为降低边缘计算系统中卸载任务的调度代价,以满足任务延时需求和系统代价最小为目标,建立边缘计算系统代价优化模型.证实该问题是NP-hard问题,提出一种基于两阶段代价优化任务调度算法TTSCO.根据改进BF算法求解初步任务调度方案,降低首阶段的非必要代价,生成调度最终解.分析算法时间复杂度,通过仿真实验验证了算法性能,实验结果表明,与LINGO理论最优解相比,调度解的近似比在95％情况下均小于1.2,该算法可以进一步降低系统代价.

退役动力电池参数一致性差,串并联后用于储能,木桶效应明显且故障率高.为了解决上述问题,提出了基于模块化多电平变换器(MMC)改进的模块化多电平H桥变换器(MMHC)分布式储能可解决上述问题.但MMHC储能变换器子模块众多,阻容感网络复杂,采用传统的并网模型对其进行研究并不能精确地描述其动态行为和静态行为,基于其传统的并网模型设计的整数阶PI控制器亦不能达到快速动态响应和较好鲁棒性的要求.针对上述问题,提出了从频域方面辨识出MMHC储能变换器的分数阶并网模型,基于该分数阶并网模型以ITAE准则和ISE准则为性能指标,采用模型匹配方法设计分数阶PIλ控制器.通过仿真和实验比较分析,分数阶模型可更好地反映复杂系统的行为,设计的分数阶控制器具有快速的动态响应和更好的鲁棒性,实现了MMHC储能变换器优越的并网运行.

纯电动汽车永磁同步电机(PMSM)控制系统是一种非线性、强耦合的复杂系统,为提高永磁同步电机的整体控制性能以及鲁棒性,文中提出了一种双闭环分数阶PID控制策略,分别对电流内环和速度外环进行相应的分数阶控制.通过建立电机的电流环模型进而得到整体系统的分数阶模型,设计出电流环以及速度环的分数阶PID控制器,并利用粒子群优化算法(PSO)对控制器参数进行寻优,以便于获得更好的控制性能.将所设计的分数阶控制器应用于车用永磁同步电机的控制中,通过与传统PID控制在不同转速、负载突变时的控制性能相对比,验证了文中所设计的分数阶PID控制器具有良好的动态特性和鲁棒性.

以能源互联网的最核心环节—能源路由器为研究背景,将前端级联H桥和双有源全桥(DAB)级分别进行独立控制时,将DAB级等效为输入独立输出并联拓扑进行研究.为提高其输出端负载情况突变及输入电压突变时输出电压的动态性能,借鉴直接功率思想,结合闭环控制和前馈控制,以H桥内的超前桥臂与滞后桥臂之间的移相角为控制变量,提出了一种复合控制策略.通过实验将所提复合控制与传统电压闭环控制进行性能对比验证.实验表明:该控制方法极大提升了变换器的动态性能,其控制实现也相对简单且对电路参数依赖性较小,有较好的兼容性和可移植性,在其他类型变换器动态响应研究中有借鉴的潜力.

退役动力电池梯次利用,不仅商业效益明显而且环境效益突出.将不同品牌、不同规格的退役动力电池进行重组用于储能时,面临电池参数差异大、"木桶效应"明显等问题,进行并网时,易导致相内不均衡.针对该问题,以MMHC(模块化多电平混合换流器)储能变换器为硬件基础,考虑电池SOC(荷电状态),采用一种SOC分层控制方法进行相内均衡控制.搭建实验平台进行验证,结果表明:提出的控制方法可以达到解决不同品牌、不同规格退役动力电池梯次利用相内均衡的要求,延长了退役动力电池的循环利用寿命.

变压器时常发生磁饱和现象.为防止磁饱和的发生,通常在磁路中加入一段气隙或减少绕组匝数.研究在磁回路中加入气隙来避免磁饱和时,气隙量对高频平面变压器特性参数的影响.以TDK铁氧体设计的高频平面变压器为研究对象,通过理论分析计算出饱和电流和气隙量.在ANSYS中建立3D仿真模型和电路简化模型,改变气隙量,分析高频平面变压器的涡流场、静电场特性以及原副边电压特性,将仿真数据用MATLAB处理后,得出其电气参数变化规律.

Cotton segmentation is the key step of cotton picking robot vision system. The accurate and effective segmenta-tion of cotton is useful to its position in three-dimensional space while cotton is in bright light or shadow complex field envi-ronment. It can get fitness function by training BP neural network and its output error, and then use selection, crossover and mutation operation in genetic algorithm to optimize neural network weights and threshold until the output error meets the requirement or it reaches a predetermined number of iterations. Finally, according to the obtained BP neural network weights and threshold, it segments cotton image. The experiment of image segmentation with 136 images photographed in cotton field environment shows that the algorithm can segment cotton image in bright light or shadow accurately and seg-mentation accuracy rate is up to 91.9%and it converges faster than BP.

Because of the direct sunlight and shadow effection,it is difficult to segment cotton image in the natural environment.In order to segment cotton exactly,a new method is proposed combined color clustering and V channel of HSV model.Firstly,the original image is preprocessed by P-M diffusion filtering;secondly,improved K-means clustering algorithm is adopted to finish initial segmentation;lastly,V channel of HSV model is choosen as the feature of cotton image,which may get rid of background effect.The experiments show that the proposed algorithm has accurate segmentation result in the direct sunlight and shadow conditions.

A new kind of cotton image segmentation algorithm based on Support Vector Machine(SVM) is proposed.The cotton image is classified into the target and the background.The sample pixel value is extracted from the two categories under OHTA color space.These samples are trained with category information by SVM.Otsu method is used to deal with the image.SVM classifier is trained to segment the cotton image after pretreatment,and the noise is removed by region label.Experimental results show that this method can segment cotton image with the complex background,and the accuracy of segmentation cotton images is as high as 92.3%.Its classify speed is better than directly using SVM to segment the image.The accuracy and stability are better than threshold segmentation method.

Because of the direct sunlight and shadow effect,it is difficult to segment cotton image in the natural environment.In order to segment cotton exactly,an improved segmentation algorithm based on watershed is proposed.In the method,a P-M diffusion pattern denoising filter is adopted in the original image.The multi-scale morphology gradient is modified by the hard threshold which gets from robust mean value estimation.Watershed transform is utilized to implement segmentation with the modifying gradient image and a region merging method based on L*a*b* space neighbor color similarity is employed to improve the segmentation result.The experiments show that the proposed algorithm has accurate segmentation result in the direct sunlight and shadow conditions.