Benchmarking of electricity networks: Practical problems with its use for regulation

Benchmarking is a technique derived for management purposes, intended to identify outliers in any population as possible problems for resolution. Several regulatory agencies have tried to use benchmarking to define the degree of inefficiency of regulated companies, by reference to some target or frontier. This paper identifies the main problem inherent in trying to adapt a management technique to a regulatory purpose – namely that it requires the exercise of subjective judgements. The resulting lack of predictability and objectivity is not conducive to the provision of efficient regulatory incentives.This paper explains where subjective judgement enters into regulatory applications of benchmarking, which is not always apparent. The choice of model and selection of data sets are two obvious areas, but the main problem arises over the assumption that any costs not explained by the model must be due to inefficiency. Such assumptions are simply unsupported by evidence and lead to cost targets (or cost reduction targets defined over several years) that are little better than subjective guesses. There are other, superior ways to set cost reduction targets, based on long-term trends in total factor productivity for the regulated sector, which are used in the US and in other countries. However, some regulators feel obliged to use benchmarking and the paper concludes with recommendations as to how benchmarking should support further and more objective investigations into the costs of regulated utilities.     

THE IMPORTANCE OF NEUTRAL EXAMPLES FOR LEARNING SENTIMENT

Most research on learning to identify sentiment ignores "neutral" examples, learning only from examples of significant (positive or negative) polarity. We show that it is crucial to use neutral examples in learning polarity for a variety of reasons. Learning from negative and positive examples alone will not permit accurate classification of neutral examples. Moreover, the use of neutral training examples in learning facilitates better distinction between positive and negative examples.     

On stability of second-order quasi-polynomials with a single delay

In this note the stability of a second-order quasi-polynomial with a single delay is studied. Although there is a vast literature on this problem, most available solutions are limited to some particular cases. Moreover, some published results on this subject appear to contain imprecise, or even wrong, conditions. The purpose of this note is to show that by accurate application of known theories, a complete explicit characterization of stability regions can be derived in a most general case. As a byproduct of the proposed analysis, we show that in the high-order case the quasi-polynomial is delay-independent unstable whenever its delay-free version has an odd number of unstable roots (or, equivalently, a negative static gain).     

95W-Ni-Fe 合金工艺缺陷的 SAM 和 XPS 分析

藉助扫描俄歇探针、Ｘ射线光电子谱等分析了９５Ｗ－Ｎｉ－Ｆｅ合金碳污染和氧化的本质。发现，碳污染试样中碳的化学状态主要为石墨，少量为铁或镍的有机化合物，这些物质的界面富集使材料脆化。钨合金氧化后，氧大量富集于钨／粘结相界面，使沿着该界面的断裂更容易发生。     

速溶奶粉中卵磷脂含量的测定

叙述了速溶奶粉中卵磷脂的测定方法，样品通过一系列处理后，测定其磷含量，回收率为９６％￣９８％。     

经纬仪竖盘检测新方法及综合检验仪

本文提出一种检定经纬仪竖直盘测角误差的新方法．该方法精度高，稳定性好，能直接与国家角度标准溯源．由它组成的“经纬仪综合检验仪”可以检测ＪＪＧ４１４—９４光学经纬仪检定规程中所有的项目及细分误差，其总不确定度为０．４″．被测仪器包括各种类型及精度的光学经纬仪、电子经纬仪、水准仪等．     

前向人工神经网络模型用于时间序列分析

介绍前向人工神经网络模型，并与传统的统计学线性模型进行比较。为了适应时间序列变量的特点，对逆传播模型进行了改进，并对太阳黑子活动的观察值这一典型的时间序列进行了分析，并检验了模型的预测能力，取得了较好的结果     

ZS1756振动筛动态分析

应用ＳＡＰ５Ｐ软件计算矿用振动筛前４０阶固有模态，从提高寿命观点看，第１阶，第３－４阶和第６阶模态应采取抑制措施；从队噪观点看，前６阶模态应给予注意。     

通信中的话音编码技术

在分析回顾现有话音编码方案基础上提出话音编码系统的五层结构模型以及“在收端利用边信息获取激励码”的概念。     

异型材挤出口模内的流动模拟

本文根据挤出机理，对口模内的流动行为进行适当的简化和假定，建立了挤出口模内流动分析的数学模型，并采用横截面／假想区域法实现了复杂异型材挤出口模内的数值模拟，数值分析结果与三维有限元解和解析解的对比，取得了满意的效果。