous research results on collapse column water inrush and floor water inrush mechanism, it gives more mature criterion for water inrush.3. It is put forward the index system and model of monitoring and early warning for mine water inrush. Considering of easily acquisition,functionality and practical experience, early warning and monitoring indicators are proposed, such as including: water coal rock pillar of mine resistivity, water pressure and temperature, rock stress and strain, routine hydrochemistry index: TDS?Na++K+?Ca2+?Mg2+?CL-?SO42-?HCO3-?PH etc..According to monitoring indicators, early warning models can be classified into physical and chemical modes, developing a real-time monitoring system consisted of dynamic comprehensive monitoring for impermeable layer resistivity parameter and multi parameter monitoring for single point in key position, the principle and method of monitoring and early warning is raised for two physical and chemical modes.4. As key parameter of monitoring and early warning criterion, the real-time monitoring and inversion method of effective thickness of coal and rock pillar resisting water is proposed. The change law of apparent resistivity is comprehensively analyzed in damage zone of coal and rock pillar resisting water caused by mining in two case of fill water and without groundwater, so inversion method of mining damage range is formed using real-time monitoring apparent resistivity data, namely contrasting and analyzing resistivity change trend from before mining, mining and after mining, interpreting damage zone of coal and rock pillar resisting water, and obtaining effective thickness of the pillar, this method is verified in the field.5. Determining method of key position of monitoring and early warning is studied, software system based on GIS components and ANN coupling technology is developed. The performed evaluation index system mainly includes: water pressure of filling water source, water quantity, thickness of coal and rock pillar resisting water, fault influence index, collapse column influence index and poor sealing borehole influence index, calculation formulas for three indexes are constructed. Using mentioned water inrush criterion, solving method of the relevant parameters in the formula is also given, including quantification methods of all evaluation indexes. To determine key positions for monitoring and early warning, the system based on GIS components and ANN coupling technology is completed, composited of 6 subsystems: map control subsystem, map display and output subsystem, map tool subsystem, database management subsystem, information index system, forecast and analysis subsystem. The whole system can delineate high risk region of water inrush and confirm key positions of monitoring and early warning by powerful spatial analysis functions of GIS and nonlinear function of ANN.6. Based on the spatial analysis function of GIS,the method of dynamically constructing the standard water sample library is realiazed, which improves the precision of mine water source identification and correct rate of mine water chemical warning. Construction of water source of sample library is the key to chemical warning of mine water inrush. This paper changes the traditional standard construction method of water samples, namely putting all the known sources of water samples into a standard water sample library, and one mine only establishes one standard library. Based on GIS components spatial analysis function, this paper has realized the dynamic construction of standard library, namely according to different water points, considering aquifer lithology of water samples at the bursting water points, geological structures, runoff areas, effects of mining activities, standard sampling time and many other factors. Using GIS spatial analysis function to find out standard water samples similar to hydrological geological conditions of water points and near the collection time, then constructing standard water library to achieve building different standard water sample libraries according to different dynamic water points. Taking Xinglong Coal Mine water samples numbered 496 as an example, the test based on the dynamic construction methods of the standard water sample library shows that the method of dynamic standard sample library can effectively improve the accuracy of the library back substitution.7. Based on the study and compare of the mine water identification method, the algorithm which achieve the recognition of mine single water source and mixed water source is designed, provids the basis for early warning criterion for mine water chemistry. Mine water recognition comparative analysis shows that the bayesian discriminant model can be applied to all possible information of mine water bursting and can achieve small sample test analysis with the smallest error risk of identification. Towards the multiple factors, relevance, overall complex features to the formation of mine water quality, Gray Correlation Analysis can find out the closeness between the various indicators of mine water and achieve water source identification, and this method is also very suitable for small sample analysis. This paper has designed and achieved mine water recognition algorithms of bayesian discriminant analysis and gray correlation analysis, which can realize single water source recognition and mixed water recognition. Mixed water recognition uses two or three kinds of mixed water source samples library to achieve two or three kinds of source of identification, and can give the approximate percentage of each water source.The mine water source identification results can provide the basis for mine chemical warning.8. The physical early-warning criteria of the mine water inrush is researched, qualitative and quantitative physical early-warning criteria is summarized and proposed. According to the changes of physical early-warning indicators monitored and previous instances of the water inrush, five kinds of qualitative early-warning criteria is proposed: The stress of the monitoring position suddenly reduces, the strain increases sharply,water temperature and pressure change and gradually close to the target water source, the resistivity of water resisting layer near the target water source decreases and the range gradually increases, the resistivity of the water channel continues to reduce and extends to the working face or the target water source.According to the relationship of the effective thickness monitoring and the theory or criterion of effective thickness Lv,developing the quantitative physical early-warning criteria, when the effective thickness monitoring is located in [1.7Lv, + ?), [1.4Lv, 1.7Lv), [Lv, 1.4Lv), [0, Lv), which are respectively no warning, blue alert, yellow and red alert, the calculation methods of different water damage types Lv are given, and also combining with the related rules of the prevention and control of mine water,it provides the necessary supplement for the quantitative early-warning criteria.9. The water chemistry warning criteria of the mine water inrush of a single water source and mixed water source is proposd. When the recognition result is a single water source, it is no warning if the water source is a weak water-filled, it is blue alert if the water source is a medium-fat water-filled, it is yellow alert if the water source is strong water-filled and there is no hazard risk, it is orange alert if the water source is a strong water-filled and there is a hazard risk, it is red alert if the water source is highly water-filled or the goaf area water. When the recognition result is the mixed water source, it will release the corresponding alarm when the percents of the strongest water-filled water source reach to at least 30% and meet the above early-warning criteria.10. The physical and chemical early-warning system of mine water inrush are Designed and developed, and are applied in Zhaozhuang mine. The hardware and software architecture of the physical and chemical early-warning system of the mine water inrush is designed; the corresponding software system is developed; the physical and chemical early-warning system of mine water inrush is researched. Putting the 5303 working face of Zhaozhuang mine as the research object, the physical early-warning system is arranged and the reasonable early-warning criteria is set up. The monitoring and early-warning of the water inrush are achieved,which ensure the safe mining of the working face.