Evaluation on sampling point densities for assessing indoor air quality

In Hong Kong, the Environmental Protection Department (HKEPD) has launched an indoor air quality (IAQ) certification scheme to promote an acceptable IAQ in workplaces. However, the associated uncertainties and measurement efforts have not been detailed for practical measurement in indoor spaces. In this study, indoor carbon dioxide (CO₂) concentration is selected as an indicator of the IAQ to investigate the probable errors and measurement efforts in different sampling schemes regarding the sampling point density. In particular, a one-year measurement for sample-spatial average indoor CO₂ concentration at 17 sampling locations in a typical large office (floor area=1200 m²) has been used to evaluate the probable errors of the sample-spatial average concentrations using different sampling point densities. The result shows that the measured concentration at a single sampling point could not be representative for the space and more than one sampling locations would be required in order to increase the measurement accuracy. Mathematical expressions for the sample-spatial average concentration at a confidence level at certain sampling point densities are proposed. When doubled the required measurement points, it was found that the probability of obtaining a measured CO₂ concentration at the confidence level of 95% could be increased from 70% to 90%, as compared with the current sampling practice. It is recommended to specify the measurement uncertainties in future codes so that effort for IAQ measurements in indoor spaces could be determined for practical strategies.     

Perceived indoor air quality and its relationship to air pollutants in French dwellings

Perception of indoor air quality (PIAQ) was evaluated in a nationwide survey of 567 French dwellings, and this survey was combined with measurements of gaseous and particulate matter (PM₁₀ and PM_2.5) indoor air pollutants and indoor climate parameters. The perception was assessed on a nine‐grade scale by both the occupants of the dwellings and the inspectors who performed the measurements. The occupants perceived the air quality in their homes as more pleasant than the inspectors. The inspectors perceived the air quality as more unpleasant in dwellings in which the residents smoked indoors. Significant associations between PIAQ and indoor air pollutant concentrations were observed for both the inspectors and, to a lesser extent, the occupants. Introducing confounding parameters, such as building and personal characteristics, into a multivariate model suppressed most of the observed bivariate correlations and identified the tenure status of the occupants and their occupation as the parameters that most influenced their PIAQ. For the inspectors, perceived air quality was affected by the presence of smokers, the season, the type of ventilation, retrofitting, and the concentrations of acetaldehyde and acrolein.     

室内空气品质预测新方法

分析了常用的室内空气品质预测方法中存在的问题,根据室内空气品质不满意率和化学污染物对数评价指标间的线性关系,在室内污染物浓度场预测的基础上提出了一种预测室内空气品质的新方法。该方法简单可行,且能综合体现主、客观评价结果。     

A model for predicting failure of oil pipelines

Oil and gas pipelines transport millions of dollars of goods everyday worldwide. Even though they are the safest way to transport petroleum products, pipelines do still fail generating hazardous consequences and irreparable environmental damages. Many models have been developed in the last decade to predict pipeline failures and conditions. However, most of these models were limited to one failure type, such as corrosion failure, or relied mainly on expert opinion analysis. The objective of this paper is to develop a model that predicts the failure cause of oil pipelines based on factors other than corrosion. Two models are developed to help decision makers predict failure occurrence. Regression analysis and artificial neural networks (ANNs) models were developed based on historical data of pipeline accidents. The two models were able to satisfactory predict pipeline failures due to mechanical, operational, corrosion, third party and natural hazards with an average validity of 90% for the regression model and 92% for the ANN model. The developed models assist decision makers and pipeline operators to predict the expected failure cause(s) and to take the necessary actions to avoid them.     

通风房间内污染物浓度的预测

对 3种预测空调或自然通风房间室内外气相和颗粒相有害物浓度间演化关系的典型模式进行了分析 ,由于各种方法建立的基础和条件不同 ,讨论并指出它们的适用范围和应用中可能出现的问题 ,提出改进意见     

Evaluation on four sampling schemes for assessing indoor air quality

Continuous sampling is one of the common approaches for assessing indoor pollutant level. It is believed that the longer the measurement time, the higher the accuracy and confidence level of the measurement can be achieved. In 2003, the Hong Kong Environmental Protection Department (HKEPD) launched an Indoor Air Quality (IAQ) certification scheme to promote an acceptable IAQ in workplaces. However, measurement efforts and uncertainties associated with the sampling method have not been addressed. Alternative sampling schemes taking shorter measurements in the sampling period were proposed in some circumstances. In this study, the average carbon dioxide (CO₂) concentration of a workplace is selected as an indicator of the indoor air quality to investigate the probable errors and measurement efforts in four sampling schemes, regarding the sampling period: in Scheme A, it is from a continuous sampling throughout the measurement; in Scheme B, it is from two sampling periods of two equal sessions of the measurement; in Scheme C, it is from two structural sampling periods of the two sessions; and in Scheme D, the average concentration is from four sampling periods in four equal sessions of the measurement. In particular, a year-round indoor CO₂ concentration at 17 locations in a typical office in Hong Kong was used to evaluate the probable errors using these four sampling schemes. At certain confidence levels, the required measurement times of the alternative sampling schemes (Schemes B, C, D) were evaluated and compared with that of an 8-hour continuous one (Scheme A). It was found that Scheme C would offer a reduction of measurement effort up to 30%. It is recommended to specify the uncertainties and efforts of measurement in future codes, and to consider these sampling schemes in determining practical strategies for IAQ measurement.     

Energy saving potential and repercussions on indoor air quality of demand controlled residential ventilation strategies

Ventilation is ambiguously related to the energy saving rationale originating from the mitigation of global warming, the reaching of peak oil or health concerns related to fossil fuel burning. Since it makes up for about half of the energy consumption in well-insulated buildings, it is an attractive target for energy saving measures. However, simply reducing ventilation rates has unwanted repercussions on the indoor air quality. Two main strategies have been developed to reconcile these seemingly opposing interests: heat recovery and demand control ventilation. This paper focuses on the energy saving potential of demand controlled mechanical exhaust ventilation in residences and on the influence such systems may have on the indoor air quality to which the occupants of the dwellings are exposed. The conclusions are based on simulations done with a multi-zone airflow model of a detached house that is statistically representative for the average Belgian dwelling. Four approaches to demand based control are tested and reported. Within the paper exposure to carbon dioxide and to a tracer gas are used as indicators for indoor air quality. Both energy demand and exposures are reported and compared to the results for a standard, building code compliant, exhaust system, operating at continuous flow rates. The sensitivity of the control strategies to environmental and user variations is tested using Monte-Carlo techniques. Under the conditions that were applied, reductions on the ventilation heat loss of 25–60% are found, depending on the chosen control strategy (with the exclusion of adventitious ventilation and infiltration).     

Comparison of indoor air quality in electrified and un-electrified dwellings in rural South African villages

A feasibility study was undertaken to assess the suitability of South African rural villages due to be electrified, for the purposes of undertaking a large-scale study of the impact of reductions in indoor air pollution on acute lower respiratory infections. As part of the feasibility study, quantitative assessments of indoor air pollution in non-electrified and electrified dwellings were performed. Concurrent measurements were made of levels of respirable particulate matter (RSP-stationary), and carbon monoxide (CO) (personal on children <18 months), as well as a stationary co-located with RSP) over a 24-h period in 52 un-electrified and 53 electrified dwellings. The proportion of dwellings with a detectable 24-h concentration of RSP was significantly higher in un-electrified (48.1%) than electrified dwellings (24.5%) (chi(2) = 6.30 on 1 d.f., P = 0.012). In addition a Kruskal-Wallis test (adjusted for ties) showed that the distribution of RSP differed between un-electrified and electrified areas (Kruskal-Wallis chi(2) = 8.20 on 1 d.f., P = 0.014). In those dwellings where some RSP was detected, the amount was on average higher in the un-electrified areas (mean 162 microg/m(3), median 107 microg/m(3)) than in the electrified areas (mean 77 microg/m(3), median 37.5 microg/m(3)). Stationary (kitchen CO) levels in un-electrified dwellings ranged from 0.36 to 20.95 p.p.m. However, in electrified dwellings, kitchen levels ranged from 0 to 11.8 p.p.m. When mean concentrations of CO were compared between electrified and un-electrified dwellings using a two-sample t-test (on log-transformed data), there was overwhelming evidence (P = 0.0004) that the mean level of log (CO) in the kitchen was higher in the un-electrified areas (1.25 vs. 0.69) and also overwhelming evidence (P < 0.0001) that the mean level of log (CO) on the child was higher in the un-electrified areas (0.83 vs. 0.34). Of importance in terms of both policy and for a potential future large-scale study, is that measurable significant differences in indoor pollutants between electrified and un-electrified dwellings during summer were found in spite of only partial transition to electricity use for cooking in electrified villages. PRACTICAL IMPLICATIONS: It is estimated that at least two-thirds of all households in the developing world are still primary dependent on biomass fuels and coal. This situation applies to 59% of rural households in South Africa. In the last decade a program of providing electricity to three million homes has been underway in South Africa. Among others this intervention aims to reduce exposure to pollutants from burning biomass fuels and reduce detrimental health effects, especially in young children. This study provides scientific evidence that electrified homes in South African villages have lower levels of air pollution (RSP and CO) relative to their non-electrified counterparts.     

办公楼室内环境品质控制与优化

室内办公环境品质直接影响办公室中人员的工作效率和健康,需要进行有效地控制和优化,以保证其舒适性、低能耗和健康的要求。以某办公室为对象,研究了室内环境品质各参数和控制量之间的关系,建立了室内环境品质双线性模型。利用实验数据进行了模型验证,结果表明:模型输出能较好拟合实际情况;并基于模型预测方法进行室内环境品质控制,通过夏季工况仿真实验证明了提出方法比传统控制方式具有更小的稳态误差和较小的超调性,且更加节能。     

Presentation of machine learning methods to determine the most important factors affecting road traffic accidents on rural roads

The purpose of this research was to develop statistical and intelligent models for predicting the severity of road traffic accidents (RTAs) on rural roads. Multiple Logistic Regression (MLR) was used to predict the likelihood of RTAs. For more accurate prediction, Multi-Layer Perceptron (MLP) and Radius Basis Function (RBF) neural networks were applied. Results indicated that in MLR, the model obtained from the backward method with the correct percent of 84.7% and R² value of 0.893 was the best method for predicting the likelihood of RTAs. Also, MLR showed that the variables of not paying attention to the front not paying attention to the frontroad ahead, followed byand then vehicle-motorcycle/bike accidents were the greatest problems. Among the models, MLP had a better performance, so that the prediction accuracy of MLR, MLP, and RBF were 84.7%, 96.7%, and 92.1%, respectively. MLP model, due to higher accuracy, showed that the variable of reason of accident had the highest effect on the prediction of accidents, and considering MLR results, the variables of not paying attention to the front and then vehicle-motorcycle/bike accidents had the most influence on the occurrence of accidents. Therefore, motorcyclists and cyclists are more prone to accidents, and appropriate solutions should be adopted to enhance their safety.     

Perceptions in the U.S. building industry of the benefits and costs of improving indoor air quality

How building stakeholders (e.g. owners, tenants, operators, and designers) understand impacts of Indoor Air Quality (IAQ) and associated energy costs is unknown. We surveyed 112 stakeholders across the United States to ascertain their perceptions of their current IAQ and estimates of benefits and costs of, as well as willingness to pay for, IAQ improvements. Respondents' perceived IAQ scores correlated with the use of high‐efficiency filters but not with any other IAQ‐improving technologies. We elicited their estimates of the impacts of a ventilation–filtration upgrade (VFU), that is, doubling the ventilation rate from 20 to 40 cfm/person (9.5 to 19 l/s/person) and upgrading from a minimum efficiency reporting value 6 to 11 filter, and compared responses to estimates derived from IAQ literature and energy modeling. Minorities of respondents thought the VFU would positively impact productivity (45%), absenteeism (23%), or health (39%). Respondents' annual VFU cost estimates (mean = $257, s.d. = $496, median = $75 per person) were much higher than ours (always <$32 per person), and the only yearly cost a plurality of respondents said they would pay for the VFU was $15 per person. Respondents holding green building credentials were not more likely to affirm the IAQ benefits of the VFU and were less likely to be willing to pay for it.     

岩体变形模量偏最小二乘回归与神经网络关联性研究

偏最小二乘回归不直接考虑因变量与自变量回归问题，而直接提取与系统有关的新的综合变量，并能利用交叉原理确定成分个数，尤其在分析单因变量与多自变量间关系上。其所得结果更为满意；人工神经网络具有自适应学习和记忆能力，尤其是3层BP网络模型。把这两者相关联。以岩体变形模量为因变量，岩石比重等7个因素为自变量，分析研究了焦作—晋城高速公路试验资料，所得拟合值与实测值误差最大为5．43％，较偏最小二乘法回归分析19．07％的误差好了许多。该方法为参数选取找到了一条新的途径。     

Approximately predicting the cost and duration of school reconstruction projects in Taiwan

Regression and neural network models have been developed to predict the cost and duration of projects for the reconstruction of schools which must be quickly rebuilt. Data for the school reconstruction projects in central Taiwan, which received the most serious damage from the Chi‐Chi Earthquake, were collected and analysed. The analytical results demonstrate that the floor area provides a good basis for estimating the cost and duration of school reconstruction projects, and suggest that the neural network model with back‐propagation learning technique is a feasible approach that yields better prediction results than the regression model for school reconstruction projects.     

基于偏最小二乘回归的半参数模型在大坝安全监测系统中的应用研究

半参数模型能有效地分离系统误差,故受到测绘学界的广泛重视。针对大坝安全监测系统中的多元回归、逐步回归和偏最小二乘回归三种常用统计模型的不足,引入半参数模型,提出了基于偏最小二乘回归的半参数模型,并通过实例验证该方法具有良好的适用性。     

Prediction of the crack condition of highway pavements using machine learning models

Departments of Transportation regularly evaluate the condition of pavements through visual inspections, nondestructive evaluations, image recognition models and learning algorithms. The above methodologies, though efficient, have drawn attention due to their subjective errors, uncertainties, noise effects and overfitting. To improve on the outcomes of the shallow learning models already used in pavement crack prediction, this paper reports on an investigation of the use of recursive partitioning and artificial neural networks (ANN; deep learning frameworks) in predicting the crack rating of pavements. Explanatory variables such as the average daily traffic and truck factor, roadway functional class, asphalt thickness, and pavement condition time series data are employed in the model formulation. Overall, it is observed that the recursive partitioning (regression tree – R² > 0.8 and classification tree – R² > 0.6) and ANN (continuous response – R² > 0.8 and categorical response – R² > 0.6) are compelling machine learning models for the prediction of the crack ratings based on their goodness-of-fit statistics, mean absolute deviation (MAD < 0.4) and the root mean square errors (RMSE between 0.30 and 0.65).     

Using extreme learning machines for short-term urban water demand forecasting

This study explores the ability of various machine learning methods to improve the accuracy of urban water demand forecasting for the city of Montreal (Canada). Artificial Neural Network (ANN), Support Vector Regression (SVR) and Extreme Learning Machine (ELM) models, in addition to a traditional model (Multiple linear regression, MLR) were developed to forecast urban water demand at lead times of 1 and 3 days. The use of models based on ELM in water demand forecasting has not previously been explored in much detail. Models were based on different combinations of the main input variables (e.g., daily maximum temperature, daily total precipitation and daily water demand), for which data were available for Montreal, Canada between 1999 and 2010. Based on the squared coefficient of determination, the root mean square error and an examination of the residuals, ELM models provided greater accuracy than MLR, ANN or SVR models in forecasting Montreal urban water demand for 1 day and 3 days ahead, and can be considered a promising method for short-term urban water demand forecasting.     

Application of surrogate artificial intelligent models for real‐time flood routing

Developing a robust flood forecasting and warning system (FFWS) is essential in flood‐prone areas. Hydrodynamic models, which are a major part of such systems, usually suffer from computational instabilities and long runtime problems, which are particularly important in real‐time applications. In this study, two artificial intelligence models, namely artificial neural network (ANN) and adaptive neuro‐fuzzy inference system (ANFIS), were used for flood routing in an FFWS in Madarsoo river basin, Iran. For this purpose, different rainfall patterns were transformed to run‐off hydrographs using the Hydrologic Engineering Center (HEC)‐1 hydrological model and routed along the river using HEC river analysis system RAS hydrodynamic model. Then, the simulated hydrographs with different lag times were used as inputs for training of ANN and ANFIS models to simulate flood hydrograph at the basin outlet. Results showed that the simulations obtained from ANN and ANFIS coincided with the results simulated by the HEC‐RAS, and application of such models is strongly suggested as a backup tool for flood routing in FFWSs.     

Development of an England-wide indoor overheating and air pollution model using artificial neural networks

With the UK climate projected to warm in future decades, there is an increased research focus on the risks of indoor overheating. Energy-efficient building adaptations may modify a buildings risk of overheating and the infiltration of air pollution from outdoor sources. This paper presents the development of a national model of indoor overheating and air pollution, capable of modelling the existing and future building stocks, along with changes to the climate, outdoor air pollution levels, and occupant behaviour. The model presented is based on a large number of EnergyPlus simulations run in parallel. A metamodelling approach is used to create a model that estimates the indoor overheating and air pollution risks for the English housing stock. The performance of neural networks (NNs) is compared to a support vector regression (SVR) algorithm when forming the metamodel. NNs are shown to give almost a 50% better overall performance than SVR.     

我国目前室内空气品质改善的对策与措施

强调改善室内空气品质应该有一种紧迫感，不应纠缠一些复杂的理论问题；当务之急是尽快建立起我国的室内空气品质标准和评价方法；尽管将成熟有效的措施和方法应用于实际。从这三方面介绍了应采取的对策和措施。