典型优化目标函数下源参数反演性能对比研究

基于外场实验数据,从反演高估率、准确性、稳定性角度系统评估了8种典型目标函数在不同未知源参数反演情形下的反演性能差异.研究发现,不同目标函数反演性能差异显著.仅反演单参数源强时（Q）,对数变换目标函数高估率最大（79.4%）,偏差平方和目标函数准确性最高（P_ARD<50%=82.3%,ARD=（35.3±9.1）%）,目标函数稳定性无明显差异（CV<0.01）.三参数反演（Q,x,y）时,标准化均方根误差目标函数源强高估率最大（98.5%）,对数变换目标函数准确性、稳定性最高（P_ARD<50%=91.1%,ARD=（48.4±9.8）%;CV=0.01）;位置方面,偏差平方和目标函数准确性最高（AD=（36.12±11.39） m）,对数变换目标函数稳定性最强（CV=0.0018）.四参数反演（Q,x,y,z）时,标准化均方根误差目标函数源强高估率最大（98.5%）,对数变换目标函数准确性、稳定性最优（P_ARD<50%=61.7%,ARD=（55.2±16.5）%;CV=0.03）;位置方面,相关系数总体表现最优（AD=（34.37±10.72） m;CV=0.011）.整体上,随反演参数变化,对数变换目标函数源强反演性能最稳定.     

基于无人机多光谱的沼泽植被识别方法

为了探究高分辨率无人机多光谱数据对沼泽植被群丛识别能力,在洪河国家级自然保护区的核心区、缓冲区和实验区分别建立典型样区,通过低空无人机搭载的RGB及多光谱相机获取研究区正射影像,构建多维数据集并确立4种分类方案.采用面向对象的随机森林（RF）算法,对输入的多维数据集进行变量选择和参数（mtry、ntree）调优,构建适合沼泽植被群丛尺度识别模型.结果表明：优化的面向对象的RF算法对沼泽湿地植被具有较高的识别能力,在95%的置信区间内,核心区方案四（结合了光谱波段、纹理特征、几何特征、位置特征、地表高程信息和植被指数）获得最高总体精度为87.12%,kappa系数为0.850,比方案二（结合了光谱波段、几何特征和位置特征）总体精度高12.27%,kappa系数高0.140;对于单一典型沼泽湿地植被识别精度中,芦苇获得最高的用户精度高于88%,生产者精度高于90%,小叶章的生产者精度高于85%,但是在核心区用户精度较低,仅为78%.该方法可以作为沼泽植被群丛识别的有效方法,为研究沼泽湿地生态环境变化提供更准确的数据支持.     

黑碳仪测量气溶胶吸收系数的校正算法和影响因素研究进展

黑碳仪作为常用的吸收系数测量方法之一,广泛应用于气溶胶光吸收特性的实验室和外场观测研究.然而,受测量方法的限制,吸收系数测量的准确性受到颗粒物负载效应,以及滤膜和颗粒物的多重散射效应的影响.本文综述了黑碳仪的测量原理及现有校正算法,探究了其吸收系数测量的影响因素,分析了黑碳气溶胶的混合状态对其吸收系数测量偏差（C_ref’）的影响,探讨了C_ref’与气溶胶单次散射反照率（SSA）的关系以及C_ref’的波长依赖关系,并提出了今后相关研究的建议.     

化工企业开展清洁生产的主要途径

结合化工企业的生产实际,提出了实现清洁生产的主要途径,阐述了开展清洁生产审核应注意的问题.     

基于遗传算法的门限自回归模型在海温预测中的应用

提出了建立门限自回归模型（ＴＡＲ）的一套简便通用的方法。用作者提出的改进遗传算法，可同时优化门限值和自回归系数，成功地解决了ＴＡＲ建模过程所涉及的大量复杂寻优工作这一难题，为ＴＡＲ模型的广泛应用提供了强有力的工具。实例计算的结果说明了这套方法的可行性和有效性，同时也说明了，通过门限值的控制作用，ＴＡＲ模型可以有效地利用如海洋资源所隐含的时序分段相依性这一重要信息，限制了模型误差，从而保证了ＴＡＲ模型预测性能的稳健性，提高了预测精度。该方法具有通用性，在各种非线性时序预测中具有重要的理论意义和应用价值     

基于ACE法和模拟退火算法的土层等效参数反演

提出了一种基于ACE法和模拟退火算法的土层参数反演方法:一方面先用均匀试验构造出输入数据与输出数据,再利用ACE非参数回归方法构造两者之间的高效响应面,然后通过响应面关系进行插值运算来得出对应参数的响应值;另一方面根据响应面的关系构造出反演目标函数,利用模拟退火法对目标函数进行寻优计算,从而能避免寻优过程中的目标函数值...     

湿地对福州市热岛效应影响遥感分析

以4期Landsat TM/OLI影像为基础,利用大气校正后的单窗算法对福州市地表温度进行反演并验证.进而利用ArcGIS的空间分析工具和土地利用/覆被类型数据,进行湿地对福州市热岛效应影响分析.结果显示:(1)通过地表温度实测数据对地表温度反演结果进行验证,二者之间R2值为0.9475,满足研究需要;(2)各土地利用...     

Development of a framework to implement a recycling program in an elementary school

This study developed a system to plan, evaluate, implement, and monitor a waste reduction and source-separation program, which was then tested through a field study in a New York City elementary school. Despite efforts to implement integrated waste management in elementary schools, appropriate planning tools that document performance and support program development are lacking. To fill this gap, a system for analyzing existing programs was developed. After identifying the phases required for a successful program, a scoring system was developed to quantify the performance of existing programs. To assess existing protocol, recommendations from leading municipalities were reviewed, and the performance of existing programs was quantified within the scoring system. The existing protocol was found to contain similarities, but an overarching approach was needed to connect the provisions with programs that have been successfully deployed. From this analysis, a scoring system was developed to take into consideration the major elements needed for a waste reduction and source-separation plan, colloquially referred to as a recycling plan, as well as the specifics of implementation. This system allows for the review of programs already developed, and can pinpoint limitations prior to implementation. The framework was evaluated through the test case school, with both waste audit data and diagrams. From the lessons learned, successful techniques and recommendations for utilizing this framework were created.     

实施清洁生产 创建绿色企业——开展清洁生产工作的几点体会

文章阐述了辽河油田公司开展清洁生产工作的基本情况,结合管理现状,分析总结了开展清洁生产工作的管理经验,阐述了通过科技进步所取得的清洁生产成绩,归纳整理出可借鉴的清洁生产审核经验,为清洁生产审核工作提供了有益的建议。     

An improved grey wolf optimizer algorithm for identification and location of gas emission

Identification of the leakage of hazardous gases plays an important role in the environment protection, human health and safety of industry production. However, lots of current optimization algorithms, such as particle swarm optimization (PSO) and Grey Wolf Optimizer (GWO), suffer from poor global optimization capability and estimation accuracy. In this work, a hybrid differential evolutionary and GWO (DE-GWO) algorithm is proposed. Tested by simulation cases and Prairie Grass emission experimental data, DE-GWO shows higher estimation accuracy than GWO. Compared with the other four optimization algorithms, DE-GWO exhibits finer robust stability under different population sizes, fewer iterations, as well as higher estimation accuracy with fewer search agents. Importantly, simulation results demonstrate that DE-GWO is more suitable to apply in the scene with a small number of sensors. Therefore, the proposed in this paper outperforms other optimization algorithms for the gas emission inverse problem. DE-GWO can provide reliable estimation towards gas emission identification and positioning, which shows huge potential as the data analysis module of real-time monitoring and early warning system.