| Tailings pond is one of major hazard installations in China, its dam-break can bring disasters by mining and seriously influence the safety of ambient people's lives and their property, so it is urgently to study the stability of tailings dam. Through some recently safety assessment projects of tailings ponds and the key project of safety science & technology in Hunan Province (study on theory and method of safety assessment of tailings pond (HN08-27) ) which executed by the author, the statistical theory, Gray theory, Neural Network theory, Reliability theory, Fuzzy math, and Unascertained mathematical theory etc., are applied to analyze varies of method, phreatic surface Matrix Method is proposed; Focus on its own dynamic variability and uncertainty, based on general regression neural network theory and the gray theory, coupled model is establishment on the basis of reliability theory and the fuzzy theory, tailings Dam fuzzy random reliability model is also proposed; then, systematically and firstly raised unascertained measure comprehensive evaluation model of tailings dam stability. The main conclusions as follows:1) phreatic surface is one of the most important factors to influence tailings dam security degree. Based on analysis of affecting factors of phreatic surface in tailings dam, the superimposed influence function is established, and the phreatic surface matrix prediction method is bring forward to predict the phreatic surface. According to the daily observation of phreatic surface and the water-level of ponds, the phreatic surface matrix is attained by regression analysis. Finally, a tailings dam projects in Hunan is given to illustrate stability analysis method of tailings dam based on the saturation line matrix method.2) Due to the extremely complicated seepage boundary and dynamic changing conditions of tailings dam. In-depth analysis of factors affected tailings dam phreatic surface, phreatic surface prediction coupling model based on Generalized Regression Neural Network model (GRNN) and GM (1,1) was established. A tailings dam project is tested using this model. It shows that this coupling model the combination the advantages of both grey prediction model and GRNN, which use the both of a grey system required a sample of data small, simple principle, operator convenience, short-term forecasts of high precision, the advantages of testing can also play a nerve Network parallel computing, fault-tolerant capability and adaptive capacity, and other advantages, the model is less overcome the raw data, data volatility of the accuracy of the forecast, the forecast also enhance the adaptability.3) The deficiencies of traditional safety factor calculation methods , including it is impossible to give practical engineering design variables accurate means, and they mean a large extent, by many objective factors or man-made role of the Impact of these effects have greater degree of ambiguity. But the damage to the dam from the safety limits of difficult to clear, reliable and failure in an intermediate between the transitional state of existence, which is a fuzzy area. Based on the above two issues resolved of the first time, this paper rely on fuzzy theory applied to the stability of the tailings dam, both consider the tailings dam destruction fuzzy, and consider the main variables and parameters of the fuzzy. The fuzzy and the randomness of the integration of tailings dam project on the fuzzy random reliability analysis methods. Results shows that the results and the actual situation in good agreement, it is more scientific and rational than dam safety factor and the traditional calculation methods, which reflect the real situation for the stability of the tailings dam and it also provide a new way to analysis stability of tailings dam.4) Fuzzy theory and reliability theory are introduced to study the seismic stability of the tailings dam. Considered saturation line or not (for floods, dam dried and dam closed), Fuzzy random reliability model of the tailings dam has been studied systematically in the seismic stability of the tailings dam. a tailings dam earthquake fuzzy random reliability method is proposed in this e field . The establishment of the model is applied to the tailings dam reliability analysis of a tailings dam in Panzhihua City of Sichuan Province, it shows that, in line with the actual situation well and seismic fuzzy random reliability analysis model to better reflect the real seismic loading, the true stability of the tailings dam Situation.5) Based on the unascertained mathematical theory, a tailings dam stability unascertained comprehensive evaluation model is established for the first time, a comprehensive evaluation index system is built; the evaluation of the unascertained measure function is formed. Based on the information Entropy theory, the weight of evaluation index is determined, according the confidence level criteria to determine level of stability of the tailings dam. In order to test the model, which is applied in a tailings dam in Hengyang City of Hunan Province, evaluation results show that the model can solve the factors of uncertainty Problems of evaluation, the methods is scientific and rational; it can be promoted to projects application.
|
PDF Full Text Request