Development of machine learning based prediction models for hazardous properties of chemical mixtures

Lower flammability limit (LFL), upper flammability limit (UFL), auto-ignition temperature (AIT) and flash point (FP) are crucial hazardous properties for fire and explosion hazards assessment and consequence analysis. In this study, a comprehensive prediction model set was constructed by using expanded chemical mixture databases of chemical mixture hazardous properties. Machine learning based gradient boosting quantitative structure-property relationship (GB-QSPR) method is implemented for the first time to improve the model performance and prediction accuracy. The result shows that all developed models have significantly higher accuracy than other regular QSPR models, with the 5-fold cross-validation RMSE of LFL, UFL, AIT, and FP models being 1.06, 1.14, 1.08, and 1.17, respectively. All developed QSPR models can be used to estimate reliable chemical mixture hazardous properties and provide useful guidance in chemical mixture hazard assessment and consequence analysis.     

The determination and matching analysis of pinch point temperature difference in evaporator and condenser of organic rankine cycle for mixed working fluid

Exergo-economic analysis of the pinch point temperature difference (PPTD) in both evaporator and condenser of sub-critical organic Rankine cycle system (ORCs) are performed based on the first and second laws of thermodynamics. Taking mixture R13I1/R601a as a working fluid and the annual total cost per net output power Z as exergo-economic performance evaluation criterion, the effects of PPTD in evaporator ΔT_e, and the PPTD ratio of condenser to evaporator y, on the exergo-economic performance of ORCs are analyzed. Moreover, how some other parameters influence the optimal PPTD in evaporator ΔT_e,opt and the optimal PPTD ratio of condenser to evaporator y_opt are also discussed. It has been found that the exergo-economic performance of ORCs is remarkably influenced by ΔT_e and y, and there exists ΔT_e,opt and y_opt. In addition, ΔT_e,opt and y_opt are affected by heat transfer coefficient ratio of condenser to evaporator ß, the temperature of working fluid at dew point in condenser T_1a, and composition of R13I1/R601a: larger ß and T_1a lead to lower ΔT_e,opt and y_opt; by contraries, larger mass fraction of R13I1 makes ΔT_e,opt and y_opt increase, and y_opt increases linearly. The effects of the temperature of working fluid at bubble point in evaporator T_3a, mass flow rate of exhaust flue gas m_g, and inlet temperature of exhaust flue gas T_gi on ΔT_e,opt and y_opt are very slight. For comparison, three additional working fluids, namely R601a, R245fa, and 0.32R245fa/0.68R601a, are also taken into account.     

基于遗传神经网络的除湿机故障诊断与寿命预测

目的识别除湿机的性能状态和预测吸附剂的剩余寿命。方法针对除湿机故障过程缓变的特点,提出一种基于数据驱动的遗传神经网络模型。首先,为解决设备失效程度划分模糊的问题,由5个热力参数组成反映吸附剂劣化程度的特征向量,关联分析得到除湿机的5类故障模式。其次,利用遗传神经网络建立状态参数和故障模式的映射关系。最后,对表征设备吸附能力的主参数进行外推预测。结果训练好的诊断网络可准确地识别出设备的劣化程度及其演变过程,预测网络的预测精度非常高。结论该方法可有效地实现对除湿机的故障诊断与预测。。     

目标分解法在清洁生产中的应用

间苯二甲酸装置扩产改造后排污量增加较多,翻倍的排污量严重影响到装置发展,针对装置的排污问题,利用目标分解法查找关键原因并进行技术改造,改造后装置排污量大大减少,提高了装置的竞争力。     

红外系统对点源目标作用距离的优化计算

作用距离是红外系统核心性能指标之一,受大气、背景等多方面因素的影响。针对传统作用距离计算方法存在的局限,综合考虑多个影响因素,提出了一种优化计算方法,首次精确计算了作用距离与实际距离之差的大气衰减值,并对计算方法进行了验证。结果表明,优化计算方法具有较好的计算精度,可用于红外系统性能预测和整体设计。     

基于服务式GIS的流域水质目标管理技术平台架构设计

针对流域水质目标管理与GIS集成技术,对比分析传统模式与服务式GIS模式的优劣,提出了基于服务式GIS的流域水质目标管理技术平台的层次结构、总体架构及建设步骤;基于具体建设需求分析,构建了辽河流域水质目标管理技术平台的详细架构。     

南京市郊设施蔬菜重金属含量及健康风险分析

采用火焰原子吸收光谱法测定了南京市郊2个设施蔬菜栽培基地蔬菜中的重金属含量,并与国内蔬菜标准进行比较;运用目标危害系数(the target hazard quotient,THQ)方法评价了当地居民摄食蔬菜导致的健康风险,并比较了成人和儿童居民THQ值大小.结果表明,蔬菜中Pb、Cd、Zn和Cu的平均含量分别是0.1...     

建阳市农业面源污染调查与防治对策研究

就目前建阳市农业面源污染的现状进行调查,并提出防治对策。     

Experimental study on filtration and continuous regeneration of a particulate filter system for heavy-duty diesel engines

This study investigated the filtration and continuous regeneration of a particulate filter system on an engine test bench, consisting of a diesel oxidation catalyst（DOC） and a catalyzed diesel particulate filter（CDPF）. Both the DOC and the CDPF led to a high conversion of NO to NO2 for continuous regeneration. The filtration efficiency on solid particle number（SPN） was close to100%. The post-CDPF particles were mainly in accumulation mode. The downstream SPN was sensitively influenced by the variation of the soot loading. This phenomenon provides a method for determining the balance point temperature by measuring the trend of SPN concentration.     

西安市PM_(2.5)污染源位置确定的模型应用研究

针对空气中PM2.5污染物的传播问题,以西安市的情况为实例,合理考虑风力等天气和季节因素的影响,建立能够刻画该地区PM2.5的发生和演变规律的有风高斯点源扩散数学模型,代入相关数据后,利用MATLAB软件求解出PM2.5污染源由一区扩散到另一区的浓度,最后由综合的各区总浓度得出PM2.5污染源的位置,并与该市实际检测到PM2.5数据进行对比分析。结果表明,确定的位置比较准确,该方法能为城市污染治理提供参考。