Influencing Factors of Total Factor Energy Efficiency in Bohai Rim Region Based on DEA-Tobit Model

Bohai Rim region is an important economic development area and a large carbon emission area in China.It is of great significance to explore the total factor energy efficiency and its influencing factors for the low carbon transformation and high-quality development of the Bohai Rim region.Based on the total factor energy efficiency framework,the DDF-DEA model was used to calculate the total factor energy efficiency,and the internal and external differences of the total factor energy efficiency were further analyzed.The internal and external influencing factors were determined by ML index method and classical endogenous growth theory,and then the Tobit panel model was used to empirically analyze the action mechanism of all influencing factors of total factor energy efficiency in the Bohai Rim region.The results show that the pure technical efficiency,scale efficiency and technological progress among the internal influencing factors contribute to the improvement of energy efficiency in the Bohai Rim region.Industrial structure,industrial internal structure and ownership structure inhibit the improvement of energy efficiency.Energy consumption structure and energy endowment also have a negative impact on energy efficiency.Therefore,measures such as promoting technological progress,adjusting economic structure and optimizing energy structure will effectively improve total factor energy efficiency in the Bohai Rim region.

Spatiotemporal Dynamics of Green Total-factor Water-use Efficiency and Its Influencing Factors in China

In this study,we developed an evaluation index system for green total-factor water-use efficiency(GTFWUE)which reflected both economic and green efficiencies of water resource utilization.Then we measured the GTFWUE of 30 provinces/municipalities/autonomous regions(hereafter provinces)in China(not including Tibet,Hong Kong,Macao,Taiwan as no data)from 2000 to 2018 using a minimum distance to the strong frontier model that contained an undesirable output.We further analyzed the regional differences and spatial correlations of GTFWUE using these values based on Global and Local Moran’s I statistics,and empirically determined the factors affecting GTFWUE using a spatial econometric model.The evaluation results revealed that the GTFWUE differed substantially between the regions.The provinces with high and low GTFWUE values were located in the coastal and inland areas of China,respectively.The eastern region had a significantly higher GTFWUE than the central and western regions.The GTFWUEs for all three regions(eastern,central,and western regions)decreased slowly from 2000 to 2011(except 2005),remained stable from 2012 to 2016,and rapidly increased in 2017 before decreasing again in 2018.We found significant spatial correlations between the provincial GTFWUEs.The GTFWUE for most provinces belonged to the high-high or low-low cluster region,revealing a significant spatial clustering effect of provincial GTFWUEs.We also found that China’s GTFWUE was highly promoted by economic growth,population size,opening-up level,and urbanization level,and was evidently hindered by water endowment,technological progress,and government influence.However,the water-use structure had little impact on GTFWUE.This study fully demonstrated that the water use mode would be improved,and water resources needed to be used more efficiently and green in China.Moreover,based on the findings of this study,several policy recommendations were proposed from the aspects of cross-regional cooperation,economy,society,and institution.

Factors influencing oil recovery by surfactant-polymer flooding in conglomerate reservoirs and its quantitative calculation method

This study aims to clarify the factors influencing oil recovery of surfactant-polymer(SP)flooding and to establish a quantitative calculation model of oil recovery during different displacement stages from water flooding to SP flooding.The conglomerate reservoir of the Badaowan Formation in the seventh block of the Karamay Oilfield is selected as the research object to reveal the start-up mechanism of residual oil and determine the controlling factors of oil recovery through SP flooding experiments of natural cores and microetching models.The experimental results are used to identify four types of residual oil after water flooding in this conglomerate reservoir with a complex pore structure:oil droplets retained in pore throats by capillary forces,oil cluster trapped at the junction of pores and throats,oil film on the rock surface,isolated oil in dead-ends of flow channel.For the four types of residual oil identified,the SP solution can enhance oil recovery by enlarging the sweep volume and improving the oil displacement efficiency.First,the viscosity-increasing effect of the polymer can effectively reduce the permeability of the displacement liquid phase,change the oil-water mobility ratio,and increase the water absorption.Furthermore,the stronger the shear drag force of the SP solution,the more the crude oil in a porous medium is displaced.Second,the surfactant can change the rock wettability and reduce the absorption capacity of residual oil by lowering interfacial tension.At the same time,the emulsification further increases the viscosity of the SP solution,and the residual oil is recovered effectively under the combined effect of the above two factors.For the four start-up mechanisms of residual oil identified after water flooding,enlarging the sweep volume and improving the oil displacement efficiency are interdependent,but their contribution to enhanced oil recovery are different.The SP flooding system primarily enlarges the sweep volume by increasing viscosity of solution to start two kinds of residual oil such as oil droplet retained in pore throats and isolated oil in dead-ends of flow channel,and primarily improves the oil displacement efficiency by lowing interfacial tension of oil phase to start two kinds of residual oil such as oil cluster trapped at the junction of pores and oil film on the rock surface.On this basis,the experimental results of the oil displacement from seven natural cores show that the pore structure of the reservoir is the main factor influencing water flooding recovery,while the physical properties and original oil saturation have relatively little influence.The main factor influencing SP flooding recovery is the physical and chemical properties of the solution itself,which primarily control the interfacial tension and solution viscosity in the reservoir.The residual oil saturation after water flooding is the material basis of SP flooding,and it is the second-most dominant factor controlling oil recovery.Combined with the analysis results of the influencing factors and reservoir parameters,the water flooding recovery index and SP flooding recovery index are defined to further establish quantitative calculation models of oil recovery under different displacement modes.The average relative errors of the two models are 4.4%and 2.5%,respectively;thus,they can accurately predict the oil recovery of different displacement stages and the ultimate reservoir oil recovery.

Extreme gradient boosting algorithm based urban daily traffic index prediction model:a case study of Beijing,China

The exhaust emissions and frequent traffic incidents caused by traffic congestion have affected the operation and development of urban transport systems.Monitoring and accurately forecasting urban traffic operation is a critical task to formulate pertinent strategies to alleviate traffic congestion.Compared with traditional short-time traffic prediction,this study proposes a machine learning algorithm-based traffic forecasting model for daily-level peak hour traffic operation status prediction by using abundant historical data of urban traffic performance index(TPI).The study also constructed a multi-dimensional influencing factor set to further investigate the relationship between different factors on the quality of road network operation,including day of week,time period,public holiday,car usage restriction policy,special events,etc.Based on long-term historical TPI data,this research proposed a daily dimensional road network TPI prediction model by using an extreme gradient boosting algorithm(XGBoost).The model validation results show that the model prediction accuracy can reach higher than 90%.Compared with other prediction models,including Bayesian Ridge,Linear Regression,ElatsicNet,SVR,the XGBoost model has a better performance,and proves its superiority in large high-dimensional data sets.The daily dimensional prediction model proposed in this paper has an important application value for predicting traffic status and improving the operation quality of urban road networks.

Leakage Rate Model of Urban Water Supply Networks Using Principal Component Regression Analysis

To analyze the factors affecting the leakage rate of water distribution system, we built a macroscopic "leakage rate–leakage factors"(LRLF) model. In this model, we consider the pipe attributes(quality, diameter,age), maintenance cost, valve replacement cost, and annual average pressure. Based on variable selection and principal component analysis results, we extracted three main principle components—the pipe attribute principal component(PAPC), operation management principal component, and water pressure principal component. Of these, we found PAPC to have the most influence. Using principal component regression, we established an LRLF model with no detectable serial correlations. The adjusted R2 and RMSE values of the model were 0.717 and 2.067, respectively.This model represents a potentially useful tool for controlling leakage rate from the macroscopic viewpoint.

甘肃省农业绿色全要素生产率测算及影响因素研究

提升农业绿色全要素生产率对于实现农业可持续发展具有重要的意义。运用SBM-GML指数法对甘肃省各市州2011-2020年的农业绿色全要素生产率及其区域差异进行分析,最后运用Tobit模型对甘肃省农业绿色全要素生产率影响因素进行分析,并在此基础上提出相应的对策建议。研究结果显示,(1)甘肃省农业绿色全要素生产率总体呈上升趋势,上升的主要动力来源于技术进步。(2)甘肃省农业绿色全要素生产率存在一定的空间差异性,总体呈现出“西低东高”的分布特点。(3)在甘肃省农业绿色全要素生产率的影响因素中,经济发展水平、城镇化水平、财政支农水平和农业产业结构对甘肃省农业绿色全要素生产率的提升具有显著的正向影响,受灾率对甘肃省农业绿色全要素生产率的提升具有显著的负向影响,且不同区域的农业绿色全要素生产率的影响因素也各不相同。最后,在以上分析的基础上提出了相应的对策建议。

Using geospatial technologies to delineate Ground Water Potential Zones(GWPZ)in Mberengwa and Zvishavane District,Zimbabwe

The main objective of the study was to delineate Ground Water Potential Zones(GWPZ)in Mberengwa and Zvishavane districts,Zimbabwe,utilizing geospatial technologies and thematic mapping.Various factors,including geology,soil,rainfall,land use/land cover,drainage density,lineament density,slope,Terrain Ruggedness Index(TRI),and Terrain Wetness Index(TWI),were incorporated as thematic layers.The Multi Influencing Factor(MIF)and Analytical Hierarchical Process(AHP)techniques were employed to assign appropriate weights to these layers based on their relative significance,prioritizing GWPZ mapping.The integration of these weighted layers resulted in the generation of five GWPZ classes:Very high,high,moderate,low,and very low.The MIF method identified 3%of the area as having very high GWPZ,19%as having high GWPZ,40%as having moderate GWPZ,24%as having low GWPZ,and 14%as having very low GWPZ.The AHP method yielded 2%for very high GWPZ,14%for high GWPZ,37%for moderate GWPZ,37%for low GWPZ,and 10%for very low GWPZ.A strong correlation(ρof 0.91)was observed between the MIF results and groundwater yield.The study successfully identified regions with abundant groundwater,providing valuable target areas for groundwater exploitation and highvolume water harvesting initiatives.Accurate identification of these crucial regions is essential for effective decision-making,planning,and management of groundwater resources to alleviate water shortages.

恩施州乡村旅游地农户返贫风险评估及其影响因素研究

有效预防返贫风险是后脱贫时代脱贫攻坚和乡村振兴战略有效衔接的关键.该研究将风险的概率纳入返贫风险评估体系,从发生概率和破坏程度两方面综合评估其返贫风险,系统反映返贫风险的可能性和破坏性,进而探究其影响因素.研究发现:1)农户各维度的返贫风险水平从高到低依次为个体风险、家庭风险、社会风险和自然风险,农户受家庭风险和个体风险影响较大,存在明显的内生性、个体化特征.2)依据风险的发生概率和破坏程度,返贫风险可分为“高概率-高破坏”型、“高概率-低破坏”型和“低概率-低破坏”型3种类型,其中,“高概率-高破坏”型风险包括非农就业能力和教育负担,返贫风险指数最高,对农户返贫的威胁最大.3)抚养人数和户主的文化程度等家庭特征是影响返贫风险的主要因素,乡村旅游扶贫开发模式不同导致农户的返贫风险存在显著差异.在返贫防治实践中,应因户施策,重点防治内生性返贫,同时要根据返贫风险发生概率及破坏程度的差异,制定合理的返贫风险预警机制及防治措施.

变化环境下渭河流域土壤水储量模拟及影响因素

[目的]随着气候变化和人类活动影响加剧,流域气候和下垫面条件发生改变,为准确模拟土壤水分变化过程并分析其影响因素对于地区的水资源管理和植被建设的重要意义。[方法]以黄土高原地区渭河流域为研究对象,基于ABCD水文模型,采用EFAST法分析模型参数敏感性,构建常参数和时变参数模型,比较不同参数模型的径流深和土壤水储量模拟精度,探讨气候变化和植被恢复对渭河流域土壤水储量变化的影响,明确土壤水储量变化的主导因素。[结果]土壤水层补给地下水的比例参数c最为敏感,地下水储放系数d、流域实际蒸散发量与土壤蓄水量之和的上限参数b次之,表征土壤完全饱和前径流的倾向性参数a最不敏感;与ABCD常参数模型相比,时变参数模型将率定期径流深模拟结果的NSE、KGE、R^(2)分别提高19%,10%,19%,验证期的NSE、KGE、R^(2)分别提高7%,7%,9%,时变参数模型显著改善径流深模拟效果;基于时变参数模型模拟的渭河流域土壤水储量与ERA5-Land 0—100 cm土层土壤水储量的相关关系最强,二者的变化过程基本相同;不同影响因素对渭河流域土壤水储量变化的影响程度由强到弱依次为潜在蒸散发>降水>NDVI。[结论]研究成果为该地区水资源规划和管理、植被建设提供科学依据,也为其他类似地区的土壤水储量研究提供参考。

连片特困地区县域交通扶贫效率测度及影响因素——以滇西边境山区为例

交通扶贫效率测度可为乡村振兴阶段区域交通规划方案制定提供重要的理论依据。以滇西边境山区56个县域四年间交通与经济发展面板数据为样本,构建交通扶贫效率投入产出指标体系,运用DEA模型、Malmquist指数测度滇西边境山区各县交通扶贫效率,并分析其时空分布特征。同时,构建Tobit回归模型,探究县域交通扶贫效率的影响因素。研究表明:滇西边境山区县域交通扶贫效率整体呈逐步增长态势,在乡村振兴阶段仍然有较大提升空间,且交通扶贫综合效率主要受到规模效率的影响;滇西边境山区县域交通扶贫效率大幅提升;依据交通扶贫综合效率的平均值和Malmquist指数值两个维度,滇西边境山区56个县可划分为潜力型、发展型、成熟型及巩固型4种类型,而发展型和成熟型是滇西边境山区交通扶贫效率的主要形态;县域规模、产业结构对交通扶贫效率的提升具有反向抑制作用,就业支持、消费水平、区位优势、投资规模均在不同程度上对滇西边境山区交通扶贫效率产生正向促进作用;各影响因素对不同发展阶段的县域交通扶贫效率在影响强度和作用路径上存在差异。因此,应进一步优化县城交通基础设施空间布局,强化劳动力就业等因素与交通运输发展的“联动效应”,针对不同交通扶贫发展阶段及影响因素短板,采取差异化、合理化发展策略。

中国省域科技成果转化效率评价及影响因素分析——基于超效率SBM-Malmquist-Tobit模型

[研究目的]提升省域科技成果转化效率,对提高国家科技发展水平和竞争力具有重要意义。[研究方法]以我国31个省(区、市)为研究对象,运用SBM-Malmquist模型对省域科技成果转化效率进行静态与动态分析,进一步结合Tobit回归模型分析科技成果转化效率的影响因素。[研究结论]我国省域科技成果转化效率整体表现良好,区域间呈现出“中部>东部>西部>东北部”的态势,且各区域之间差异较大。我国科技成果转化效率增长主要得益于技术进步,技术进步对科技成果转化效率的拉动作用抵消了技术效率变化和规模效率变化衰退的影响,市场需求和地区产业结构对科技成果转化效率具有显著正向作用,经济发展水平和财政支持对科技成果转化效率的影响不显著。推进我国科技成果转化效率的提高,仍需要从区域协同、投入产出结构优化、技术进步等方面发力。

住院老年慢性病共病患者营养不良的影响因素及预测模型构建

目的 探讨住院老年慢性病共病患者营养不良的影响因素及预测模型构建。方法 采用便利抽样法选取2023年1月—2024年2月在苏州大学附属苏州九院老年医学科住院的老年慢性病共病患者426例为研究对象。以微型营养评定简表(MNA-SF)评分<8分且白蛋白<34.0 g/L或前白蛋白<200 mg/L为营养不良依据,将患者分为营养不良组和无营养不良组。比较2组一般资料、口腔状况[采用口腔健康评估量表(OHAT)评估]、膳食炎症指数(DII,采用食物频率问卷表评估)、日常活动能力[采用Barthel指数(BI)评估]。采用多因素Logistic回归分析对老年慢性病共病患者营养不良的影响因素进行探讨,并构建模型公式。采用R软件运行梯度提升机(GBM)算法构建GBM预测模型。采用受试者工作特征(ROC)曲线分析2个模型的预测效能,采用Delong检验比较2个模型的曲线下面积(AUC)的差异。结果 92例被诊断为营养不良(营养不良组),334例患者无营养不良(无营养不良组)。营养不良组与无营养不良组年龄、慢性病共病种类、多重用药种类、OHAT评分、DII、BI评分比较,差异有统计学意义(P<0.05)。年龄大、慢性病共病种类多、多重用药种类多、OHAT评分高、DII高、BI评分低均是老年慢性病共病患者营养不良的影响因素(P<0.05)。ROC曲线显示,GBM模型的AUC为0.901,Logistic回归模型的AUC为0.874。Delong检验提示,GBM模型的预测效能优于Logistic回归模型(P<0.05)。结论 住院老年慢性病共病患者营养不良与年龄、慢性病共病种类、多重用药种类、OHAT评分、DII、BI评分有关,以此构建GBM模型可有效评估患者营养不良的发生风险。

重庆主城区降雨径流污染特征及其社会影响因素分析

为解析重庆主城区社会影响因素对降雨径流水质的影响,对2010—2021年政策影响下降雨径流水质的年际变化和降雨径流污染物指标进行相关性分析,采用对数平均迪氏指数法量化人口规模、技术创新、经济发展、产业结构4种社会影响因素对降雨径流水质的影响。结果表明:重庆主城区降雨径流水质呈改善趋势,且污染物指标相关性增强,2019年屋面降雨径流水质和2021年路面降雨径流水质均提升到“较清洁与污染”之间,固体悬浮物与COD、NH_(3)-N、TN、TP呈显著相关关系(p<0.01),屋面降雨径流中相关系数r>0.75,路面降雨径流中相关系数r>0.50;空气治理政策对屋面径流污染改善效果显著,海绵城市和排水系统建设政策的有效实施则对路面径流污染控制效果显著;社会影响因素中技术改善效应和产业结构效应对水质改善起主导作用,经济发展效应和人口规模效应对水质改善呈负向作用。

塔里木河流域水-耕地-粮棉系统耦合关系及影响因素分析

作为典型的干旱区和生态脆弱区,塔里木河流域是我国粮食、棉花的重要生产基地。水资源、耕地资源与粮棉资源的良性协调发展成为流域可持续发展的关键。通过构建塔里木河流域水-耕地-粮棉系统,测算其发展指数,并以耦合协调度模型测量了系统的耦合程度,在此基础上利用Fractional Logit模型分析了水-耕地-粮棉系统协调发展的影响因素。结果表明:(1)塔里木河流域各系统的发展指数呈现水系统>水-耕地-粮棉系统>耕地系统>粮棉系统。五地州中,巴音郭楞蒙古自治州各系统发展指数排名与流域一致,喀什地区和阿克苏地区在耕地和粮棉系统发展水平领先,水资源系统发展水平靠后,而克孜勒苏柯尔克孜自治州与和田地区则是水系统发展指数位居前列,耕地和粮棉系统发展较落后。(2)流域内水-耕地-粮棉系统耦合协调度介于0.475~0.680之间,处于勉强协调与初级协调之间。五地州中,喀什地区最高,克孜勒苏柯尔克孜自治州最低。这一指标经过缓慢提升后在2017年后明显下降。五地州中,阿克苏地区下降最为明显。(3)水库数量、一般公共预算支出及人口数量是影响水-耕地-粮棉系统协调发展状况的关键因素,三者每增加1个单位将分别引起耦合协调度1.0%、21.0%、35.6%的提升。

原水2-MIB影响因素分析及数据AI应用探索

《生活饮用水卫生标准》(GB 5749—2022)将2-甲基异莰醇(2-MIB)列为扩展指标,并提出0.00001 mg/L(10 ng/L)的要求,部分水厂现有工艺难以满足要求,需积极应对。文章以深圳市SZ水库、TG水库、XL水库、SY水库2021年—2022年的2-MIB数据为基础,分析2-MIB变化规律及其影响因素。结果表明,不同水库2-MIB浓度水平差异较大,2-MIB整体波动明显并呈季节性波动特征,其浓度受水温、藻密度、高锰酸盐指数、气温、风速等因素影响较大。深圳地区部分常规处理工艺水厂需通过投加粉末活性炭以保障出水2-MIB达标,为预判2-MIB浓度每日变化趋势以及时调整粉末活性炭投加量,文章基于2-MIB变化规律及影响因素分析结果,结合AI算法构建原水2-MIB预测预警模型,该模型可预测未来15 d内2-MIB逐日浓度,平均误差达到4.83 ng/L,有效助力2-MIB稳定达标。

基于孔隙结构分形维数的水驱气藏见水井产能预测模型

基于渗流理论和分形介质理论,利用分形维数与相渗曲线之间的幂函数关系,建立了气水两相流动条件下考虑孔隙结构分形维数与有效非达西系数的产能预测模型。研究表明:水驱气藏产水气井产能随着渗透率和有效厚度的增大而增大,但增大幅度随着参数值的增大逐渐变缓;随着分形维数、生产水气比、表皮系数、供给半径的增大,产水气井产能明显降低。在本文参数取值范围下,渗透率和有效厚度对产水气井产能影响最显著,是产能主控因素。模型可以准确预测水驱气藏产水气井产能,从而为调整气井生产制度、延缓气藏水侵、提高气藏开发效果提供理论指导与技术支持。

混合式抽水蓄能电站水温影响研究

【目的】常规电站水库水温结构的研究主要集中在气候变化的影响上,没有考虑抽水对水库水温结构的影响,且对抽水蓄能电站水库水温结构的研究也较少,为明晰抽蓄对水库水温的影响,【方法】以金沙江上游叶巴滩混合式抽水蓄能电站为研究对象,运用CE-QUAL-W2模型对水库表层水温、下泄水温和底层水温进行模拟,并分析混蓄前后上、下库的水温变化,以阐明混合式抽水蓄能水电站对水温的影响。【结果】结果显示:叶巴滩表层水温抽蓄前后最大变幅为5.8℃,底层水温最大变幅为2.8℃,下泄水温最大变幅为0.3℃;拉哇表层水温抽蓄前后最大变幅为1.8℃,底层水温最大变幅为0.1℃,下泄水温最大变幅为0.2℃。【结论】结果表明:(1)抽蓄对表层水温呈坦化作用,加强中层水体的掺混,增大底层水温的变幅;(2)混蓄对上下库水温的影响受到库区距离影响;(3)混蓄对上库叶巴滩和下库拉哇的下泄水温影响不显著。研究结果可为混合式抽水蓄能电站的水环境保护提供科学依据。

黏粒含量及含水比对红黏土强度特性的影响

【目的】为了研究不同黏粒含量及不同含水比对红黏土强度特性的影响,【方法】以桂林市红黏土为研究对象,根据净水沉降原理从所取红黏土中提取黏粒,以制备不同黏粒掺量的红黏土试样,进行固结不排水三轴试验,对不同黏粒含量及不同含水比红黏土强度指标、应力-应变关系进行分析。【结果】结果显示:随着黏粒掺量的增加,红黏土的黏聚力从30.57~37.71 kPa增加至43.36~59.66 kPa,而内摩擦角从18.0°~22.0°减少至14.5°~15.5°;随着含水比的增加,红黏土的黏聚力从37.71~59.66 kPa减少至30.57~43.36 kPa,而内摩擦角从14.5°~18.0°增加至15.5°~22.0°。【结论】结果表明:黏粒含量和含水比对红黏土的黏聚力和内摩擦角的影响明显,随着黏粒的不断添加,黏聚力会不断上升,内摩擦角会不断减小,但随着含水比的增大,黏聚力呈线性减小,内摩擦角呈线性增加,黏粒含量较含水比对黏聚力和内摩擦角的影响更为显著;黏粒含量和含水比对红黏土应力-应变关系曲线的影响基本属于硬化型,黏粒含量越大,硬化性越好,含水比越大,土体的主应力差越小,但土体的硬化性保存不变。可见,红黏土的抗剪强度指标受黏粒含量和含水比的影响明显。

滴灌条件下温室黄瓜土壤蒸发的影响因素及模拟

明确土壤蒸发的变化过程和影响因素,建立简便且具有较高精度的计算模型,对提高水资源利用效率,减少无效水分消耗具有重要指导意义。以温室滴灌黄瓜为试验对象,对土壤蒸发进行了连续测量,引入Priestley-Taylor模型的土壤蒸发模块,将蒸发过程分为2个阶段,探讨了3个限制系数(Deardorff,1977:f_(sw)-1;Yao et al.,2013:f_(sw)-2;Ershadi et al.,2014:f_(sw)-3)对模型精度的影响。结果表明:全生育期温室滴灌黄瓜的土壤蒸发在0.13~9.15 g/d之间变化,平均值为3.15 g/d,总体表现为“增加-降低-增加”的变化趋势;土壤蒸发与表层含水率和LAI均呈e指数函数关系,与含水率呈正比,与LAI呈反比;土壤蒸发系数与含水率变化过程相似,全生育期在0.49~1.26之间变化;3个模型在第2阶段的模拟精度较高,且f_(sw)-1的精度优于f_(sw)-2和f_(sw)-3,MAE和RMSE分别为0.14和0.21 mm/d。因此,采用PT-f_(sw)-1模型模拟滴灌条件下的土壤蒸发具有较高精度,可为精确掌握温室滴灌黄瓜水分消耗提供依据。

基于机器学习的新疆植被覆盖变化及其影响因子之间的关系

以新疆植被覆盖变化及其影响因子之间的关系作为研究对象,通过比较多元线性回归、随机森林回归、XGBoost、支持向量机回归,优选出精度最高的模型,依据最优模型计算出的属性重要程度,对15个影响因子进行重组并分析,探索了气温、降水量、辐射量、潜在蒸发、经度、纬度、高程、地貌类型、坡度、坡向、人类影响指数、径流、土壤类型、土壤湿度、植被类型15个影响因子对植被覆盖变化的影响。结果表明,XGBoost模型对归一化植被指数(NDVI)预测准确率最高,随机森林回归次之。在研究区内对NDVI影响程度最大的影响因子是土壤湿度、径流、植被类型、经度、潜在蒸发、气温、辐射量、地貌类型、降水量。在影响因素类型方面,气候条件因素对NDVI影响程度最大,土壤特征因素影响程度次之,地形地貌因素比前两者低。