随着电力市场和碳市场改革的不断深化,梯级水电在碳交易-电力交易(碳-电)耦合市场中的竞价问题亟待解决.如何在考虑自身非凸发电特性的同时解耦交易规则、应对市场不确定性是解决问题的关键.为此,构建了一种计及国家核证自愿减排量(CCER)市场供需不确定性的碳-电耦合市场下梯级水电站竞价的双层优化模型.上层模型以市场出清价为参数,决策梯级水电在电力市场和CCER市场中的发电计划和竞价曲线;下层模型以市场竞价信息为参数,计算不同时段、不同场景下的CCER市场出清结果.所提模型考虑了梯级水电非凸运行特性和碳-电耦合市场规则,采用多种可能的供需关系综合描述CCER市场的不确定性.运用强对偶理论、大M法、分段线性法等技术,将双层模型转化为混合整数线性规划模型并求解.以中国澜沧江干流两座梯级水电站为背景进行分析,结果表明考虑CCER市场不确定性可有效提高竞价计划的稳定性,耦合市场机制可引导发电企业根据两类市场需求调整发电计划,发电企业在提高自身收益的同时促进了系统资源合理配置.

中国云南省的电源结构以水电为主,水电消纳问题异常突出,且云南水电外送的受端调峰压力巨大.为充分挖掘云南电网"西电东送"通道潜力,最大限度消纳云南水电,同时响应多电网的调峰需求,文中构建了兼顾多电网调峰与水电消纳的跨流域梯级水电站调度的混合整数线性规划模型.模型采用目标权重法将多电网调峰问题转化成单目标问题,并通过调整各电网余荷距平绝对值的平均值权重,使得目标兼顾了多电网的调峰需求与水电消纳.同时,通过引入特殊顺序集约束,对最小持续时间约束提出了一种简易的线性化方法,此方法可应用于其他同类型的约束中.最后,以澜沧江干流与金沙江干流的18座水电站为例,验证了模型的可行性与有效性,结果表明所提方法可满足多电网调峰与水电消纳的需求.

间质性肺疾病(interstitial lung disease,ILD)是一组原因不同的慢性肺疾病,先天性表面活性物质功能缺陷是婴幼儿ILD的重要原因之一,在法国和英国分别占婴幼儿ILD的17.5％[1]和7.5％[2].编码肺表面活性物质蛋白(surfactant protein,SP)的基因(SFTP)及其相关代谢通路的基因,包括STPB、SFTPC、ATP结合盒转运子A3 (ATP-binding cassette transporter A3,ABCA3)出现突变,则会引起表面活性物质功能障碍,不仅可引起足月新生儿出生后即出现呼吸窘迫综合征(respiratory distress syndrome,RDS),也可引起儿童期ILD并出现持续性低氧血症等表现[3].现对2018年6月在首都医科大学附属北京儿童医院确诊的ABCA3基因单亲二倍体纯合突变导致的1例婴幼儿ILD患儿临床特点进行总结,并对ABCA3基因突变所致ILD的临床表型与基因型关系进行文献复习.

Objective:To summarize the clinical features of children with autosomal dominant hyper-IgE syndrome (AD-HIES) and the differential diagnosis of hyper-IgE syndrome and allergic diseases as well.Methods:All clinical data, including general information, clinical features, and genetic changes, from 7 children with AD-HIES who were diagnosed in Beijing Children′s Hospital Affiliated to Capital Medical University from April 2016 to June 2020 were analyzed retrospectively.The diagnostic criteria are based on the National Institutes of Health′s (NIH)′s hyper-IgE syndrome score and combined with the results of gene detection, shown as follows: (1) NIH score over 40, with signal transducer and activator of transcription 3 gene ( STAT3) pathogenic mutation; (2) NIH score between 20 and 40, with reported STAT3 pathogenic mutation; (3) NIH score less than 20 points was excluded. Results:There were 3 males and 4 females.The onset age of 7 cases was within 2 months after birth, and the mean age at diagnosis was 3 years old.All seven cases had recurrent skin or lung infections, with 4 cases having skin and lung infections, 1 case of skin abscesses at the BCG vaccination site, and 2 cases without skin infection suffering from recurrent pneumonia.The mean onset age of skin abscess in 5 cases was 1.5 years, and pus culture of 3 cases were Staphylococcus aureus.Four cases developed bullae and 6 cases had lung infections.Four cases had otitis media, and oral thrush was seen in 4 cases.One case of skin and lung infection developed liver abscess and sepsis.Seven cases had eczema, which was disco-vered in the neonatal period for 6 cases.Four cases had the symptoms of eczema for the first visit.Two cases had food allergy, and 1 case had recurrent wheezing within 1 year old.The serum IgE level and blood eosinophil count in 7 children were elevated.All children had heterozygous pathogenic mutations in STAT3.Six patients had de novo mutations.There were 6 different mutation sites.The 4 mutation sites were reported: c.1145G>A, c.1144C>T, and c. 1699A>G were missense mutations, and c. 1139+ 5G>A was splicing mutation.Two mutation sites had not been reported: c.1031A>C was missense mutation, and c. 2050G>T was nonsense mutation.The pathogenic grade of them were likely pathogenic, and the NIH score of 2 cases were above 40 score, which was consistent with the clinical diagnosis of hyper-IgE syndrome. Conclusions:Eczema is a common and early clinical manifestation of hyper-IgE syndrome, along with elevated IgE levels and eosinophil counts that need to be differentiated from allergic diseases.On the contrary, it often had recurrent skin abscesses or pneumonia, which was prone to bullae.The clinical manifestations of young children were atypical, and genetic testing was helpful for early diagnosis.

Objective:To investigate the etiology of pleural effusion in hospitalized children in Beijing Children′s Hospital.Methods:Clinical information of children with pleural effusion admitted to Beijing Children′s Hospital Affiliated to Capital Medical University from January 2016 to December 2018 was retrospectively analyzed.According to the etiology, the children were divided into infection group (parapneumonic pleural effusion, tuberculous pleurisy and empyema) and non infection group.According to the age, the children were further divided into ≤ 3 years old, >3-7 years old and > 7 years old groups.Classification of statistics was performed, and the etiology of pleural effusion were retrospectively analyzed.Results:Among the 1 165 children with pleural effusion, 746 cases(64.0%) were infected with pleural effusion, 697 cases (697/746, 93.4%) of who were parapneumonic effusion.In patients with parapneumonic effusion, 457 cases (61.3%) had Mycoplasma pneumonia (MP) infection.Infectious pleural effusion was more common in children >7 years old(339/479 cases, 70.8%), while non-infectious pleural effusion was prevalent in children under 3 years old(188/324 cases, 58.0%). The difference was statistically significant ( χ2=96.33, P<0.05). Among the patients with non-infectious pleural effusion, 239 cases (239/419 cases, 57.0%) had multi-system diseases and 97 cases (97/419 cases, 23.2%) had malignant pleural effusion.All the 18 deaths were non-infectious pleural effusion. Conclusions:The leading reason for pleural effusion in children is infection.The most prevalent symptom is parapneumonic effusion, which is mainly caused by MP.

目的 总结临床药师在呼吸科病房解答的部分药物咨询问题,为日后临床工作提供参考.方法 对2018年1月至12月间223例有临床药师参与、于我院呼吸科就诊并住院的患者用药相关问题进行回顾性分析.结果 223例问题来源按咨询问题者职称分布可分为主任医师15例(6.73％)、副主任医师26例(11.66％)、主治医师38例(17.04％)、住院医师124例(55.61％)、护士20例(8.97％).咨询问题内容按咨询药物种类分为抗菌药物107例(47.98％)、抗病毒药物14例(6.28％)、止咳平喘化痰药14例(6.28％)、保肝药10例(4.48％)、抗过敏药9例(4.04％)、抗凝药物7例(3.14％)、利胆药6例(2.69％)、其他类别56例(25.11％).咨询问题内容按咨询问题类别分为用法、用量类101例(45.29％)、药物选择类19例(8.15％)、药物不良反应类15例(6.73％)、其他类别占比88例(39.46％).结论 儿童呼吸科用药尤其是抗菌药物的合理用药是临床工作的重点.

目的 评价呼吸道病原体多重核酸检测方法在社区儿童获得性肺炎(community acquired pneumonia,CAP)诊断中的价值.方法 纳入2018年5月至12月于首都医科大学附属北京儿童医院住院的CAP患儿354例,采集痰液等标本,采用呼吸道病原体多重核酸检测方法,并分别评价其在肺炎支原体肺炎(mycoplasma pneumoniae pneumonia,MPP)和病毒性肺炎中诊断的灵敏度和特异性,以及分析其与七项呼吸道病原检测方法的一致性.结果 155例MPP组患儿,152例病原体核酸检测MP阳性,即灵敏度为98.1％;以病毒性肺炎和细菌性肺炎患儿为非MPP对照组,病原体核酸检测结果显示MP阳性4例,特异性为98.0％(195/199).158例病毒性肺炎患儿,病原体核酸检测出病毒141例,灵敏度为89.2％.以MPP和细菌性肺炎患儿为非病毒性肺炎对照组,196例患儿病原体核酸检测检出病毒39例(19.9％),特异性为80.1％.158例病毒性肺炎患儿中99份标本其病原体核酸检测结果与病毒抗原检测结果完全一致.59例不一致结果以病毒抗原检测为单一感染而病毒核酸检测为混合感染为主(30例),且主要来源于病毒抗原检测方法所不涵盖的其他病毒感染.结论 呼吸道病原体多重核酸检测在MPP诊断中具有较高的灵敏度和特异性;与传统的病毒抗原检测相比,虽存在一定的假阴性,但具有高通量检测的优势,可同时完成更多种病毒类型的检出;具有快速、灵敏度高等特点,可以为儿童呼吸道病原体的诊断以及鉴别诊断提供很好的证据.

目的 探讨儿童腺病毒肺炎后闭塞性细支气管炎的发生率以及相关危险因素.方法 收集2012年1月至2017年12月在首都医科大学附属北京儿童医院诊断为腺病毒肺炎的206例患儿的临床资料及随访资料,根据恢复期是否发生闭塞性细支气管炎分为闭塞性细支气管炎组和非闭塞性细支气管炎组,并对相关因素进行logistic回归分析.结果 47.57％(98/206)的腺病毒肺炎患儿在恢复期发生闭塞性细支气管炎.单因素分析发现,热程、急性期肺部喘鸣音、呼吸机使用时间、全身糖皮质激素开始使用时间、住院天数、既往喘息病史、急性期乳酸脱氢酶、发病年龄等8个指标两组间差异有显著性.多因素logistic回归分析显示有3个自变量纳入最佳回归方程,分别是呼吸机使用时间、既往喘息病史、急性期肺部喘鸣音,提示其为儿童腺病毒肺炎后闭塞性细支气管炎的独立危险因素.结论 在腺病毒肺炎患儿恢复期闭塞性支气管炎的发生率较高,早期预测其后遗症的发生非常重要.急性期应用呼吸机时间、肺部闻及喘鸣音以及既往有喘息病史的患儿易发生闭塞性细支气管炎.

我国已经初步建立起以中长期电能交易为主的多元电力市场,发电企业如何应对多尺度市场条件下多市场电价、履约耦合和市场风险成为亟待解决的关键问题.该文以水电占主导地位的云南电力市场为背景,提出了计及风险的多尺度电力市场条件下梯级水电月度发电计划制定方法.该方法从市场结构和结算规则出发,构建了收益–风险兼顾的发电计划模型,并引入广义析取规划及线性化技术将原始模型转化为混合整数线性规划模型予以求解.在澜沧江流域梯级水电站的实例应用表明,所提方法能够引导梯级水电响应市场价格变化,通过优化电量分配获取高额收益,并利用组合市场有效规避风险.该文为梯级水电应对新市场问题提供了一种有效解决途径.

To produce selectable marker-free (SMF) transgenic rice resistant to chewing insects, the Bacillus thuringiensis cryIA(c) gene (Bt) was introduced into two elite japonica rice varieties by using two Agrobacterium-mediated co-transformation systems. One system is with a single mini-twin T-DNA binary vector in one Agrobacterium strain, which consists of two separate T-DNA regions, one carrying the Bt while the other contains the selectable marker gene, hygromycin resistant gene (HPT). The other system uses two separate binary vectors in two separate Agrobacterium cultures, containing the Bt or HPT gene on individual plasmids. A lot of independent transgenic rice lines harboring both Bt and selectable marker genes were obtained. The results showed that the co-transformation frequency of the Bt gene and HPT gene was much higher by using the mini-twin T-DNA vector system (29.87%) than that by the two separate binary vector systems (4.52%). However, the frequency of the SMF transgenic rice plants obtained from the offspring of co-transgenic plants (21.74%) was lower for the mini-twin T-DNA vector system than that for the latter (50–60%). The data of ELISA implied that the expressed Bt proteins were quantitated as 0.025–0.103% of total leaf soluble proteins in the transgenic plant. Therefore, several elite transgenic rice lines, free of the selectable marker gene, were chosen. The results from both in vitro and in vivo insect bioassays indicated that the SMF transgenic rice was shown to be highly resistant to the striped stem borer and rice leaf folder. Moreover, in a natural field condition without any insecticide applied, all the transgenic rice plants were found to be not injured by the rice leaf folder, whereas the wild types were impaired seriously.

General object and activity recognition is a fundamental problem in computer vision, which has been the subject of much research. Traditional approaches include model-based and appearance template-based methods. Recently, inspired by methods from the text retrieval literature, local visual feature-based models have shown a lot of success for recognition of objects or activities with large within-class geometric variability. There are several challenges in this approach, namely feature selection and target modeling using these features. This thesis proposes a local-global visual feature-based framework for general object and activity recognition with novel methods for these problems: (1) Combinatorial and statistical methods for selecting informative parts to build statistical models for part-based object recognition. First a combinatorial optimization formulation is used for clustering on a weighted multipartite graph. Second, a statistical method for selecting discriminative parts from positive images is used to localize objects. (2) An entropy based vocabulary selection method for “bag-of-words” models for activity recognition. (3) Integrating both spatial and temporal information with appearance features for human activity recognition. This method models the human motions with the distribution of local motion features and their spatial-temporal arrangements. The effectiveness of the proposed methods is demonstrated by several object recognition and activity recognition data sets, which include human facial expressions and hand gestures, etc. This thesis also covers an interesting project regarding a framework of applying Discrete Fourier Transform to detect salient regions in images and video sequences. This framework generalizes the previous saliency detection methods and can be applied for saliency detection in the video sequences.

This paper presents a spatiotemporal pyramid representation for recognizing facial expressions and hand gestures. This approach works by partitioning video sequence into increasingly fine subdivisions in the space and time domains and modeling the distribution of the local motion features inside each subdivision such that the set of motion features are mapped into spatial and temporal multi-resolution histograms. This spatiotemporal pyramid is built by weighting the histograms from the different layers of the subdivisions. The proposed approach is an extension of the orderless "bag-of-words" model by approximately capturing geometric and temporal arrangements of the local motion features. The experiments on facial expression and hand gesture data sets have demonstrated the significantly improved performance over state of art results on human activity recognition tasks by using our representation.

This paper presents an approach for human action recognition by finding the discriminative key frames from a video sequence and representing them with the distribution of local motion features and their spatiotemporal arrangements.In this approach, the key frames of the video sequence are selected by their discriminative power and represented by the local motion features detected in them and integrated from their temporal neighbors.In the key frame's representation, the spatial arrangements of the motion features are captured in a hierarchical spatial pyramid structure.By using frame by frame voting for the recognition, experiments have demonstrated improved performances over most of the other known methods on the popular benchmark data sets.

This paper presents an approach for human activity recognition by representing the frames of the video sequence with the distribution of local motion features and their spatiotemporal arrangements. In this approach, the local motion features used for the representation of a frame are integrated from the ones detected in this frame and its temporal neighbors. The features' spatial arrangements are captured in a hierarchical spatial pyramid structure. By using frame by frame voting for the recognition, experiments have demonstrated improved performances over most of the other known methods on the popular benchmark data sets while approaching the best known results.

In object recognition tasks, where images are represented as constellations of image patches, often many patches correspond to the cluttered background. In this paper, we present a two-stage method for selecting the image patches which characterize the target object class and are capable of discriminating between the positive images containing the target objects and the complementary negative images. The first stage uses a combinatorial optimization formulation on a weighted multipartite graph. The following stage is a statistical method for selecting discriminative patches from the positive images. Another contribution of this paper is the part-based probabilistic method for object recognition, which uses a common reference frame instead of reference patch to avoid possible occlusion problems. We also explore different feature representation using principal component analysis (PCA) and 2D PCA. The experiment demonstrates our approach has outperformed most of the other known methods on a popular benchmark dataset while approaching the best known results.

In order to obtain marker-free transgenic rice with improved disease resistance, the AP1 gene of Capsicum annuum and hygromycin-resistance gene (HPT) were cloned into the two separate T-DNA regions of the binary vector pSB130, respectively, and introduced into the calli derived from the immature seeds of two elite japonica rice varieties, Guangling Xiangjing and Wuxiangjing 9, mediated by Agrobacterium-mediated transformation. Many cotransgenic rice lines containing both the AP1 gene and the marker gene were regenerated and the integration of both transgenes in the transgenic rice plants was confirmed by either PCR or Southern blotting technique. Several selectable marker-free transgenic rice plants were subsequently obtained from the progeny of the cotransformants, and confirmed by both PCR and Southern blotting analysis. These transgenic rice lines were tested in the field and their resistance to disease was carefully investigated, the results showed that after inoculation the resistance to either bacterial blight or sheath blight of the selected transgenic lines was improved when compared with those of wild type.

In an object recognition task where an image is represented as a constellation of image patches, often many patches correspond to the cluttered background. If such patches are used for object class recognition, they will adversely affect the recognition rate. In this paper, we present a two stage method for selecting image patches which characterize the target object class and are capable of discriminating between the positive images containing the target objects and the complementary negative images. The first stage selection is done using a novel combinatorial optimization formulation on a weighted multipartite graph representing similarities between images patches across different instances of the target object. The following stage is a statistical method for selecting those images patches from the positive images which, when used individually, have the power of discriminating between the positive and negative images in the evaluation data. The individual methods have a performance competitive with the state of the art methods on a popular benchmark data set and their sequential combination consistently outperforms the individual methods and most of the other known methods while approaching the best known results.

The field of computer vision witnesses recently a great interest focused on solving the generic object recognition problem. Traditionally, object recognition has been at the center of the computer vision field. The explosion of digital imaging and digital video that we are witnessing makes a huge urgent demand for systems and applications that are able to understand images and videos at the semantic level. The goal of this reading seminar is to catch up with the state of the art in this field and understand the achievements and challenges towards solving the generic object recognition problem.

In an object recognition task where an image is represented as a constellation of image patches, often many patches correspond to the cluttered background. If such patches are used for object class recognition, they will adversely affect the recognition rate. In this paper, we present a statistical method for selecting the image patches which characterize the target object class and are capable of discriminating between the positive images containing the target objects and the complementary negative images. This statistical method select those images patches from the positive images which, when used individually, have the power of discriminating between the positive and negative images in the evaluation data. Another contribution of this paper is the part-based probabilistic method for object recognition. This Bayesian approach uses a common reference frame instead of reference patch to avoid the possible occlusion problem. We also explore different feature representation using PCA an 2D PCA. The experiment demonstrates our approach has outperformed most of the other known methods on a popular benchmark data set while approaching the best known results.

The Principal Component Analysis (PCA) is a useful statistic technique that has found applications in fields such as recognition, classification and image data compression. It is also a common technique in extracting features from data in a high dimensional space. By linearly transforming the images into eigenspace, we project the images into a new N Dimensional space, which exhibits the properties of the samples most clearly along the coordinate axes. The most significant features /information of the images will be in the first few principal components. We can do image data compression by only keeping the first few principal components. In this project, we will show that we can compress the walking people images and later reconstruct the original images without much loss from the reduced data.