Assessing and testing prediction uncertainty for single tree-based models: A case study applied to northern hardwood stands in southern Québec, Canada

Estimating prediction uncertainty for a single tree-based model is hindered by the complex structure of these models. In this paper, we addressed this issue with a case study applied to northern hardwood stands in Québec, Canada. SaMARE is a stochastic single tree-based model that was designed for these types of stands. Using a Monte Carlo approach, the model can provide a mean predicted value and its confidence limits for some plot-level attributes.The mean predicted values were compared to observed values in terms of bias and accuracy. In addition to these common statistics, we compared nominal coverage of Monte Carlo-simulated confidence intervals with real (observed) coverage to verify the adequacy of the simulated uncertainty. A comparison was made using several plot-level attributes, which exhibited an increasing discriminative complexity. This complexity ranges from coarse attributes, such as all-species basal area, up to more complex ones, such as basal area for stems of a particular species and with sawlog potential.The results showed that in terms of absolute value, biases were small, but could be relatively high with respect to the average observed value when the discriminative complexity of the attribute increased. The comparison between nominal and real coverage of confidence intervals gave satisfactory results for all-species plot-level attributes. However, for some species-specific attributes, the Monte Carlo-simulated confidence intervals overestimated the real coverage.     

煤矿顶板冒落预报的分形研究

应用分形理论，研究了煤矿顶板运动过程中所表现的分形特征，提出了用顶板下沉速度的分数维值的变化预报顶板来压、冒落的方法，使得预报顶板来压、冒落的指标定量化成为可能，预报更加科学，把预报的理论与方法推进到新阶段。这对保障煤矿生产安全具有重要的应用价值。     

黑道凶日研究

本文提出的优势期和优势值是灾害年内出现机会最多的日期和强度，其中优势期就是黑道凶日。统计表明，黑道凶日位于节气日和节气之间的黄金分割日及其附近，在时空分布上具有许多特点。     

煤体注水湿润半径预测的神经网络模型

合理的注水半径一直是煤体注水防尘技术中难以确定的参数。笔者基于对影响煤体注水半径因素的分析和神经网络理论的原理之上 ,设计网络模型为 3层 ,输入层为 7个节点 ,应用BP网络算法 ,建立了煤体注水湿润半径的预测模型 ,并对其参数进行了讨论。然后 ,用平顶山矿务局和水城矿务局 13个矿 19个回采工作面的统计资料对BP网络进行自适应学习 ,并取η =0 .9,α =0 .82 ,控制网络总误差E≤ 10 6。经过 2 12 34次迭代后 ,网络趋于稳定。用训练好的网络对平顶山矿务局的某矿的 3层煤的注水湿润半径进行预测 ,预测结果与实测值很接近。其误差分别为 0 .5 %,0 .6 %和 0 .7%。     

杨沸火灾监测预报方法研究（I）－理论基础

本文针对在扬沸前期征兆中表现出显著燃烧微爆噪音,提出了通过监测扬沸前兆噪音、联合温度信息,对扬沸火灾进行诊断、预报的方法。本文的第一部分主要讨论了该方法的理论基础,阐明了其基本原理和技术途径。     

Validation of the propensity for angry driving scale

Problem: This study examined the validity of the Propensity for Angry Driving Scale (PADS; DePasquale, J. P., Geller, E. S., Clarke, S. W., and Littleton, L. C. (2001). Measuring road rage: Development of the Propensity for Angry Driving Scale. Journal of Safety Research, 32, 1–16) in predicting aggressive driving. Method: The PADS and the Driving Anger Scale (DAS; Deffenbacher, J. L., Oetting, E. R., and Lynch, R. S. (1994). Development of a driving anger scale. Psychological reports, 74, 83–91.) were administered to 232 college student volunteers with measures of aggressive and risky driving. Results: Convergent and discriminant validity of the PADS were supported through relationships among measures of similar constructs. The PADS significantly (p<.05) predicted moving tickets, minor accidents, aggressive driving, risky driving, and maladaptive driving anger expression, above and beyond gender, miles driven per week, and trait anger. Discussion: Findings suggest that the PADS is a useful predictor of aggressive driving and has some advantages over the DAS. Impact on Industry: The PADS is an effective predictor of aggressive driving that complements established measures like the DAS and provides researchers with another valuable tool for the assessment of aggressive driving.     

改进BP算法在煤与瓦斯突出预测中的应用

为了正确预测煤与瓦斯突出的趋势与危险性 ,基于反向BP神经网络 ,笔者提出了一种改进的BP网络模型 :为了加快BP网络的收敛速度 ,增强其跳出局部极小点的能力 ,采用了自适应变步长法和改进模拟退火法 (SA法 )相结合的方法。实际应用表明 ,该模型收敛速度快 ,准确性高 ,具有较高的可靠性和实用性 ,是一种十分有效的煤与瓦斯突出危险性预测方法。     

Evaluation of leachate composition by multivariate data analysis (MVDA)

Landfills generate emissions in the form of gas and leachate. The emissions are often measured within monitoring programmes. It is likely that the requirements of such monitoring programmes can be extended in the future, particularly in light of the increased interest in specific organic substances. Multivariate data analyses (MVDA) have been used to evaluate the possibility of predicting the content of specific organic substances from more common analyses. The results indicate that this is possible for a specific leachate. MVDA can also be used to reduce the number of analyses performed within existing monitoring programmes while retaining information about all the variables formerly included in the programmes.     

Prediction of effluent concentration in a wastewater treatment plant using machine learning models

Of growing amount of food waste, the integrated food waste and waste water treatment was regarded as one of the efficient modeling method. However, the load of food waste to the conventional waste treatment process might lead to the high concentration of total nitrogen(T-N) impact on the effluent water quality. The objective of this study is to establish two machine learning models—artificial neural networks(ANNs) and support vector machines(SVMs), in order to predict 1-day interval T-N concentration of effluent from a wastewater treatment plant in Ulsan, Korea. Daily water quality data and meteorological data were used and the performance of both models was evaluated in terms of the coefficient of determination(R~2), Nash–Sutcliff efficiency(NSE), relative efficiency criteria(d rel). Additionally, Latin-Hypercube one-factor-at-a-time(LH-OAT) and a pattern search algorithm were applied to sensitivity analysis and model parameter optimization, respectively. Results showed that both models could be effectively applied to the 1-day interval prediction of T-N concentration of effluent. SVM model showed a higher prediction accuracy in the training stage and similar result in the validation stage.However, the sensitivity analysis demonstrated that the ANN model was a superior model for 1-day interval T-N concentration prediction in terms of the cause-and-effect relationship between T-N concentration and modeling input values to integrated food waste and waste water treatment. This study suggested the efficient and robust nonlinear time-series modeling method for an early prediction of the water quality of integrated food waste and waste water treatment process.     

Predicting geogenic groundwater fluoride contamination throughout China

Endemic fluorosis exists in almost all provinces of China. The long-term ingestion of groundwater containing high concentrations of fluoride is one of the main causes of fluorosis. We used artificial neural network to model the relationship between groundwater fluoride concentrations from throughout China and environmental variables such as climatic, geological. and soil parameters as proxy predictors. The results show that the accuracy and area under the receiver operating characteristic curve of the model in the test dataset are 80.5% and 0.86%, respectively, and climatic variables are the most effective predictors. Based on the artificial neural network model, a nationwide prediction risk map of fluoride concentrations exceeding 1.5 mg/L with a 0.5 × 0.5 arc minutes resolution was generated. The high risk areas are mainly located in western provinces of Xinjiang, Tibet, Qinghai, and Sichuan, and the northern provinces of Inner Mongolia, Hebei and Shandong. The total number of people estimated to be potentially at risk of fluorosis due to the use of untreated high fluoride groundwater as drinking water is about 89 million, or 6% of the population. The high fluoride groundwater risk map helps the authorities to prioritize areas requiring mitigation measures and thus facilitates the implementation of water improvement and defluoridation projects.