One-step synthesis of DME (dimethyl ether) from syngas on hybrid catalyst is studied.In the experiment,the effects of variables including composition of feed gas,temperature and pressure are studied under the conditio...One-step synthesis of DME (dimethyl ether) from syngas on hybrid catalyst is studied.In the experiment,the effects of variables including composition of feed gas,temperature and pressure are studied under the condition as follows:the catalyst was composed of C302 methanol synthesis catalyst and CM-3-1 methanol dehydration catalyst (mixing ratio 1∶1).Liquid paraffin is used as inert solvent.The reaction temperature is in the range of 230-270?℃ and the pressure is in the range of 3.0-5.0?MPa.It turns out that with the temperature and pressure increasing the conversion of carbon,the selectivity and production rate of dimethyl ether increase,while the selectivity and production rate of methanol decrease.Furthermore,both the conversion of carbon and the selectivity of dimethyl ether are high for the syngas of appropriate CO 2.展开更多
文摘One-step synthesis of DME (dimethyl ether) from syngas on hybrid catalyst is studied.In the experiment,the effects of variables including composition of feed gas,temperature and pressure are studied under the condition as follows:the catalyst was composed of C302 methanol synthesis catalyst and CM-3-1 methanol dehydration catalyst (mixing ratio 1∶1).Liquid paraffin is used as inert solvent.The reaction temperature is in the range of 230-270?℃ and the pressure is in the range of 3.0-5.0?MPa.It turns out that with the temperature and pressure increasing the conversion of carbon,the selectivity and production rate of dimethyl ether increase,while the selectivity and production rate of methanol decrease.Furthermore,both the conversion of carbon and the selectivity of dimethyl ether are high for the syngas of appropriate CO 2.
文摘极限学习机(Extreme learning machine,ELM)是一种单隐层前馈神经网络(SLFNs),它随机选择网络的隐含层节点及其参数,训练时仅需调节输出层权值,因此ELM以极快的学习速度获得良好的推广性。考虑到ELM的特征映射函数未知时,可以将核矩阵引入到ELM中。针对模型未知的强非线性连续搅拌反应釜(Continuous Stirred Tank Reactor,CSTR),提出一种基于核极限学习机(Extreme Learning Machine with Kernels,KELM)的NARX模型辨识方法。以仿真的CSTR过程实例进行辨识实验,建立基于NARX-KELM的辨识模型。实验结果表明,在相同条件下,与带动量因子的BP神经网络、模糊神经网络(FNN)、GAP-RBF、MGAP-RBF神经网络、回声状态网络(ESN)、ELM等方法相比,KELM能够有效地改进辨识精度,而且性能更好,这表明了所提方法的有效性和应用潜力。
