基于HMM-SVR的船舶动力设备故障模式识别与状态预测研究

船舶动力设备因故障监测信号样本少、变化缓慢且数据特征呈非线性,使得设备故障模式的准确识别和状态预测比较难。鉴于此,文章研究了基于隐马尔科夫模型的故障模式识别方法,利用该模型将微弱变化的信号特征转换为变化较大的对数似然概率对故障模式实现有效识别。在此基础上进一步提出基于HMM-SVR的设备状态预测模型,将遗传算法用于支持向量回归模型参数寻优,并结合隐马尔科夫模型,实现对设备状态的预测。对船用柴油机进行仿真,结果表明上述模型具有较高的识别率,能准确预测船舶动力设备的当前状态。     

On Safety in Dismantling Temporary Strut of Super-large Tunnels with Double-layer Primary Support北大核心CSCD

More and more multiple-track tunnels and super-large section tunnels have been built, and disman- tling of temporary strut is a weak point of the whole structure during force transfer when the secondary lining is con- structed. It is significant to guarantee structure safety during dismantling of temporary strut. Little systematic re- search on safety in dismantling of temporary strut of the super-large section tunnel with double-layer primary support has been conducted, so the internal force and security of the two-layer primary support of the Xinkaotang tunnel were analyzed by a numerical analysis and site measurement, and it proves the effect of two-layer primary support on the safety during strut dismantling. The research results indicate that: (1) with constant support thickness and one-time longitudinal dismantling length, the safety factor of secondary primary support is larger than that of the first primary support, and the safety factor of the first primary support is larger than that of the single-layer primary support. Change range of safety factor for the first primary support is smaller than that of the single-layer primary support, and the safe factor for the single-layer primary support is smaller than that of the secondary primary support; (2) with the same support pattern, the safety factors increase firstly and then decrease with an increase of the onetime dismantling length. The calculated results of various cases show that the reasonable one-time dismantling length for this project is about 9 m. © 2018, Editorial Office of "Modern Tunnelling Technology". All right reserved.     

Analysis on the Characteristics and Development Trends of the Support Technology of High Ground Stress Tunnels in China北大核心CSCD

Combining the present situation and development trend of different tunnel support technologies at home and abroad, this paper analyzes the problems of rockburst in hard rock tunnels and large deformation in soft rock tunnels caused by high ground stress. It is concluded that: 1) regarding the rockburst problem, the current support technology is mainly influenced by the rock burst mechanism which is dominated by static factors, and so the used support components are generally of smaller deformation performance and "passive support" properties; 2) as the rockburst is the result of dynamic-static stress coupling, and only the anchor bolt has the "active support" attribute in the current "shotcrete+anchor bolt+wire net" support system, so the best support system should have the two functions of active support and energy release in terms of the rockburst problem, and the key focus of the research and development is anchorage members; 3) there are three main support types for large deformation in soft rock tunnels, e.g. the heavy support, layered support and yielding support. Among them, the heavy support system in underground cavern with large deformation is easy to induce excessive surrounding rock pressure, and so the applicable conditions are limited. The layered support system is still not the best choice due to its immature theoretical study, difficult determination of the thickness value and the installation time of each support layer and the interference to construction progress. With the characteristics of timely support and yielding while supporting, the yielding support system can give full play to the performance values of surrounding rocks and supporting materials, and make both of them reach the optimal state, so it is the best choice for supporting the soft rock tunnels with large deformations. © 2018, Editorial Office of "Modern Tunnelling Technology". All right reserved.     

基于AIGA-WLSSVM的埋地管道腐蚀速率预测方法

为了降低埋地管道腐蚀影响因素之间的复杂相关性,提高腐蚀预测精度,文中提出一种基于自适应免疫遗传算法-加权最小二乘支持向量机(AIGA-WLSSVM)的埋地管道腐蚀速率预测建模方法,并采用AIGA优化模型参数,进一步提高模型的学习能力和稳定性。最后通过实例分析验证了AIGA-WLSSVM建模方法在埋地管道腐蚀速率预测中的可行性和有效性,为埋地管道的检修与更换提供参考。     

制动器在水力式升船机中的应用

依托景洪升船机工程,探讨水力式升船机制动器的应用。通过对比分析电力驱动和水力驱动式升船机制动器的布置特点及工作方式,阐述制动器在水力式升船机中面临的新的技术问题,据此对水力式升船机制动器的制动力、同步性以及制动方式提出设计指标和要求,并探讨制动器的使用范围。建议水力式升船机制动器采用常开式,升船机正常运行时,制动器在船厢上、下游对接时投入使用。     

计及弹性支撑效应的独立液舱晃荡数值分析研究

文章针对弹性支撑结构作用下的液舱晃荡问题,采用基于Hilber-Hughes-Taylor(HHT)格式的隐式直接积分法求解液舱运动,应用VOF法求解流体晃荡,并结合部分单元参数的概念处理棱形液舱边界,建立了液舱运动与流体晃荡双向耦合迭代算法。以独立液舱为研究对象,通过耦合求解弹性支撑液舱晃荡,并与非弹性支撑液舱晃荡进行比较,分析了在不同刚度的弹性支撑结构作用下液舱液体运动及晃荡载荷的变化规律。该文建立的分析方法为含弹性支撑的独立液舱晃荡载荷预报提供了一种快速、有效的分析手段。     

基于小波和仿射不变矩特征融合的舰船型号识别

针对不变矩对仿射形变目标描述的不足,为提高舰船型号的识别精度,提出一种基于小波和仿射不变矩特征融合的舰船型号识别方法.首先对二值舰船图像进行归一化处理,并分别提取归一化舰船图像的小波矩特征值和仿射不变矩特征值;然后通过计算样本特征均值与标准差的比值,选择出鲁棒性好、稳定性高的特征,通过归一化方法进行融合;最后构造五类舰船的样本集,采用支持向量机(SVM)作为分类器识别测试样本的型号,分析不同矩特征、样本集大小、SVM参数、本文方法对识别精度、稳定性的影响.实验结果表明,文中给出的算法提高了识别精度,并且在训练样本集较小时仍能获得88%以上的识别率.     

双月池对某援潜救生船运动性能及波面升高的影响

介绍某援潜救生船配设双月池的船型特点,规划双月池和不带月池2种方案来评估双月池开口对援潜救生船船体运动性能的影响。采用三维势流理论对2个方案的运动性能进行计算,得到船体运动响应函数,对船舶运动性能进行短期预报,研究总结双月池效应对援潜救生船运动的影响规律。对比双月池内水面和舷外波面升高的情况,比较两个位置处波面升高的差异性,总结双月池波面升高对潜水钟收放的影响。     

基于支持向量机的发动机气路故障预诊断

为实现航空发动机气路故障在线预诊断,分析了地空数据链系统中发动机气路参数报文的协议格式,建立了基于支持向量机算法的发动机气路参数在线预测模型。以便携式地空数据链收发系统为硬件基础,构建发动机报文并行处理系统,获取建模所需的训练样本。利用最终误差预报准则确定样本数据嵌入维数,实现时序样本数据的相空间重构。提出自适应网格搜索法优化支持向量机建模参数,获得气路参数在线预测模型,与航路飞机建立地空数据链通信,预测气路参数趋势。预测结果表明:参数低压转子转速、高压转子转速、尾气温度与燃油流量的相对预测误差分别为2.5%、2.1%、1.9%与2.3%,因此,支持向量机模型具有较高预测精度。     

浅谈BAOM污水处理与回用系统在山西省高速公路附属设施的应用

以山西南部地区某高速公路服务区为例,根据山西省高速公路运营阶段附属设施污水特点及现有处理工艺存在问题,介绍了BAOM污水处理与回用系统的工艺流程、结构单元,对该系统在服务区的运行效果以及经济成本进行了详细的分析,证明了该系统在处理高速公路附属设施污水的技术和经济可行性,这对今后推进山西省高速公路污水处理设施实现标准化、规范化具有较好的借鉴意义。