基于同方差的递增时间序列自回归预测的改善方法

提出了"利用反双曲正弦函数变换提高数据列光滑程度"的新结论,获得了递增时间序列改善的自回归预测新方法。     

Study of automatic designing of line heating technique parameters

1Introduction Lineheatingisashellplateheatprocessingmethod usedinshipbuildingproduction.Itisacomplexand difficulttechnique,mainlyusedindoublebending shellplateforming.Becausetherearemanyinfluence factorsinplateforming,theinfluencedisciplineis hardtopredic…     

半潜式起重船波浪载荷预报

采用数值分析软件Sesam/HydroD对某半潜式起重船在航行状态和起重作业状态下的波浪载荷进行长期预报和短期预报.在航行状态下,预报得到的最大波浪弯矩和剪力分别为规范值的1.4倍和2.0倍,用于进行结构疲劳评估(超越概率为10-2)的波浪弯矩最大预报值约为用于进行结构强度评估(超越概率为10-8)的波浪弯矩最大预报值...     

混凝土冻融破坏机理分析及寿命预测

为提高高寒地带混凝土的耐久性,对混凝土的冻融破坏机理进行了分析,并对受到冻融破坏后的混凝土寿命预测方法进行了简介。分析表明,目前混凝土冻融破坏应用比较成熟的理论为静水压理论和渗透压理论,静水压理论和渗透压理论的主要区别就在于未冻水的迁移方向。混凝土抗冻性影响因素主要有水胶比、含气量、粗集料和掺合料。每种影响因素作用的机理不同,但都共同影响着混凝上在冻融环境下的耐久性。既有研究建立的混凝土冻融寿命预测方面还仅停留于理论层面,能够指导工程实践应用的寿命预测模型仍有待进一步研究。     

上海市防汛能力建设调查及评价

上海位于长江入海口,低洼的地势使得上海遭受台风、暴雨、高潮、洪水(俗称“四碰头”)的袭击后易成灾,多年来上海已建设抵御灾害的防汛体系,且发挥了积极的作用。在对上海防汛能力建设体系开展调查的基础上,分析存在问题和不足。同时,结合科学的层次分析法理论构建评价指标体系,为后续防汛能力建设的进一步优化提供依据。     

基于偏最小二乘回归的疏浚产量预测方法

针对绞吸式挖泥船产量预测困难的问题,对挖泥船作业实时反馈的数据进行研究。利用Relief权重算法提取出影响挖泥船产量的主要工艺参量,并在此基础上采用偏最小二乘回归,建立主要工艺参量与产量之间的数学模型,实现对挖泥船产量的预测。结果表明,利用偏最小二乘回归建立的数学模型能够很好地对挖泥船的产量进行预测,可为预测挖泥船的产量提供一种有效的方法。     

Assessment of factors associated with travel time reliability and prediction: an empirical analysis using probabilistic reasoning approach

Significant efforts have been made in modeling a travel time distribution and establishing measures of travel time reliability (TTR). However, the literature on evaluating the factors affecting TTR is not well established. Accordingly, this paper presents an empirical analysis to determine potential factors that are associated with TTR. This study mainly applies the Bayesian Networks model to assess the probabilistic association between road geometry, traffic data, and TTR. The results from this model reveal that land use characteristics, intersection factors, and posted speed limits are directly associated with TTR. Evaluating the strength of the association between TTR and the directly related variables, the log odds ratio analysis indicates that the land use factor has the highest impact (0.83) followed by the intersection factor (0.57). The findings from this study can provide valuable resources to planners and traffic operators in their decision-making to improve TTR with quantitative evidence.     

基于LSTM网络的驾驶意图识别及车辆轨迹预测

自动驾驶汽车需具备预测周围车辆轨迹的能力,以便做出合理的决策规划,提高行驶安全性和乘坐舒适性。运用深度学习方法,设计了一种基于长短时记忆（LSTM）网络的驾驶意图识别及车辆轨迹预测模型,该模型由意图识别模块和轨迹输出模块组成。意图识别模块负责识别驾驶意图,其利用Softmax函数计算出驾驶意图分别为向左换道、直线行驶、向右换道的概率;轨迹输出模块由编码器-解码器结构和混合密度网络（MDN）层组成,其中的编码器将历史轨迹信息编码为上下文向量,解码器结合上下文向量和已识别的驾驶意图信息预测未来轨迹;引入MDN层的目的是利用概率分布来表示车辆未来位置,而非仅仅预测一条确定的轨迹,以提高预测结果的可靠性和模型的鲁棒性。此外,将被预测车辆及其周围车辆组成的整体视为研究对象,使模型能够理解车-车间的交互式行为,响应交通环境的变化,动态地预测车辆位置。使用基于真实路况信息的NGSIM（Next Generation SIMulation）数据集对模型进行训练、验证与测试。研究结果表明：与传统的基于模型的方法相比,基于LSTM网络的轨迹预测方法在预测长时域轨迹上具有明显的优势,考虑交互式信息的意图识别模块具备更高的预判性和准确率,且基于意图识别的轨迹预测能降低预测轨迹与真实轨迹间的均方根误差,显著提高轨迹预测精度。     

Time taken for residents to adopt a new public transport service: examining heterogeneity through duration modelling

When a new public transport service is introduced it would be valuable for public authorities, financing organisations and transport operators to know how long it will take for people to start to use the service and what factors influence this. This paper presents results from research analysing the time taken for residents living close to a new guided bus service to start to use (or adopt) the service. Data was obtained from a sample of residents on whether they used the new service and the number of weeks after the service was introduced before they first used it. Duration modelling has been used to analyse how the likelihood of starting to use the new service changes over time (after the introduction of the service) and to examine what factors influence this. It is found that residents who have not used the new service are increasingly unlikely to use it as time passes. Those residents gaining greater accessibility benefits from the new service are found to be quicker to use the service, although the size of this effect is modest compared to that of other between-resident differences. Allowance for the possibility that there existed a proportion of the sample that would never use the new service was tested using a split population model (SPD) model. The SPD model indicates that 36% of residents will never use the new service and is informative in differentiating factors that influence whether Route 20 is used and when it is used.

Estimating the robustness of questionnaire results: lessons from a mixed-mode survey of expectations for tele-working and road-based business travel

The robustness of questionnaire results to various forms of bias are explored in the context of a dual-mode (web and hardcopy) survey of employers’ anticipations of levels of employee commuting and business travel activity under a range of future ICT scenarios. The questionnaire incorporated several innovative features which, together with the dual-mode format, allowed an unusually wide range of analyses. For example: the robustness of respondents’ opinions was tested by examining the effect of incorporating alternative versions of a briefing text, one being very positive and one very negative, about the role of ICT; instrument bias was identified via detailed comparison of the results from the two versions of the questionnaire; and the impact of exogenous factors which are often ignored or taken as constant was assessed via special supplementary questions. Analysis showed that the robustness of opinions and expectations varied and was influenced by respondent characteristics, and that results from the two versions of the questionnaire differed significantly. It is concluded that opinions and expectations are less robust, and questionnaire results are more subject to bias and myopic interpretation, than is generally recognised and that web-based surveys seem particularly vulnerable to sampling bias. Methods are suggested for measuring robustness, for reducing bias and for validating and contextualising results. The use of contrasting briefing texts is recommended as a means of establishing the robustness of opinions and expectations while supplementary questions are recommended for validating and contextualising SP and SE exercises.