用改进的前向神经网络预测铁路货运量

对影响铁路货运量的因素进行了分析。根据影响铁路货运量的诸因素的特点，介绍了一种改进的前向神经网络预测方法，并建立了铁路货运量前向神经网络预测模型。算例表明，其预测精度高于常规预测方法。     

Short-term speed predictions exploiting big data on large urban road networks

Big data from floating cars supply a frequent, ubiquitous sampling of traffic conditions on the road network and provide great opportunities for enhanced short-term traffic predictions based on real-time information on the whole network. Two network-based machine learning models, a Bayesian network and a neural network, are formulated with a double star framework that reflects time and space correlation among traffic variables and because of its modular structure is suitable for an automatic implementation on large road networks. Among different mono-dimensional time-series models, a seasonal autoregressive moving average model (SARMA) is selected for comparison. The time-series model is also used in a hybrid modeling framework to provide the Bayesian network with an a priori estimation of the predicted speed, which is then corrected exploiting the information collected on other links. A large floating car data set on a sub-area of the road network of Rome is used for validation. To account for the variable accuracy of the speed estimated from floating car data, a new error indicator is introduced that relates accuracy of prediction to accuracy of measure. Validation results highlighted that the spatial architecture of the Bayesian network is advantageous in standard conditions, where a priori knowledge is more significant, while mono-dimensional time series revealed to be more valuable in the few cases of non-recurrent congestion conditions observed in the data set. The results obtained suggested introducing a supervisor framework that selects the most suitable prediction depending on the detected traffic regimes.     

Parametrical studies of a new numerical model for controlled ship motions in extreme astern seas

Parametrical studies based on numerical simulations were carried out for very steep regular waves to assess possible improvements in the state-of-the art numerical modelling of the control and capsizing behaviour of ships in following and quartering seas. A nonlinear 6-DOF numerical model has been developed with the inclusion of frequency-dependent terms, the so called memory effects, and a flexible axis system that allows straightforward combination of seakeeping and manoeuvring models while accounting for extreme motions. The previously undertaken validation analyses using extensive model test data provided qualitatively good agreement, whereas the comparison with numerical models without coupling of the vertical motions and frequency-dependent hydrodynamic terms embodied in radiation forces identified improvements in the accuracy. However, to broaden the assessment of the numerical model, further parametrical numerical analyses were carried out using two ships, which had previously been tested in the validation analyses, for various operational and environmental conditions. These parameters were changed in accordance with the recommendations from international organisations and experience from model tests to realise and avoid dangerous conditions that often result in capsizing, such as broaching associated with surf riding and low-cycle resonance. As a result of the parametric analysis, we discuss the sensitivity of the improvements in the numerical model for various critical operational and design parameters and its possible use to provide a link between the ship's behavior and these parameters.     

混沌时间序列局域零阶预测法性能比较

利用计算机仿真比较了均值、距离加权和指数加权3种混沌局域零阶预测方法的预测精度、抗噪声及多步预测性能．在无噪声或噪声干扰较小时，距离加权预测法的性能最好；当噪声干扰较大时，指数加权预测法预测性能最优．指数加权与均值预测法几乎具有相同的多步预测能力，距离加权预测法的短期预测性能最佳．对于标准的离散混沌时间序列，3种预测方法多步预测误差达到一定值后，不再随预测步长的增加而增加；对于由连续系统抽样得到的混沌时间序列，多步预测误差呈现一定周期性变化．     

地表沉陷信息处理系统研究

建立了地表沉陷信息处理系统的理论模型，推导出地表下沉、倾斜变形、曲率变形、水平移动和水平变形等预报公式以及预报参数的解算公式及解释解。最后，利用观测站实测资料进行了验证，验证结果满足设计及工程要求。     

邮政汽车运输生产量的预测模型与程序设计研究

介绍了邮政汽车运输在采用计算机管理信息系统后，邮运量和邮件周转量的计算方法和技巧。采用温特（Winter）加法模型，建立了省一级邮政汽车运输系统中生产量的预测模型，并从生产实际的应用情况出发，研究了对这一模型的自适应校正方法。最后介绍了软件的设计要点，并给出了流程图和计算实例。     

用神经网络方法预测股票短期走势

在技术分析基础上，建立了神经网络的股票短期走势预测模型，对深圳和上海股票市场中的几种股票价格走势进行预测，取得了满意的结果。同时，还对神经网络的输入参数（开盘价、收盘价、最高价、最低价和成交量）的不同组合产生的预测结果进行了对比分析。     

基于多体动力学和有限元法的车体结构疲劳寿命仿真

提出一种多体动力学仿真和有限元法相互结合进行结构疲劳寿命预测的方法,并以机车车体结构为例进行了疲劳寿命计算。利用SIMPACK的多体仿真技术获得车体结构的动载荷历程;在ANSYS中利用准静态应力/应变分析法计算结构危险节点应力影响因子;根据模态分析技术确定车体结构固有频率和模态振型以及危险点位置。最后,基于动应力历程以及Palmigren-Miner损伤理论,利用FE-FATIGUE软件的基于应力的结构安全因子分析法对车体结构进行疲劳寿命预测,其中包括应力应变的循环计数、损伤预测和最终寿命估计。     

Broaching prediction in the light of an enhanced mathematical model,with higher-order terms taken into account

 We have attempted to develop a more consistent mathematical model for capsizing associated with surf-riding in following and quartering waves by taking most of the second-order terms of the waves into account. The wave effects on the hull maneuvring coefficients were estimated, together with the hydrodynamic lift due to wave fluid velocity, and the change in added mass due to relative wave elevations. The wave effects on the hydrodynamic derivatives with respect to rudder angles were estimated by using the Mathematical Modelling Group (MMG) model. Then captive ship model experiments were conducted, and these showed reasonably good agreements between the experiments and the calculations for the wave effects on the hull and the rudder maneuvring forces. It was also found that the wave effects on restoring moments are much smaller than the Froude–Krylov prediction, and the minimum restoring arm appears on a wave downslope but not on a wave crest amidship. Thus, an experimental formula of the lift force due to the heel angle of the ship is provided for numerical modelling. Numerical simulations were then carried out with these second-order terms of waves, and the results were compared with the results of free-running model experiments. An improved prediction accuracy for ship motions in following and quartering seas was demonstrated. Although the boundaries of the ship motion modes were also obtained with both the original model and the present one, the second-order terms for waves are not so crucial for predicting the capsizing boundaries themselves. Received: June 20, 2002 / Accepted: October 10, 2002 Acknowledgments. This research was supported by a Grant-in-Aid for Scientific Research of the Ministry of Education, Culture, Sports, Science and Technology of Japan (No. 13555270). The authors thank Prof. N. Rakhmanin of the Krylov Ship Research Institute for providing the Russian literature, as well as Mr. H. Murata of NHK (Japan Broadcasting Corporation) for translating it into Japanese. Address correspondence to: N. Umeda (e-mail: umeda@naoe.eng.osaka-u.ac.jp)     

机电产品价格的预测方法简介

依据历史价格,利用多元回归分析的方法预测机电产品的价格。在预测过程中,针对产品历史价格空缺年度较多的现象,引入价格指数,把产品价格预测的问题转化为对价格指数的预测;在回归分析中,以时间和价格政策作为价格指数的显著性因素。最后,讨论价格指数拟合值的性质。