A certain UUV exhibits a high degree of coupling between its states and control variables, making it difficult for classical reduced-order methods to be applied. Meanwhile, considering that the control variables must satisfy the physical constraints of the UUV, a hierarchical model predictive control method for the UUV is proposed. Firstly, through state transformation, the UUV controlled system is decomposed step by step into an intermediate system and a terminal system. Between them, the intermediate system is directly driven by the control variables, and a quasi-infinite horizon model predictive control method is employed to design the control law, ensuring that the control variables meet the physical constraints of the UUV. The terminal system is driven by the states of the intermediate system and directly reveals the dynamic characteristics of the UUV's states. By designing the form of the state transformation, the stability of the terminal system is guaranteed. Ultimately, by integrating the intermediate and terminal systems, stable tracking of the desired states of the UUV is achieved. To further enhance the robustness of the proposed method, an LSTM neural network is employed to observe and estimate disturbances. The numerical simulation results demonstrate that the proposed hierarchical model predictive control method for UUV can effectively control the UUV to track the desired states. Meanwhile, the designed LSTM-based disturbance observer is capable of accurately predicting disturbances. Compared with the hierarchical model predictive control method without disturbance observation, the LSTM-based hierarchical model predictive control approach significantly enhances the smoothness of the control process.

The performance of staged combustion cycle engines is affected by various uncertain factors, where even minor disturbances may cause the engine's operating conditions to deviate from the designed parameters, compromising system stability. To investigate the influence of uncertain factors (e.g., manufacturing tolerances, turbine and pump performance) on the performance of staged combustion cycle engines, 120-ton LOX/kerosene engine is selected as the research object. A combined approach of single-factor sensitivity screening and Monte Carlo-based multi-factor coupled sensitivity analysis is employed to assess the impact of these disturbances. The results indicate that volume-related factors (e.g., gas generator and thrust chamber) have negligible effects on engine system performance. In contrast, turbine efficiency and nozzle diameter exhibit moderate influence. Under multi-factor coupling conditions, the impact of nozzle diameter on system performance parameters is relatively minor, whereas the turbine stator flow area significantly affects gas generator chamber pressure. Additionally, turbine efficiency and the pump heads of both fuel and oxidizer pumps also influence system performance parameters, albeit with distinct patterns. These findings demonstrate that the combined single-factor/multi-factor sensitivity analysis method provides valuable engineering guidance. Special attention should be paid during design and manufacturing to critical factors such as turbine stator flow area and turbopump efficiency.

A novel smart bolt combined with fiber EFPI/FBG micro-structures is presented to monitor the state of tightening force of a bolted aerospace structure under high temperature circumstance. In this smart fiber bolt, the fiber EFPI micro-structure is applied to obtain the length variation in the direct of bolt axis under the effect of tightening force. And the fiber Bragg grating micro-structure is applied to obtain the bolt temperature. With the parameters obtained by the fiber EFPI/FBG micro-structures, the tightening force can be obtained through the approach presented. Test samples which are 50mm long M8 bolts combined with fiber EFPI/FBG micro-structures are made and tested by a universal testing machine with a heating chamber. The test results show that the smart fiber bolt presented can be used to obtain maximum 10kN tightening force under the temperature of 500 degree centigrade with maximum relative error 3.1%. This smart fiber bolt can be used to monitor the state of structure connection under high temperature circumstance.

To meet the unmanned operational requirements before rocket launch and improve the working efficiency of gas supply system, the automatic gas charging/discharging technology for gas cylinders has become a key focus in the renovation of rocket gas supply system at aerospace launch sites. For the gas charging/discharging processes of launch vehicle cylinders, a novel automated gas control system​based on an orifice plate design​has been​developed. This study employed AMESim software to model and investigate the automatic charging/discharging processes of cylinders. The results demonstrate that by strategically selecting different orifice plate and configuring manual charging valves with gas circuits, the proposed scheme satisfies the technical requirements for charging/discharging of gas cylinders.​Furthermore, it achieves both automatic and manual gas charging functions for different gas during testing and pre-launch phases.

Liquid launch vehicles, as the primary means for human space access currently and in the foreseeable future, have their testing and launch modes closely tied to the technical status of the launch vehicle, the construction of launch site conditions, and the achievement of high-efficiency launch capability. Through a comprehensive analysis of the main characteristics and underlying reasons at each developmental stage of global liquid launch vehicle test-and-launch modes, the evolution of these modes is divided into three phases: reliability-and-safety-driven, test-efficiency-driven, and flexibility-and-cost-driven. In each phase, the test-and-launch mode of liquid rockets is refined and improved based on the previous stage, with current developments emphasizing safety, reliability, efficiency, and low cost simultaneously. The key factors influencing the selection of test-and-launch modes for liquid launch vehicles are analyzed, including technical feasibility and heritage, launch-site adaptability, test-and-launch efficiency and flexibility, and economic considerations. Major trends in the future development of test-and-launch modes are proposed, providing useful references for the construction of future space launch sites.

To realize multiple starts of LOX/CH₄ rocket engines, a fuel rich torch igniter fed with gaseous oxygen and gaseous methane is designed, in which methane is used as film cooling refrigerant. Through CFD simulation and CFD simulation, comparative study on the ignition reliability and thermal protective effect is conducted with three types of igniters. It is found that ignition mixture ratio affects the ignition success rate and thermal protective effect directly, and relative distances between methane injectors and oxygen injectors is an important parameter which would affect the working characteristics of the ignitors. The selected ignitor with double rows of CH₄ injectors can ignite reliably while the mixture ratio is between 0.6 and 1.3. During the second round test in which the flow rate is 25~50 g/s while the mixture rate is 0.6~1.0, the ignitor ignited successfully 11 times in a row and no structure burned damage is found. The reliability of the film-cooling igniter and multiple starts is verified. With simulation results, it is showed that the relative distance between the injectors has an important influence on the thermal protective effect of film cooling. Using multi-rows of methane injectors is an optimized scheme consideration to ignition reliability and thermal protective effect.

In response to the strategic needs of becoming a leading space power, and against the backdrop of hot topics such as "Airline-Flight-Based transportation of reusable rocket" and the successful in-situ recovery at launch tower during the fifth test flight of SpaceX's Super Heavy-Starship, this study benchmarks against world-class space launch sites, and combines the actual development of Chinese space launch site construction. It refines the basic characteristics that an "Airline-Flight-Mode" launch site should possess, including multi-station parallel operation, fast turnaround launch, high safety and reliability, and low-cost launch. The current gaps in China's space launch site in adapting to "Airline-Flight-Mode" launch are analyzed, and a demand system for building the "Airline-Flight-Mode" launch capability of China's launch site is proposed. At the organizational management level, China's launch site should develop from a single-mode construction and operation to a diversified mode involving national, civilian and commercial entities, while adhering to the concept of a large industrial ecosystem, integrated deployment and collaborative design of satellites, rockets and launch sites should be promoted to achieve "launch upon delivery and reuse upon recovery". At the key technology level, emphasis should be placed on the development of pre-launch rapid refueling, post-launch rapid recovery, special robots, in-situ recovery at launch towers, modular construction, etc., to achieve efficient turnaround, high-frequency launches, and multi-station implementation. The research can support the high-quality development of space launch site planning and construction.

Adapter separation is a critical phase during the launch tube ejection process of cold-launch missiles. Its operational state influences relative positional relationships between adapter groups, as well as between adapters and the missile itself, directly impacting launch safety. Despite its importance, systematic studies on modeling the separation process and quantifying its states remain underdeveloped. This study aims to investigate the separation time and velocity of the adapter by adopting a method that combines numerical calculation, dynamic simulation, and experimental data comparison to analyze the separation process and force conditions. By establishing theoretical models for the separation process, the Runge-Kutta numerical algorithm is applied to derive solutions for separation time and lateral velocity. Dynamic simulations are further conducted to replicate the separation behavior, while experimental data analysis interpreting real-world conditions. Results demonstrate strong agreement between numerically calculated separation times/lateral velocities, simulation outputs, and experimental measurements, validating the feasibility of theoretical framework. Building on this foundation, the effects of critical parameters are explored—including angle of attack, lateral wind speed, and initial installation force of the pin assembly spring—on separation time and lateral velocity. Key influencing factors are identified, providing actionable insights for optimizing adapter design in cold-launch systems.

High-accuracy core loss models are an essential foundation for loss evaluation and efficiency optimization in servo power converters. A PWM excitation core loss model is proposed based on the DC power method to address the issue of large errors in core loss evaluation under PWM excitation using the Modified Steinmetz Equation (MSE) in traditional methods. This method directly establishes a loss model based on the measurement results of magnetic core loss under PWM excitation obtained by the DC power method, which can avoid the errors introduced by the MSE model in the waveform equivalent transformation stage and has better model accuracy. The experimental results show that the maximum relative error of the proposed model for loss evaluation under different excitation waveforms, frequencies, duty cycles, peak AC magnetic flux densities, and temperatures is -5.029%. The average absolute value of the relative error is only 1.72%, much lower than the model error of MSE (maximum relative error: 87.663%, average absolute value of relative error: 38.04%). It verifies the accuracy of the PWM excitation core loss model proposed under a wide range of operating conditions and can provide a high-precision calculation model for loss evaluation of power converter magnetic components under PWM excitation.

To study the breakup atomization and development process of kerosene propellant in crossflow under subsonic conditions, and to realize the accurate numerical simulation of fuel atomization, a kerosene jet in crossflow with average flux-to-momentum ratio q ranging from 7.9 to 46, and average We_g ranging from 5.6 to 120 is selected to enter the mainstream of the air under the velocity conditions of 11 subsonic uniformly incoming air main streams and the phenomenon of primary atomization under the action of crossflow is investigated by the Volume of FluidMethod and Adaptive Mesh Refinement method. The volume of the FluidMethod (VFM) method and Adaptive Mesh Refinement (AMR) method are used to investigate the primary atomization phenomenon under the jet's action in crossflow and verify the accuracy of the numerical simulation method by comparing it with the experimental results. The results show that jet fragmentation of kerosene propellant is mainly caused by surface tension, especially the Rayleigh-Taylor (R-T) instability plays a dominant role. In contrast, the surface fragmentation triggered by the Kelvin-Helmholtz (K-H) stability precedes the liquid column fragmentation caused by the R-T instability. The atomization process is negatively correlated with the mean gas Weber number, positively correlated with and strongly influenced by the mean flux-to-momentum ratio q, and finally an empirical equation for the jet penetration depth profile suitable for atomization in crossflow jet is obtained.