In the era of global digitization, how to improve efficiency and reduce costs in the practice of high-end complex equipment systems engineering is critical. This study proposed that the essence of digital engineering practice was an extension of the application of systems engineering in the whole life cycle of the system and the digital engineering practice could improve the efficiency, research, development, procurement, operation, and maintenance of the complex system, and reduce the costs. Based on the practical application of the U.S. Army, this paper organized a digital engineering practice methodology using one core, driven by three types of practices and guided by four steps. The digital engineering practice method outlined the collection of practices at each stage of the equipment life cycle, dominating the system development from the model-based system demonstration, requirements analysis, prototype design and even equipment operation and maintenance. The all-round and whole-process management of the complex system was realized.

Digital mission engineering (DME) is an advanced approach in the digital era. Based on the long-term research accumulations in Unmanned intelligent system-of-systems (SoS) design optimization, weaponry SoS modeling and simulation and digital decision engineering for strategic management, this paper has studied and explored the origin and progress of foreign military DME research, fusingwith recent domestic research practices in SoS engineering and SoS simulations. Finally, a technical conception of DME research methodsfor kill chain SoS from two aspects was proposed: overall requirement architecture and simulation deduction technology.

Focus on the digital research and development needs of the solid attitude orbit control engine, the application of digital prototype technology in the solid attitude orbit control engine needs to be improved. In this study, the geometric modelling method of a solid attitude orbit control engine utilizing knowledge rules was constructed. The high-precision simulation model of the internal flow field of the solid attitude orbit control engine based on the SST k-ω turbulence model was established, and the uncertainty analysis method of the performance of the solid attitude orbit control engine was proposed. The research on the integration and validation method of the multilevel digital model of solid attitude orbit control engine was conducted. The results show that the pressure and thrust steady state performance deviation of the constructed digital prototype of a solid attitude-orbit control engine is respectively 4% and 3.6%, and the pressure and thrust performance deviation of the digital prototype during the continuous thrust adjustment process is respectively 4.98% and 6% indicating a high-performance prediction accuracy. Besides, the research results can be applied in the overall design of the engine program and the program performance rapid assessment, which can promote the development of China's solid attitude-orbit control engine and its performance. The research results can also be applied to the rapid evaluation of the performance of the program, which can promote the change of the R&D mode of the domestic solid attitude-orbit control engine, powering the innovative development of the solid engine design technology, driving the overall strength of the solid engine industry, and building the technical foundation for the development of intelligent solid engine technology in the future.

The inconsistencies and inadequacies of traceability in conventional systems engineering facing complexity have prompted the search for alternative methodologies. Model-Based Systems Engineering (MBSE) has been proposed as a promising solution, gaining widespread recognition in academia and industry. Leveraging mathematical models and graphical representations, MBSE supports a systematic unified description and analysis, shifting the paradigm from document-driven to model-driven. This approach enhances traceability, testability, and verifiability. This paper has comprehensively analysed MBSE practices from four perspectives: time, logic, knowledge, and organization, thus empowering engineers to better understand and apply MBSE in real-world engineering projects.

Given the problem of increasing the number of radar emitters and pulse density in a current complex electromagnetic environment, this study proposed an intelligent recognition algorithm using a non-parametric model to achieve high-precision radar emitters recognition. The proposed method was composed of three vital steps: static clustering of radar emitters, selection of representative points for radar emitter classes, and oriented-data stream incremental recognition. Compared to conventional clustering analysis methods, the proposed method utilized the clustering idea of Dirichlet Process Mixture Models (DPMM), which adaptively realized radar emitters’ class sorting. To reduce the data storage burden while maximizing the preservation of class distribution, a gradient-based representative point selection strategy for emitter classes was established. Besides, to satisfy the requirement of incremental processing of the data stream, a label propagation-based classification inference method was applied to achieve incremental radar emitters recognition. Finally, the proposed algorithm was validated using simulated radar emitters’ data in two scenarios. The evaluation results show that the proposed method is more robust and accurate than the other seven advanced emitters recognition clustering algorithms.

To solve the problems of high cost, high risk, and difficulty in flight performance evaluation of hypersonic vehicle flight tests, a multi-physics field virtual flight method was proposed in this study. By establishing a high-fidelity multi-physics field model to simulate the flight test, fast and accurate flight performance analysis was acquired. Considering the coupling effects between aerodynamics, structure, control, and trajectory disciplines, a loosely coupled method was adopted to build a dynamic model of an elastic hypersonic vehicle and conduct virtual flight simulations. The piston theory was used to calculate unsteady aerodynamic loads, and the modal superposition method was employed to calculate structural dynamic responses. A multi-physics field virtual flight simulation process was constructed by integrating time-varying aerodynamic loads and structural vibration deformations into a six-degree-of-freedom flight dynamics model. A virtual flight simulation study was conducted on a three-dimensional hypersonic vehicle, analyzing the influence of structural dynamic deformation and unsteady aerodynamic dynamic response on its flight trajectory and control system. The results show that the virtual flight model can quickly and accurately obtain the dynamic, controlled, and load characteristics of the vehicle during flight, providing an efficient and low-cost solution for verifying the flight performance in hypersonic vehicle design.

In the scenario of complex naval escort missions, the current tactical decision support functions of anti-aircraft missile defense systems face issues such as high dependency on enemy models, poor accuracy in interception decisions, inability to effectively utilize historical battlefield data, and simplistic research objects. To resolve the above problems, a deep learning-based anti-missile interception intelligent decision-making model was proposed in this study. Firstly, a battlefield simulation platform was established to model the combat units accordingly. Then, an anti-missile interception intelligent decision-making model was designed using Long Short Term Memory neural networks. After that, a pre-battle model was trained using simulated data acquired from a constant proportional guidance particle model. Finally, the pre-battle model was transferred to the battlefield model and fine-tuned with real-time data enhanced with actual battlefield data through small-sample online training. Experiment results show that the proposed anti-missile interception intelligent decision-making model can effectively reduce dependency on enemy models and improve the accuracy of air defense missile decision-making.

To resolve the multi-agent multi-target assignment problem in complex environments, a multi-agent target assignment method using reinforcement learning has been proposed in this study. Initially, a battlefield environment model for the execution of tasks by the aircraft was established, extracting and quantifying information on the aircraft's combat capabilities and environmental conditions. Then, a reinforcement learning algorithm was employed to assign multiple groups of combat tasks. Finally, considering aircraft combat losses, dynamic task reassignment was conducted to enhance task completion in response to sudden battlefield situations. Simulation results show that significant effectiveness of the proposed algorithm is achieved.

Facing the rapid development of electronic warfare, in order to solve the problem of how to quickly obtain the battlefield situation,the effectiveness evaluation of airborne fire-control radar systems in complex electromagnetic environments was proposed in this paper. Based on the analysis of mission requirements and targets for airborne fire-control radars in actual combat scenarios, this study has decomposed their various capabilities and indicators, constructed an evaluation index system framework for airborne fire-control radar effectivenessin complex electromagnetic environments, and focused on analyzing effective evaluation indicators for clutter suppression and anti-jamming capabilities, along with detailed designs for their quantification methods. In experimental simulations, the study utilizedthe analytic hierarchy process (AHP) to acquire the weights of evaluation indicators and employed comprehensive evaluation methods to calculate the valuesof various capabilities, which verified the practicality and effectiveness of the model. This study provides support for relevant decision-making regarding airborne fire-control radars in operational environments.

Given the problems of the difficulty in multi-platform kill chain simulation, unclear kill assessment indexes, and the lack of close integration with the actual scenarios, the practical implementation of the kill chain in combat situation simulations was studied using the U.S. Army's NIFC-CA kill chain. Firstly, the characteristic information of target nodes was designed, and customized reconnaissance node effectiveness evaluation indexes were designed for E-2D AWACS, F-35 fighter jets and Aegis ship platforms in the NIFC-CA kill chain. It associated the NIFC-CA kill chain with the reconnaissance nodes and also connected the reconnaissance node effectiveness evaluation indexes with the target node information, and evaluated the level of the reconnaissance node's effectiveness in the detection of different targets. Secondly, the influence nodes were introduced to characterize the interference of the reconnaissance node in the combat scenario and the anti-interference effectiveness level of the reconnaissance node itself. Then, the target node, influence node, and reconnaissance node were correlated using a state transfer matrix, and the reconnaissance node, command node, and strike node were correlated using dependency analysis of the Functional Dependency Network Analysis (FDNA), giving the quantitative value of the overall killing capability of different killing chains. Finally, the simulation design of the kill chain model was carried out in the form of an adjacency table.

The scale of the equipment is widely applied in various tasks. The existing AWACS is charging the system human-machine interaction; the traffic sign is compressed; the operation dialogue is overlapping; the interface design style is different; the menu function is not reasonable while the function menu is inadequate, making it not ideal for the function of the preset police machine in the information warfare. This paper proposed one kind of warning machine allegation system machine interaction optimization design, allowing existing problems to be optimized and improved. The GRA-TOPSIS (Gray Relational Analysis-Technique for Order Preference by Similarity to Ideal Solution) evaluation method was used to analyze and calculate human-machine interaction efficiency and compare various performance indicators of the original system. The results show a significant improvement in the human-machine interaction efficiency of the control system after optimization, providing valuable insights and references for the development and upgrade of the AWACS control system.

To improve the coordination and accuracy of large aircraft component manufacturing and assembly, this study proposed a method for controlling the quality of complex assemblies. Based on the research of the key enabling technologies of digital measurement, such as integrated measurement process control using multi-field integration, informatization of the whole process of tooling and product assembly measurement, virtual assembly driven by measured data, etc., an integrated platform for online measurement and control management was developed, supporting dynamic control of geometric position, assembly relationship and its quality in the whole assembly cycle. It helped users adjust and improve assembly schemes online and conducted engineering applications in the assembly process of multiple products, where significant improvement in assembly quality and efficiency was achieved.

Verifying missile motion simulation models using flight test data is an effective way to create simulation models. This paper systematically introduced a verification method which combined qualitative analysis using dynamic performance consistency checking and quantitative analysis using static performance consistency checking. Based on the above, verification indicators and reliability calculation methods for missile motion simulation models were confirmed. The results show that although flight test samples are limited, as long as the methods are appropriate, reliable validation results have been obtained.

The discrete manufacturing process of aerospace products, characterized by intricate assembly procedures, complex manufacturing techniques, and frequent technical status changes, brings a series of challenges for data management, especially difficulties in integrating multi-source heterogeneous data and inaccuracies in product status detection. To overcome these challenges, this paper has proposed a normalized automated product quality analysis process using multi-source, multi-modal data. Utilizing bill of material (BOM) as the data organization framework, leveraging data platform components enables real-time monitoring and early warning of product quality status across multiple batches via successful envelope analysis of single indicators and joint analysis of multiple indicators, thus allowing product quality and consistency to be enhanced precisely. This study provides theoretical support for the prediction and controlling of product qualities and offers guidance for subsequent quality analysis and reinspection.

Digital twin technology presents physical entities digitally and provides effective conditions for breaking down data barriers between testing sites and management via the interaction and integration of the physical world and the virtual world. Using the five-dimensional digital twin theory, the Shanghai Institute of Space Power-sources has realised functions such as testing commissioning and scheduling, testing data collection of product, scene modeling, real-time monitoring of testing processes, and data analysis. This study has improved the efficiency of testing management, explored the feasibility of virtual testing, and laid a solid foundation for reducing model testing costs and upgrading refined testing management.