Research On The Operational Reliability Of Depleted Oil And Gas Reservoirs

As one of the most important means of natural gas peaking shaving,the reliability of underground gas storage have great influence on natural gas pipeline network system operation.Combining the process flow calculation and the failure mechanism assessment,this paper gives the quantitative evaluation results of the dynamic reliability of both units and the whole system.In this paper,the process calculation is simplified to adopt the requirement of underground gas storage reliability assessment.For formation part,three calculation methods are put forward to estimate formation pressure and the applicability of the three methods is analyzed in detail.To obtain the gas storage capacity,the parical swarm optimization alogorition is used to porcess field datas.The relevant process parameters of wellbore system are obtained by nodal analysis method.According to the calculation results of gas wellhead process parameters and the operation characteristics of ground equipment,the process parameters of equipment on the ground are obtained.PIPESIM multiphase flow simulator is used to model the underground gas storage system and analyze the variation rules of the operation parameters.In the reliability assessment prcess of underground gas storage units or equipment,the failure mechanism and the operation parameters of units or equipment are considered.The dynamic reliability of units or equipment under the different conditions is obtained by Monte Carlo simulation technique.For some equipement whose failure records are easy to obtain,some statistical methods such as the time series Monte Carlo simulation,Bayesian Inference and Principal Component Analysis are used to establish the relationship between reliability and operation time.Considering the maintainability of some equipment and units,the distribution of mean time between failure and repair time of equipment is studied by the correlation analysis method.In order to solve the time consuming problem of Monte Carlo simulation,this paper applies the artificial neural network algorithm in units and equipment reliability assessment.To improve the prediction accuracy of the artificial neural network,some methods such as imulated annealing algorithm,limits learning,genetic algorithms are integrated to optimize the traditional modeling process of artificial neural network.The trained models obtaibed by the improved artificial neural network(ANN)method are used to predict the reliability of the underground gas storage units.Comparing with the results obtained by traditional Monte Carlo simulation,the trained ANN model increases the reliability calculation speed significantly.Finally,a two-layer Monte Carlo simulation framework that integrated the process calculation and the reliability models of underground gas storage units is established to evaluate the dynamic reliability of underground gas storage.And the framework is used to evaluate the reliability of an underground gas storage under the different operation conditions.By these studies,the underground gas storage reliability assessment methods are established and can be used in guiding the production process.