摘要：采用实际城市污水在序批式反应器(SBR)中试装置中成功培养出好氧颗粒污泥.颗粒污泥的沉降速率>21 m·h-1,显示了良好的沉降性能.颗粒污泥对化学需氧量(COD)、氮(N)、磷(P)具有同步去除的能力.在3 h的运行周期中,反应器出水COD<50 mg·L-1,氨氮(NH+4-N)<5.0 mg·L-1,总氮(TN)<15 mg·L-1,对TN和总磷(TP)的去除率达到50%.采用共聚焦激光扫描显微镜(CLSM)对好氧颗粒污泥的结构分析表明,好氧颗粒污泥的胞外多聚物(EPS)在整个颗粒污泥中均匀地分布,构成颗粒污泥的骨架.对颗粒污泥的X射线衍射(XRD)分析显示,在好氧颗粒污泥的内部存在许多无机矿物质,在颗粒污泥形成初期,废水中的无机成分可能具有"晶核"的作用.

摘要：提出近似重复矢量(Approximate Repeat Vector,ARV)模型用于DNA序列冗余片段的描述.通过将数据生物信息学特征引入压缩预处理,并使用ARV矢量构造编码码本,提出了非对称DNA序列压缩算法BioLZMA-2.算法引入基于粒子群优化的Memetic改进方法CLIPSO-MA用于压缩码本的智能优化设计,有效提升了编码性能.在标准测试序列上的实验结果表明,BioLZMA-2可获得比现有DNA序列数据压缩方法更高的压缩率.

摘要：多频测试使模拟电路响应的故障状态和正常状态差异最大化,而神经网络具有解决复杂分类问题的能力。结合两者优点,提出一种基于多频测试和神经网络的故障诊断方法:通过灵敏度分析指导多频测试矢量生成,选择最优测试激励;提取各测试节点响应的故障信息,利用神经网络对各种状态下的特征向量进行分类决策,实现对故障元件的检测和定位。实验结果表明,该方法对模拟电路故障诊断非常有效,具有很强的实用性。

摘要：The development of highly active DFT catalysts for an electrocatalytic N_(2)reduction reaction(NRR)under mild conditions is a difficult *** this study,a series of atom‐pair catalysts(APCs)for an NRR were fabricated using transition‐metal(TM)atoms(TM=Sc−Zn)doped into g‐CN *** electrochemical mechanism of APCs for an NRR has been reported by well‐defined density functional theory *** calculated limiting potentials were−0.47 and−0.78 V for the Fe_(2)@CN and Co_(2)@CN catalysts,*** to its high suppression of hydrogen evolution reactions,Co_(2)@CN is a superior electrocatalytic material for a N_(2)*** Fe_(2)@CN may be a strongly attractive material for an NRR with a relatively low overpotential after an improvement in the *** two‐way charge transfer affirmed the donation‐acceptance procedure between N_(2)and Fe_(2)@CN or Co_(2)@CN,which play a crucial role in the activation of inert N≡N *** study provides an in‐depth investigation into atom‐pair catalysts and will open up new avenues for highly efficient g‐CN‐based nanostructures for an NRR.