Four assessment methods(two pollution indexes and two fuzzy mathematical models) were employed to investigate the environmental quality of four soils around a ferroalloy plant in Nanjing City.Environmental quality was...Four assessment methods(two pollution indexes and two fuzzy mathematical models) were employed to investigate the environmental quality of four soils around a ferroalloy plant in Nanjing City.Environmental quality was assessed as class IV(moderately polluted) for each soil with single-factor index method,and was identified to be classes IV,III(slightly polluted),III,and III for soils A,B,C,and D,respectively,with the comprehensive index model.In comparison with the single-factor index method,the comprehensive index model concerned both dominant parameter and average contribution of all factors to the integrated environmental quality.Using the two fuzzy mathematical methods(single-factor deciding and weighted average models),the environmental risks were determined to be classes IV,III,II(clean),and II for soils A,B,C,and D,respectively.However,divergence of the membership degree to each pollution class still occurred between the two methods.In fuzzy mathematical methods,membership functions were used to describe the limits between different pollution degrees,and different weights were allocated for the factors according to pollution contribution.Introduction of membership degree and weight of each factor to fuzzy mathematical models made the methods more reasonable in the field of environmental risk assessment.展开更多
The purified terephthalic acid (PTA) petrochemical wastewater molecula r toxicity detected by use of Mouse Genome 430A 2.0 GeneChip was conducted in th is research. The toxic dose to male mice was 0.03 g/(kg·d) o...The purified terephthalic acid (PTA) petrochemical wastewater molecula r toxicity detected by use of Mouse Genome 430A 2.0 GeneChip was conducted in th is research. The toxic dose to male mice was 0.03 g/(kg·d) of PTA in the wa stewater. The mice liver total RNA was isolated as the temple for synthesis of cDNA and then the cDNA as the temple for synthesis of cRNA. Hybridizing the cRN A with the target genes on the gene chip, there were 232 genes expression levels up-regulated and 74 genes down-regulated discovered obviously. The foremost 40 genes for both the highest and the lowest expression levels involved endogenetic steroid and hormone metabolism, immune system, the leukocyte activity and infla mmation, detoxification in liver, reproduction and growth hormone, regulation im mune factors of anti-tumor and anti-infection and cancer to the mice sampled. Th e data suggest the PTA wastewater contained over 5 aromatics and their toxicitie s integrated were much higher than the pure chemical PTA. And the pure chemical PTA toxicities data cannot be used to evaluate the toxicity of the PTA wastewate r instead.展开更多
作为磨矿过程的主要生产质量指标,磨矿粒度是实现磨矿过程闭环优化控制的关键.将磨矿粒度控制在一定范围内能够提高选别作业的精矿品位和有用矿物的回收率,并减少有用矿物的金属流失.由于经济和技术上的限制,磨矿粒度的实时测量难以实现...作为磨矿过程的主要生产质量指标,磨矿粒度是实现磨矿过程闭环优化控制的关键.将磨矿粒度控制在一定范围内能够提高选别作业的精矿品位和有用矿物的回收率,并减少有用矿物的金属流失.由于经济和技术上的限制,磨矿粒度的实时测量难以实现.因此,磨矿粒度的在线估计显得尤为重要.然而,目前我国所处理的铁矿石大多数为性质不稳定的赤铁矿,其矿浆颗粒存在磁团聚现象,所采集的数据存在大量异常值,使得利用数据建立的磨矿粒度模型存在较大误差.同时,传统前馈神经网络在磨矿粒度数据建模过程中存在收敛速度慢、易于陷入局部最小值等缺点,且单一模型泛化性能较差,现有的集成学习在异常值干扰下性能严重下降.因此,本文在改进的随机向量函数链接网络(random vector functional link networks,RVFLN)的基础上,将Bagging算法与自适应加权数据融合技术相结合,提出一种基于鲁棒随机向量函数链接网络的集成建模方法,用于磨矿粒度集成建模.所提方法首先通过基准回归问题进行了实验研究,然后采用磨矿工业实际数据进行验证,表明其有效性.展开更多
基金the PhD Fund of the National Education Ministry of China (No20030284038)the Interna-tional Foundation for Science (NoW/4215)
文摘Four assessment methods(two pollution indexes and two fuzzy mathematical models) were employed to investigate the environmental quality of four soils around a ferroalloy plant in Nanjing City.Environmental quality was assessed as class IV(moderately polluted) for each soil with single-factor index method,and was identified to be classes IV,III(slightly polluted),III,and III for soils A,B,C,and D,respectively,with the comprehensive index model.In comparison with the single-factor index method,the comprehensive index model concerned both dominant parameter and average contribution of all factors to the integrated environmental quality.Using the two fuzzy mathematical methods(single-factor deciding and weighted average models),the environmental risks were determined to be classes IV,III,II(clean),and II for soils A,B,C,and D,respectively.However,divergence of the membership degree to each pollution class still occurred between the two methods.In fuzzy mathematical methods,membership functions were used to describe the limits between different pollution degrees,and different weights were allocated for the factors according to pollution contribution.Introduction of membership degree and weight of each factor to fuzzy mathematical models made the methods more reasonable in the field of environmental risk assessment.
基金The Ph. D Fund of the National Education Ministry of China (No. 20030284038) and the Hi-Tech Research and DevelopmentProgram(863) of China(No. 2001AA214191)
文摘The purified terephthalic acid (PTA) petrochemical wastewater molecula r toxicity detected by use of Mouse Genome 430A 2.0 GeneChip was conducted in th is research. The toxic dose to male mice was 0.03 g/(kg·d) of PTA in the wa stewater. The mice liver total RNA was isolated as the temple for synthesis of cDNA and then the cDNA as the temple for synthesis of cRNA. Hybridizing the cRN A with the target genes on the gene chip, there were 232 genes expression levels up-regulated and 74 genes down-regulated discovered obviously. The foremost 40 genes for both the highest and the lowest expression levels involved endogenetic steroid and hormone metabolism, immune system, the leukocyte activity and infla mmation, detoxification in liver, reproduction and growth hormone, regulation im mune factors of anti-tumor and anti-infection and cancer to the mice sampled. Th e data suggest the PTA wastewater contained over 5 aromatics and their toxicitie s integrated were much higher than the pure chemical PTA. And the pure chemical PTA toxicities data cannot be used to evaluate the toxicity of the PTA wastewate r instead.
文摘作为磨矿过程的主要生产质量指标,磨矿粒度是实现磨矿过程闭环优化控制的关键.将磨矿粒度控制在一定范围内能够提高选别作业的精矿品位和有用矿物的回收率,并减少有用矿物的金属流失.由于经济和技术上的限制,磨矿粒度的实时测量难以实现.因此,磨矿粒度的在线估计显得尤为重要.然而,目前我国所处理的铁矿石大多数为性质不稳定的赤铁矿,其矿浆颗粒存在磁团聚现象,所采集的数据存在大量异常值,使得利用数据建立的磨矿粒度模型存在较大误差.同时,传统前馈神经网络在磨矿粒度数据建模过程中存在收敛速度慢、易于陷入局部最小值等缺点,且单一模型泛化性能较差,现有的集成学习在异常值干扰下性能严重下降.因此,本文在改进的随机向量函数链接网络(random vector functional link networks,RVFLN)的基础上,将Bagging算法与自适应加权数据融合技术相结合,提出一种基于鲁棒随机向量函数链接网络的集成建模方法,用于磨矿粒度集成建模.所提方法首先通过基准回归问题进行了实验研究,然后采用磨矿工业实际数据进行验证,表明其有效性.
