基于LSTM 神经网络模型的灞河流域径流预报研究

径流预报是缓解洪水的一种重要方法。基于1978-2010 年的水文资料,结合长短期记忆神经网络（Long-Short Term Memory,LSTM）,构建了灞河流域径流预测模型,并且评价了模型对同一流域不同特征水文站的差异及不同季度的预测效果差异。结果表明：不同神经元的组合,对LSTM 模型预测效果会产生影响,利用最佳的神经元组合可以更加有效预测径流量变化,大峪河大峪（三）站的最佳组合为第一层神经元128 个,第二层神经元32 个;灞河罗李村（四）站的最佳组合为第一层神经元128 个,第二层神经元8 个;灞河马渡王站的最佳组合为第一层神经元8 个,第二层神经元2 个。不同站点的LSTM 最佳模型都能较为有效的预测三个水文站2006-2010 年的径流量变化,其中大峪河大峪（三）站效果最佳,其余两个站点效果相对较差。LSTM 模型对各个季度的预测效果有差异,各个站点大部分第三季度的均方根误差都较大,而对第一、四季度的径流预测相对较准确。     

浅论日常生活用品设计

当前的设计方向,正在前所未有的强调其人文关怀。设计,本没有那么高高在上,而是要为生活服务。日常用品是如此的平凡普通,以至于我们通常漠视它们的存在。在日常生活中寻找设计,感悟生活,这是当代设计师应该具备的素质,因为人们生活在设计的海洋中,生活本身就是设计的起源地,而设计,归根结底就是对生活的发言。     

福建龙岩土在日用细瓷坯料中的应用

本文阐述了福建龙岩土在日用细瓷坯料中的应用研究。确定了龙岩土在坯料配方中的理想使用范围为20～40％,制定了理想的坯料配方,研制成高白度、低吸水率、热性能良好、合格率高的日用细瓷。     

基于日常生活的城市设计策略探析——以重庆为例

日常生活是城市空间的基本功能之一,以日常生活空间为核心的城市设计能够在保障城市空间开放性的基础上创造出更具地域人文特色的城市整体形态。本文以重庆为例,针对当前城市日常生活空间存在的主要问题,从宏观、中观、微观三个层面提出了优化城市生活格局、完善生活空间体系、创造生活性空间等城市设计策略,并对相应的城市设计控制机制提出了初步见解。     

Application of switched adaptive system to load forecasting

In this paper, we propose an approach for next day peak load forecasting for electrical companies. First a nonlinear model for the peak load is proposed taking into account the historical load and the temperature. Based on this model time-varying local models are obtained for some temperature intervals. The peak load forecasting system is constructed based on these local models which parameters are estimated using an on-line recursive algorithm. We remark that in this methodology it is not necessary to know precisely the temperature of the days since the proposed system is based on an interval for the future temperature instead of a number. An application example illustrates the proposed approach.     

Classification of daily solar radiation distributions using a mixture of Dirichlet distributions

In order to characterize the fluctuating nature of solar radiation in tropical climate, we classify daily distributions of the clearness index k_t by estimating a finite mixture of Dirichlet distributions without assuming any parametric hypothesis on these daily distributions. The method is applied to solar radiation measurements performed in Guadeloupe (16°2N, 61W) where important fluctuations can be observed even within a period of a few minutes. The results exhibit four distinct classes of distributions corresponding to different types of days. The sequence of such classes can be of interest for future weather prediction.     

AMSR-E亮温监测中国近地表冻融循环算法研究

为更加精细地获取日内地表冻融循环信息,联合使用AMSR-E升轨(13:30)和降轨(1:30)一天两次过境的逐日亮温数据,采用双指标算法,分别判定升轨和降轨过境时刻的地表冻融状态,综合两次过境信息将地表细分为日内完全冻结、日内完全融化、日内冻融循环和日内逆向冻融循环4种类型。经过验证,日内完全冻结地表分类精度达90%,日内完全融化地表和日内冻融循环地表分类精度约为70%。在日内地表冻融循环分类基础上,将年冻融循环周期划分为稳定冻结期、稳定融化期、春季冻融过渡期和秋季冻融过渡期。根据2004年分类结果得出,1月份,日内完全冻结地表和日内完全融化地表大体以秦岭淮河为界,随着温度逐渐升高,日内完全冻结地表南界、日内完全融化地表北界和日内冻融循环地表的南北界从东南向西北移动,温度降低,各边界线从西北向东南移动;稳定冻结期主要出现在1月和2月,稳定融化期出现在7月和8月,春季冻融过渡期是3~5月,秋季冻融过渡期是9~11月。     

基于日常生活视角的农转非社区公共空间使用研究 ——以重庆市礼嘉街道嘉和社区为例

城镇化进程蚕食乡村地区所引发的传统 居住空间变迁，使城市边缘区的农转非社区面临 社会空间重构。这类社区的公共空间完全按照 城市居住景观来修建，但居民延续了很多农村原 有的生活习惯与社区文化，与周边城市在空间上 的同质化和社区居民与“城里人”的异质性导致 按城市住区景观设计的农转非社区公共空间与 社区居民日常活动的不适配。因此，本文首先回 顾了我国农转非社区的形成、社会异质性与空间 同质化特征，并梳理了国内外异质社区公共空间 适配性的相关研究；其次，基于对重庆市嘉和社 区的田野调研，从公共空间属性与使用两个层 面，分析了农转非社区居民的日常活动特征与公 共空间的运行机制，评析了各项公共空间资源配 置与利用率、聚集性以及活动类型之间的适配 关系，探讨了影响农转非社区公共空间配置与设 计与居民日常活动需求失衡的原因，为国内农转 非社区更新，尤其是农转非社区公共空间质量的提升提供科学依据和政策建议。     

【城市笔记：之17】迟到的回望 城市笔记人

同济大学西校门边上的中德学院楼是由青年建筑师庄慎于1999年至2002年间负责设计与施工的。建筑师庄慎通过对建筑布局的不断调试，对流线内外与上下的巧妙切换。对空间大小与明暗的控制。对建筑体量与校园关系的多层次过渡，不仅满足了设计任务书的种种要求。还成功地在建筑语言、空间序列、构造表达等层面上融进了庄慎当年对现代建筑大师勒·柯布西耶作品的个人理解。中德学院楼自落成到现在被媒体介绍和报道得非常少。在这个意义上。这篇对庄慎的访谈，也算是对这一完成水平不错的作品的迟到回望。     

Analysis of the problem of optimal placement and capacity of the hydrogen energy storage system in the power system

Nowadays the trend of increasing the generation units based on renewable energy sources in the electric power system can be observed. Obviously, this is due to the intensifying level of consumer load and demand for electricity. However, renewable generation is characterized by intermittent energy production, which can cause and potential imbalance between generation and demand, especially during off-peak periods. Therefore, in order to ensure a reliable power supply to consumers, it is necessary to use a maneuverable reserve of capacity, such as energy storage systems, in conjunction with the renewable energy source unit. Over the past 10 years, the energy storage market has grown by almost 50%: the installed capacity of energy storage system in the world is about 5 GW. Analysis of the literature on the subject determines the need to study the impact of these devices on the parameters of electric power systems and one of the primary tasks is to determine the optimal location and capacity of energy storage system in the power system. This paper presents the result of solving the task of determining the optimal parameters of a hydrogen energy storage system using the particle swarm optimization method for example a test scheme radial distribution system – 33 bus IEEE. The choice of the type of energy storage is based on such advantages of a hydrogen energy storage system as environmental friendliness, high energy capacity and the ability to store electricity for a long period of time. In addition, compared to lithium-ion batteries, hydrogen energy storage systems have a long life time of about 25 years, during this period of time there is no degradation and significant deterioration of its properties. All these advantages of hydrogen as an energy carrier allow to take into account not only the criterion of total value of active power losses and its maximum reduction respectively, but the possibility and economic efficiency of partial use of the stored hydrogen for other needs when determining the optimal scenario of their operation in the process of discharge.