Three-part auctions versus self-commitment in day-ahead electricity markets

We examine the economic consequences of a bid-based security-constrained centralized unit commitment paradigm based on three-part offers, which is the prevalent day-ahead market-clearing mechanism in restructured electricity markets in the United States. We then compare this approach with an energy-only auction with self-commitment (such as in Australia) addressing efficiency and pricing as well as the tradeoff between coordination losses and incentives to bid truthfully.     

改进粒子群算法的两阶段梁式结构损伤识别

针对实际工程监测时损伤识别误差大的问题,提出一种基于改进粒子群算法的两阶段识别方案。第1阶段利用D-S证据理论融合算法进行损伤定位;第2阶段利用改进的粒子群算法,对定位结果进行修正,同时准确定量损伤。仿真算例和实验分析结果表明:由于第1阶段损伤定位减少了可能损伤单元的数量,第2阶段基于改进粒子群算法的搜索范围减小,能更准确地识别多损伤和小损伤的位置和程度,且抗噪性能良好。     

A new prediction model of electricity load based on hybrid forecast engine

In this paper, a new prediction model is introduced based on hybrid forecast engine and new feature selection. In this model, the load signal is filtered by feature selection to filter out the best candidates. Then, the proposed forecast engine is predicted the output of feature selection. In this model, the weights of proposed forecast engine are optimised by an intelligent algorithm to increase its accuracy. Effectiveness of the proposed method is applied over real-world engineering test case and compared with other different well-known methods. Obtained results proof the validity of the proposed method.     

Assessment of total suspended solids (TSS) event load and its uncertainties in combined sewer system from continuous turbidity measurements

This paper presents a methodology for assessing TSS wet weather event load along with the associated uncertainties from continuous turbidity measurements. The proposed method is applied to dozens of rainfall events sampled on various urban catchments in the cities of Paris and Nantes. This method takes into account the dependency between the residual errors made by converting the turbidity values into TSS concentrations using an average TSS-turbidity relationship. Results obtained indicate that the inter-event variations of TSS-turbidity relationships induce significant systematic errors on TSS event loads which can reach ±30% of the average TSS event load.     

基于荷载效应的结构抗风设计方法研究

对等效静风荷载的基本理论进行分析,指出GB 50009-2001《建筑结构荷载规范》的风振系数法给出的等效静风荷载在只考虑1阶振型的前提下,能够实现各种荷载效应的等效;需考虑多阶振型的结构,只有在各阶振型完全相关的条件下,才能得到可满足全部荷载效应等效的静风荷载;一般情况下,包括阵风荷载因子法在内的计算方法,给出的等效静风荷载只能保证单个或多个响应目标的等效,其他荷载效应存在较大不确定性,可能造成据此进行的抗风设计偏于保守或不安全。为克服这些传统抗风设计方法的局限,提出了直接以荷载效应包络值进行结构抗风设计的基本思路：结合风洞试验得到各种荷载效应在不同风向角下可能出现的上、下限值,再将其与其他荷载作用下的效应直接组合进行结构设计。通过工程实例分析表明,采用阵风荷载因子法得出的等效静风荷载值可能导致其他荷载效应被大大高估,而直接采用荷载效应包络值用于结构设计,不但其值准确,且易于操作。     

基于微粒群算法和BP网络的结构损伤识别

提出了基于BP网络的结构损伤识别方法,同时引入了微粒群算法对BP网络权值和阀值进行优化,建立了完整模型和损伤模型,进行了仿真,经检测样本证明,微粒群算法优化后的BP网络能很好的识别出结构损伤的位置和程度,是一种快速、有效的结构损伤识别方法。     

基于全年动态负荷分析的冷热源方案设计

以北京某酒店为例,对该建筑进行全年逐时动态负荷计算分析。根据计算分析结果,确定制冷季梯级利用离心式冷水机组、螺杆机组和空气源热泵机组作为空调系统冷源,供热季利用板式换热器和空气源热泵机组作为空调系统热源,以达到系统节能的目的。     

一种基于遗传算法的动载荷识别方法

提出了一种基于遗传算法的时域内载荷识别方法,并通过工程实例加以分析,将遗传算法应用到载荷识别过程中,将结构动力学的反问题转化为正计算,避免了传统方法的不足,进一步证明了该算法的有效性。     

基于负荷预测的中央空调节能控制

结合对空调冷负荷进行预测的必要性,通过具体工程实例介绍了基于负荷预测的中央空调节能控制,阐述了冰蓄冷空调系统优化控制的一般过程,给出了进行负荷预测的方法及基于负荷预测的中央空调优化控制方案。     

Resource-constrained critical path analysis based on discrete event simulation and particle swarm optimization

The absence of a valid resource-constrained critical path method (CPM) not only hampers the widespread use of mainstream project scheduling software in construction management practice, but also destabilizes the very foundation of any sophisticated, CPM-based time or cost analysis in construction scheduling research. This has motivated us into developing an innovative, fully-automated solution to resource-constrained CPM called the Simplified Simulation-based Scheduling system (short as S3). S3 takes advantage of the simplified discrete event simulation approach (SDESA) and the evolutionary optimization technique called particle swarm optimizer (PSO) to automate the formulation of a resource-constrained schedule with the shortest total project duration. We clarify basic issues of resource scheduling, elaborate on the formation of a CPM simulation model by SDESA, present PSO algorithms, and discuss the PSO solution formulation and simulation–optimization interaction in relation to the development of S3 software. In order to introduce S3 to construction schedulers, we also reference the relevant functionalities and features of Primavera Project Planner (P3) and Microsoft Project, which are applied alongside S3 in two case studies. The first case is a classic textbook example while the second case is based on a real drainage project in Hong Kong. In both cases, S3 eclipses the current CPM software with respect of (1) shortening the total project duration; (2) optimizing provisions of resources of various types; and (3) producing valid total float values to guide schedule implementation.     

基于微粒群算法和模糊神经网络的火灾探测研究

针对火灾探测的特点,将模糊系统和神经网络有机结合,实现模糊系统设计参数的自动调整。采用符合国家标准明火、阴燃火以及厨房环境下的干扰火等作为模糊神经网络的训练样本和测试样本,依据模糊神经网络算法要求,完成了网络结构的设计,并给出相应的计算模型,利用微粒群算法对网络的权值进行学习与训练。结果表明,该算法在探测国家标准火的火灾状态方面具有有效性和可行性。     

基于负荷密度模型的长沙城区负荷预测

建立了负荷密度模型，分析了使用负荷密度模型进行负荷预测需要考虑的问题，介绍了负荷密度模型参数的取定方法，并以长沙城区为例进行分区负荷预测。     

基于建筑全年动态冷负荷的冷水机组优化配置方案

提出一种在设计阶段对冷水机组方案进行优化配置的方法.首先,冷水机组的能耗计算简化为制冷机的实际制冷量和冷却水进口温度两个独立变量的函数.进一步,通过建筑动态负荷计算获得全年冷负荷频率特性以及相应的室外湿球温度分布,其中湿球温度决定了冷却水最低进口温度.最终,计算出各种冷水机组配置方案的全年以及不同冷负荷需求工况下的运行电耗,并得出最优化的节能方案.     

Traffic load modelling based on structural health monitoring data

Live and fatigue load models are foundations for the life-cycle design of highway bridges. Many highway bridges are now equipped with structural health monitoring (SHM) systems, which provide valuable data to establish load models. In this paper, traffic load models of the Binzhou Yellow River Highway Bridge are developed based on the field measurement of vehicles by an existing SHM system. The probabilistic distribution model and extreme value distribution of gross vehicle weight are statistically analysed using the monitoring data. The results indicate that they follow the bimodal lognormal and Gumbel distributions, respectively. The fatigue load spectrum is also studied. The logistic model is employed to predict the long-term traffic volume, and its parameters of the logistic model are updated using the monitored traffic volume. The combination of the fatigue load spectrum and the traffic volume forecast using the updated logistic model provides a load model for estimation of fatigue damage evolution of bridges.     

基于动静载试验的某市政桥梁性能评价

针对某市政桥梁的成桥状态和性能进行了检测和评价,通过利用重车对桥梁进行静载和动载试验,测量了桥梁跨中截面的竖向挠度和应力(应变)分布、构件的开裂情况、控制截面的冲击系数、结构的动力特性等指标,进而对该桥梁的性能进行评估,研究结果表明该桥承载能力及工作状况符合设计要求和满足正常的使用要求。     

基于指数荷载传递模型的单桩承载力特性分析

在桩基模拟试验的基础上提出一种指数型桩身荷载传递模型,该指数模型所需参数只有两个,且物理意义明确。引入该指数模型,并基于桩端荷载-位移关系为双曲线的假设,推求了桩-土共同作用的微分方程,经过对方程进行求解,得到了单桩桩身荷载与沉降关系的解析解,进而得到桩顶的荷载-位移曲线,可用于确定桩的承载力。工程实测结果与本文解析解结果的对比表明,较为吻合。     

基于APDL语言的结构动载荷响应分析综述

运用振动分析中载荷响应的基本原理,研究并运用ANSYS有限元计算软件提供的APDL语言处理了结构动载荷的加载情况,介绍了各种动载荷作用下结构动态响应分析的简便、通用的方法,实践证明该方法也同时可进行结构的模态分析。     

粒子群优化协同克里金法在确定山地斜坡土层厚度中的应用

克里金法是广泛应用的空间插值方法,但仅考虑单一因素的普通克里金法在确定山地斜坡土层厚度中存在较大误差。针对普通克里金法中的不足之处,提出了一种确定土层厚度的基于粒子群优化的协同克里金法。该方法首先用粒子群优化算法拟合半变异函数,然后将该函数用于以高程值作为辅助变量的协同克里金法中,并根据均方根误差来评价土层厚度的不确定性。将该方法应用于重庆万盛某边坡土层厚度的确定,通过交叉验证,结果表明:与普通克里金插值法相比较,考虑高程的协同克里金法插值的均方根误差降低了39.32%;基于粒子群优化的普通克里金法和协同克里金法的均方根误差分别降低了28.79%和48.45%。基于粒子群优化的协同克里金插值法对提高土层厚度的插值精度有较大作用。     

基于模糊-RBF BP神经网络的天然气负荷预测

针对天然气负荷变化的特点,综合考虑各种影响因素,提出一种基于模糊-RBF P神经网络的天然气负荷预测模型,并编制了负荷预测软件。在RBF网络模型中采用了最新邻聚类算法,实现了网络结构和参数的双重调节,提高了训练的速度和预测的精度。将该软件应用于实例计算,并与相关模型结果比较,证明该模型可以更加快速准确地预测天然气的负荷值。