Deep learning architecture for air quality predictions

With the rapid development of urbanization and industrialization, many developing countries are suffering from heavy air pollution. Governments and citizens have expressed increasing concern regarding air pollution because it affects human health and sustainable development worldwide. Current air quality prediction methods mainly use shallow models; however, these methods produce unsatisfactory results, which inspired us to investigate methods of predicting air quality based on deep architecture models. In this paper, a novel spatiotemporal deep learning (STDL)-based air quality prediction method that inherently considers spatial and temporal correlations is proposed. A stacked autoencoder (SAE) model is used to extract inherent air quality features, and it is trained in a greedy layer-wise manner. Compared with traditional time series prediction models, our model can predict the air quality of all stations simultaneously and shows the temporal stability in all seasons. Moreover, a comparison with the spatiotemporal artificial neural network (STANN), auto regression moving average (ARMA), and support vector regression (SVR) models demonstrates that the proposed method of performing air quality predictions has a superior performance.     

Pollination ecology of <Emphasis Type="Italic">Eomecon chionantha</Emphasis> Hance (Papaveraceae), an endemic species in China

Eomecon chionantha Hance, the only species in the genus Eomecon, is an angiosperm species endemic to China. Here we investigated the pollination syndrome, pollinator generalization of E. chionantha. Our results showed that the individual flower longevity of E. chionantha was 3–5 days. The flowering phenology of E. chionantha can be divided into five phases. We found significant differences in the pollination efficiency of the different insects visiting E. chionantha with bees and flies being the most efficient pollinators. This, in addition to floral morphology and phenology, suggests that E. chionantha is generalized on bees and flies and possesses an outcrossing, partially self-compatible and pollinator-dependent breeding system.     

洞庭湖区血吸虫病疫水人水相互作用关系及防控方案研究

研究洞庭湖区血吸虫病人水相互作用关系及传染防控,是推进湖区社会经济稳定健康发展的重要保障。从水文情势、产业发展、居民行为、城乡建设四方面探讨了洞庭湖区血吸虫病疫水人水相互作用关系,并运用系统动力学模型,设置教育优先方案、生态保护优先方案和城镇化推进方案对其防控方案进行系统仿真模拟和对比分析,研究结果表明:(1)水文等生态因素对于人水接触的影响逐步减少,而社会经济因素则显得越来越重要。(2)系统动力学模型能有效展示血吸虫病各影响因素之间的关系,是研究血吸虫病防控的有效方法之一。(3)依据仿真模拟结果,城镇化推进方案是经济增长、城乡协调、环境友好的最优血防方案。     

A highly resolved temporal and spatial air pollutant emission inventory for the Pearl River Delta region, China and its uncertainty assessment

A highly resolved temporal and spatial Pearl River Delta (PRD) regional emission inventory for the year 2006 was developed with the use of best available domestic emission factors and activity data. The inventory covers major emission sources in the region and a bottom–up approach was adopted to compile the inventory for those sources where possible. The results show that the estimates for SO₂, NO_x, CO, PM₁₀, PM_2.5 and VOC emissions in the PRD region for the year 2006 are 711.4 kt, 891.9 kt, 3840.6 kt, 418.4 kt, 204.6 kt, and 1180.1 kt, respectively. About 91.4% of SO₂ emissions were from power plant and industrial sources, and 87.2% of NO_x emissions were from power plant and mobile sources. The industrial, mobile and power plant sources are major contributors to PM₁₀ and PM_2.5 emissions, accounting for 97.7% of the total PM₁₀ and 97.2% of PM_2.5 emissions, respectively. Mobile, biogenic and VOC product-related sources are responsible for 90.5% of the total VOC emissions. The emissions are spatially allocated onto grid cells with a resolution of 3 km × 3 km, showing that anthropogenic air pollutant emissions are mainly distributed over PRD central-southern city cluster areas. The preliminary temporal profiles were established for the power plant, industrial and on-road mobile sources. There is relatively low uncertainty in SO₂ emission estimates with a range of −16% to +21% from power plant sources, medium to high uncertainty for the NO_x emissions, and high uncertainties in the VOC, PM_2.5, PM₁₀ and CO emissions.     

降低FCC再生烟气NOx排放助剂的实验室评价

通过采用ACE装置与烟气NOx分析仪器联用的实验室评价方法,可在更接近实际催化裂化反应-再生过程的条件下,评价助剂对再生烟气中NOx的催化转化性能,同时还可考察助剂的加入对催化裂化产品分布的影响。采用该方法对几种降NOx助剂的性能进行了评价,结果表明,在催化剂体系中含有Pt基CO助燃剂的情况下,加入4％的RDNO;助剂后,烟气NOx降低幅度约30％～40％,且催化裂化产品分布基本不受影响。     

负载TiO2浮石对污水处理厂二沉池出水的矿化与灭菌研究

城市污水处理厂第二沉淀池出水的污染物浓度较低,经过矿化及灭菌等深度处理后可以回用,从而减轻对淡水资源的需求。本实验采用溶胶-凝胶法制备了负载TiO2的浮石光催化剂,对城市污水处理厂第二沉淀池出水进行了矿化及灭菌处理研究。采用低压汞灯对处理水样照射2h,有机物降解率可以达到50％,灭菌效率达到100％。处理后的水可以作为农作物灌溉、城市绿化和娱乐用水。负载在浮石表面的TiO2薄膜经过10次的光催化循环实验后,没有明显破损,可以再次循环使用。TiO2浮石具有质轻、机械性能好和价格低等优点,该方法处理废水的适用浓度低,可以广泛应用于微污染水和城市污水处理厂出水的深度处理。     

碱激发胶凝材料及其固化Pb^2＋的试验研究

对碱激胶凝材料（碱-偏高岭土、碱-矿渣和碱-粉煤灰）与水-水泥体系固化Ph^2＋进行了试验研究,其中水-水泥体系为对比样。结果表明：对于相同稠度的碱激发胶凝材料,即使Pb^2＋含量达到2．4％,除了碱-粉煤灰的抗压强度略低外,其他都达到30MPa以上,满足填埋或做建筑基材的使用要求;与水泥相比,碱激发胶凝材料能显著降低重金属离子（Ph^2＋）浸出浓度,其规律性与其NH4交换容量大小的规律性一致,与其固化体的抗压强度的大小没有相关性。