不同声音偏好的游客声景观主观评价差异——以丽江大研古镇为例*

古镇作为集自然景观与人文景观一体的景区，具有丰富的声音类型。本研究以大研古镇为例，通过实地调研探究不同声音偏好的游客对古镇声景观的主观评价差异。基于游客的声音偏好，将游客分为偏爱自然声和偏爱人工声两大类。通过因子分析提取游客声景观主观评价的5个主因子：大小、音质、效价、偏好和唤醒。进一步分析发现，这5个因子具有一定的层级性，其中从大小到唤醒代表声景观评价从声音的物理属性向声音的联想评价逐级提升。其中在低层级评价(大小、音质)中，偏好自然声的游客和偏好人工声的游客无显著差异，低层级评价具有稳定性；而在高层级评价(效价、偏好和唤醒)中，偏好人工声的游客更关注古镇声景观的淳朴性和遗产性。因此，游客对声景观的主观评价可视为一种指标，判断景区声景所处的评价阶段，从而为景区声景观改善提供更有针对性的建议。     

Astrocyte and ions metabolism during epileptogenesis:A review for modeling studies

As a large group of cells in a central nervous system, astrocytes have a great influence on ion and energy metabolism in a nervous system. Disorders of neuronal ion and energy metabolism caused by impaired astrocytes play a key role in the pathogenesis of epilepsy. This paper reviews the existing computational models of epileptogenesis resulting from impaired astrocytes and presents several open perspectives with regard to ion and energy metabolism-induced epileptogenesis in a neuron-astrocyte-capillary coupled model.     

Origins of Boosted Charge Storage on Heteroatom-Doped Carbons

Although tremendous efforts have been devoted to understanding the origin of boosted charge storage on heteroatom-doped carbons, none of the present studies has shown a whole landscape. Herein, by both experimental evidence and theoretical simulation, it is demonstrated that heteroatom doping not only results in a broadened operating voltage, but also successfully promotes the specific capacitance in aqueous supercapacitors. In particular, the electrolyte cations adsorbed on heteroatom-doped carbon can effectively inhibit hydrogen evolution reaction, a key step of water decomposition during the charging process, which broadens the voltage window of aqueous electrolytes even beyond the thermodynamic limit of water (1.23 V). Furthermore, the reduced adsorption energy of heteroatom-doped carbon consequently leads to more stored cations on the heteroatom-doped carbon surface, thus yielding a boosted charge storage performance.     

基于深度学习的ELM实时识别研究

基于深度学习的方法，在HL-2A装置上开发出了一套边缘局域模(ELM)实时识别算法。算法使用5200次放电数据(约24.19万数据切片)进行学习，得到一个深度为22层的卷积神经网络。为衡量算法的识别能力，识别了HL-2A装置自2009年实现稳定ELMy H模放电以来所有历史数据(约26000次放电数据)，共识别出1665次H模放电，其中误识别35次，误报率为2.10%。在实际的1634次H模放电中，漏识别4次，漏识别率为0.24%。该误报率和漏报率可以满足ELM实时识别的精度要求。识别算法在实时控制环境下，对单个时间点的平均计算时间为0.46ms，可以满足实时控制的计算速度要求。     

On the output function in a Ginsburg’s machine

We give the form of the output function in Ginsburg’s machine in which the input and output dictionaries are abelian groups and the transition function is of a special form.