PREDICTIVE REAL TIME CONTROL OF SURCHARGED INTERCEPTORS: IMPACT OF SEVERAL CONTROL PARAMETERS1

ABSTRACT: A global predictive real time control strategy minimizing overflow volumes from combined sewers during rainfall was applied to control flows entering the Marigot interceptor (Laval, Canada) for 23 rain events that occurred in this urban area during the summer of 1989. Different surcharge intensities were allowed in the sewer. The duration of the control horizon as well as the location and number of control regulators were varied to assess the impact of these parameters on total overflow volumes and on control safety. Due to the high propagation speed of flow waves in pressurized conduits, it was found that five‐ minute control horizons were sufficient to ensure control performance and safety when important surcharges were permitted in the interceptor and when the controlled regulators were located where pressurized flow occurred. It was also found that it is possible to reduce the number of controlled regulators by placing them at locations intercepting the largest volumes of water during rain periods.     

基于灰色理论的特种设备安全事故预测研究

应用灰色理论,借助于EXCEL及MATLAB软件工具,并以我国近年实际数据为基础,对我国“十二五”期间特种设备安全事故的绝对死亡人数和万台设备死亡人数进行了预测.结果表明,基于灰色理论的特种设备事故的死亡人数和万台设备死亡人数的模拟值和真实值较吻合,预测模型精度检验为好；预测指标可为特种设备安全政策制订和科学监管决策提供依据及参考.     

WATER QUALITY MANAGEMENT IN THE HSINTIEN RIVER IN TAIWAN1

ABSTRACT: The objective of cost effectiveness has led to the use of mathematical decision models to implement the best water quality control program in a river from the various alternatives available at a time. The paper presents the water quality control program in the Hsintien River in Taiwan by the use of probabilistic programming technique.     

TWO-DIMENSIONAL DIFFUSION-PROBABILISTIC MODEL OF A SLOW DAM BREAK1

ABSTRACT: A two-dimensional model of a dam-break flood wave is developed by simplifying the St. Venant equations to eliminate local acceleration and inertial terms and combining the simplified equations with continuity to form a diffusion type partial differential equation. This model is cascaded with a two point probability estimate scheme to account for uncertainty in the dam break flood hydrograph and channel roughness. The development and application of the probabilistic model is the main contribution of this paper. The approach is applied to a hypothetical dam break of Long Valley Dam on the Owens River above Bishop, California.     

Applying a Large,Statewide Database to the Assessment,Stressor Diagnosis,and Restoration of Stream Fish Communities

In this report, predictions of the species that were expected to occur at stream sites were generated and probable stressors to fish species that were predicted to occur but were absent were diagnosed. Predictions were generated based on the hierarchical screening method of Smith and Powell (1971, Am. Mus. Novit. 2458, 1–30), using fish abundance in conjunction with 25 environmental variables at 895 sites. The sites were sampled throughout Maryland and represent the entire range of environmental quality from severely degraded to minimally degraded. Stressor variable values that exceeded tolerance thresholds for species that were expected to occur, but were absent, were considered to be probable stressors. This method was tested for efficacy in stream site assessments and stressor diagnosis using an independent data set. Sites that were classified as degraded according to the IBI and to non-biological criteria had fewer predicted species present compared to minimally influenced sites, indicating that the proportion of predicted species present accurately represents the biological integrity of a stream site. The nine stressors that were applied to the test data set accounted for species absences in 43.7% of degraded sites. Impervious land cover was the most common stressor identified. In addition to assessing stream biological integrity and identifying stressors to fish species, this approach also provides tolerance thresholds for predicted fish species that are useful endpoints necessary to plan effective restoration of fish species in Maryland.     

人的失误理论研究进展

人的失误理论研究已进入结合认知心理学并以人的失误动态过程为研究热点的阶段。笔者回顾了人的失误理论研究的进展,讨论了相关的人的认知行为类型、认知失误的基本概念和人的失误模型。例如适用于不同情景和应用条件的几种模型:刺激-调制-响应(S-O-R)模型;失误的决策阶梯(Step-ladder)模型;通用GEMS模型以及Worledge认知模型。由于人的行为的复杂性和难以预测性造成了人的可靠性分析(HRA)的困难,因此,对人的行为的深入了解必须从人的行为特性及其规律性入手,将人的可靠性分析与行为科学理论结合起来,揭示人的失误发生的内在规律。与此同时,重点分析了概率安全评价技术(PSA)中如何对人的失误事件进行定量估计;如何考虑人的心理因素影响的几种重要的人的失误理论模型,并对今后这该领域中的研究方向进行了讨论。     

SOIL MOISTURE PREDICTION USING SUPPORT VECTOR MACHINES1

ABSTRACT: Herein, a recently developed methodology, Support Vector Machines (SVMs), is presented and applied to the challenge of soil moisture prediction. Support Vector Machines are derived from statistical learning theory and can be used to predict a quantity forward in time based on training that uses past data, hence providing a statistically sound approach to solving inverse problems. The principal strength of SVMs lies in the fact that they employ Structural Risk Minimization (SRM) instead of Empirical Risk Minimization (ERM). The SVMs formulate a quadratic optimization problem that ensures a global optimum, which makes them superior to traditional learning algorithms such as Artificial Neural Networks (ANNs). The resulting model is sparse and not characterized by the “curse of dimensionality.” Soil moisture distribution and variation is helpful in predicting and understanding various hydrologic processes, including weather changes, energy and moisture fluxes, drought, irrigation scheduling, and rainfall/runoff generation. Soil moisture and meteorological data are used to generate SVM predictions for four and seven days ahead. Predictions show good agreement with actual soil moisture measurements. Results from the SVM modeling are compared with predictions obtained from ANN models and show that SVM models performed better for soil moisture forecasting than ANN models.     

PSA application for a PCB detoxification plant

 Polychlorinated biphenyls (PCBs), a type of persistent organic pollutants (POPs), are considered to be endocrine disrupters. According to the Stockholm Convention on POPs, PCB detoxification is being accelerated globally. We have developed an environmentally sound chemical PCB detoxification plant using the ultraviolet ray/catalyst method. The purpose of this paper is to check the design methodology for the PCB detoxification plant by the application of probabilistic safety analysis. First, possible hazardous events were determined; second, the weakest points in these hazardous events were established; and third, the impact when failures arise at the weak points in the system were studied. We clarified the preventive measures as follows. To prevent PCBs leaking into the environment, select leak-tight valves for the piping containing PCBs, and reduce the number of valves. To prevent fire or explosions due to leaks of an inflammable mixture, select leak-tight valves, reduce the number of valves installed in the piping, and improve the reliability of the suppression tank in the isopropyl alcohol (IPA) recovery unit. Received: February 27, 2001 / Accepted: June 17, 2002     

Guild management: an evaluation of avian guilds as a predictive tool

The use and applicability of the guild concept to management is evaluated and questioned. Ecological problems are never as simple as implied in using one or two guild axes. A close examination of bird communities in a ponderosa pine forest reveals little relationship between guilds or guild blocks and the responses exhibited by individual bird species or bird species groups. Response guilds changed from year to year without any obvious changes in vegetation. A three-year composite analysis shows a clearer picture of the responses of ponderosa pine forest birds to the overall interactions between structure, weather, competition, and so on. The six response groups in the composite analysis are species that (1) were absent in 1973 on most or all study plots and showed no preference for any forested site; (2) had their highest densities on the medium cut and light cut plots; (3) were absent in 1973 on most or all study plots and had their highest densities on the medium cut and light cut plots; (4) had their highest densities on the untreated, light cut, and medium cut plots; (5) had their highest densities on the untreated and light cut plots and were either absent or had greatly reduced densities on all other plots; and (6) were present only on the clearcut, except for the Rock Wren which was also on the medium cut and heavy cut plots. The overall correlation between species density and guild density was significantly higher for response guilds ( <0.05) than for any of the structural or functional guilds. The whole concept of guild management needs much more research and development before it can be recommended as a management tool.