Development and Operational Testing of a Super‐Ensemble Artificial Intelligence Flood‐Forecast Model for a Pacific Northwest River

Coastal catchments in British Columbia, Canada, experience a complex mixture of rainfall‐ and snowmelt‐driven contributions to flood events. Few operational flood‐forecast models are available in the region. Here, we integrated a number of proven technologies in a novel way to produce a super‐ensemble forecast system for the Englishman River, a flood‐prone stream on Vancouver Island. This three‐day‐ahead modeling system utilizes up to 42 numerical weather prediction model outputs from the North American Ensemble Forecast System, combined with six artificial neural network‐based streamflow models representing various slightly different system conceptualizations, all of which were trained exclusively on historical high‐flow data. As such, the system combines relatively low model development times and costs with the generation of fully probabilistic forecasts reflecting uncertainty in the simulation of both atmospheric and terrestrial hydrologic dynamics. Results from operational testing by British Columbia's flood forecasting agency during the 2013‐2014 storm season suggest that the prediction system is operationally useful and robust.     

Kinetics characteristics of coal low-temperature oxidation in oxygen-depleted air

The longwall gob (mined-out) area is one of the main places that are prone to coal spontaneous combustion and most of the residual coal in it is in oxygen-depleted air as it is a semi-enclosed space. A DSC (Differential scanning calorimetry) test system was designed to accurately test the heat generation of coal oxidation in different oxygen concentration atmosphere, based on which the kinetics parameters (activation energy and pre-exponential factor) of coal low-temperature oxidation in oxygen-depleted air were solved out. The results show that the kinetics parameters present obvious stage features and the lower the oxygen concentration is, the smaller is the difference of the kinetics parameters that in different oxidation stages. When the oxygen concentration is lower than 5% and 3% for jet coal and meagre coal respectively, the kinetics parameters of slow oxidation start to be greater than that of rapid oxidation. Both in the slow oxidation and rapid oxidation stage, with the decrease of oxygen concentration, kinetics parameters present significant decline on the whole while in different oxygen concentration range, the decline rate is different. It's concluded that when assessing the residual coal's self-heating risk, we need to use the corresponding kinetics parameters of coal oxidation in the oxygen concentration of the location where the residual coal is and the safety factor will be greater to only use the kinetics parameters of coal oxidation in slow oxidation stage. This study is of great significance for the assessment and control of the self-heating risk of coal that in different oxygen concentration atmospheres of the longwall gob areas.     

Modelling and Prediction of Diesel Engine Performance using Relevance Vector Machine

Diesel engines are being increasingly adopted by many car manufacturers today, yet no exact mathematical diesel engine model exists due to its highly nonlinear nature. In the current literature, black-box identification has been widely used for diesel engine modelling and many artificial neural network (ANN) based models have been developed. However, ANN has many drawbacks such as multiple local minima, user burden on selection of optimal network structure, large training data size, and over-fitting risk. To overcome these drawbacks, this article proposes to apply an emerging machine learning technique, relevance vector machine (RVM), to model and predict the diesel engine performance. The property of global optimal solution of RVM allows the model to be trained using only a few experimental data sets. In this study, the inputs of the model are engine speed, load, and cooling water temperature, while the output parameters are the brake-specific fuel consumption and the amount of exhaust emissions like nitrogen oxides and carbon dioxide. Experimental results show that the model accuracy is satisfactory even the training data is scarce. Moreover, the model accuracy is compared with that using typical ANN. Evaluation results also show that RVM is superior to typical ANN approach.     

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.     

Serious BTEX pollution in rural area of the North China Plain during winter season

Atmospheric BTEX compounds(benzene, toluene, ethylbenzene and xylenes) in a rural site of the North China Plain(NCP) were preliminarily investigated in winter, and the outdoor concentrations(25.8–236.0 μg/m3) were found to be much higher than those reported in urban regions. The pollution of BTEX inside a farmer's house was even more serious, with combined concentrations of 254.5–1552.9 μg/m3. Based on the ratio of benzene to toluene(1.17 ± 0.34) measured, the serious BTEX pollution in the rural site was mainly ascribed to domestic coal combustion for heating during the winter season. With the enhancement of farmers' incomes in recent years, coal consumption by farmers in the NCP is rapidly increasing to keep their houses warm, and hence the serious air pollution in rural areas of the NCP during winter, including BTEX, should be paid great attention.     

不同有机废水厌氧产酸过程的研究

主要研究了不同类废弃有机物厌氧产酸过程中产生的挥发性脂肪酸的产量和组成比例。研究发现,废糖蜜溶液、蛋白废液和淀粉废液在达到最高产酸量时,超过50%的VFAs是乙酸,其次是丙酸,还含有少量的丁酸和戊酸。废甘油溶液在达到最高产酸量时,产生的VFAs的主要成分是丙酸。4种有机废液中,相同初始COD浓度条件下,淀粉废液产生的VFAs产量最多,初始COD为5000mg/L时最高产量为2.321g/L,其次是废甘油溶液,最高产量是2.244g/L。     

有机化学品不同温度下(过冷)液体蒸气压预测模型的建立与评价

(过冷)液体蒸气压(PL)是评价化学品在环境中分配、迁移和归趋行为的重要参数。PL具有较强的温度依附性。发展一种能够精确预测不同环境温度下化学品PL的方法,有助于填补化学品生态风险评估的大量数据缺失。本研究收集整理了661种有机化合物在不同温度下(200~830 K)共计10 478个log PL值。在此基础上,采用偏最小二乘(PLS)回归和支持向量机(SVM)方法,构建了PL的线性和非线性预测模型。结果表明:2种模型均具有良好的拟合度、稳健性及预测能力,SVM模型的预测性能略高于PLS模型(PLS:R2adj.tra=0.912,RMSEtra=0.477,Q2ext=0.910;SVM:R2adj.tra=0.997,RMSEtra=0.092,Q2ext=0.967)。机理分析表明,温度是影响PL的主要因素,温度越高,蒸气压越大;其次,X1sol也影响PL大小,X1sol用来描述分子间的色散作用,分子间色散力越小,蒸气压越大;此外,化合物的氢键个数、极性和分子构型等因素也影响PL大小。采用Wiliams plot方法表征了PLS模型应用域。所建立的模型可用来预测烷烃、烯烃、醇、酮、羧酸、苯、酚、联苯、卤代芳香烃、含N化合物及含S化合物在不同温度下的PL数据。     

Study on acceleration and suppression properties of coal gangue with different void fractions on gas explosion propagation

Gas explosion is one of the main disasters in coal mining. Plenty of coal gangue are generally distributed in the disaster areas in gob. Experiments were carried out to explore the propagation law of the gas explosion distributed by coal gangue. The variation characteristics of the overpressure, pressure rise rate, and flame shape with void fractions were analyzed. The results showed that the effect of the coal gangue on the explosion intensity changed from suppression to acceleration with the increase of void fraction, the flame front upstream blockage area changed from laminar state to turbulent divergent state, and a reverse flame was formed. When the void fraction of the coal gangue was 0.50–0.65, the maximum overpressure downstream of the blocked area were positively correlated with the void fraction and the critical suppression range was between 0.50 and 0.55. When the void fraction was lower than 0.50, the flame was quenched in the coal gangue, neither the flame nor the pressure could pass through the blocked area. It is helpful to guide the improvement of coal recovery process to avoid the expansion of the explosion impact in coalmine gob.     

基于无人机影像的城市植被精细分类

针对传统植被资源调查方法工作量大、成本高、效率低的问题,利用高分辨率无人机遥感影像,联合地物光谱-纹理-空间信息,构建了一种适用于描述城市不同植被种类的多维特征空间,在此基础上对三种应用广泛的分类算法(基于像素的、面向对象的支持向量机及深度学习Mobile-Unet语义分割模型)开展了对比分析研究.结果表明:本文提出的联合地物光谱-纹理-空间信息的特征空间构建方法能够有效地描述城市不同类型植被的特征差异,提升影像分割、植被分类的精度;在分类精度上,基于像素和面向对象的支持向量机分类结果的总体精度均超过90%,深度学习方法的总体分类精度为84%;在算法效率上,传统机器学习方法也优于深度学习方法.因此,得出结论针对城市小区域、小样本的植被精细分类,传统机器学习分类方法比深度学习方法效果更好.     

碳中和趋势下数学模拟在污水处理系统中的发展与综合应用

数学模拟技术在污水处理方面被广泛应用,为了系统总结相关技术,本文回顾了污水处理系统中数学模拟技术的发展历程;综述了活性污泥模型(ASM)与机器学习(ML)在水质预测及参数工况优化领域中的应用;重点探究了污水处理系统中温室气体排放模型,以及多目标优化模型在污水处理系统中温室气体排放(GHG)、出水质量(EQI)和运行成本(OCI)的权衡问题;归纳了数学模拟技术在实现污水厂能量自给与资源回收的应用发展.研究结果表明数学模拟技术能准确预测出水水质、快速优化工艺参数、权衡温室气体排放、出水水质与运行成本之间的关系、以及提高资源回收效率等.因此,数值模拟技术可有效指导污水处理工艺的运行优化以及管理,为污水处理行业减污降碳协同增效提供技术支撑.