The global adaptive H∞ synchronization is intensively investigated for the general delayed complex dynamical networks. The network under consideration contains unknown but bounded nonlinear coupling functions, time-v...The global adaptive H∞ synchronization is intensively investigated for the general delayed complex dynamical networks. The network under consideration contains unknown but bounded nonlinear coupling functions, time-varying delay, and external disturbance. Based on the Lyapunov stability theory, linear matrix inequality （LMI） optimization technique and adaptive control, several global adaptive H∞ synchronization schemes are estab- lished, which guarantee robust asymptotical synchronization of noise-perturbed network as well as a prescribed robust H∞ per- formance level. Finally, numerical simulations have shown the feasibility and effectiveness of the proposed techniques.展开更多
We propose a self-organized optimization mechanism to improve the transport capacity of complex gradient networks. We find that, regardless of network topology, the congestion pressure can be strongly reduced by the s...We propose a self-organized optimization mechanism to improve the transport capacity of complex gradient networks. We find that, regardless of network topology, the congestion pressure can be strongly reduced by the self-organized optimization mechanism. Furthermore, the random scale-free topology is more efficient to reduce congestion compared with the random Poisson topology under the optimization mechanism. The reason is that the optimization mechanism introduces the correlations between the gradient field and the local topology of the substrate network. Due to the correlations, the cutoff degree of the gradient network is strongly reduced and the number of the nodes exerting their maximal transport capacity consumedly increases. Our work presents evidence supporting the idea that scale-free networks can efficiently improve their transport capacity by self- organized mechanism under gradient-driven transport mode.展开更多
Electrochemical micromachining （EMM） technology for fabricating micro structures is presented in this article. By applying ultra short pulses, dissolution of a workpiece can be restricted to the region very close to...Electrochemical micromachining （EMM） technology for fabricating micro structures is presented in this article. By applying ultra short pulses, dissolution of a workpiece can be restricted to the region very close to the electrode. First, an EMM system for meeting the requirements of the EMM process is established. Second, sets of experiments is carried out to investigate the influence of some of the predominant electrochemical process parameters such as electrical parameters, feed rate, electrode geometry features and electrolyte composition on machining quality, especially the influences of pulse on time on shape precision and working end shape of electrode on machined surface quality. Finally, after the preliminary experiments, a complex microstructure with good shape precision and surface quality is successfully obtained.展开更多
固体充填采煤作为一种兼顾资源回收与生态保护的绿色开采方法,其核心环节煤矸分选是井下采选充一体化技术高效运行的前提,而煤矸识别作为实现煤矸精准分选的关键技术,面临着井下复杂工况中特征提取困难、边界定位模糊等挑战。为此,以固...固体充填采煤作为一种兼顾资源回收与生态保护的绿色开采方法,其核心环节煤矸分选是井下采选充一体化技术高效运行的前提,而煤矸识别作为实现煤矸精准分选的关键技术,面临着井下复杂工况中特征提取困难、边界定位模糊等挑战。为此,以固体充填采煤井下煤矸分选为研究背景,提出一种融合多模态大模型(Multimodal Large Language Model,MLLM)的固体充填采煤井下分选煤矸下落瞬态图像识别方法。首先,自主设计并搭建固体充填采煤井下分选煤矸下落瞬态图像采集实验平台,以模拟井下低照度、高粉尘的复杂工况,利用高速相机采集不同工况下的煤矸下落瞬态图像;对采集的煤矸图像进行预处理,运用优化算法提升低照度图像亮度并改善粉尘环境图像质量,同时进行标注和数据扩充,构建用于煤矸识别模型训练和测试的数据集;其次,针对传统SegFormer模型在煤矸图像边界识别中的缺陷,引入高效通道注意力机制(ECA)并优化损失函数,构建ECSegFormer模型;进一步提出将MLLM融合到ECSegFormer模型,形成MLLM-ECSegFormer煤矸识别模型融合架构,利用多模态大模型Qwen-VL(7B)提取煤矸目标中心坐标,通过高斯热图生成空间注意力掩膜,分阶段融入ECSegFormer编码器,实现多模态先验知识与图像特征的动态交互。试验结果表明,融合多模态大模型后,各经典图像识别模型性能均显著提升。其中,MLLM-ECSegFormer的MIoU提升至95.50%、MPA提升至98.92%、准确率提升至98.87%,在识别精度、模型复杂程度和识别效率方面均显著优于经典图像识别模型,且与其他图像识别模型相比,MLLM-ECSegFormer在复杂工况下的边缘识别连续性更强,尤其在粉尘干扰、煤矸形态不规则场景中,对目标区域的分割精度显著优于传统模型。研究成果为煤矸精准识别提供了新方法,提升了固体充填采煤技术的智能化水平,对煤炭资源的绿色智能开采具有重要意义。展开更多
基金Supported by the National Natural Science Foundation of China (60904060,61104127)the Open Foundation of Hubei Province Key Laboratory of Systems Science in Metallurgical Process (C201010)
文摘The global adaptive H∞ synchronization is intensively investigated for the general delayed complex dynamical networks. The network under consideration contains unknown but bounded nonlinear coupling functions, time-varying delay, and external disturbance. Based on the Lyapunov stability theory, linear matrix inequality （LMI） optimization technique and adaptive control, several global adaptive H∞ synchronization schemes are estab- lished, which guarantee robust asymptotical synchronization of noise-perturbed network as well as a prescribed robust H∞ per- formance level. Finally, numerical simulations have shown the feasibility and effectiveness of the proposed techniques.
基金Supported by the Education Foundation of Hubei Province under Grant No D20120104
文摘We propose a self-organized optimization mechanism to improve the transport capacity of complex gradient networks. We find that, regardless of network topology, the congestion pressure can be strongly reduced by the self-organized optimization mechanism. Furthermore, the random scale-free topology is more efficient to reduce congestion compared with the random Poisson topology under the optimization mechanism. The reason is that the optimization mechanism introduces the correlations between the gradient field and the local topology of the substrate network. Due to the correlations, the cutoff degree of the gradient network is strongly reduced and the number of the nodes exerting their maximal transport capacity consumedly increases. Our work presents evidence supporting the idea that scale-free networks can efficiently improve their transport capacity by self- organized mechanism under gradient-driven transport mode.
基金National Natural Science Foundation of China (50635040)National High-tech Research and Development Program (2009AA04Z302)Jiangsu Provincial Natural Science Foundation (BK2008043)
文摘Electrochemical micromachining （EMM） technology for fabricating micro structures is presented in this article. By applying ultra short pulses, dissolution of a workpiece can be restricted to the region very close to the electrode. First, an EMM system for meeting the requirements of the EMM process is established. Second, sets of experiments is carried out to investigate the influence of some of the predominant electrochemical process parameters such as electrical parameters, feed rate, electrode geometry features and electrolyte composition on machining quality, especially the influences of pulse on time on shape precision and working end shape of electrode on machined surface quality. Finally, after the preliminary experiments, a complex microstructure with good shape precision and surface quality is successfully obtained.
文摘固体充填采煤作为一种兼顾资源回收与生态保护的绿色开采方法,其核心环节煤矸分选是井下采选充一体化技术高效运行的前提,而煤矸识别作为实现煤矸精准分选的关键技术,面临着井下复杂工况中特征提取困难、边界定位模糊等挑战。为此,以固体充填采煤井下煤矸分选为研究背景,提出一种融合多模态大模型(Multimodal Large Language Model,MLLM)的固体充填采煤井下分选煤矸下落瞬态图像识别方法。首先,自主设计并搭建固体充填采煤井下分选煤矸下落瞬态图像采集实验平台,以模拟井下低照度、高粉尘的复杂工况,利用高速相机采集不同工况下的煤矸下落瞬态图像;对采集的煤矸图像进行预处理,运用优化算法提升低照度图像亮度并改善粉尘环境图像质量,同时进行标注和数据扩充,构建用于煤矸识别模型训练和测试的数据集;其次,针对传统SegFormer模型在煤矸图像边界识别中的缺陷,引入高效通道注意力机制(ECA)并优化损失函数,构建ECSegFormer模型;进一步提出将MLLM融合到ECSegFormer模型,形成MLLM-ECSegFormer煤矸识别模型融合架构,利用多模态大模型Qwen-VL(7B)提取煤矸目标中心坐标,通过高斯热图生成空间注意力掩膜,分阶段融入ECSegFormer编码器,实现多模态先验知识与图像特征的动态交互。试验结果表明,融合多模态大模型后,各经典图像识别模型性能均显著提升。其中,MLLM-ECSegFormer的MIoU提升至95.50%、MPA提升至98.92%、准确率提升至98.87%,在识别精度、模型复杂程度和识别效率方面均显著优于经典图像识别模型,且与其他图像识别模型相比,MLLM-ECSegFormer在复杂工况下的边缘识别连续性更强,尤其在粉尘干扰、煤矸形态不规则场景中,对目标区域的分割精度显著优于传统模型。研究成果为煤矸精准识别提供了新方法,提升了固体充填采煤技术的智能化水平,对煤炭资源的绿色智能开采具有重要意义。
