考虑异质性的定频空调负荷聚合建模及功率跟踪策略

为便于建立空调负荷聚合模型,目前研究往往假设空调个体参数一致,忽略了空调受控群内部参数差异较大的实际情况,亦未计及空调最小停运时间与整数设定温度的影响,因此该文考虑空调负荷的异质性,提出一种改进型聚合建模方法。该方法根据空调额定制冷量分组,并通过模糊C-均值聚类算法提高受控群电功率的模拟精度。引入最小停运时间以及设定温度的整数约束,使仿真结果更接近实际定频空调运行工况。同时,在实际运算中采用坐标格式存储稀疏矩阵,提高运算速度和模型的实用性。基于该改进型聚合模型,进一步提出一种功率跟踪策略,该策略计算各聚类群每一时刻的需求响应能力,通过求解0—1整数规划问题实现调控对象最优组合。仿真算例验证所提策略对目标功率跟踪的良好效果。     

纳米三氧化二铝在水热条件下的聚集生长模式研究

通过对产物形貌及晶粒度测试,掌握水热温度、水热时间对水热产物影响的基本规律,运用临界成核理论及颗粒成长理论解释提高水热温度、延长水热时间导致的颗粒形态逆转的原因,结合晶体成长阶段性理论,揭示了提高水热温度或延长水热时间以及增大前躯体过饱和状态条件下诱发的两类聚集生长方式,阐述了第Ⅰ类聚集和第Ⅱ类聚集生长的根本原因,其中第Ⅰ类聚集生长是因为过饱和度的增大使得部分AlOOH颗粒在水热溶液中溶解形成的Al^3＋聚集体在另一些颗粒表面的叠合,而发生第Ⅱ类聚集生长的本质是由于pH酸碱环境使得多个AlOOH颗粒通过某些结构相容晶界间的键联而形成的具有特殊生长形态的颗粒。     

Hourly wind power variations in the Nordic countries

Studies of the effects that wind power production imposes on the power system involve assessing the variations of large‐scale wind power production over the whole power system area. Large geographical spreading of wind power will reduce variability, increase predictability and decrease the occasions with near zero or peak output. In this article the patterns and statistical properties of large‐scale wind power production data are studied using the data sets available for the Nordic countries. The existing data from Denmark give the basis against which the data collected from the other Nordic countries are benchmarked. The main goal is to determine the statistical parameters describing the reduction of variability in the time series for the different areas in question. The hourly variations of large‐scale wind power stay 91%–94% of the time within ±5% of installed capacity in one country, and for the whole of the Nordic area 98% of the time. For the Nordic time series studied, the best indicator of reduced variability in the time series was the standard deviation of the hourly variations. According to the Danish data, it is reduced to less than 3% from a single site value of 10% of capacity. Copyright © 2004 John Wiley & Sons, Ltd.     

融合图像轮廓矩和傅立叶描叙子的手势识别

针对手势识别中人手是复杂变形体,手部特征描叙容易受到环境因素影响的特点,提出了一种基于混合轮廓特征的手势识别的新方法.首先根据手的颜色信息将手从复杂的背景中分离出来,然后提取出手轮廓的矩不变量和Fourier描叙子特征,然后将2种特征融合构成混合特征,最后将新特征输入BP神经网络进行识别.实验表明:此方法具有较好的鲁棒性和较高的识别率.     

Normalization of discrete planar objects

The case is made for normalization of discrete planar objects prior to comparison with test objects and an expression for normalization is derived using a Euclidean distance function based on the underlying continuous boundaries of the objects and their prototypes. The results are given in both the spatial and the frequency domain; an analysis of errors due to the quantization introduced by using a discrete grid is also given.     

冶金工业英文超级科技词表叙词分类的研究

阐述了《冶金工业英文超级科技词表》编制过程中的叙词分类原则,分析了典型的分类案例,得出如下结论：正确的分类要以《中图分类法》为依据;准确的分类要以专业知识为基础;批量分类是提高效率和一致性的有效方法;分类是一个不断完善的过程。     

MIMO系统中STTC译码的一种改进型算法

本文对MIMO系统中用于STTC编码器的VA译码算法进行了改进,建立了节点度量的递推方程,并给出了分枝转移的度量方法。改进后的VA算法适用于已经估计出MIMO系统中信道转移阵的条件下。相比于标准的STTC译码算法,改进后的算法可以在一根接收天线上独立进行,并且节点度量可以通过递推方程进行累积。当多根天线独立译码后,可以进一步进行后续比较以再次降低误码率。仿真结果显示：改进VA在比特误码性能上与标准VA算法接近,但是在分段度量计算时收敛速度要远快于标准VA。