The digital twin workshop is of great significance for the design of the new workshop and the upgrading of the original workshop, but the characteristics of “human–machine–environment” dynamic and static integration bring challenges to the rapid modeling of the digital twin workshop. The existing methods have problems such as low efficiency of workshop 3D scene modeling, difficulty in realizing dynamic and static fusion modeling, and lack of quantitative evaluation of dynamic and static fusion. For the above problems, under the existing concept of digital twin workshop, combined with three-dimensional scanning technology, indoor positioning technology, sensor technology, etc., a rapid modeling method of digital twin workshop based on dynamic and static fusion of human–machine–environment is proposed, and the corresponding modeling framework is established. The rapid reconstruction of the static environment of the digital twin workshop based on three-dimensional laser scanning, the dynamic object perception of the workshop based on multi-sensor fusion, and the dynamic and static fusion driven by real-time data are studied. The evaluation index of dynamic and static fusion is designed to quantitatively evaluate the fusion effect of the constructed digital twin workshop. Finally, the robot manufacturing workshop of Zhengzhou University of Light Industry is tested to verify the feasibility and effectiveness of the proposed method.

Digital twin technology is a key technology for achieving intelligent manufacturing and industrial digital transformation, and has received widespread attention and research from both academia and industry. With the rapid development of data collection technology, computer performance, and related intelligent algorithms in the past decade, significant progress has been made in the research and application of digital twin technology. In response to the diversity of digital twin technology application fields and the complexity of object hierarchy, this article reviews the development process of the basic concepts and models of digital twin, and summarizes the key technologies and development achievements of digital twin technology in the manufacturing industry. On this basis, this article elaborates the current application status of digital twin technology in the manufacturing industry at three levels and the new paradigm of empowering intelligent manufacturing with digital twins. Finally, this article discusses the future development trends of digital twin technology in the manufacturing industry, providing reference for subsequent research.

The digital twin model of industrial robot can simulate the behavior and performance of industrial robot in the real world, but its simulation accuracy will be reduced due to the influence of service conditions such as scene update and equipment wear. In this paper, an update method of digital twin model of industrial robots based on deep reinforcement learning is proposed. In this method, the simulation tool Coppeliasim is used to establish the digital twin model of industrial
robots. At the same time, the key parameters of the digital twin model such as PID parameters and joint damping are optimized based on the depth deterministic strategy gradient (DDPG) algorithm, so as to realize the parameter update of the model and improve the model accuracy. Finally, the simulation experiment of ABB–IRB2400 industrial robot is carried out to verify the effectiveness of the proposed method.

Combination of digital twin and manufacturing field is increasingly close. Digital twin can help machine tools achieve autonomous monitoring, evaluation and reasoning in an intelligent manufacturing environment and significantly improve intelligence. In this paper, we develop a digital twin intelligent monitoring platform for the milling process of machine tools, which can realize real-time interaction between the physical manufacturing process and its virtual data model. In order to realize the bidirectional connection between different modules and between the physical machine and the virtual model, the data acquisition module and the digital twin monitoring platform are built based on the idea of hardware abstraction layer development and integrated with the open CNC system. The digital twin monitoring platform has the features of loose coupling, modularity and easy expansion, which integrates the data analysis twin model built by dynamics analysis model and artificial intelligence algorithm. It can realize the functions of milling chatter identification and suppression, surface topography prediction, etc. The machining experiments have been verified to demonstrate the good performance of the monitoring platform in the actual manufacturing environment.

Complex manufacturing processes face challenges posed by scenario complexity and multiple dynamic events. In order to improve the accuracy of complex product quality prediction, an event-driven product digital twin system framework is proposed by combining digital twin and event-driven, and an event-driven product manufacturing multidimensional twin model is established, which is utilized to simulate various scenarios in the actual manufacturing process and combined with the key event information to achieve a more accurate prediction of product quality. Then, the product quality prediction model based on hybrid neural network is constructed by combining convolutional neural network(CNN), bidirectional gated recurrent unit (BiGRU) and self-attention mechanism for the time-dependent relationship in the event sequence. Finally, the event-driven digital twin operation model for quality prediction of transmission assembly is illustrated by taking dual clutch transmission (DCT) assembly as an example; and the accuracy of the proposed quality prediction model is verified by comparing with the traditional single-model prediction method.

Intelligent process design is the core of process design in the digital twin environment, and part process knowledge modeling is a prerequisite for achieving intelligent process design based on digital twins. To address the issues of low structured and difficult to reuse machining process data for complex thin-walled parts in the aerospace field, a onstruction and quality evaluation method for a typical knowledge graph of machining process for complex thin-walled parts is proposed. Firstly, the composition and structure of machining process knowledge are analyzed. Secondly, the visualization of process knowledge was realized through ontology modeling, knowledge extraction, knowledge storage, and other related technologies, and the knowledge retrieval of machining process was realized based on Neo4j database. Finally, the analytic hierarchy process is used to evaluate the constructed knowledge map, and the machining process knowledge of frame and segment parts is taken as the verification object, the comprehensive accuracy of the sub-map is 92.28%. The experimental results show that the process knowledge modeling method based on knowledge map is feasible, which can help to realize the effective organization and reuse of process knowledge, and lay the foundation for digital twin intelligent process design.

In recent years, composite materials have been widely used in complex structural parts instead of metal materials. Aiming at the problem of unstable braiding angle caused by the mandrel when radius of the mandrel changes in the braiding process of special-shaped preforms, this paper proposes a control method for the take-up speed of the robot. Firstly, the special-shaped structure mandrel is discretized to obtain the radius of each segment, and then optimize the take-up speed of the robot, so that the adjustment of the braiding angle can be realized and the relationship between the advancing distance and the actual braiding length can be got. Finally, the braiding length is compensated by the second advance of the robot, and the next segment of the mandrel is braided based on the dynamic convergence length. The experimental results show that the control method can effectively reduce the braiding angle error of the special-shaped structure mandrel and keep it within ± 3°, which is of great significance for the strict control of the braiding angle in actual production and the braiding of mandrels with arbitrary curvature.

The effect of shot peening on the surface integrity and fatigue properties of aluminum-clad 7B04–T7451 aluminum alloy was studied, and the effect of the aluminum-clad layer on the shot peening effect was clarified. Vickers hardness tester, in-situ nanometer testing system, laser confocal microscope, X-ray diffractometer were used to analyze the changes of microhardness, surface profile, surface roughness and residual stress of the samples after shot peening. The fatigue life of shot peening specimens was tested by fatigue testing machine and the characteristics of fatigue fracture were analyzed by scanning electron microscope. The results show that the hardness of the surface layer of the shot peened aluminum-clad 7B04–T7451 sample increases gradually from the surface to the interior until the hardness of the base material, and the surface hardness is slightly higher than that of the original sample; The larger the size, the larger the surface roughness is; The residual compressive stress layer with a depth of about 150 μm is introduced by shot peening, but its residual compressive stress value and stress peak value are both small. The fatigue properties of the shot peening clad 7B04–T7451 aluminum alloy did not improve. Due to the significant increase in surface roughness and plastic deformation, the fatigue crack source was initiated faster on the surface of the aluminum-clad layer, and the fatigue life decreased.

Due to the advantages of high connection strength, little damage to composite materials, single-sided connection and smooth fracture without grinding, blind bolted rivets are widely used for fastening composite structures in the aerospace field. At present, blind bolted rivets are mainly imported from foreign countries, lacking of stable manufacturing methods domestically. This paper focused on the strength calculation method for a certain type of blind bolted rivets, firstly introduced its installation process and analyzed mechanical property. Then based on the Fourth failure criteria (Von mises yield criterion), expressions for the maximum stress at key positions of the screw (fracture groove and first thread) were established. By establishing a finite element model of the screw with two different UNJF threads, the stress concentration coefficients in expressions were calculated. Finally, the influence of various factors on the maximum stress at critical positions was analyzed. This research would be meaningful and valuable in improving the connection reliability and economic efficiency through this established strength calculation method.

In order to optimize the milling parameters in milling of TA15 titanium alloy using carbide end cutter, and to control the surface integrity and improve the fatigue life of components, the effects of machined surface integrity on the fatigue life of TA15 titanium alloy were investigated through milling experiments, surface integrity tests, and fatigue tests. The results show that the surface roughness R_a of TA15 titanium alloy is 0.148–0.245 μm after milling, the maximum residual compressive stress is detected on the surface, the depth of the residual compressive stress affected layer is approximately 30 μm, and the depth of the hardened layer is approximately 70 μm. The fatigue life increases with the decrease of surface roughness and the increase of residual compressive stress. The optimal combination of milling parameters: milling speed 40 m/min, feed per tooth 0.03 mm/z, and the maximum fatigue life is 4.044×10⁵ times. The tensile fracture of the machined TA15 titanium alloy is ductile fracture with deep dimples.

In this paper, the stability of milling in vertical machining center is predicted by numerical integration method. Firstly, the single-degree-of-freedom and two-degree-of-freedom milling dynamic models affected by chatter are represented by time-delay differential equation, and the delay period is discretized; Secondly, the time-delay differential equation is solved by numerical integral method, the milling stability is determined according to Floquet, and the stability lobe diagram is obtained. Finally, in order to verify the correctness of the numerical integration method, the milling force coefficient and modal parameters are obtained by parameter identification experiments, the machining parameters are selected from the obtained lobe diagram for experimental verification. The numerical simulation results show that the numerical integration method is superior to the 1st-SDM method and the 1st-FDM method in terms of computational efficiency and accuracy. The numerical integration method solution results are consistent with the experimental results, which can guide the selection of milling process parameters and ensure the stability of milling processing.