Back-propagation neural network for long-term tidal predictions

During the recent years, the availability of accurate ocean tide models has become increasingly important, as tides are the main contributor to disposal and movement of sediments, tracers and pollutants, and to a whole range of offshore applications in engineering, environmental observations, exploration and oceanography. Tides can be conventionally predicted by harmonic analysis, which is the superposition of many sinusoidal constituents with amplitudes and frequencies determined by a local analysis of the measured tide. However, accurate predictions of tide levels could not be obtained without a large number of tide measurements by the harmonic method. An application of the back-propagation neural network using short-term measuring data is presented in this paper. On site tidal level data at Taichung Harbor in Taiwan will be used to test the performance of the present model. Comparisons with conventional harmonic methods indicate that the back-propagation neural network mode also efficiently predicts the long-term tidal levels.     

Development of a regional neural network for coastal water level predictions

This paper presents the development of a Regional Neural Network for Water Level (RNN_WL) predictions, with an application to the coastal inlets along the South Shore of Long Island, New York. Long-term water level data at coastal inlets are important for studying coastal hydrodynamics sediment transport. However, it is quite common that long-term water level observations may be not available, due to the high cost of field data monitoring. Fortunately, the US National Oceanographic and Atmospheric Administration (NOAA) has a national network of water level monitoring stations distributed in regional scale that has been operating for several decades. Therefore, it is valuable and cost effective for a coastal engineering study to establish the relationship between water levels at a local station and a NOAA station in the region. Due to the changes of phase and amplitude of water levels over the regional coastal line, it is often difficult to obtain good linear regression relationship between water levels from two different stations. Using neural network offers an effective approach to correlate the non-linear input and output of water levels by recognizing the historic patterns between them. In this study, the RNN_WL model was developed to enable coastal engineers to predict long-term water levels in a coastal inlet, based on the input of data in a remote NOAA station in the region. The RNN_WL model was developed using a feed-forwards, back-propagation neural network structure with an optimized training algorithm. The RNN_WL model can be trained and verified using two independent data sets of hourly water levels.The RNN_WL model was tested in an application to Long Island South Shore. Located about 60–100 km away from the inlets there are two permanent long-term water level stations, which have been operated by NOAA since the1940s. The neural network model was trained using hourly data over a one-month period and validated for another one-month period. The model was then tested over year-long periods. Results indicate that, despite significant changes in the amplitudes and phases of the water levels over the regional study area, the RNN_WL model provides very good long-term predictions of both tidal and non-tidal water levels at the regional coastal inlets. In order to examine the effects of distance on the RNN_WL model performance, the model was also tested using water levels from other remote NOAA stations located at longer distances, which range from 234 km to 591 km away from the local station at the inlets. The satisfactory results indicate that the RNN_WL model is able to supplement long-term historical water level data at the coastal inlets based on the available data at remote NOAA stations in the coastal region.     

Wind-wave variability in a shallow tidal sea—Spectral modelling combined with neural network methods

In this paper the wind-wave variability in the tidal basins of the German Wadden Sea is modelled with combined numerical and neural-network (NN) methods. First, the wave propagation and transformation in the study area are modelled with the state-of-the-art third-generation spectral wave model SWAN. The ability of SWAN to accurately reproduce the phenomena of interest in nonstationary conditions governed by highly variable winds, water levels and currents is shown by comparisons of the modelled and measured mean wave parameters at four stations. The principal component analysis of the SWAN results is then used to reveal the dominating spatial patterns in the data and to reduce their dimensionality, thus enabling an efficient and relatively straightforward NN modelling of mean wave parameters in the whole study area. It is shown that the data produced with the approach developed in this work have statistical properties (discrete probability distributions of the mean wave parameters etc.) very close to the properties of the data obtained with SWAN, thus proving that this approach can be used as a reliable tool for wind wave simulation in coastal areas, complementary to (often computationally demanding) spectral wave models.     

基于小波变换和LSTM神经网络的格陵兰岛近海海域海平面变化预测

海平面变化包含多种不同时间尺度信息，传统的预测方法仅对海平面变化趋势项、周期项进行拟合，难以利用海平面变化的不同时间尺度信号，使得预测精度不高。本文基于深度学习的预测模型，提出一种融合小波变换（wavelet transform，WT）与LSTM （long short-term memory，LSTM）神经网络的海平面异常组合预测模型。首先利用小波分解得到反映海平面变化总体趋势的低频分量和刻画主要细节信息的高频分量；然后通过LSTM神经网络对代表不同时间尺度的各个分量预测和重构，实现海平面变化的非线性预测。基于该模型的海平面变化预测的均方根误差、平均绝对误差和相关系数分别为12.76 mm、9.94 mm和0.937，预测精度均优于LSTM和EEMD-LSTM预测模型，WT-LSTM组合模型对区域海平面变化预测具有较好的应用价值。     

Modelling of a medium-term dynamics in a shallow tidal sea, based on combined physical and neural network methods

The paper presents an approach towards a medium-term (decades) modelling of water levels and currents in a shallow tidal sea by means of combined hydrodynamic and neural network models. The two-dimensional version of the hydrodynamic model Delft3D, forced with realistic water level and wind fields, is used to produce a two-year-database of water levels and currents in the study area. The linear principal component analysis (PCA) of the results is performed to reveal dominating spatial patterns in the analyzed dataset and to significantly reduce the dimensionality of the data. It is shown that only a few principal components (PCs) are necessary to reconstruct the data with high accuracy (over 95% of the original variance). Feed-forward neural networks are set up and trained to effectively simulate the leading PCs based on water level and wind speed and direction time series in a single, arbitrarily chosen point in the study area. Assuming that the spatial modes resulting from the PCA are ‘universally’ applicable to the data from time periods not modelled with Delft3D, the trained neural networks can be used to very effectively and reliably simulate temporal and spatial variability of water levels and currents in the study area. The approach is shown to be able to accurately reproduce statistical distribution of water levels and currents in various locations inside the study area and thus can be viewed as a reliable complementary tool e.g., for computationally expensive hydrodynamic modelling. Finally, a detailed analysis of the leading PCs is performed to estimate the role of tidal forcing and wind (including its seasonal and annual variability) in shaping the water level and current climate in the study area.     

基于最小二乘与径向基函数神经网络的海平面变化预测

在对海平面变化规律进行深入分析的基础上,应用最小二乘神经网络组合模型对海平面变化趋势进行预测;对卫星测高海平面异常序列中的周期项及线性趋势项利用最小二乘模型进行拟合,残差部分则采用径向基函数神经网络模型进行预测。对中国近海海域卫星测高海平面异常序列的预测表明,连续1个月的预测精度为0.52 cm, 3个月的预测精度为0.65 cm,证明了该组合模型在海平面变化短期预测方面的可靠性,其在海平面变化预测领域具有较高的应用价值。     

基于异地潮位资料和BP神经网络的潮位推算研究

文章通过BP神经网络模型,利用西沙站的实测潮位推算三亚站潮位,研究用一地点的潮位资料去推算另一地点(异地)潮位的方法。文章比较了不同隐含层节点数和输入因子对潮位推算结果的影响,采用预测时间(t)之前N个小时(t–N+1,…,t–1,t)西沙站的实测潮位数据作为输入因子,输入因子数目在2~10之间,隐含层分别采用节点数3、4、5、10和15建模,分多种情况进行推算。结果显示,对文中使用的特定情形,隐含层为4个节点的效果最好,隐含层为15个节点的效果最差;输入层为2个节点的效果最好,输入因子增多会使得推算效果变差。隐含层为4个节点、输入因子为t–1、t时刻潮位的仿真验证的结果最好,推算值和实测值之间的相关系数为0.9901,均方根误差为0.06m,误差在–0.16~0.15m之间。结果表明,如果两个地点的潮位具有物理上的关联,通过BP神经网络模型,用一地点的实测潮位推算另一地点潮位的方法是可行的。     

System identification for nonlinear maneuvering of large tankers using artificial neural network

This paper deals with the application of nonparametric system identification to a nonlinear maneuvering model for large tankers using artificial neural network method. The three coupled maneuvering equations in this model for large tankers contain linear and nonlinear terms and instead of attempting to determine the parameters (i.e. hydrodynamic derivatives) associated with nonlinear terms, all nonlinear terms are clubbed together to form one unknown time function per equation which are sought to be represented by the neural network coefficients. The time series used in training the network are obtained from simulated data of zigzag maneuvers and the proposed method has been applied to these data. The neural network scheme adopted in this work has one middle or hidden layer of neurons and it employs the Levenberg–Marquardt algorithm. Using the best choices for the number of hidden layer neurons, length of training data, convergence tolerance etc., the performance of the proposed neural network model has been investigated and conclusions drawn.     

Development of wavelet network model for accurate water levels prediction with meteorological effects

Accurate water levels modeling and prediction is essential for safety of coastal navigation and other maritime applications. Water levels modeling and prediction is traditionally developed using the least-squares-based harmonic analysis method that estimates the harmonic constituents from the measured water levels. If long water level measurements are not obtained from the tide gauge, accurate water levels prediction cannot be estimated. To overcome the above limitations, the current state-of-the-art artificial neural network has recently been developed for water levels prediction from short water level measurements. However, a highly nonlinear and efficient wavelet network model is proposed and developed in this paper for water levels modeling and prediction using short water level measurements. Water level measurements (about one month and a week) from six different tide gauges are employed to develop the proposed model and investigate the atmospheric changes effect. It is shown that the majority of error values, the differences between water level measurements and the modeled and predicted values, fall within the −5 cm and +5 cm range and root-mean-squared (RMS) errors fall within 1–6 cm range. A comparison between the developed highly nonlinear wavelet network model and the harmonic analysis method and the artificial neural networks shows that the RMS of the developed wavelet network model when compared with the RMS of the harmonic analysis method is reduced by about 70% and when compared with the RMS of the artificial neural networks is reduced by about 22%. It is also worth noting that if the atmospheric changes effect (meteorological effect) of the air pressure, the air temperature, the relative humidity, wind speed and wind direction are considered, the performance accuracy of the developed wavelet network model is improved by about 20% (based on the estimated RMS values).     

基于人工神经网络的赤潮预测模型

本文利用非线性时间序列预测模型,将海洋预报和人工神经网络BP算法相结合,提出了基于神经网络的海洋预报模型;运用改进的三层BP(Back Propagation)神经网络模型对海洋气象进行赤潮灾害监测和预报;同时针对仿真结果进行分析,结果表明该模型具有较好的预测能力。     

Rudder roll stabilization for fishing vessel using neural network approach

This paper presents a neural network (NN) controller for a fishing vessel rudder roll system. The aim of this study is to build a NN controller which uses rudder to regulate both the yaw and roll motion. The neural controller design is accomplished with using the classical back-propagation algorithm (CBA). Effectiveness of the proposed NN control scheme is compared with linear quadratic regulator (LQR) results by simulations carried out a fishing vessel rudder roll stabilizer system.     

Prediction of maximum wave-induced liquefaction in porous seabed using multi-artificial neural network model

In the last few decades, considerable efforts have been devoted to the phenomenon of wave-induced liquefactions, because it is one of the most important factors for analysing the seabed and designing marine structures. Although numerous studies of wave-induced liquefaction have been carried out, comparatively little is known about the impact of liquefaction on marine structures. Furthermore, most previous researches have focused on complicated mathematical theories and some laboratory work. In the present study, a data dependent approach for the prediction of the wave-induced liquefaction depth in a porous seabed is proposed, based on a multi-artificial neural network (MANN) method. Numerical results indicate that the MANN model can provide an accurate prediction of the wave-induced maximum liquefaction depth with 10% of the original database. This study demonstrates the capacity of the proposed MANN model and provides coastal engineers with another effective tool to analyse the stability of the marine sediment.     

基于集合经验模态分解与BP组合模型的短期余水位预测

水位在忽略观测误差的前提下,可分解为潮位和余水位,后者具有较强的空间相关性以及非平稳特征,是影响水位预报精度的主要因素。港口工程、航运计划编制等方面对实时高精度水位预报具有重要需求,这对余水位预报模型构建提出了更高要求。另外,利用高精度余水位预报模型可减少验潮站布设数量。针对余水位短期预测模型精度不高的现状,本文对余水位进行集合经验模态(EEMD)分解,获得余水位在时间序列上的本征模函数(IMF);使用快速傅立叶变换(FFT)分析各本征模函数的频谱特征;再利用BP神经网络对各个本征模函数进行训练,预测了未来6 h、12 h、24 h的余水位值。对哥伦比亚河下游河口处的3组典型验潮站的余水位数据的预测结果表明,在未来6 h、12 h内的余水位的预测精度达到厘米级,在24 h内接近厘米级,证明了该组合模型在余水位短期预测方面的可行性。     

海洋环境中平台钢腐蚀速率的三层BP 神经网络预测

利用三层BP神经网络预测海洋环境因素对材料的腐蚀速率的影响。结合实测的pH值、温度、溶解氧、盐度、生物附着等影响因素,分析了上述环境因素对平台钢腐蚀的影响,建立环境因素与腐蚀速率之间的映射关系,预测了平台钢在海洋环境中的腐蚀速率。结果表明,全浸区腐蚀速率预测误差为6.95%,潮差带腐蚀速率预测误差为4.2%,预测精度较高。说明利用三层BP神经网络预测钢在海水中腐蚀速率技术可行,具有较高的预测精度和应用价值。     

A forecasting model for wave heights based on a long short-term memory neural network

To explore new operational forecasting methods of waves, a forecasting model for wave heights at three stations in the Bohai Sea has been developed. This model is based on long short-term memory(LSTM) neural network with sea surface wind and wave heights as training samples. The prediction performance of the model is evaluated,and the error analysis shows that when using the same set of numerically predicted sea surface wind as input, the prediction error produced by the proposed LSTM model at S...     

Prediction of the dynamic response of a slender marine structure under an irregular ocean wave using the NARX-based quadratic Volterra series

The time series of the dynamic response of a slender marine structure was predicted in approximate sense using a truncated quadratic Volterra series. The wave-structure interaction system was identified using the NARX (Nonlinear Autoregressive with Exogenous Input) technique, and the network parameters were determined through supervised training using prepared datasets. The dataset used for network training was obtained by nonlinear finite element analysis of the slender marine structure under random ocean waves of white noise. The nonlinearities involved in the analysis were both large deformation of the structure under consideration and the quadratic term of the relative velocity between the water particle and structure in the Morison formula. The linear and quadratic frequency response functions of the given system were extracted using the multi-tone harmonic probing method and the time series of the response of the structure was predicted using the quadratic Volterra series. To check the applicability of the method, the response of a slender marine structure under a realistic ocean wave environment with a given significant wave height and modal period was predicted and compared with the nonlinear time domain simulation results. The predicted time series of the response of structure with quadratic Volterra series successfully captured the slowly varying response with reasonably good accuracy. This method can be used to predict the response of the slender offshore structure exposed to a Morison type load without relying on the computationally expensive time domain analysis, especially for screening purposes.     

基于人工神经网络的半潜平台动力定位系统推力分配策略研究

针对带有禁止角的半潜平台动力定位系统推力分配算法功率较大的问题,提出了一种基于人工神经网络拟合桨—桨干扰推力损失函数的序列二次规划推力分配算法。该方法考虑了半潜平台桨—桨干扰造成的推力损失,引入推力系数来表达推力损失。利用人工神经网络拟合推力系数,将推力损失加入到推力分配的数学模型中,取消了禁止角。采用序列二次规划求解推力分配数学模型。最后以某半潜式钻井平台为例,选取三种浪向角工况进行推力分配仿真模拟,结果显示该算法在高效分配定位所需推力的同时有效减小了功率消耗,应用前景广泛。     

激光单分子BTEX异常的人工神经网络评价技术

采用BP(Back-Propagation Network)神经网络方法,根据试验区BTEX指标实测数据,结合油气化探、地质、地球物理等资料建立BTEX异常综合评价指标体系及评分标准,完成BTEX异常的BP神经网络综合评价模型并对试验区进行含油气远景评价,研究结果表明该技术具有较好的应用前景。