Remaining Useful Life Prediction Method for Multi-Component System Considering Maintenance:Subsea Christmas Tree System as A Case Study

Maintenance is an important technical measure to maintain and restore the performance status of equipment and ensure the safety of the production process in industrial production,and is an indispensable part of prediction and health management.However,most of the existing remaining useful life(RUL)prediction methods assume that there is no maintenance or only perfect maintenance during the whole life cycle;thus,the predicted RUL value of the system is obviously lower than its actual operating value.The complex environment of the system further increases the difficulty of maintenance,and its maintenance nodes and maintenance degree are limited by the construction period and working conditions,which increases the difficulty of RUL prediction.An RUL prediction method for a multi-omponent system based on the Wiener process considering maintenance is proposed.The performance degradation model of components is established by a dynamic Bayesian network as the initial model,which solves the uncertainty of insufficient data problems.Based on the experience of experts,the degree of degradation is divided according to Poisson process simulation random failure,and different maintenance strategies are used to estimate a variety of condition maintenance factors.An example of a subsea tree system is given to verify the effectiveness of the proposed method.

Nitrogen removal influence factors in A/O process and decision trees for nitrification/denitrification system

In order to improve nitrogen removal in anoxic/oxic(A/O) process effectively for treating domestic wastewaters, the influence factors, DO(dissolved oxygen), nitrate recirculation, sludge recycle, SRT(solids residence time), influent COD/TN and HRT(hydraulic retention time) were studied. Results indicated that it was possible to increase nitrogen removal by using corresponding control strategies, such as, adjusting the DO set point according to effluent ammonia concentration; manipulating nitrate recirculation flow according to nitrate concentration at the end of anoxic zone. Based on the experiments results, a knowledge-based approach for supervision of the nitrogen removal problems was considered, and decision trees for diagnosing nitrification and denitrification problems were built and successfully applied to A/O process.

Cost of Multicast Logical Key Tree Based on Hierarchical Data Processing

How to design a multicast key management system with high performance is a hot issue now. This paper will apply the idea of hierarchical data processing to construct a common analytic model based on directed logical key tree and supply two important metrics to this problem： re-keying cost and key storage cost. The paper gives the basic theory to the hierarchical data processing and the analyzing model to multieast key management based on logical key tree. It has been proved that the 4-ray tree has the best performance in using these metrics. The key management problem is also investigated based on user probability model, and gives two evaluating parameters to re-keying and key storage cost.

Application of Decision Tree Algorithm in Stamping Process

Various process parameters exert different effects in stamping process. In order to study the relationships among the process parameters of box stamping process, including the blank holder force, friction coefficient, depth of drawbead, offset and length of drawbead, the decision tree algorithm C4.5 was performed to generate the decision tree using the result data of the box stamping simulation. The design and improvement methods of the decision tree were presented. Potential and valuable rules were generated by traversing the decision tree, which plays an instructive role on the practical design. The rules show that the correct combination of blank holder force and setting of drawbead are the dominant contribution for controlling the cracking and wrinkling in box stamping process. In order to validate the rules, the stamping process for box was also performed. The experiment results show good agreement with the generated rules.

Finite family trees of continuous time birth and death processes for evaluating the transmitting speed of information on communication networks

A finite random graph generated by continuous time birth and death processes with exponentially distributed waiting times was investigated, which is similar to a communication network in daily life. The vertices are the living particles, and directed edges go from mothers to daughters. The size of the communication network was studied. Furthermore, the probability of successfully connecting senders with receivers and the transmitting speed of information were obtained.

An Improved Treed Gaussian Process

Many black box functions and datasets have regions of different variability. Models such as the Gaussian process may fall short in giving the best representation of these complex functions. One successful approach for modeling this type of nonstationarity is the Treed Gaussian process [1], which extended the Gaussian process by dividing the input space into different regions using a binary tree algorithm. Each region became its own Gaussian process. This iterative inference process formed many different trees and thus, many different Gaussian processes. In the end these were combined to get a posterior predictive distribution at each point. The idea was that when the iterations were combined, smoothing would take place for the surface of the predicted points near tree boundaries. We introduce the Improved Treed Gaussian process, which divides the input space into a single main binary tree where the different tree regions have different variability. The parameters for the Gaussian process for each tree region are then determined. These parameters are then smoothed at the region boundaries. This smoothing leads to a set of parameters for each point in the input space that specify the covariance matrix used to predict the point. The advantage is that the prediction and actual errors are estimated better since the standard deviation and range parameters of each point are related to the variation of the region it is in. Further, smoothing between regions is better since each point prediction uses its parameters over the whole input space. Examples are given in this paper which show these advantages for lower-dimensional problems.

Learned Index和B-Tree在不同分布数据上的性能对比及优化

Learned Index是一种通过训练模型来建立输入数据和存储位置之间映射关系的索引,它能学习到数据间分布的信息,而不同的数据分布将影响模型训练准确率和模型复杂度之间的平衡。为了探索Learned Index适用的场景,使用不同分布、不同数据量的数据对它和加以优化的可更新的自适应学习索引(ALEX)进行性能测试,并与B-Tree进行对比,最终发现Learned Index构建大批量数据的索引时间比B-Tree短,读操作性能、存储空间大小有明显的优势,但写操作性能较差,因此得出Learned Index更适用于大数据情景下的在线分析处理(OLAP)数据库,用于静态数据的存储和查询操作的结论。基于B-Tree的索引结构,对初版Learned Index的结构进行了优化和调整,最终使优化后Learned Index在大批量数据的读写操作性能上有明显提高,其中读操作最高达到原版Learned Index的2倍,写操作最高达到原版的3倍。

基于藤蔓树降维剪枝的多部件系统剩余寿命预测

随着多部件系统日益复杂化和智能化,系统中多个部件间协同工作,其退化的相互影响往往不可忽略。因此,在考虑多部件系统部件间相互影响的基础上,提出了一种构造藤蔓树降维剪枝二元Copula函数来对高维变量间相关结构进行建模的剩余寿命预测方法。首先通过Lévy过程来建模多部件系统的退化模型,利用Akaike信息准则进行Copula函数的最优遴选。其次通过Copula函数来表征高维多变量下两两部件间的双向随机相关性,采用藤蔓树降维剪枝的方法来对高维变量间相关结构进行降维,并进行多部件系统部件的实时剩余寿命预测建模;接着考虑单个部件异质性的影响下,利用贝叶斯参数估计和极大似然估计结合的方法来实现对随机模型以及随机超参数的动态更新。最后通过商用模块化航空推进系统仿真涡轮发动机模型来验证所提出模型的合理性和有效性。

江汉平原引种45年生黑杨派品种生长过程比较

以江汉平原引种的3个45年生黑杨派品种‘I-63’、‘I-69’和‘I-72’为研究对象,通过截取不同高度的圆盘进行树干解析,比较不同品种胸径、树高及材积生长过程,旨在为江汉平原黑杨大径材培育提供数据和技术支撑。结果表明,这3个品种的去皮胸径分别达到58.35、61.70、53.60 cm,树高分别为24.84、22.14、27.79 m,去皮材积分别为3.92、3.93、3.11 m^(3)。3个品种的胸径连年生长量最大值分别出现在第5年、第6年、第5年,树龄超过17年后胸径生长明显放缓;树高连年生长量最大值分别出现在第8年、第4年、第6年,树龄超过20年后树高生长接近峰值;材积年平均生长量最大值分别出现在第18年、第15年、第16年。对应3个品种的材积成熟龄,统计发现,‘I-63’、‘I-69’、‘I-72’材积速生期分别在第8~18、6~15、7~16年。整体来看,3个品种都具有较长的寿命和生长期,在江汉平原表现出极好的适应性,相对来说‘,I-69’干型较好,生长表现更为优良,推广价值更高。

基于MBD的高层建筑施工工艺结构化表征

针对高层建筑施工过程中的庞大内容和复杂的组织关系,本文首先深入分析了二维纸质工艺图纸和文件作为技术交底形式所带来的抽象性、理解困难、规范性差及数据一致性难以保障等问题。基于对高层建筑施工工艺特性和信息集成需求的分析,本文规范化定义了施工工艺的要素和属性,提出了基于工艺结构树的施工工艺信息组织管理方法,并构建了多层级工艺信息模型以实现工艺信息的结构化表达。借鉴制造业的MBD理念,本文以工艺结构树为数据组织核心,结合BIM模型与工艺信息卡片,提出了集成几何设计与施工工艺信息的三维施工工艺结构化表征框架。最后,通过开发交互式作业指导系统并结合实际案例验证,完成了项目主体、工艺和各类资源的有序整合,为智能建造提供了理论与方法支持。

茶树精油微胶囊包埋工艺优化及表征

为抑制茶树精油挥发,增强其缓释效果,以茶树精油为芯材、海藻酸钠为壁材,采用离子凝胶法制备茶树精油微胶囊,在明确了海藻酸钠质量分数、吐温-80体积分数、氯化钙质量分数及芯壁质量比等单因素对包埋率影响的基础上,采用响应面法对其制备工艺进行优化。通过扫描电子显微镜(SEM)、傅里叶红外光谱(FTIR)及X-射线衍射仪(XRD)对茶树精油微胶囊进行表征,并对其缓释性能进行测定。结果表明:茶树精油微胶囊的最佳制备工艺条件为海藻酸钠质量分数1.5%、吐温-80体积分数0.2%、氯化钙质量分数1.4%、芯壁质量比2∶1,在此条件下制备的茶树精油微胶囊外形圆润、无拖尾、强度高、形态饱满,包埋率可达79.12%,同时具有良好的缓释效果;FTIR及XRD分析结果显示海藻酸钠通过静电作用与茶树精油结合并将其成功包埋,起到保护作用。本研究所采用的微胶囊制备工艺简单、包埋效果好,极大地提高了茶树精油的稳定性。

考虑综合性能最优的非短视快速天基雷达多目标跟踪资源调度算法

合理有效的资源调度是天基雷达效能得以充分发挥的关键。针对天基雷达多目标跟踪资源调度问题,建立了综合考虑目标威胁度、跟踪精度与低截获概率(LPI)的代价函数;考虑目标的不确定、天基平台约束以及长远期期望代价,建立了多约束下的基于部分可观测的马尔可夫决策过程(POMDP)的资源调度模型;采用拉格朗日松弛法将多约束下的多目标跟踪资源调度问题转换分解为多个无约束的子问题;针对连续状态空间、连续动作空间及连续观测空间引起的维数灾难问题,采用基于蒙特卡罗树搜索(MCTS)的在线POMDP算法—POMCPOW算法进行求解,最终提出了一种综合多指标性能的非短视快速天基雷达多目标跟踪资源调度算法。仿真表明,与已有调度算法相比,所提算法资源分配更合理,系统性能更优。

具有串并行异类工序约束的多柔性车间联合调度

为了减少具有串并行异类工序约束多车间联合调度的总延期时间,提出了基于知识牵引遗传算法的调度求解方法。首先,采用扩展工艺树描述串并行异类工序约束,并基于无向图描述机器在多车间的分布;针对染色体初始化和进化过程中的扩展工艺树约束,定义了紧前工序数和剩余紧前工序数的概念,基于剩余紧前工序数设计了染色体初始化和进化方法;为了提高遗传算法的进化能力,将种群进化能力和最优个体进化能力作为知识,用于牵引算法的进化方式和方向,从而提出了知识牵引遗传算法的求解方法。经实验验证,知识牵引遗传算法调度的总延期时间均值最小,为30.8 h,说明该算法在多车间调度中具有最好的优化性能;且总延期时间盒须图长度最小,说明知识牵引遗传算法的稳定性也较好。

因天气影响的航空器绕飞管制安全评估及策略分析

为降低航空器在绕飞情况下管制指挥事故的风险,采用事故树分析法(fault tree analysis, FTA)和层次分析法(analytic hierarchy process, AHP)相结合的方式对管制员在进行绕飞管制时发生指挥事故建立综合风险评估模型并制定风险管控措施,以及验证所提管控措施的有效性。首先对各风险因素进行逐层分析,找出其关联性,建立绕飞管制指挥事故的事故树模型。以此为基础,结合层次分析法,确定各层风险指标权重。同时依据各指标总权重的排序结果,提出了在绕飞情况下管制指挥事故的管控措施,最后以长沙区域管制空域为实例对所提方案进行验证。结果表明:在本文方案下虽然雷雨、风切变、协调程序过于复杂和航空器偏离计划航路仍然是导致绕飞管制事故的重要因素,但其重要程度下降了6.2%,且成熟管制员人数不足、扇区流量大和预案频繁发生变化的状况得到有效解决。证明了该方案能够有效抑制导致绕飞管制指挥事故发生的事件的重要程度,减少因极端天气和扇区负荷过大而出现的安全风险事故,整体提高管制系统的稳定性和可靠性。

高铁枢纽与城市轨道交通换乘流线仿真与优化

针对高铁枢纽与城市轨道交通换乘流线优化问题,以南京南站作为研究对象,建立全过程仿真模型,识别换乘瓶颈,采用梯度提升决策树法(Gradient Boosting Decision Tree,GBDT)确定参数重要度,提出改善方案.首先,分解换乘过程的行人与设施流线,分析高速铁路到达客流与城市轨道交通客流的分布特征.其次,使用AnyLogic软件建立高铁枢纽换乘城市轨道交通的全过程仿真模型,分析现状仿真结果,识别空间瓶颈.然后,设计不同优化类型下的措施参数及调整范围,形成不同参数组合方案.采取梯度提升决策树算法,识别不同措施参数的相对重要度,并据此确定改善措施的优先级.最后,依据措施的优先级,确定不同类型下的优化组合方案,选择机器学习全局可解释性方法对其进行优化效果分析,为不同场景下的服务改善提出建议.研究结果表明:换乘瓶颈主要集中于楼/扶梯通道设施以及闸机、售检票机等服务设施处;乘客换乘城轨的购票比例对平均换乘时间和单位时间最大换乘人数均起到重要影响,对于平均换乘时间,城轨自动售票机数量、购票时间、城轨进站服务时间、城轨进站闸机数量的影响相对较大,对于单位时间最大换乘人数,购票时间与城轨进站闸机服务时间的影响相对较大.为提高高铁枢纽换乘效率,建议推广电子客票和多种支付方式,优化购票及检票设施.

基于CART决策树的调度算法研究

以往的作业车间存在大量的离线加工数据。基于数据挖掘、调度规则和算法优化相关知识,提出基于贝叶斯优化的改进CART算法来对车间数据挖掘利用,根据车间数据的属性逐步划分节点,生成树状结构,剪枝,最后生成加工规则。通过不同数据算例的实验结果表明,经过贝叶斯优化后的CART算法相较于传统CART算法提高了对数据划分的能力并提升了生成的决策树的准确度。

拐枣汁加工工艺优化及质量检测

为了研发一款风味独特、营养丰富的拐枣汁,以感官评价为指标,通过单因素实验优化了拐枣汁加工工艺,分别研究了出汁率、脱涩和澄清等关键工艺,在此基础上,以拐枣汁为原材料,白砂糖、柠檬酸及食用盐为辅料做调配实验,通过正交试验进行配方优化。确定最佳配比是添加白砂糖3%,柠檬酸0.10%,食用盐0.03%;确定提高出汁率果胶酶的最佳添加量为0.09%,脱涩单宁酶的最佳添加量为0.15%,澄清拐枣汁所需的明胶最佳添加量为1.2%。此时产品澄清透明,酸甜适宜,拐枣风味浓郁,稳定性最好。

基于故障树的迈腾发动机节气门体实验研究

针对电子节气门体故障引发的迈腾发动机转速不能跟进的故障现象,通过构建符合实际控制逻辑的故障树分析电子节气门体的实验过程。首先根据ETC控制逻辑和GX3电路原理剖析了从体现驾驶员意愿的踏板位置信号传输给J623到G187和G188采集信号向J623反馈节气门开度的整个闭环控制过程,然后建立“迈腾B8L2.0T发动机转速不能跟进”故障树。再依据底事件重要度,采用发动机分析仪FSA740等依次对G186驱动信号、G187和G188共用工作电压、信号电压、信号波形进行测试,最后通过分析波形图,提炼出了占空比信号控制特点、J623提供5 V电源的基本条件、信号电压成互补的方式输出等。实验研究表明:基于故障树选定的节气门体测试项目数据能客观反映电子节气门内在运行规律,为准确判断G186、G187和G188的具体故障部位提供真实的数据支撑。

不同品种和工艺条件对白茶品质形成机制的影响研究进展

白茶属微发酵茶,不同茶树品种制作工艺白茶鲜醇可口和花香浓郁的品质有所差异,在不同工艺条件下,茶中多酚类物质、蛋白质、游离氨基酸、生物碱和可溶性糖等内含物质发生不同程度变化,导致白茶外形、色泽及冲泡后汤色、滋味和香气发生不同程度的变化,进而赋予了独特的品质特征。白茶的品质形成十分复杂,其成分的组成、含量及所表现的香气特征受茶树品种和工艺条件的不同而有所差异。鉴于此,为推动工艺白茶产业的发展,挖掘白茶品质形成机制,本文综述了近几年白茶在萎凋和干燥过程中品质形成机制,分析了不同茶树品种和工艺条件对白茶品质影响原因并进行了展望,旨在为提升白茶品质、优化白茶加工工艺及利用和选育更多不同茶树品种供理论依据。

茶树花的品质成分及功效研究进展

茶树花是茶叶生产产生的附加产物,含有丰富的营养成分,具有多种保健功能。为提高茶树花资源利用效率,本文综合文献资料,总结分析茶树花的主要成分及其功效,其中成份包括水溶性糖、茶皂素、蛋白质、氨基酸、黄酮类化合物和超氧化物歧化酶等,讨论不同茶树花之间有效部分的差异及其潜在的研究应用价值,包括水浸出物、茶多糖、茶氨酸和茶多酚,并对不同茶树花在食品领域、日用领域和医疗领域的选择性深度生产加工进行展望。