Ship motion extreme short time prediction of ship pitch based on diagonal recurrent neural network

A DRNN （diagonal recurrent neural network） and its RPE （recurrent prediction error） learning algorithm are proposed in this paper . Using of the simple structure of DRNN can reduce the capacity of calculation. The principle of RPE learning algorithm is to adjust weights along the direction of Gauss-Newton. Meanwhile, it is unnecessary to calculate the second local derivative and the inverse matrixes, whose unbiasedness is proved. With application to the extremely short time prediction of large ship pitch, satisfactory results are obtained. Prediction effect of this algorithm is compared with that of auto-regression and periodical diagram method, and comparison results show that the proposed algorithm is feasible.     

基于串联灰色神经网络的船用柴油机状态预测研究

文章首先对灰色预测模型进行了改进,建立了GM（1,1）综合预测模型;其次针对灰色理论和神经网络的互补性,创建了以等维新息GM（1,1）模型＋BP神经网络的串联型灰色神经网络预测方法,在此基础上提出了3＊3＊1＊1结构的预测模型,并对其建模过程和算法进行了分析讨论;最后以某型船用柴油机铁谱监测数据为样本,对这种新预测模型的可行性与准确性进行了验证。     

港口集装箱吞吐量的灰色神经网络预测模型研究

为了降低港口集装箱吞吐量的预测误差,提高预测精度,文章通过分析传统的灰色预测模型和 BP 神经网络预测模型的优缺点,构建了灰色神经网络港口集装箱吞吐量预测模型,该模型充分发挥了灰色模型所需初始数据少和 BP 神经网络非线性拟合能力强的特点。以实际数值作为初始数据,各种灰色模型的预测值为神经网络的输入值,神经网络的输出值为组合预测结果。通过实例分析,结果表明：灰色神经网络预测模型提高了预测精度,预测结果比较理想,优于单一预测模型,因此,该模型用于港口集装箱吞吐量预测是可行的、有效的。     

Numerical prediction of wave-induced long-term extreme load effects in a flexible high-speed pentamaran

The theory and procedure established by Wu and Moan in 1996 and 2005 and Wu and Hermundstad in 2002 were applied to a high-speed transatlantic pentamaran containership. Nonlinear time-domain simulations of ship motions and load effects were carried out in different sea states. The simulated responses were validated against model tests with satisfactory results. The short-term probabilities of exceedance were estimated by using different stochastic analysis procedures. The long-term probabilities of exceedance were obtained based on the short-term results. These served as information about loading in a reliability-based design approach. The load effects in a semiprobabilistic design were also calculated at an appropriate probability of exceedance level.     

基于PCA和RBF神经网络的绞吸挖泥船实时产量预测

绞吸挖泥船在实际作业过程中的动态特性非常复杂,影响产量的控制因素众多.若这些控制因素全部参与产量预测比较耗时.为了实时训练网络及预测产量,先对影响绞吸挖泥船产量的控制因素进行主成分分析(PCA),再根据分析结果约减控制因素;在系统仿真建模中,分别以全部因素和约减后因素作为径向基(RBF)神经网络的输入变量,以产量作为输...     

基于遗传算法优化的小波神经网络对船舶受扰力及力矩建模预报

首先构造小波神经网络,建立船舶横向和纵向运动模型及其运动受扰力模型,采用小波神经网络对船舶受扰力建模预报,并用遗传算法对其进行优化,对预报结果进行分析比较,给出超前5秒、15秒的船舶运动受扰力与力矩预报结果.     

混沌理论和神经网络相结合的舰船摇荡运动极短期预报

舰船摇荡运动具有混沌特性,因而可以应用混沌理论对其进行预报.介绍了混沌时间序列预测原理;建立了基于混沌理论相空间重构技术的RBF神经网络模型,并将其用于舰船摇荡运动预报;通过对某实船纵摇时历的预报计算,证明了采用混沌和神经网络相结合的预报方法,能有效提高预报精度和延长预报时长.     

基于混沌理论与RBF神经网络的船舶运动极短期预报研究

文章基于混沌动力系统相空间重构理论,利用关联维数法和最大Lyapunov指数法,对船舶运动时间序列的混沌特性进行了判定。并利用RBF神经网络较强的非线性映射功能,结合相空间重构理论建立了船舶运动极短期直接多步预报模型。实例预报结果表明,所建立的预报模型应用于船舶运动极短期预报取得了令人满意的预报精度,预报时间可达10 s。     

The application of modeling and prediction with MRA wavelet network

As there are lots of non-linear systems in the real engineering, it is very important to do more researches on the modeling and prediction of non-llnear systems. Based on the muhi-resolution analysis (MRA) of wavelet theory, this paper combined the wavelet theory with neural network and established a MRA wavelet network with the scaling function and wavelet function as its neurons. From the analysis in the frequency domain, the results indicated that MRA wavelet network was better than other wavelet networks in the ability of approaching to the signals. An essential research was carried out on modeling and prediction with MRA wavelet network in the non-linear system. Using the lengthwise sway data received from the experiment of ship model, a model of offline prediction was estab lished and was applied to the short-time prediction of ship motion. The simulation results indicated that the forecasting model improved the prediction precision effectively, lengthened the forecasting time and had a better prediction results than that of AR linear model. The research indicates that it is feasible to use the MRA wavelet network in the short -time prediction of ship motion.     

基于BP神经网络模型下的桩群阻力预测

在30m长、3m宽、0.26m深的循环水槽中,应用阻力相似理论进行物理模型桩群模拟,以试验中的实测桩群阻力数值作为期望值,建立基于BP神经网络的桩群阻力预测模型。应用该模型进行桩群阻力预测,通过对比实测数据,发现预测值相对误差很小,预测结果合理可信。由此可以认为,以物理模型试验数据为基础,依托神经网络进行桩群阻力预测的方法值得推广和探讨。     

Long-term prediction method for the green water load and volume for an assessment of the load line

 To develop a practical prediction method for the green water load and volume on the bow deck in irregular waves, model tests were conducted using a tanker and a cargo ship on a domestic Japanese voyage. The relation between green water load and relative water height at the stem was considered. Based on the finding that the maximum value of the green water load is proportional to the square of the maximum value of the water elevation over the bow top, the probability density functions of the green water load and volume in short-term predictions were proposed. It was verified that the proposed functions show good agreement with the measured distributions, especially in the tails, and were better than conventional functions. Using these functions, long-term predictions of the green water load were carried out. It was confirmed that the present method is more rational than the conventional one for estimating the long-term probability of the green water load. An assessment of the bow height of a domestic Japanese ship from the viewpoint of deck wetness was carried out using these prediction methods. This research was used as the technical background for the revision of domestic rules on load lines, which was enforced in October 2001. Received: July 19, 2002 / Accepted: October 30, 2002 Acknowledgment. Some of the present study was carried out as part of a cooperation project (RR45) with the Shipbuilding Research Association of Japan, supported by the Nippon Foundation. Address correspondence to: Y. Ogawa (e-mail: ogawa@nmri.go.jp)     

Time series prediction based on LSTM neural network for top tension response of umbilical cables

The umbilical cable is an essential component of offshore oil and gas extraction systems. The severe marine environment poses a high challenge to the safety of the umbilical cable structure during operation. The analysis of an umbilical cable requires complex and resource-demanding finite element time-domain simulations to obtain their nonlinear dynamic response. Therefore, in order to solve the problem of structural safety monitoring and real-time assessment of remaining life of umbilical cables under extreme sea states, there is a great need to predict the dynamic response of umbilical cables quickly and accurately during operation, for ease of making fast decisions for system operation and maintenance before the arrival of extreme sea states. Given the strong nonlinear function-approximation ability of the neural network, this study proposes an efficient method for the prediction of the time series of umbilical cable top tension response based on LSTM (long short-term memory) neural network. We use LSTM neural network and ARIMA (autoregressive integrated moving average) model in a real engineering case for time series prediction of the top tension response of the umbilical cable, and the results of the two models are analyzed and compared, and the efficiency and accuracy of the LSTM neural network model are verified. Furthermore, the hyperparameter, dataset and generalization ability of LSTM model are discussed. The results indicate that feasibility of the tension response prediction of umbilical cables under dynamic load in complex marine environments.     

基于小波神经网络和疲劳曲线的结构疲劳寿命及可靠度预测

基于疲劳曲线,考虑应力范围的随机性,获得结构疲劳寿命所服从的分布,计算出期望寿命的可靠度.以非线性Morlet小波基作为激励函数形成神经元,结合小波与神经网络的优点,建立小波神经网络模型,从而得到应力范围、结构疲劳寿命及可靠度之间的关系,用于预测某应力范围下所期望可靠度的疲劳寿命,或者所期望疲劳寿命的可靠度.     

神经网络算法在海底声源非线性特性的数学建模

基于声源探测及目标检测跟踪系统在海底开发领域中的应用越来越广泛,而由于海水温度的不平衡以及海面多经衰落等影响,声波信号在海底的传输呈现非线性特性。本文在研究海下声波传输特性的基础上,利用神经网络对海底声波传输模型进行数学描述,并利用参数阵来建立声波非线性模型,最后对算法进行仿真分析。     

Strain prediction for critical positions of FPSO under different loading of stored oil using GAIFOA-BP neural network

FPSO (floating, production, storage and offloading) units are widely used in the offshore oil and gas industry. Generally, FPSOs have excellent oil storage capacity owing to their huge oil cargo holds. The volume and distribution of stored oil in the cargo holds influence the strain level of hull girder, especially at critical positions of FPSO. However, strain prediction using structural analysis tools is computationally expensive and time consuming. In this study, a prediction tool based on back-propagation (BP) neural network called GAIFOA-BP is proposed to predict the strain values of concerned positions of an FPSO model under different oil storage conditions. The GAIFOA-BP combines BP model and GAIFOA which is a combination of genetic algorithm (GA) and an improved fruit fly optimization algorithm (IFOA). Results from three benchmark tests show that the GAIFOA-BP model has a remarkable performance. Subsequently, a total of 81 sets of training data and 25 sets of testing data are obtained from experiment using fiber Bragg grating (FBG) sensors installed on the surface of an FPSO model. The numerical results show that the GAIFOA-BP is capable of predicting the strain values with higher accuracy as compared with other BP models. Finally, the reserved GAIFOA-BP model is utilized to predict the strain values under the inputs of a 10-day time series of volume and distribution of stored oil. The predicted strain results are further used to calculate the fatigue consumption of measurement points.     

酸性环境下混凝土寿命预测模型的建立及应用

采用模拟加速试验的方法,研究酸性腐蚀条件下混凝土性能劣化规律,旨在建立一个合理的酸性环境下混凝土寿命预测模型。预测酸性环境下混凝土寿命主要通过3种方式:1)根据灰色系统理论,利用GM(1,1)模型预测其强度变化规律;2)根据Arrhenius定理构建预测模型,利用试验强度变化数据预测混凝土寿命;3)通过研究酸性环境下混凝土中性化深度变化规律来达到辅助预测的效果。结果表明:根据灰色系统理论或Arrhenius定理构建的预测模型,对酸性环境下混凝土进行寿命预测的准确性较高;同时提出了结合中性化深度变化规律进行综合分析的设想。     

混合型专家系统在船舶动力装置故障诊断系统中的应用

介绍了一种基于人工神经网络的混合型专家系统及其在船舶动力装置故障诊断系统的应用。具体描述了混合型专家系统的基本原理,论述了船舶动力装置故障诊断系统的结构与功能。最后用一个实例说明混合型专家系统在船舶动力装置故障诊断系统上的应用。