基于指数平滑技术的装备维修备件需求预测

以装备维修备件的需求量预测问题为研究对象,运用指数平滑预测技术,构建了装备维修备件需求量预测模型,并借助Excel环境对某装备维修备件的需求量进行了预测。预测结果表明,该方法为科学预测装备维修备件的需求量提供了一种可行的方法。     

基于PCA-LSSVM的厌氧废水处理系统出水VFA在线预测模型

采用IC厌氧废水处理系统处理人工合成废水,并利用PCA-LSSVM模型对系统出水挥发性脂肪酸(VFA)进行预测.首先利用主成分分析法(PCA)分析影响厌氧废水出水VFA浓度的多个变量的相关性并降低输入变量维数,然后用网格搜索结合10倍交叉验证优化LSSVM模型参数sig2和gam,最后利用建立的模型对实验数据进行仿真预测.仿真结果表明,稳态LSSVM模型对稳态条件下厌氧废水处理系统出水VFA具有很好的仿真预测能力,相对误差在4.72%以内,平均相对百分比误差(MAPE)为1.61%,均方根误差(RMSE)为1.08,相关系数达0.9996;稳态干扰LSSVM模型对厌氧废水处理系统出水VFA的仿真预测精度有所降低但仍然具有较好的预测能力,平均相对百分比误差(MAPE)为15.83%,均方根误差(RMSE)为15.45,相关系数为0.9984,该方法可为厌氧出水VFA在线预测和厌氧废水处理系统的优化控制提供指导.     

Oxidation of diesel soot on binary oxide CuCr(Co)-based monoliths

Binary oxide systems(Cu Cr2O4, Cu Co2O4), deposited onto cordierite monoliths of honeycomb structure with a second support(finely dispersed Al2O3), were prepared as filters for catalytic combustion of diesel soot using internal combustion engine’s gas exhausts(O2, NOx, H2 O, CO2) and O3 as oxidizing agents. It is shown that the second support increases soot capacity of aforementioned filters, and causes dispersion of the particles of spinel phases as active components enhancing thereby catalyst activity and selectivity of soot combustion to CO2. Oxidants used can be arranged with reference to decreasing their activity in a following series: O3 NO2> H2 O > NO > O2> CO2. Ozone proved to be the most efficient oxidizing agent: the diesel soot combustion by O3 occurs intensively(in the presence of copper chromite based catalyst) even at closing to ambient temperatures.Results obtained give a basis for the conclusion that using a catalytic coating on soot filters in the form of aforementioned binary oxide systems and ozone as the initiator of the oxidation processes is a promising approach in solving the problem of comprehensive purification of automotive exhaust gases at relatively low temperatures, known as the "cold start" problem.     

Oxidation of diesel soot on binary oxide CuCr(Co)-based monoliths

Binary oxide systems (CuCr₂O₄, CuCo₂O₄), deposited onto cordierite monoliths of honeycomb structure with a second support (finely dispersed Al₂O₃), were prepared as filters for catalytic combustion of diesel soot using internal combustion engine's gas exhausts (O₂, NO_x, H₂O, CO₂) and O₃ as oxidizing agents. It is shown that the second support increases soot capacity of aforementioned filters, and causes dispersion of the particles of spinel phases as active components enhancing thereby catalyst activity and selectivity of soot combustion to CO₂. Oxidants used can be arranged with reference to decreasing their activity in a following series: O₃ ? NO₂ > H₂O > NO > O₂ > CO₂. Ozone proved to be the most efficient oxidizing agent: the diesel soot combustion by O₃ occurs intensively (in the presence of copper chromite based catalyst) even at closing to ambient temperatures. Results obtained give a basis for the conclusion that using a catalytic coating on soot filters in the form of aforementioned binary oxide systems and ozone as the initiator of the oxidation processes is a promising approach in solving the problem of comprehensive purification of automotive exhaust gases at relatively low temperatures, known as the “cold start” problem.     

Leak location of pipelines based on transient model and PSO-SVM

An improved and integrated approach of support vector machine and particle swarm optimization theory (PSO-SVM) is first used to detect the leak location of pipelines and overcome the problem of multiple leaks. The calibration and predictive ability of improved PSO-SVM is investigated and compared with that of other common method, back-propagation neural network (BPNN). Two conditions are evaluated. One with a leak involves a set of 20 samples, while another with two leaks has 127 samples. Both internal and external validations are performed to validate the performance of the resulting models. The results show that, for the two conditions, the values calculated by improved PSO-SVM are in good agreement with those simulated by transient model, and the performances of improved PSO-SVM models are superior to those of BPNN. This paper provides a new and effective method to inspect the multiple leak locations, and also reveals that improved PSO-SVM can be used as a powerful tool for studying the leak of pipeline.     

Fuzzy modelling to identify key drivers of ecological water quality to support decision and policy making

Water quality modelling is an effective tool to investigate, describe and predict the ecological state of an aquatic ecosystem. Various environmental variables may simultaneously affect water quality. Appropriate selection of a limited number of key-variables facilitates cost-effective management of water resources. This paper aims to determine (and analyse the effect of) the major environmental variables predicting ecological water quality through the application of fuzzy models. In this study, a fuzzy logic methodology, previously applied to predict species distributions, was extended to model environmental effects on a whole community. In a second step, the developed models were applied in a more general water management context to support decision and policy making. A hill-climbing optimisation algorithm was applied to relate ecological water quality and environmental variables to the community indicator. The optimal model was selected based on the predictive performance (Cohen’s Kappa), ecological relevance and model’s interpretability. Moreover, a sensitivity analysis was performed as an extra element to analyse and evaluate the optimal model. The optimal model included the variables land use, chlorophyll and flow velocity. The variable selection method and sensitivity analysis indicated that land use influences ecological water quality the most and that it affects the effect of other variables on water quality to a high extent. The model outcome can support spatial planning related to land use in river basins and policy making related to flows and water quality standards. Fuzzy models are transparent to a wide range of users and therefore may stimulate communication between modellers, river managers, policy makers and stakeholders.     

A review on political factors influencing public support for urban environmental policy

Environmental quality is important to urban residents’ physical, social and psychological well-being. Governments have been formulating and implementing policies to mitigate environmental deterioration in cities. To effectively implement policies and achieve policy goals, gaining sufficient public support is an essential prerequisite; the policies would be in ruins and the government may also suffer from setbacks in other policy areas in the absence of public support. Therefore, to understand what contributes to public support is a crucial task for policymakers. Though current literature on socio-demographic and attitudinal/psychological factors provides fruitful accounts for policy support, a comprehensive examination of political determinants has yet to be established. In response to this, this review paper explores political factors that influence level of policy support based on existing literature. An integrated framework is proposed to explain policy support for urban environmental policy. This paper also discusses the possible directions of future research.     

基于支持向量机的洪水灾情综合评价模型

在阐述支持向量机的基本原理、二值分类和多值分类技术及各自特性的基础上，结合洪水灾情综合评价中受自然环境、社会经济等诸多因素的影响且实测样本资料较少的特点，以及目前已有评价模型不能或难以解决的小样本、“过学习”、局部最小等实际难题，提出了基于支持向量机的洪水灾情综合评价模型，并应用实例进行了验证。研究结果表明，此模型和传统的灾情评估法、人工神经网络评价模型一样有效合理，并且模型运算时间比人工神经网络模型要短。因此，不仅为洪水灾情综合评估提供了一种新的模型，而且由于支持向量机遵循统计学习理论中结构风险最小化的原理，具有解决有限样本、非线性及高维识别中的优势，必将比其他传统的评价模型得到更广泛的应用和发展。〖     

A Structural Equation Model of Residents＇ Support for Mega Event

The aim of this article is to develop a structural equation model to assess key factors of residents' support for hosting mega event based on previous literature. The model consisted of five latent constructs and eight path hypotheses. A survey was conducted in Shanghai before 2010 World Expo. It was found that the support for mega events is affected directly and/or indirectly by four determinants factors: perceived benefits, perceived costs, personal benefits and community attachment, and support relies heavily on perceived benefits rather than costs. This study contributes to the existing body of knowledge in an attempt to understand local residents' support for a mega event in different economic and cultural settings.