Multi-objective optimization of double suction centrifugal pump using Kriging metamodels

This paper proposes a new multi-objective optimization method for a family of double suction centrifugal pumps with various blade shapes, using a Simulation-Kriging model-Experiment (SKE) approach. The Kriging metamodel is established to approximate the characteristic performance functions of a pump, namely, the efficiency and required net positive suction head (NPSHr). Hence, the two objectives are to maximize the efficiency and simultaneously to minimize NPSHr. The Non-dominated Sorting Genetic Algorithm II (NSGA II) and Multi-Objective Evolutionary Algorithm based on Decomposition (MOEA/D) have been applied to the multi-objective optimization problem, respectively. The Pareto solution set is obtained by a more effective and efficient manner of the two multi-objective optimization algorithms. A tradeoff optimal design point is selected from the Pareto solution set by means of a robust design based on Monte Carlo simulations, and the optimal solution is further compared with the value of the physical prototype test. The results show that the solution of the proposed multi-objective optimization method is in line with the experiment test.     

A novel adaptive convolutional neural network for fault diagnosis of hydraulic piston pump with acoustic images

As an essential part of hydraulic transmission systems, hydraulic piston pumps have a significant role in many state-of-the-art industries. Thus, it is important to implement accurate and effective fault diagnosis of hydraulic piston pumps. Owing to the heavy reliance of shallow machine learning models on the expertise and experience of engineers, fault diagnosis based on deep models has attracted significant attention from academia and industry. To construct a deep model with good performance, it is necessary and challenging to tune the hyperparameters (HPs). Since many existing methods focus on manual tuning and use common search algorithms, it is meaningful to explore more intelligent algorithms that can automatically optimize the HPs. In this paper, Bayesian optimization (BO) is employed for adaptive HP learning, and an improved convolutional neural network (CNN) is established for fault feature extraction and classification in a hydraulic piston pump. First, acoustic signals are transformed into time–frequency distributions by a continuous wavelet transform. Second, a preliminary CNN model is built by setting initial HPs. The range of each HP to be optimized is identified. Third, BO is employed to select the optimal combination of HPs. An improved model called CNN-BO is constructed. Finally, the diagnostic efficiency of CNN-BO is analyzed using a confusion matrix and t-distributed stochastic neighbor embedding. The classification performance of different models is compared. It is found that CNN-BO has a higher accuracy and better robustness in fault diagnosis for a hydraulic piston pump. This research will provide a basis for ensuring the reliability and safety of the hydraulic pump.     

基于改进关联分类的两次学习方法

关联分类通常产生大量的分类规则,导致在分类新实例时经常产生规则冲突问题。针对这种规则冲突问题,提出了一种基于改进关联分类的两次学习框架。利用频繁且互关联的项集产生分类规则改进关联分类算法,有效减少了规则数。应用改进的关联分类算法产生的一级规则一次性分离出训练集中规则冲突的所有实例。然后,在冲突实例上应用改进的关联分类算法进行第二次学习得到二级规则。分类新实例时,首先利用第一级规则进行分类。如果出现规则冲突,则利用第二级规则分类该实例。实验结果表明,基于改进关联分类的两次学习方法降低了规则冲突比率,并且显著提高了分类准确率。     

基于双重遗传的聚类分析算法研究

针对影响k-means聚类效果的聚类数目和初始中心点两大因素,提出了基于双重遗传的kmeans算法。它用外层遗传算法控制聚类数目,用内层遗传算法控制聚类的初始中心点,并采用类间距离和类内距离以及二者之间的比值来评价聚类结果的好坏,在算法终止后,可同时求得较优的聚类数目和某聚类数目下的较优初始中心点。此外,根据内外层遗传算法的特殊性,采用不同的编码策略适应算法需求,为保留优质个体,采用精英个体保留策略。通过UCI数据集测试实例证明此算法有很好的实用性,对数据挖掘技术有一定参考价值。     

A double line SIW cavity backed antenna for WLAN applications

A compact double line substrate integrated waveguide (DLSIW) cavity backed antenna is realized using half mode SIW technology for WLAN applications. The existing single line SIW antennas for WLAN applications have low gain and less efficiency. To overcome this limitation, DLSIW structure is proposed. The new DLSIW structure simultaneously achieves better gain, radiation efficiency, and good front to back ratio (FTBR) with compact size. To improve the FTBR, ground extension is made. The size reduction of the proposed design is implemented with half mode SIW topology. The gain and efficiency improved new DLSIW antenna is fabricated using FR4 material and it resonates at 5.27GHz WLAN frequency. The size of antenna is 44 mm × 18.75 mm × 1.6 mm and it has the gain of 5.824 dB. The radiation efficiency and FTBR of the antenna are 69.13% and 13.65 dB, respectively. The design is experimentally tested and compared with earlier WLAN antennas. There is a better accordance between simulated and measured results.     

A PZT insulin pump integrated with a silicon microneedle array for transdermal drug delivery

Many of the compounds in drugs cannot be effectively delivered using current drug delivery techniques (e.g., pills and injections). Transdermal delivery is an attractive alternative, but it is limited by the extremely low permeability of the skin. As the primary barrier to transport is located in the upper tissue, Micro-Electro-Mechanical-System (MEMS) technology provides novel means, such as microneedle array and PZT pump, in order to increase permeability of human skin with efficiency, safety and painless delivery, and to decrease the size of the pump. Microneedle array has many advantages, including minimal trauma at penetration site because of the small size of the needle, free from condition limitations, painless drug delivery, and precise control of penetration depth. These will promote the development of biomedical sciences and technology and make medical devices more humanized. So far, most of the insulin pumps being used are mechanical pumps. We present the first development of this novel technology, which can assemble the PZT pump and the microneedle array together for diabetes mellitus. The microneedle array based on a flexible substrate can be mounted on non-planar surface or even on flexible objects such as a human fingers and arms. The PZT pump can pump the much more precision drug accurately than mechanical pump and the overall size is much smaller than those mechanical pumps. The hollow wall straight microneedle array is fabricated on a flexible silicon substrate by inductively coupled plasma (ICP) and anisotropic wet etching techniques. The fabricated hollow microneedles are 200 μm in length and 30 μm in diameter. The microneedle array, which is built with on-board fluid pumps, has potential applications in the chemical and biomedical fields for localized chemical analysis, programmable drug-delivery systems, and very small, precise fluids sampling. The microneedle array has been installed in an insulin pump for demonstration and a leak free packaging is introduced.The support from Ministry of Science and Technology of the People’s Republic of China with contract number of 2005AA40420.     

一种双链结构的多目标进化算法DCMOEA

提出一种双链结构的多目标进化算法(DCMOEA).该算法采用双链结构表示个体,执行过程中无需设置外部归档集合,并采用ε支配策略保持解群的多样性.DCMOEA与MOEA/D、NSGA-II、SPEA2和PAES一同在4个2-目标ZDT函数和4个3-目标DTLZ问题上进行实验,并从算法所获解集的收敛性、分布均匀性和宽广性3个方面进行比较,仿真实验结果表明了DCMOEA的综合性能最好,是一种颇具竞争力的多目标进化算法.     

非平衡输入状态下矩阵式变换器的双电压控制策略与仿真

本文分析了当输入不对称或不平衡时,矩阵式变换器的双电压瞬时值控制的原理和在该控制策略下输出电压的合成规律,证明该控制策略具有良好的抗干扰性能;并分析了在非平衡状态下,输入电流的畸变及其消谐方法.最后搭建了与实际电路相近的仿真模型,通过仿真分析,验证了理论分析的正确性,为矩阵式变换器进一步应用研究提供了参考.     

二级倒立摆在常规滑模变结构控制实验中的应用

二级倒立摆是一个高度非线性系统,具有强耦合和快速性的特点,必须采用有效的非线性系统的控制方法才可使其稳定在平衡位置.随着智能控制的发展,大多数研究人员和学者将变结构在倒立摆控制系统中的应用,作了大量的理论研究以及仿真实验,为实际应用作了铺垫.但是仿真实验在实际应用过程存在着较强的限制性,其表现为控制器结构越复杂,仿真结...     

基于自适应阈值投影的双行车牌分割方法

Aiming at the problem that there are few algorithms and low correct segmentation rate of double row license plate number in China, an adaptive projection method for double row license plate segmentation is proposed. Firstly, the image of license plate in HSV (Hue Saturation Value) space is binarized. Secondly, the double line license plate is segmented into a single line license plate by the adaptive projection method. Finally, the characters on the license plate number are segmented. Experiments show that this method can effectively segment the double line license plate with white characters on blue background, red characters on white background, black characters on green background, and black characters on yellow background, and the correct segmentation rate of the double line electric bicycle brand is higher. At the same time, this method overcomes the shortcomings of the deep learning method such as long training time and a large number of data sets. The proposal t is a simple and efficient segmentation method.     

一种动态双种群果蝇优化算法求解FJSP

为了有效解决柔性作业车间调度问题(FJSP),提出了一种具有较强进化机制的动态双种群果蝇优化算法(DDFOA),该算法采用自适应移动步长,并动态地将种群划分为先进子种群和后进子种群,其中先进子种群侧重局部搜索,后进子种群负责全局搜索。同时针对柔性作业车间调度问题,设计了合适的编码转化方案。最后,对算法的收敛性进行了证明,并选用经典算例对其进行仿真实验,仿真结果验证了DDFOA求解FJSP的有效性。     

基于两次优先级排序的虚拟网络映射算法

为解决现有的虚拟网络映射算法忽略网络本身属性,仅按照请求到达的顺序分配资源而导致物理资源利用率低的问题,利用时间窗模型,提出了基于两次优先级排序的虚拟网络映射算法。在第一次排序中,粗化虚拟网络请求的同时根据业务类型、属性参数计算请求优先级,初步确定窗口中虚拟网络映射顺序;在第二次排序中,综合考虑链路带宽资源需求和节点途径跳数,通过链路权重来确定优先级,计算最佳映射路径。仿真结果表明,该算法降低了虚拟网络请求的平均等待时间,提高了请求接受率及收益开销比。     

A novel ESD power supply clamp circuit with double pull-down paths

Electrostatic-discharge(ESD)protection design is one of the key challenges of advanced CMOS processes.RC-triggered and MOSFET-based power supply ESD clamp circuits have been widely used to obtain the desired ESD protection ability.In this paper,a MOSFET-based ESD power clamp circuit with only10 ns RC time constant for 0.18-μm process is presented.A double pull-down path is proposed to avoid false triggering,reject power supply noise and reduce energy consumption.The performance of the novel clamp circuit is excellent,consuming very small layout area.The simulation results show that this clamp circuit can be used in industry.     

一种基于聚合度模型的WSNs双簇头分簇路由协议

在无线传感器网络（WSNs）中能量负载不均衡问题,影响了网络的生命周期。提出一种基于聚合度模型的WSNs双簇头分簇路由协议（DCHP）,DCHP协议将节点聚合度与剩余能量作为考虑因素引入阈值计算,从而使高剩余能量且聚合度高的节点优先选为第一簇头。在此基础上,根据簇内节点能量选出第二簇头,完成簇间多跳路由转发数据。仿真实验表明：DCHP协议能更好平衡网络能量负载问题,延长网络生命周期。     

大功率白光LED驱动电路的双环检测方法

基于CSMC0.5μm标准CMOS工艺,设计了一种带有双环检测的大功率LED恒流驱动芯片。仿真结果表明,芯片可在2MHz频率下工作,驱动电流最高可达1.5A,在24V电源电压时,电源效率可达95%。当电源电压在6V跳变±10%,驱动1W350mA的LED时,LED电流精度达0.02%。对比单环检测模式,该电路的LED驱动电流响应时间缩短了近2/3。此模式在保持高精度恒流和高效率的同时,有效地缩短了LED驱动电流的响应时间。     

基于骨干多边形的传感器网络分区双连通恢复算法

针对现有算法恢复分区连通性存在容错性差的问题,提出了分区双连通性恢复算法DCRA。该算法旨在网络中心区域构建骨干多边形,分区以两条互不相交的路径与多边形连接,从而实现分区间的双连通。仿真实验表明,与现有一些双连通算法相比,所提算法不仅减少了部署中继节点的数量,而且算法的运行时间能够减少60%左右,可以快速确定部署的中继节点位置,从而快速恢复分区连通性。     

一种新型滑模控制双幂次趋近律

针对滑模控制中传统趋近律存在收敛速度慢、时间长和抖振严重等不足,提出一种利用双幂次趋近律提高系统状态收敛速度的设计方案.该双幂次趋近律无论在远离滑动模态还是在接近滑动模态的空间内均具有快速收敛能力.理论分析表明,该双幂次趋近律具有二阶滑模特性,当系统存在不确定性时,系统状态及其导数可以快速收敛到平衡零点的邻域内.仿真结果表明,双幂次趋近律与传统幂次趋近律、指数趋近律、快速幂次趋近律相比,具有更快的收敛速度和更好的运动品质.     

A triangular MIMO array antenna with a double negative metamaterial superstrate to enhance bandwidth and gain

Multiple‐input‐multiple‐output (MIMO) array antenna integrated with the double negative metamaterial superstrate is presented. The triangular metamaterial unit cell is designed by combining two triangular elements positioned in complementary on the same plane at different sizes. Such design with more gaps is used to excite rooms for more capacitance effects to shift the resonance frequency thus enlarging the bandwidth of the MIMO antenna. The unit cell is arranged in 7 × 7 periodic array created a superstrate metamaterial plane where the C_stray exists in parallel between the two consecutive cells. It is found that the existence of C_stray and gaps for each unit cells significantly influenced the bandwidth of the MIMO antenna. The higher value of the capacitance will lead to the negativity of permittivity. The superstrate plane is then located on top of the 4 × 2 MIMO with a gap of 5 mm. The integration resulted in improving the bandwidth to 12.45% (5.65‐6.4GHz) compared to only 3.49% bandwidth (5.91‐6.12GHz) of the MIMO antenna itself. Moreover, the negative permeability characteristic is created by a strong magnetic field between the complementary unit cells to have 14.05‐dBi peak gain. Besides that, the proposed antenna managed to minimize the mutual coupling and improve the mean effective gain, envelope correlation coefficient, and multiplexing efficiency.     

一类离散时间代数Riccati矩阵方程对称解的双迭代算法

利用逆矩阵的Neumann级数形式,将在线性二次优化问题中遇到的含未知矩阵之逆的离散时间代数Riccati矩阵方程(DTARME)转化为高次多项式矩阵方程,然后采用牛顿算法求高次多项式矩阵方程的对称解,并采用修正共轭梯度法求由牛顿算法每一步迭代计算导出的线性矩阵方程的对称解或者对称最小二乘解,建立求DTARME的对称解的双迭代算法。双迭代算法仅要求DTARME有对称解,不要求它的对称解唯一,也不对它的系数矩阵做附加限定。数值算例表明双迭代算法是有效的。