In order to meet the requirements of high-precision, low-damage, low-stress assembly performance and technical indicators of the composite main bearing structure of the new-generation large aircraft, this paper takes the coupling analysis and control of various key features of the product as the design clue, takes the engineering application research as the main line, and systematically expounds the digital high-performance assembly coordination technical system with key feature identification, tolerance simulation analysis, deformation prediction in assembly process, preassembly analysis of measured data and high-precision assembly compensation as the main technical paths. Taking the assembly of a typical composite outer wing box as the object, the above method is applied to a complete whole-process engineering application research and experimental verification, which solves the problems of assembly coordination and product compensation  trategy design of the composite outer wing box components. The product assembly quality meets the design requirements.

In response to the problems of multiple data sources, inconsistent description forms, and low utilization in helicopter assembly production lines, research was conducted on multi-source data perception and fusion application technology for helicopter assembly production lines. Firstly, the connotation and overall framework of multi-source data perception and fusion technology in production line operation were defined, including the construction of a multi-source data collection system for production lines, the construction of a multi-source data fusion model, the construction of a multisource data analysis and prediction model for assembly processes, and the application module for multi-source data analysis in production lines. Secondly, based on the multi-source data from actual assembly lines, we planned to adopt standardized production Internet of things (IOT) technology, ontology based semantic description methods, and multi-source data based quality mapping technology to provide feasible solutions for perception and fusion technology in production lines. Finally, the application measures of the above technologies in virtual operation process, including analysis of production line operation status, monitoring of equipment usage status, and improvement of assembly quality were elaborated, which provided a data foundation for the digital twin construction and virtual efficient operation of helicopter assembly production lines.

To address the issues of large theoretical model deviations and long finite element model consumption in traditional aircraft docking simulation, a data-driven aircraft docking process simulation technology research is carried out. Firstly, by using virtual–real fusion technology, a six-degree of freedom platform kinematic model and a body–attitude adjustment platform pose transformation model were constructed. Real-time simulation data transmission was achieved through data exchange to drive the virtual model. The measurement data and virtual model were used for fuselage docking simulation, and the docking process pose parameters were determined, providing a data basis for adjusting the attitude adjustment platform parameters. Afterwards, in order to improve the efficiency of data solving, based on the finite element model for solving the deformation of the fuselage, the deformation amount is calculated and converted into the change in fuselage position. The attitude angle and position change of the fuselage are used as input and output values to construct a surrogate model, and the effectiveness of the method is verified. Finally, a simulation system for the docking process of the fuselage was developed. Taking the docking process of the fuselage test piece as an example, the results verified that the system was feasible.

Composite materials are widely used in the aerospace field due to their excellent comprehensive properties. The application range of composite materials is gradually expanding from secondary load-bearing structures to main load-bearing structures. The traditional metal assembly panels constituting the fuselage structure are gradually been replaced by composite integral panels. The composite panels with assembly processing characteristics that are different from traditional metal panels, therefore, new requirements for assembly methods and assembly processes are proposed. Aiming at the assembly process of aircraft composite fuselage panels, the large-scale measurement technology, the assembly positioning and attitude adjustment technology, and the advanced hole connection of composite fuselage panels are introduced. Research progress and application situation at home and abroad in recent years are systematically summarized, and the future research and application development direction of aircraft large-scale composite structure assembly technology is expounded.

Spindle is a typical multi-bearing rotor system, it is difficult to analyze its accuracy due to the extensive parallel connection in its structure and the mutual coupling of bearing and component errors. To address this problem, this study introduces a novel approach for analyzing the rotation accuracy of aviation manufacturing spindles. Firstly, the small displacement torsor theory and the skin model method were used to model the form and location tolerance and rotation trajectory of the bearings to construct a full tolerance model of the components. Secondly, the full parallel connection in the spindle was analyzed, and a method for analyzing the rotation accuracy of the spindle based on the optimization method solving the error propagation of components was proposed. Finally, the tolerance analysis and sensitivity calculation of a certain type of grinding spindle are carried out using the proposed method, and the results show that the method can effectively analyze the rotation accuracy of the spindle and the pass rate of accuracy increases from 74.3% to 88.2% through improved tolerance design, demonstrating an effectively improved tolerance design level of the spindle.

Aiming at the automation measurement problem of aircraft skin seam gaps and step differences, a skin seam autonomous tracking method is studied based on line structured light visual detection technology. By combining traditional image morphology based processing methods with optical flow methods, real-time tracking of skin seam features is achieved, thereby determining the measurement positions of gaps and step differences. The accuracy of seam tracking was improved by correcting the eye-in-hand calibration parameters based on measured data. The experimental results show that when the tracking speed is 7 mm/s, the position accuracy of seam tracking is higher than 0.500 mm, and the attitude accuracy of seam tracking is higher than 0.5°. This tracking system has good accuracy and stability and can meet the requirements of aircraft skin seam tracking.

Augmented reality (AR) enhances an operator’s capacity to perceive information by superimposing visual assembly process instructions onto the workbench. This technology significantly improves the quality and efficiency of complex product assembly, such as in aerospace industries, while also reduces cognitive load on operators. This paper initially examines and discusses key technologies in current AR assembly, including tracking, human–computer interaction, and display. Subsequently, it systematically reviews the research and application status of AR assembly for existing aerospace products from three perspectives: Head-mounted AR, handheld AR, and projection AR, and this paper also explores the impact of artificial intelligence technology on AR assembly. Lastly, it summarizes the existing challenges and future development trends of AR assembly technology, with the aim of providing a reference for the research and  engineering application of intelligent assembly of complex products in aerospace industry.

Augmented reality-based assembly guidance systems superimpose digital information onto physical scenes to effectively guide complex assembly tasks. However, the gap between humans and the physical world in the assembly environment is huge, and the information to be fused into the physical world needs to be prepared in advance and triggered manually during the assembly process. The study of real-time and ubiquitous prompts is a hot research topic for complex assembly in augmented reality environment. In this paper, we propose an augmented reality assembly method based on the assistance of large language models (LLMs), the core of which is to use LLMs as another brain in the assembly process, providing ubiquitous assembly guidance and prompts support for technique information. Firstly, a system of LLMs-assisted augmented reality assembly method is established, and the elements and interrelationships of the system are analysed. Secondly, a matching process information model is constructed for LLMs environment. Then, the assisted guidance decision-making method and process based on LLMs are given. Finally, combined with a cable assembly expertise, a professional question answering system is developed to realize the smart assisted guidance of LLMs. Results show that the assembly pass rate increased by 15% and the effectiveness of the method is verified through several cases.

The assembly process of the variable stator vane (VSV) adjusting mechanism of aero-engine requires manual detection of the anti-loosening wire assembly correctness of the connecting rod, which is inefficient and errorprone. An intelligent fault-detection method based on multi-model cascade is proposed to replace the manual detection operation. The method is a model integration of multiple convolutional neural networks, which consists of three parts: detection module, classification module, and post-processing of comparison & fault detection. Firstly, the depthwise separable convolution with lightweight decoupling head mixing different sizes of convolutional kernels is proposed on the detection module to improve YOLOv5s, and the improved YOLOv5s achieves an average accuracy of 97.9% on the test set, which is improved by 3.4% and 1.5% compared to YOLOv5s and YOLOv8s, respectively. Secondly, the ConvNeXt classification head is improved by using 7×7 deep convolution instead of global average pooling on the classification module, and the performance is improved, reaching an accuracy of 97.5% and 95.4% on the connecting rod dataset and the thread dataset, respectively. Finally, the results of the two classification models are matched in the post-processing module to obtain the assembly detection result. The intelligent fault-detection method is verified by the image dataset collected from the field assembly workshop, and the results show that the average precision of the proposed method reaches 92.7%, which further verifies the reliability of the proposed method.

Carbon fiber reinforced polymer (CFRP) is widely used in aircraft panel structures. Due to molding process constraints, there are varying degrees of assembly gaps at component interfaces. In engineering practice, when the gap exceeds a certain threshold, gap-compensation measures are required. The panel assembly experiences complex deformation and stress concentration after gap compensation, affecting subsequent aircraft operational performance. Therefore, simulation analysis and experimental verification on the impact of bolt fastening sequences on CFRP panel assembly quality after gap filling are conducted. Analysis reveals that the panel offsets with a rotational deformation centered at the upper right endpoint towards the lower left. Normal deformations occur on both sides of the panel and around the bolt holes, with the maximum normal deformation value decreasing by approximately 63.4% when changing the bolt fastening sequence. The fastening sequence has a minor influence on the overall stress distribution of the panel, while significantly affects the stress distribution around the bolt holes and the distribution of panel assembly deformations, favoring symmetric joining over sequential joining. The stress distribution around the holes is closely related to the gap span and amount, increasing in magnitude and range as the gap amount rises.

In order to solve the problem that the riveting process of carbon fiber reinforced polymer (CFRP) and metal material laminated components is more complicated than that of a single material, a method for sensitivity analysis of the riveting process parameters of laminated composites based on the method of moment-independent PAWN is proposed. Firstly, taking CFRP and aluminum alloy (Al) heterogeneous laminated composite structure as the research object, the simulation and analysis model of riveting CFRP/Al laminated structure is established by using the finite element method, and the correctness of the simulation and analysis model is verified by experiments. Secondly, a proxy model of riveting process parameters based on the random forest method is established with the riveting interference amount as the target; furthermore, data extension is carried out through the proxy model, and a global sensitivity analysi s based on the moment-independent PAWN method is carried out with 17 riveting process parameters as input variables and the riveting interference amount as the output variable, and the sensitivity coefficients based on this method are given. Among the global sensitivity coefficients of riveting process parameters on riveting interference amount, the sensitivity coefficients of elasticity modulus and thickness of the heterogeneous laminated composites are 0.16921 and 0.11837, respectively, followed by the coefficients of 0.012 for the displacement load and 0.011 for the rivet diameter. The values of other process parameters are small, and the sensitivity coefficients of some parameters that have insignificant effects on the riveting interference amount are close to zero.

In order to explore the influence of assembly clearance on the vibration fatigue performance of singlelap structure, an elastoplastic finite element model of metal riveting joint of local structure in aircraft engine air intake is established by using ABAQUS software. The Johnson–Cook failure criterion is used to simulate the progressive failure behavior of rivets and connected parts, and the stress distribution and vibration fatigue life of riveting joint considering different assembly clearance amounts are obtained. The vibration fatigue test is carried out on single-lap riveting joint, and the experimental results are in good agreement with the simulation results in terms of fatigue life, which verifies the accuracy of the numerical model. Compared with the model without clearance, the vibration fatigue life of riveting joint with assembly clearance is significantly reduced by 4.7%–18.0%. The existence of clearance also leads to the transfer of extrusion load between adjacent rivets during riveting process, which is generally expressed as the stress surge at the contact interface between rivets and connected plates, leading to differences in residual stress fields of riveting joint with and without clearance, thereby affecting the vibration fatigue performance of riveting joint.