Application Of Probabilistic Risk Assessment In The Tailing Dam Of

As the indispensable facility of mine concentrator, tailings pond has been one of the major sources of danger, and its security has been paid more and more attention. Dam break accident is known as one of the main disasters, but also the study focus of safety journalists and scholars. This paper takes system theory as the guidance, and tailings dam as a synthesis of silt buildings and drainage structures. From different functional views, analyze the different levels, subsystems and factors, and use Probabilistic Risk Assessment (PRA) method for the qualitative and quantitative analysis. Then taking Haikou phosphorite Caojiagou tailings pond as for instance, combined with its own peculiarities, focus on the analysis of the dam break probability leaded by flood control facilities failures. The main research work and achievements are as follows:The probabilistic risk assessment method is applies to the analysis of tailings dam break accident, and the tailings pond is seen as a whole system whose a variety of factors influence and interact each other. The paper taking system theory as guidance uses main logic diagram, event tree and fault tree to analyze the different levels, subsystems and factors of tailings pond. The main logic diagram is used to find the initial events leading to the tailings dam break, and event tree to deduct the development of initial events, and get the event sequence groups and accident sequence groups, which can further clarify the evolution of initial events. At last, directing at the individual events among the initial events and intermediate events, use fault tree to deeply analyze its causal factors, and provide basis for the quantitative calculation behind. Through the qualitative analysis, understand the main causal factors of tailings dam break, which can provide basis for the safety management of the tailings.According to the lack of event tree and fault tree at quantitative analysis, the use of Binary Decision Diagram (BDD) can solve the problem of probability of dependent events. The paper puts forward the disjoint approach for the repeat events, and gives the accident sequence group probability model based on event tree analysis method.Based on Binary Decision Diagram method, by analyzing the minimum backtracking set of BDD corresponded with fault tree, the paper gives fault tree top event probability calculation. It is benefit for quantitative evaluation of risk factors, and provides basis for the proposed targeted control measures.Taking Caojiagou tailings pond of Haikou phosphate for instance, it proposes the process of using probabilistic risk assessment method to analyze the risk. Combining with the particularity of Caojiagou tailings pond, gives the probability of tailings dam break leaded by drainage facilities failures,and provides basis for the enterprise management improvement.