Adaptive Predictive Inverse Control of Offshore Jacket Platform Based on Rough Neural Network

The offshore jacket platform is a complex and time-varying nonlinear system,which can be excited of harmful vibration by external loads.It is difficult to obtain an ideal control performance for passive control methods or traditional active control methods based on accurate mathematic model.In this paper,an adaptive inverse control method is proposed on the basis of novel rough neural networks (RNN) to control the harmful vibration of the offshore jacket platform,and the offshore jacket platform model is established by dynamic stiffness matrix (DSM) method.Benefited from the nonlinear processing ability of the neural networks and data interpretation ability of the rough set theory,RNN is utilized to identify the predictive inverse model of the offshore jacket platform system.Then the identified model is used as the adaptive predictive inverse controller to control the harmful vibration caused by wave and wind loads,and to deal with the delay problem caused by signal transmission in the control process.The numerical results show that the constructed novel RNN has advantages such as clear structure,fast training speed and strong error-tolerance ability,and the proposed method based on RNN can effectively control the harmfid vibration of the offshore jacket platform.     

Ecosystem health assessment of Honghu Lake Wetland of China using artificial neural network approach

Honghu Lake, located in the southeast of Hubei Province, China, has suffered a severe disturbance during the past few decades. To restore the ecosystem, the Honghu Lake Wetland Protection and Restoration Demonstration Project (HLWPRDP) has been implemented since 2004. A back propagation (BP) artificial neural network (ANN) approach was applied to evaluatinig the ecosystem health of the Honghu Lake wetland. And the effectiveness of the HLWPRDP was also assessed by comparing the ecosystem health before and after the project. Particularly, 12 ecosystem health indices were used as evaluation parameters to establish a set of three-layer BP ANNs. The output is one layer of ecosystem health index. After training and testing the BP ANNs, an optimal model of BP ANNs was selected to assess the ecosystem health of the Honghu Lake wetland. The result indicates that four stages can be identified based on the change of the ecosystem health from 1990 to 2008 and the ecosystem health index ranges from morbidity before the implementation of HLWPRDP (in 2002) to middle health after the implementation of the HLWPRDP (in 2005). It demonstrates that the HLWPRDP is effective and the BP ANN could be used as a tool for the assessment of ecosystem health.     

太湖水质参数MODIS的遥感定量提取方法

本文利用MODIS数据的可见光、近红外波段和准实时的地面采样数据,分别利用线性回归模型和神经网络模型反演了太湖的叶绿素a和悬浮物浓度.结果表明,利用MODIS数据的波段组合(M2/M8)和(M2/M9)可估算太湖的叶绿素a浓度;而MODIS数据的波段组合(M12/M17)、(M13/M17)及MODIS(M4)波段能定量估算太湖的悬浮物浓度,但估算精度仍不能满足实际需要.因此,构建了一个以MODIS可见光及近红外波段为输入,以太湖水质参数为输出的2层BP神经网络模型反演太湖的水质参数,大大提高了反演精度.     

基于雷达反射率因子和雷电定位数据的深度学习雷电预报模型

利用卷积神经网络和门控循环单元（Gated recurrent units ）神经网络,基于雷达反射率因子和雷电定位数据开展了雷电预报研究。首先构建了引用注意力机制的基于卷积神经网络和门控循环单元神经网络的深度学习模型（Attention ConvGRU）;然后将雷达反射率因子数据和对应时间段（6 min）的雷电定位数据处理成图像数据后输入深度学习模型,训练出可预报雷电的模型,包括3种模型：单雷电数据模型、单雷达数据模型和雷电〖CD*2〗雷达双数据模型;最后开展了预报试验和定量评估。综合评估表明,本文建立的雷电预报模型综合预报准确率达到96.74%,虚警率35.83%,关键成功指数（Critical Success Index, CSI）为0.2072。个例分析表明,预报模型对于具有明显移动趋势的雷暴过程（A类雷暴）的预报效果优于不具有明显移动趋势的雷暴过程（B类雷暴）,且随着B类雷暴强度减弱模型预报能力逐渐减弱。     

EWT-Elman组合模型短期电离层TEC预报

针对电离层总电子含量(TEC)非线性、高噪声的特点,建立基于经验小波变换(EWT)和Elman神经网络的短期电离层组合预报模型。运用该模型对不同地磁环境的电离层TEC时间序列进行建模预报,结果表明,EWT-Elman组合模型可反映电离层TEC的变化特征,地磁平静期预测平均相对精度为93%,均方根误差为1.04 TECu;地磁扰动期预测平均相对精度为92.4%,均方根误差为2.18 TECu。单一Elman模型、EMD-Elman组合模型以及EWT-BP组合模型在地磁平静期平均相对精度最高为90.7%,均方根误差最小为1.33 TECu;地磁扰动期平均相对精度最高为90.7%,均方根误差最小为2.57 TECu。对比其他模型,本文方法预测效果最优。     

神经网络油气检测技术

多参数神经网络油气检测技术首先提取四大类２６ 个地震波特征参数,并定量描述这些参数,然后用神经网络技术对这些参数进行分析,预测油气分布,使油气识别工作定量化、计算机化。本次研究开发了一套神经网络油气识别软件,并利用其对东濮凹陷胡１９ 块进行油藏精细描述,预测沙三下砂组含油面积０－７ ｋｍ２ ,石油地质储量７０ ×１０４ 吨。     

基于神经网络的专家系统在温室控制中的应用

为了实现温室控制,针对温室环境的多输入、多输出、非线性和难以建立数学模型等特点,提出一种基于BP神经网络的专家系统并用于温室控制。该方法将传感器采集的温度、湿度等信息输入到神经网络专家系统,在获得决策结果的同时通知控制部分执行相应的决策。这种方法不仅解决了传统专家系统知识获取的瓶颈问题、推理能力差、智能化低的缺点,而且克服了神经网络不具有解释功能的问题。总之,成功实现了神经网络和专家系统功能上的互补,较好的用于温室控制。     

桥梁水平位移混沌特征识别与神经网络预测研究

跨海大桥系统受外界影响扰动,其变形伴有混沌现象发生.对桥梁变形监测数据实现了混沌识别,运用C-C法计算时间序列的延迟时间,用G-P方法求得最佳嵌入维数,通过求取的时间延迟和最佳嵌入维数对桥梁变形监测数据进行相空间重构,为混沌时间序列预测模型的建立奠定基础;基于RBF神经网络建立混沌时间序列预测模型,对实测数据进行桥梁变...     

滑坡变形预测灰色神经网络耦合模型的构建及适用性分析

结合灰色模型和神经网络的数据处理特点,提出串联、并联和混联式3种结构的灰色神经网络滑坡变形预测模型。串联式将滑坡变形位移时序分解为趋势项和随机项,采用灰色模型提取滑坡位移时序趋势,利用神经网络逼近随机波动;并联式以灰色模型和神经网络分别对滑坡预测,采用智能非线性组合,按照预测目标精度动态调整权重,从而获取最终组合预测结果;混联式通过增加灰白化层及灰模型群,对神经网络拓扑结构进行优化,达到弱化滑坡原始监测数据随机性、提高预测模型稳健性的目的。将3种模型应用于古树屋滑坡变形预测,并对其适用性进行讨论。结果表明,3种结构的灰色神经网络耦合模型均提高了预测精度,适用于复杂状况下滑坡体的变形预测。     

基于改进BP神经网络的海底底质分类

通过采用遗传算法优化神经网络初始权值的方法,将GA算法与BP神经网络有机结合,应用于海底底质分类。基于多波束测深系统获取的反向散射强度数据,应用改进的BP神经网络分类方法,实现对海底基岩、砾石、砂、细砂和泥等底质类型的快速、准确识别。通过实验比较,GA-BP神经网络分类精度明显高于BP神经网络,证明了该方法的有效性和可靠性。