APPLICATION OF GREY MODEL AND ARTIFICIAL NEURAL NETWORKS TO FLOOD FORECASTING1

The main focus of this study was to compare the Grey model and several artificial neural network (ANN) models for real time flood forecasting, including a comparison of the models for various lead times (ranging from one to six hours). For hydrological applications, the Grey model has the advantage that it can easily be used in forecasting without assuming that forecast storm events exhibit the same stochastic characteristics as the storm events themselves. The major advantage of an ANN in rainfall‐runoff modeling is that there is no requirement for any prior assumptions regarding the processes involved. The Grey model and three ANN models were applied to a 2,509 km² watershed in the Republic of Korea to compare the results for real time flood forecasting with from one to six hours of lead time. The fifth‐order Grey model and the ANN models with the optimal network architectures, represented by ANN1004 (34 input nodes, 21 hidden nodes, and 1 output node), ANN1010 (40 input nodes, 25 hidden nodes, and 1 output node), and ANN1004T (14 input nodes, 21 hidden nodes, and 1 output node), were adopted to evaluate the effects of time lags and differences between area mean and point rainfall. The Grey model and the ANN models, which provided reliable forecasts with one to six hours of lead time, were calibrated and their datasets validated. The results showed that the Grey model and the ANN1010 model achieved the highest level of performance in forecasting runoff for one to six lead hours. The ANN model architectures (ANN1004 and ANN1010) that used point rainfall data performed better than the model that used mean rainfall data (ANN1004T) in the real time forecasting. The selected models thus appear to be a useful tool for flood forecasting in Korea.     

关于铁路行车事故预测的应用研究

本文论述了事故预测对铁路行车安全的重要性,并应用灰色预测模型和概率回归估计法模型研制了预测软件,进行铁路行车事故的宏观及微观预测,为采取有针对性的防范措施提供一个可靠信息。     

几种事故预测方法的探讨

本文根据生产单位事故发生的随机性特点,论证了未来时段中事故次数及其时间分布等各项指标,是能够用多种随机数学模型和灰色系统理论来预测的。用非随机的函数模型预測事故是不合适的。给出了几种事故预測方法,并提出了几种值得研究和试用的方法。     

陕西延安地区1992—2000年干旱趋势灰色预测

本文存分析延安地区干旱特征的基础上,采用地地区40a的资料,运用灰色理论的预测方法,预测了1992～2000年间干旱发生的季节及强度,预计该时该区的旱情将持续出现,故提出必须立足抗旱,以保农业生产的稳步发展。     

灰关联分析在故障树诊断中的应用

介绍一种灰色诊断法。在故障诊断模式识别中,运用灰关联分析,依据故障树的底事件重要度,通过关联度计算及排序,对故障树分析中,各种故障模式发生的可能性大小作出了准确的判断,从而为处理事故的先后、缓急提供了依据。     

ESTIMATION OF PRE- AND POST-DEVELOPMENT NONPOINT WATER QUALITY LOADINGS1

ABSTRACT: A simple procedure for estimating pre- and post-development water quality loadings from residential communities is discussed. The procedure deals with: (a) gathering basic water quality loading numbers observed by others at several watersheds with various land uses; (b) obtaining the breakdown of proposed land uses at various phases of the community development; and (c) estimating pre- and post-development water quality loading numbers by taking the weighted average of the basic loading numbers in terms of areal coverages of different land uses at various phases of development. Results of this simplified procedure have been verified indirectly by comparing them with the estimates derived independently through a more fundamental but time-consuming approach. The procedure was used to evaluate the anticipated water quality impact of two future residential communities in South Florida by analyzing four water quality parameters: Suspended Soils (SS), Total Nitrogen (TN), Total Phosphorus (TP), and Biochemical Oxygen Demand (BOD₅). Although computation of loading numbers with mixed land uses is not an exact science at the present time, the recommended approach appears to be the best available technique to analyze quantitatively the water quality-quantity-land use interactions.     

模糊综合评价法的改进及其在水库水质评价中的应用

为了克服传统模糊综合评价法中权重赋值的片面性及最大隶属度原则导致评价结果不合理的弊端,提出了基于层次分析法和熵权法的主客观组合权重,并结合相乘相加算子及加权平均原则,建立了改进的模糊综合评价法.选取TP、TN、DO、NH3-N、CODMn和BOD5为评价指标,用改进的模糊综合评价法对重庆某水库2012年7-12月的水质进行了综合评价.结果表明:TN为主要的污染物;水质最差时段为8月,评价结果为2.655,水质属于Ⅱ级,略偏于Ⅲ级.评价结果与人工神经网络模型所得结果基本一致.     

基于MGWR的黄河流域土壤有机碳密度和影响因素分析

土壤有机碳作为最大陆地碳库,其空间分布特征和影响因素对于全球碳循环过程具有重要影响.基于土壤有机碳密度数据,结合环境因子,使用多尺度地理加权回归（MGWR）模型预测了黄河流域土壤有机碳密度（SOCD）和影响因素.结果表明：①黄河流域0~20 cm和0~100 cm的SOCD范围分别为0~14.82 kg ·m^-2和0~32.39kg ·m^-2,均值分别为3.48 kg ·m^-2和8.07kg ·m^-2,储量则分别为2.76 Pg和6.48 Pg;②各生态系统类型中,0~20 cm的SOCD从大到小依次为：森林>水体与湿地>其他>草地>农田>聚落>荒漠,0~100 cm的SOCD从大到小依次为：水体与湿地>森林>其他>草地>农田>聚落>荒漠,SOCR从大到小皆为：草地>农田>森林>荒漠>水体与湿地>聚落>其他;③黄河流域SOCD的分布主要受常数项、剖面曲率、NDVI和降水的影响,曲率和粉砂对深层的SOCD的分布也具有重要影响;此外,降水、NDVI和常数项（除森林外）是影响各生态系统的主要因素,曲率和粉砂则仅对荒漠和其他生态系统具有重要影响.研究结果得出了黄河流域SOCD的空间分布和影响因素,可为黄河流域碳平衡、土壤质量评价和生态治理恢复与巩固提升提供科学依据.     

北京市PM2.5时空分布特征及其与PM10关系的时空变异特征

PM_(2.5)时空分布特征及其与其它污染物的相关关系是PM_(2.5)时空统计分析的主要研究内容.然而,现有的方法直接从监测站点的角度对时空分布特征进行分析,难以有效地揭示PM_(2.5)浓度的聚集分布特征;同时,常用的地理加权回归在对PM_(2.5)与其它污染物间关系进行建模的过程中,缺乏同时考虑时间异质性与空间异质性,从而不能准确地描述依赖关系的时空变异特征.为此,首先借助于空间聚类分析技术,对北京市2014年PM_(2.5)浓度的聚集结构进行探测,在此基础上,通过聚集结构来分析PM_(2.5)季节性时空分布特征.然后,利用地理时空加权回归对北京市PM_(2.5)与PM_(10)季节平均浓度间关系进行建模,依据回归结果分析PM_(2.5)-PM_(10)间关系的时空变异特征.实验结果表明,春夏季节PM_(2.5)污染程度及空间变异程度均低于秋冬季节,各季节PM_(2.5)浓度均表现为北部浓度低、南部浓度高的空间分布特征;地理时空加权回归具有更好的拟合效果,由回归系数进一步可发现,春夏季PM_(2.5)-PM_(10)相关性低于秋冬季PM_(2.5)-PM_(10)相关性;各季节均表现为西北部PM_(2.5)-PM_(10)的相关性高于东南部PM_(2.5)-PM_(10)的相关性.     

中国碳排放及影响因素的市域尺度分析

评估区域碳排放及其与社会经济状况的关系对于制定碳减排措施至关重要.以中国339个地级及以上城市(不含新疆部分城市和港澳台地区)为研究对象,探究了非化石能源占比、土地开发度、常住人口城镇化率、第二产业占比、人均GDP和人均建设用地面积对人均CO₂排放量的影响.通过构建模拟人均CO₂排放量的贝叶斯信念网络,识别各因素对人均CO₂排放量的全局影响;采用多尺度地理加权回归模型,分析各因素对人均CO₂排放量的局部影响.结果表明：(1)2020年,中国地级及以上城市人均CO₂排放量呈现出由南向北递增,东部沿海向内陆递减的格局.(2)从全局来看,人均CO₂排放量对各因素的敏感性从高到低依次为：人均建设用地面积>人均GDP>常住人口城镇化率>土地开发度>第二产业占比>非化石能源占比.(3)从局部来看,各因素与人均CO₂排放量的空间关系方向与全局关系一致,关系强度上存在空间异质性.(4)清洁能源、脱碳技术、土地节约集约利用...