Study On Optimization Of Low Temperature Helium Extraction Process From Natural Gas

Helium is widely used in various industries due to its extraordinary physical and chemical properties such as difficult liquefaction,radioactive inertness,almost insoluble in water and the highest ionization potential,and its use value in many industries is irreplaceable by other substances.China uses about 15 million cubic meters of helium every year,with a market price of about 1.8 billion yuan.However,the helium resources in China are relatively small.The helium supply market security and strategic security are worrying.Therefore,it is of great significance to study the extraction process of helium,how to reduce the energy consumption cost of helium extraction and how to improve the purity of helium.Low-temperature helium extraction from natural gas is a commonly used process at present,but the low-temperature helium extraction process can only extract about 60%-70%of the crude helium,which needs to be refined.The low-temperature helium extraction process is only the first step to produce helium,but for the later steps,the concentration of crude helium extracted should be as high as possible.Therefore,the purpose of this study is to select a reasonable helium extraction process based on natural gas liquefaction,and to optimize the parameters of the process to achieve the minimum energy costs under the condition of ensuring the helium concentration.Specifically,it includes the following contents:(1)Selected the state equation of HYSYS simulation operation,treated the the raw gas of a gas field.DEA absorbent is selected to deacidify and molecular sieve is dehydrated by comparison,so that the raw gas can meet the requirements of impurity content allowed by low temperature operation process.Analyzed the physical properties of the treated raw gas and discussed the dew point of each component and raw gas,which laid a good foundation for the subsequent process construction.(2)The process flow of helium extraction from natural gas at low temperature and the process flow of natural gas liquefaction are analyzed respectively.The process flows of the two are similar,so three kinds of helium extraction process flow based on natural gas liquefaction are proposed.They are:mixed refrigerant refrigeration+nitrogen cycle refrigeration process,natural gas expansion refrigeration+nitrogen cycle refrigeration process,nitrogen expansion liquefaction refrigeration process.Under the same operating conditions,the three schemes are comprehensively compared from six aspects:crude helium concentration,helium recovery rate,comprehensive energy consumption of the device,minimum operating temperature of the device,LNG production and recovery rate,and equipment investment.and the scheme 3-nitrogen expansion liquefaction refrigeration process is determined to be the optimal combined process.(3)After analyzing the parameters of the optimal process,I determined six parameters which have great influence on the energy consumption of the process,and the influence of each parameter on the energy consumption of the device and the final crude helium concentration is analyzed respectively.The six parameters are:the feed temperature of the first tower,the feed pressure of the first tower,the reflux ratio of the first tower,the high pressure of the refrigerant,the low pressure of the refrigerant and the flow rate of the refrigerant.(4)In this paper,two methods are selected to optimize the optimal process.Firstly,BP neural network algorithm is adopted,which is a multi-layer feedforward network trained by error reverse propagation algorithm.After preparing a set of test data and a set of test data,the calculation is carried out,and the accuracy of the algorithm is determined by analyzing the error.Finally,a set of optimum parameters were obtained.Under these parameters,the crude helium concentration was 63.52%,and the energy consumption was 18.08%less than before.But BP neural network algorithm can only give the optimal group of known groups,that is,the real optimal parameters may be near the optimal solution given.So this paper uses response surface analysis method to optimize the process parameters.(5)Response Surface Analysis(RSA)is a statistical method that uses multiple quadratic regression equation to fit the functional relationship between factors and response values.Through the analysis of the equation of regression,we can find the best technological parameters and solve the multivariate problem.The regression equation of energy consumption and crude helium concentration was obtained by response surface methodology,and a set of optimal parameters were obtained.Under these parameters,the crude helium concentration was 62.53%,and the energy consumption was 21.4%less than that before optimization.In addition,the values of these parameters are near the optimal solution given by BP neural network,which also proves the accuracy of BP neural network.(6)The influence of helium content and nitrogen content of feed gas on the final crude helium product was analyzed.It was found that the change of helium content had little effect on the crude helium concentration extracted by the device;the change of nitrogen content below 25%has little effect on the crude helium concentration extracted by the device,but the effect of helium extraction decreased significantly when it was higher than 25%.