基于改进遗传算法的RBF网络的截球策略

在RoboCup2D仿真足球比赛中,球员的底层动作设计的好坏直接影响着比赛的输赢,特别是一些关键的动作,比如截球、传球、射门等动作。鉴于此,提出一种新的截球方法,即改进的遗传算法和RBF神经网络相结合的截球方法。该方法是用改进的遗传算法优化RBF神经网络的结构参数,通过优化,提高了网络的全局搜索效率。实验表明,经过改进遗传算法优化的RBF神经网络的截球成功率比单一的RBF神经网络截球成功率高很多。     

模糊控制在磁悬浮球系统实时控制中的应用

在磁悬浮球系统中采用PID控制时,动态性能较差,控制效果不理想.为了获得控制效果较好的控制器,根据模糊控制原理,设计磁悬浮模糊控制器.根据PID实时控制中的经验,确定模糊控制器的输入输出变量模糊化范围,建立合理的模糊规则表.在Matlab/Simulink中组建实时控制模块,并分别对磁悬浮球系统在无干扰和有干扰下进行实时控制.对控制效果进行分析讨论,证明模糊控制器具有良好的鲁棒性和在非线性系统控制中的有效性.     

Analytical estimation of error in flank milling of ruled surfaces

This article introduces a method to estimate geometric error during flank milling of a ruled surface. The various positioning schemes developed by researchers are intended to reduce this geometric error in order to mill with larger sized milling cutters while respecting the tolerance interval. There are two trends in positioning: either positioning is simple and right from the start it is easy to determine design of the maximum allowed milling cutter radius, or positioning is complex and determination of the maximum milling cutter dimensions can only be conducted after digital calculations of the error. It will then be necessary to choose another milling cutter radius and recommence the positioning procedure and error calculation in order to validate the tool. In the present study, a method to estimate error in the scope of complex positioning is presented. The aim is to be capable of choosing a maximum cutting tool radius that respects the tolerance interval.     

基于改进DeepLabV3+的铣床碎屑图像分割方法

针对铣床碎屑形状不规则导致图像分割中碎屑轮廓不清晰、分割精度低的问题,本文提出一种改进的DeepLabV3+铣床碎屑分割算法。首先在DeepLabV3+的Xcepetion模块中嵌入通道与空间注意力机制(convolutional block attention module, CBAM)模型,优化通道的权重和位置信息,加强碎屑图像区域的特征学习;其次将DeepLabV3+的空洞空间卷积池化金字塔(atrous spatial pyramid pooling, ASPP)模块改为密集连接(dense conolutional network, DenseNet)方式,增大碎屑图像特征点的感受野,提升铣床碎屑图像特征的复用效率;最后在解码过程中采用多尺度自适应特征融合方法,聚合多尺度特征作为解码器的输入特征,提高碎屑图像分割的精度与鲁棒性。实验结果表明,本文算法优于其他分割算法,改进后算法相比DeepLabV3+,像素准确率提高0.026,平均交并比(mean intersection over union,MIOU)提高0.020,F₁值提高了0.013。     

Wavelet CMAC for MIMO uncertain nonlinear systems using a modified social ski-driver algorithm

This study uses a Mexican hat wavelet membership function for a cerebellar model articulation controller (CMAC) to develop a more efficient adaptive controller for multiple input multiple output (MIMO) uncertain nonlinear systems. The main controller is called the adaptive Mexican hat wavelet CMAC (MWCMAC), and an auxiliary controller is used to remove the residual error. For the MWCMAC, the online learning laws are derived from the gradient descent method. In addition, the learning rate values are very important and have a great impact on the performance of the control system; however, they are difficult to choose accurately. Therefore, a modified social ski driver (SSD) algorithm is proposed to find optimal learning rates for the control parameters. Finally, a magnetic ball levitation system and a nine-link biped robot are used to illustrate the effectiveness of the proposed SSD-based MWCMAC control system. The comparisons with other existing control algorithms have shown the superiority of the proposed control system.     

Feed-direction ultrasonic vibration−assisted milling surface texture formation

The formation mechanism of surface texture for feed-direction ultrasonic vibration?assisted milling (UVAM) was investigated in this study by establishing a kinematic model and a pressing model of UVAM. The kinematic model showed that the cutter tip, which was supposed to be totally sharp, produced closed scratches by crossed trajectories. The variation trends of the interval for closed textures were of sine function. A comparative experiment was carried out by UVAM and conventional milling. A dividing line close to the X coordinate divides the surface feature into two parts. The pressing model showed that the tool minor cutting edge left clear traces with certain width because of the tool minor cutting edge angle. Scratches by tool minor cutting edge were intermittent and regularly varied when feed-direction vibration was introduced. All the surface feature changes are in the radial direction and the trajectory intersections shall always be the scratch grooves or ridges. The ratio between ultrasonic vibration frequency and spindle speed, tool radius, and the located cutter rotation angle affected the changing rule of scratches by tool minor cutting edge. The analytical models and the experimental results proved to each other reasonable.     

Mechanical activation-assisted electro-thermal explosion synthesis of the Ti2AlC phase

This study deals with the synthesis of the Ti₂AlC phase using the Electro-Thermal Explosion under Pressure with Confinement (ETEPC) technique. The effects of the ETEPC technique and the milling process parameters on the TiC_x phase content and the formation mechanism of the Ti₂AlC phase were investigated. The latter is mainly affected by the morphology of the powder mixture and aluminum melted amount. The optimization of the above parameters allowed the achievement of the desired reaction, leading to the formation of the Ti₂AlC phase with a purity of about 97?wt%. The results clearly demonstrate that the ETEPC process enables one to control both time and material synthesis temperature.     

Formulation and performance evaluation of a novel coconut oil–based metalworking fluid

Metalworking fluid (MWF) supplies a film of lubricant to abate friction, acts as a cooling media to rebate induced heat, and prevents metal pick-ups by flushing away the chips. Hence a liquid used as a cutting fluid reduces wear on the tool, reduces the energy consumption, and produces a better surface quality on the work piece. This paper describes the formulation of a novel water-soluble MWF and its performance evaluation during straight turning and end milling experiments carried out with AISI 304 stainless steel, mild steel, and cast iron as work piece materials. The MWF was prepared by mixing water with white coconut oil as the base oil and food-grade additives as surfactants. Viscosity, pH value, and biodegradability were measured and compared with a commercially available non-vegetable oil–based MWF. The surface roughness and tool surface temperature were measured throughout the machining experiments, and better performances were observed with the coconut oil–based MWF. Tool tip geometry and flank wear for straight turning machining operation were identified by observing scanning electron microscope (SEM) images.     

Application of a hot-melt granulation process to enhance fenofibrate solid dose manufacturing

Evaluation of hot-melt granulation of fenofibrate and croscarmellose sodium and its cooling time for the molten mass in a ratio of 55:45 was conducted to assess the manufacturing process capability to produce an acceptable granulation which flows well on Korsch PH300 tablet compression machine. The formation of the drug-polymer eutectic mixture was investigated by differential scanning calorimetry, scanning electron microscopy and X-ray powder diffraction. The physical properties of the hot-melt was determined by examining the milled blocks after solidification and milling after cooling periods of 10, 20 and 30 d. The milled material was assessed for the effect of hold time of the blend on the solid dose compression characteristics. The impact of cooling on the processing of the blocks was assessed after 10, 20 and 30 d of cooling. The study suggests that after the hot-melt formed the fenofibrate crystallized independently and a solid solution with croscarmellose sodium was not formed. The age of the blocks determined the hardness of the crystals, changing the processing nature of the granules with respect to compression and powder flow characteristics. The blocks processed after 20 d and beyond produced granules with a characteristic suitable for holding the blend for 14 d in the bin with no impact on flow properties and compressibility of the blend. There was no chipping, capping, sticking or picking observed and a higher compression speed was achieved.