【摘要】 变压吸附技术可对氢气进行提纯.文中通过Aspen Adsorption软件,建立了五组分气体（H₂/CO₂/CH₄/CO/N₂=38/50/1/1/10%）在以活性炭和沸石作为吸附剂的层状床的吸附传质传热模型,探究了吸附压力和活性炭与沸石层高度比对层状床变压吸附穿透曲线的影响,进而研究了不同吸附压力、p/F比、吸附剂高度比和吸附床的数量对氢气循环性能的影响.结果表明增大吸附压力和p/F比,氢气的纯度增加,回收率减少;在一定范围内,当高度比减少时,氢气的纯度增加,回收率降低;增加吸附床的数量,氢气的纯度增加,回收率略有下降.基于以上结论,采用了BP神经网络对层状床氢气纯化性能进行多目标优化,并将目标优化预测结果与Aspen模型计算结果对比,结果显示神经网络能很好的对变压吸附纯化性能进行优化.更多还原

【Abstract】 Pressure swing adsorption can be used to purify hydrogen. In this paper, the adsorption mass transfer model of five-component gas（H₂/CO₂/CH₄/CO/N₂=38/50/1/1/10 vol.%） in a layered bed packed with activated carbon and zeolite was established by Aspen Adsorption software. The effects of adsorption pressure and the height ratio of activated carbon to zeolite layer on the penetration curve of pressure swing adsorption in layered bed were explored, and then different adsorption pressures, p/F ratios and adsorbents were studied.The results show that with the increase of adsorption pressure and p/F ratio, the purity of hydrogen increases and the recovery rate decreases. In a certain range, when the height ratio decreases, the purity of hydrogen increases and the recovery rate decreases. With increasing the number of adsorption beds, the purity of hydrogen increases and the recovery rate decreases slightly. Based on the above conclusions, BP neural network was used to optimize the hydrogen purification performance of the layered bed, and the prediction results of the target optimization were compared with those calculated by Aspen model. The results show that the neural network can optimize the pressure swing adsorption purification performance well.更多还原