Research On Safety Instrument System Of Natural Gas Dehydration Device Based On SIL

Gas dehydration is an important part of the gas purification process, which directly affects the safety of the subsequent production process and the quality of the natural gas. During the process of production, the safety instrumented system (SIS) can be used to provide a safe and reliable protection layer, which can reduce the probability of the occurrence of risk failure. And the relationship between the setting of the interlocking relationship of the protection, can effectively play a protective role in the SIS system At present, the reliability block diagram, fault tree and Markov model are used to calculate the safety integrity level of the oil and chemical industry.The assessment of safety instrumented system safety integrity level is the starting point, first of all failure modes of safety instrumented system and reliability indices are analyzed, that the influence of common invalidation in different redundancy structures are different, and the common cause failure factor is modified in this paper. Secondly, if the interlocking equipment is complex and calculation is difficult, Markov and fault tree is too complicated and difficult to calculate, Bayesian network reasoning is applied to safety instrument system safety to calculate the integrity level in this paper. The main method is changing the fault tree of interlocking relationship to Bayesian network, the probabilities of elementary events are the priori probabilities of parent nodes in Bayesian network, the logic relations of each event in the fault tree are the conditional probabilities of intermediate nodes and leaf nodes, junction tree calculates safety integrity level in the Bayesian network. And the problem of ignoring the parameter uncertainty in the Markov model is calculated by combining the D-S evidence theory and the Markov model. The main interlock relationships of dewatering devices are based on four controlling points, whicn are the liquid level of the water tower, liquid level of flash tank, liquid level of purification tower and inlet flow of dehydration tower, separately their safety integrity levels are calculated by the Markov models, Bayesian networks and Markov models based on D-S evidence theory. Calculation results of the three methods are consistent. On this basis, the calculation results are analyzed, and the three methods are compared and the advantages of the Bayesian network and Markov model based on D-S evidence theory to calculate the level of security integrity are proved.