在共享经济时代，物流服务行业建立稳定的联盟至关重要。因此，在物流企业联盟内部，如何确保公平的收益分配成为一大挑战。然而，由于联盟存在信息不完备的情况，一些传统的合作博弈单值解，例如Shapley值，可能并不适用于该问题，这些解法更适用于联盟的总体收益和各博弈者的参与水平都能够被准确估计的情形。研究了信息不完备的物流企业联盟的收益分配问题，并提出了一种相关的收益分配模型。为了证明所提出模型的适用性，进行了一个案例研究。结果表明，模糊Shapley值能合理地提升多方合作后的各方收益，有效地分配物流企业联盟的收益。

产业互联网是站在整个产业链视角，充分运用沉淀的产业数据，进行痛点分析和资源匹配，向全产业提供普惠式的综合服务和基础设施建设，提升产业链供应链的稳定性和现代化。在产业链发展中，上下游企业以及企业与终端用户间的业务对接和协同已经成为了一个关键问题。而传统的信息检索方式由于存在检索方式单一、检索内容固化等原因难以有效适应产业链中对相关产品和业务的个性化检索和信息咨询需求。随着人工智能的发展，智能客服已经成为一种新的解决产业链个性化信息检索和咨询问题的关键技术手段。然而，传统智能客服的适应能力相对有限，有时对特定问题的回答还需要在系统中进行预先定义，对于不断变化的业务语境难以灵活应对。为了提供一种面向产业链上下游企业的可以灵活适配目标企业业务需求的智能客服系统，提出了一种基于大语言模型（LLM）的领域自适应智能客服构建方法，该方法利用开源的大语言模型，使得中小企业可以针对具体下游任务进行垂域适配，结合私域知识库搭建低成本高质量的客户服务，同时也能避免敏感数据泄露问题，从而能够解决产业互联网中灵活地智能客服系统构建问题，降低产业互联网中中小企业低成本系统构建需求。

In order to solve the problem of task allocation and worker fatigue in hybrid flow shop， a mixed integer programming model considering worker fatigue is proposed. According to the characteristics of the problem， the real number coding method is used to solve the problem of process sequencing and worker assignment， and an improved nutcracker optimization algorithm （INOA） is proposed to solve the problem. Firstly， reverse learning population initialization is used to increase the diversity of the population. Secondly， the parameter factors \mu and \gamma are optimized to improve the search efficiency and convergence speed of the algorithm. Finally， in order to improve the robustness of the algorithm， a population diversity monitoring mechanism is introduced， and Cauchy disturbance is used to avoid stagnation caused by too low population diversity. The INOA and other algorithms are tested on 12 functions. The test results show that the performance of INOA is significantly better than other algorithms. Through the example verification of the connection line assembly line， the results further verify the superiority of the algorithm. The proposed optimization scheme can effectively shorten the completion cycle， prevent excessive fatigue of workers， and promote the rational allocation of tasks.

To address the issues of low efficiency， excessive redundancy， and the inability to dynamically avoid obstacles in indoor robot path planning when using the A^* algorithm， this paper proposes a fusion algorithm that combines an optimized A^* algorithm with the dynamic window approach. The proposed algorithm enhances the heuristic function by incorporating the distance from the parent node to the target node. It quantifies the obstacle information to dynamically adjust the weights of the heuristic function using an obstacle rate function. Additionally， it introduces a cost for turning to reduce unnecessary turns in the path and designs a strategy to remove redundant points， ensuring a globally optimal static path. It incorporates the offside angle， flexibly selects the key nodes of the A^* algorithm as local target points within the dynamic window to avoid the path getting trapped in local optima. The experimental results demonstrate that the improved fusion algorithm enhances the search efficiency， reduces the path length， resolves the issue of the dynamic window approach being trapped in local optima and enables the real-time obstacle avoidance.

Detection on the surface defect of Pouch Cells is a key procedure of the production process. Aiming at the problems of low detection accuracy and difficult imaging of large-sized batteries in the existing detection methods， a detection method based on photometric stereo imaging and deep learning was proposed. Firstly， a Multi-Source Time-Sharing Exposure Imaging System （MSTIS） was established by combining photometric stereo and line scan camera imaging technology. After obtaining the surface images of batteries under different light sources through time-sharing exposure， photometric stereo calculation was conducted to obtain the curvature map with 3D information. Then， to solve the problem of missed detection of minor target and multi-scale defects， the YOLOv8 algorithm was improved. An edge information enhancement module （EIEM） was developed using a dual-channel convolution structure， which incorporated Sobel convolution and conventional convolution to improve feature edge extraction capabilities. The semantic and detail information fusion method （SDI） was integrated with the bidirectional feature pyramid module to boost the recognition accuracy of tiny defects. A lightweight shared convolution detection head was also implemented to reduce the algorithm's computational load.The experimental results show that the average detection accuracy of this method reaches 94.2% and the detection speed reaches 116 FPS， which can effectively detect the surface defects of pouch cells.

随着智能制造、人工智能等技术的快速发展，立体仓储和仓储机器人得到了广泛的应用，空间定位测量技术的需求提升。传统的定位测量技术存在精度低、测量范围有限的问题。为此，针对大尺寸空间三维立体位姿测量的应用场景，首先采用旋转激光扫描机构构建光电传感网络，激光平面后方交会于待测目标，然后运用EPNP算法求出物料运输机器人在激光扫描机构坐标系的3D坐标，最后利用单位四元数法求出物料运输机器人相对于世界坐标系的位姿信息。应用结果表明：该方法能够进行大尺寸空间三维立体定位测量，系统稳定，鲁棒性强，能够满足实际应用需求。

Aiming at the problems of low defect edge contrast， wide defect scale span and dense tiny defects in online steel surface quality inspection， which cause traditional lightweight models to suffer from insufficient recognition accuracy and poor adaptability to real-time production line working conditions， this paper proposes an improved steel surface defect detection algorithm named APGS-YOLO based on YOLO11. Firstly， a C3k2-AP multi-scale edge enhancement module is designed to improve the feature extraction capability for tiny defects and suppress the interference from steel plate background textures by means of feature branching and multi-directional asymmetric convolution. Secondly， the GhostBottleneckV2 lightweight attention residual module is introduced to enhance global feature fusion while reducing model computational overhead. Thirdly， a cross-scale feature fusion strategy is adopted to strengthen the extraction capability of multi-scale defect features. Finally， a CSSDIoU scale-sensitive hybrid loss function is constructed to improve the localization accuracy of tiny defects. Experimental results show that the mAP@0.5 and mAP@0.5：0.95 of the proposed APGS-YOLO reach 80.3% and 48.2%， which are 2.9% and 2.5% higher than those of the original YOLO11n. The proposed model outperforms other mainstream object detection models. Meanwhile， the model achieves a 7.0% parameter compression rate with an inference speed of 86 frames per second. The proposed algorithm balances detection accuracy and real-time inference performance， which can satisfy the automatic online quality inspection requirements of steel production lines.

The dexterous multi-fingered robotic hand， serving as a key end-effector， is pivotal for enabling robots to perform fine-grained grasping and compliant manipulation. Its advancement holds significant importance for promoting automation in manufacturing， enhancing the intelligence of service robots， and expanding applications in specialized environments. Focusing on humanoid multi-fingered dexterous hand technologies， this paper systematically reviews the current state-of-the-art and future trends. It begins by elucidating the fundamental concepts， system architecture， and typical characteristics of dexterous hands. This is followed by a comprehensive of research achievements from domestic and international teams and commercially available mainstream multi-fingered dexterous hand products， covering various degrees-of-freedom designs and their respective hardware and software implementations. Key technologies， including core hardware components， multi-modal sensory fusion， and control strategies， are critically analyzed. The paper subsequently summarizes practical applications across domains such as industrial assembly， daily life assistance， and operations in extreme environments. Current challenges， particularly in reliability， multi-modal coordination， generalization capability， human-robot safety， and integration and application， are identified. Finally， future research directions are prospected from multiple perspectives， including standard establishment， novel mechanical structures， advanced multi-modal perception and fusion， bionic evolution， and embodied intelligence， aiming to provide valuable insights for in-depth research and groundbreaking applications of dexterous hands.

Design for Assembly （DFA） is an important approach for improving manufacturability and reducing assembly costs during the early design stage of mechanical products. However， existing DFA evaluation methods mainly rely on manual experience and rule-based inspection， making it difficult to establish cross-domain associations among geometric design information， assembly process knowledge， and manufacturing resource capabilities. Consequently， assembly constraint conflicts cannot be effectively identified and traced at the early design stage. To address the aforementioned issues， this study proposes a knowledge-based reasoning and design optimization method for mechanical-product DFA under assembly constraints. Firstly， an automatic part DFA feature extraction method based on computational geometry is developed. By analyzing the B-Rep geometric topology information of STEP models， structured design features including dimensional parameters， physical properties， symmetry， and stability are extracted automatically， enabling the transformation from three-dimensional models into DFA evaluation data. Secondly， a ternary collaborative ontology model of “part-process-equipment” is established. Through knowledge graph representation， semantic associations among design attributes， assembly processes， and manufacturing resources are constructed， and dynamic mapping of manufacturing resource parameters is achieved. Furthermore， a hybrid reasoning mechanism integrating OWL-based description logic reasoning and Python-based rule reasoning is developed. Combined with a hierarchical 931 scoring evaluation system， the proposed method realizes assembly constraint conflict identification， cause tracing， constraint severity evaluation， and design optimization feedback. A butterfly valve assembly containing 58 components is selected as the experimental case. Two typical constraint problems， including manufacturing equipment capability mismatch and structural design constraints， are analyzed to validate the proposed method. In addition， modular ablation experiments are conducted to investigate the impacts of geometric feature extraction， knowledge modeling， and hybrid reasoning mechanisms on the performance of the method. The experimental results demonstrate that the proposed method can effectively support assembly constraint diagnosis during the early design stage of mechanical products， achieving design problem identification， conflict cause tracing， and optimization decision support. The proposed method provides an effective approach for manufacturing-oriented collaborative design optimization.

Currently reported weld defect detection methods suffer from issues such as missed detection of small-scale object defects and insufficient multi-scale feature representation.So， this paper proposes a YOLOv-RPW-based weld defect detection method. First， the method introduces a C3K2-RAFConv module based on receptive field attention mechanism into the backbone network of the YOLOv11 model， enhancing local receptive field feature representation to improve the extraction of fine-grained defect features like cracks and pores. Second， considering the high proportion of small-scale objects in weld defects， a lightweight P2-Lite small-object detection branch is designed in the detection head to fully utilize shallow high-resolution features， boosting small defect detection capability while controlling model complexity. Finally， the WIoU loss function is introduced to optimize the bounding box regression process， enhancing the model's localization accuracy for defect objects. Experimental results demonstrate that the proposed YOLOv-RPW model achieves superior performance in weld defect detection tasks， with P， R， mAP@0.5， and mAP@0.5：0.95 reaching 91.26%， 83.58%， 90.31%， and 67.24%， respectively， showing improvements of 4.31%， 4.31%， 4.35%， and 3.27% compared with the baseline YOLOv11 model. In summary， the proposed method effectively reduces false and missed detections of weld defects， significantly improves the model's detection accuracy and localization capability for small-scale weld defects， and provides an effective solution for automated weld defect inspection in industrial scenarios.

Predictive maintenance reduces unnecessary maintenance by timely diagnosing or predicting the state of the device， and deep learning methods can realize the above process by mapping the device feature parameters to the state of the device. The wide variety of equipment and the predictive maintenance of equipment in multiple working conditions also place higher demands for deep learning methods， and a reasonable selection of deep learning methods thus becomes crucial. Therefore， the development process and the latest application of deep learning in predictive maintenance are discussed in this paper， with a view to further promote the application of predictive maintenance in more devices， to improve the accuracy of deep learning methods in predictive maintenance applications， and to envision the future development of deep learning methods in predictive maintenance.

Currently， the fault diagnosis methods for the equipment in the insulation board production line lack systematic organization and effective utilization of the historical fault data. To address this issue， a method for constructing a fault knowledge graph for insulation board production lines is proposed. Firstly， the characteristics of selected fault knowledge were analyzed to define the ontology of fault knowledge for insulation board production lines. Secondly， multi-source fault data was preprocessed at the data level and sequentially annotated to form a dataset suitable for experimentation. Based on this data， deep learning models such as BERT-BiLSTM-CRF were introduced on entity recognition. Entities were matched based on syntactic trees to generate triplet data. Finally， the Neo4j graph database was employed to store the triplet information， completing the construction of the production line fault knowledge graph. The research results indicate that the entity recognition model achieved an accuracy of 84.84%， a recall rate of 89.88%， and an F ₁ score of 87.29%. Leveraging this， the fault knowledge graph for the insulation board production line was established， enabling effective utilization of fault knowledge and precise explanation of fault causes. These findings can provide knowledge support for fault diagnosis in insulation board production lines.

针对智能机器人提出一种基于自适应事件触发机制的鲁棒路径跟踪控制算法。首先，考虑到系统的非线性特征，采用T-S模糊模型方法逼近系统横向动力学；其次，考虑到有限网络带宽引起的通信约束，在设计鲁棒控制时引入了先进的自适应事件触发方案。然后基于李亚普诺夫函数理论，给出了一组充分条件来计算所设计控制器的事件触发权矩阵和控制增益。最后，在MATLAB/Simulink平台上进行了仿真模拟以说明所提方法的有效性和鲁棒性。

To address the issues of low efficiency and susceptibility to local optima in the layout planning and optimization of casing production workshops， a mathematical model for the layout planning and optimization of aero-engine casing workshop was developed with the objective of balancing space utilization and logistics costs， incorporating constraints from the casing production process and workshop logistics pathways. Considering the complexity of both production and logistics in casing manufacturing， as well as the characteristics of genetic algorithms and harmony search algorithms， an integrated approach combining Systematic Layout Planning （SLP） with an improved genetic-harmony search algorithm （SLP-GA-HS） was proposed. The procedural flow and specific implementation of the algorithm were designed in detail. Using the layout planning and optimization of a specific casing production workshop as a case study， a workshop layout optimization model was constructed. Analyses of logistics and non-logistics were conducted， followed by optimization algorithm solving and comparative algorithm analysis， demonstrating the effectiveness of the proposed method.

Addressing the industrial challenges of geometric-semantic fragmentation， low dynamic updating efficiency， and the trade-off between high precision and lightweight representation in 3D reconstruction of power grid equipment， this paper proposes a novel method integrating Neural Radiance Fields （NeRF） with lightweight Building Information Modeling （BIM）. Firstly， a cross-modal alignment mechanism is designed： NeRF parses multi-source data to generate a high-fidelity density field， while an innovatively introduced Attribute-Aware Attention Module dynamically embeds BIM semantic parameters into the NeRF feature space. This achieves joint representation of geometric structures and equipment attributes， resolving semantic absence. Secondly， an incremental updating algorithm is proposed： local change regions are identified via structural similarity index （SSIM） difference analysis and BIM change logs. The adaptive learning rate pruning optimizer AdaBound fine-tunes local NeRF parameters， and a Git-style incremental storage strategy reduces single-device update latency to 3.2 minutes while cutting storage overhead by 98%. Lastly， a cloud-edge-end collaborative architecture is constructed： The model is compressed to 0.3 M parameters via knowledge distillation. Coupled with a static-dynamic hierarchical rendering strategy， mobile-end loading latency of 180 ms and an interactive frame rate of 30 FPS are achieved. Experiments in a 220 kV substation scenario demonstrate superiority over traditional BIM and standard NeRF： Chamfer Distance drops to 2.03 cm， semantic recall reaches 98.7%， and transmission success rate under weak network conditions rises to 98%. This method provides key technical support for smart grid inspection and dynamic operational analysis.

在按照船级社标准，进行船用柴油机曲轴疲劳强度校核时，需要进行大量曲线图数据查询工作，包括曲轴各个位置的应力集中系数、结构细节的影响系数等参数，给曲轴疲劳强度计算带来麻烦和不稳定因素，同时也是曲轴疲劳强度计算软件开发的障碍。首先，针对图像干扰多，多曲线等问题，通过图像预处理、灰色图像二值化、图像降噪等过程，提取了曲线数据；研究了数据的曲线拟合方法，并通过分析比较多种曲线拟合方法，得出了精度高和使用方便的数据拟合公式。

智能制造是信息技术与先进制造技术深度融合形成的先进生产方式，其发展伴随着信息化的进步，经历了数字化、网络化两个阶段，目前正处于具有感知学习能力的制造系统为典型特征的智能化阶段，称之为新一代智能制造阶段。本文立足该阶段需求及特点，以人工智能的应用场景及作用为主要探讨对象，系统介绍了智能制造的三个发展阶段，阐释了人工智能技术之于新一代智能制造系统的重要作用，进而提出人工智能技术在制造环节应用的几点思索，希望为智能制造相关实践提供一些参考。

Aiming at the problem that traditional convolutional neural networks cannot effectively extract global features and some deep learning models are more complex， this paper proposes a lightweight gearbox fault diagnosis method based on multi-scale convolutional neural networks and feature fusion ViT. Firstly， a multi-scale feature extraction module is constructed， which captures the feature information of the data from multiple scales by multi-scale convolutional neural network using different scale convolution kernels， and fully exploits the local features of the input information. Then， the feature fusion ViT module is designed， which utilises an improved multi-attention mechanism to capture the global features of the fault information， and further constructs the D-MLP to reduce the number of parameters in the model using depth-separable convolution. Finally， the experimental validation is given using gearbox data from Southeast University， and the results show that， compared with the comparison methods， the proposed method has high fault diagnosis accuracy and good generalization ability under complex conditions such as variable operating conditions and variable noise.

高效电梯集群控制算法对于缩短乘客候梯时间，以及降低电梯能源消耗具有重要作用。针对高层电梯集群控制需求，首先，对传统最短距离算法进行深入分析与设计。其次，为解决该算法在外呼信号实时最优分配方面的不足，结合PLC扫描原理进行改进，实现了外呼信号的N次灵活分配，从而大幅提升了算法的灵活性和适应性，优化了系统运行性能和效率。最后，通过西门子S7-1200 PLC控制器和电梯仿真软件，验证了改进算法的显著优势。

为保证双足机器人的运动过程中机身的稳定性并能够抵抗一定程度的外部冲击力的干扰，设计了一种基于虚拟模型控制（VMC）与全身控制（WBC）的双足机器人力矩控制方法。该方法用虚拟模型控制对机器人模型进行简化，用得到的降阶模型求解机器人行走过程中的地面支反力；用全身控制对机器人控制任务进行优先级排序，并求解运动过程中的关节加速度；最后将二者代入刚体动力学方程中求解控制所需要的关节力矩。控制目标为驼鸟形双足机器人，使用的仿真平台为MIT机器人仿真平台。经过仿真验证，该控制方法可以有效地对目标进行控制，且机身的稳定性高，计算的求解效率高，运算复杂程度低，证明了该方法对于双足机器人控制的效果较好。

The machining quality of electric spindles critically determines precision， efficiency， and stability in precision manufacturing. However， the machining process faces challenges due to diverse product types， multiple operating conditions and scarce target-condition data， making consistent quality of electric spindle difficult to guarantee. To address this， this paper proposes a transfer-learning-based method for multi-operating-condition quality prediction. The method first extracts spindle time-series signals and employs the Synthetic Minority Over-sampling Technique to balance historical and target-condition data distributions. Subsequently， constructs a two-stage regression model， TrAdaboost.R2， and leverages knowledge transfer to predict spindle quality under target conditions. Finally， the proposed method is validated with electric spindle data， demonstrating its superior prediction performance. This approach provides an effective framework for the precise quality prediction of electric spindles across varying operating conditions.

针对受不确定性影响的非线性液压伺服系统，设计了一种基于高增益比例积分观测器的全局快速终端滑模控制方法。首先，采用输入输出反馈线性化方法对非线性模型进行线性化处理，使其适用于观测器与控制器的设计结构，其中比例积分观测器用于估计包括建模误差、外部干扰和测量噪声等综合不确定性，全局快速终端滑模控制器用于快速补偿观测器误差，从而进一步提高控制系统的鲁棒性。此外，根据李亚普诺夫稳定性理论证明了整个闭环系统的稳定性。最后，将所设计方法与传统PID控制以及传统滑模控制方法在相同的不确定性环境下进行了对比实验，实验结果表明所提方法具有更快的响应速度、更高的跟踪精度与抗干扰能力。