Variable thrust liquid rocket engine pintle injector optimization method and apparatus

Abstract

The application provides a variable-thrust liquid rocket engine needle injection optimization method and device, and relates to the technical field of structural design. The method comprises the following steps: obtaining each design structure parameter of a flow-adjustable needle injection applied to a variable-thrust liquid rocket engine; establishing a three-dimensional transient numerical simulation model based on each design structure parameter; performing simulation calculation on the dynamic process of continuous adjustment of the needle based on the three-dimensional transient numerical simulation model, obtaining a simulation data set, and training a response surface model using the simulation data set; after training, an intelligent agent model of the needle injection is obtained; based on the simulation data set, the intelligent agent model of the needle injection and a specified reinforcement learning optimization algorithm are combined to optimize the agent model, an intelligent agent model with high precision and fast response is obtained, the optimization of the design variables of the intelligent agent model is completed, and the design efficiency of the needle injection is improved.

Description

Technical Field

[0001] This application relates to the field of structural design technology, and in particular to an optimization method and apparatus for a needle-plug injector of a variable thrust liquid rocket engine. Background Technology

[0002] Wide-range thrust adjustment technology for variable-thrust liquid rocket engines is a key technology in the aerospace field today, and the needle-plug injector is an important actuator for thrust adjustment. Inside the needle-plug injector is a movable needle plug connected to a movable sleeve. By changing the position of the sleeve, the injection area is changed, thereby achieving thrust adjustment. Its operating characteristics directly determine the precision of the adjustment.

[0003] In existing technologies, the design methods for needle-plug injectors in variable-thrust liquid rocket engines are mainly based on traditional trial-and-error methods or experience-based optimization methods. These methods require designers to optimize the injection atomization and flow rate performance of the needle-plug injector through multiple trials and adjustments. The design process relies on empirical formulas and repeated experiments, which limits the iteration speed of needle-plug injector development and results in low design efficiency for high-performance needle-plug injectors. Summary of the Invention

[0004] The purpose of this invention is to provide an optimization method and apparatus for needle-plug injectors in variable thrust liquid rocket engines, so as to solve the technical problem of low design efficiency of high-performance needle-plug injectors.

[0005] In a first aspect, this application provides an optimization method for a needle-plug injector of a variable-thrust liquid rocket engine, the method comprising:

[0006] Obtain the design structural parameters of the flow-adjustable needle-plug injector for use in variable-thrust liquid rocket engines;

[0007] A three-dimensional transient numerical simulation model is established based on the aforementioned design structural parameters; wherein, the three-dimensional transient numerical simulation model is used to characterize the dynamic characteristics of the needle-plug injector during the needle-plug movement process, the dynamic characteristics including flow response characteristics and dynamic spray flow field characteristics;

[0008] The simulation calculation of the continuous adjustment dynamic process of the needle plunger is performed based on the three-dimensional transient numerical simulation model to obtain a simulation dataset. The response surface model is then trained using the simulation dataset to obtain an intelligent agent model for the needle plunger injector. The response surface model is used to map the design variables, which contain the design structural parameters and the operating condition data of the needle plunger injector, to the performance parameters of the needle plunger injector, so as to complete the order reduction process from high-dimensional input to low-dimensional output.

[0009] Based on the simulation dataset, the intelligent agent model of the needle-plug injector is optimized by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm, so as to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector.

[0010] In one possible implementation, establishing a three-dimensional transient numerical simulation model based on the various design structural parameters includes:

[0011] A three-dimensional geometric model is established based on the design structural parameters, and fluid domain preprocessing is performed based on the three-dimensional geometric model to obtain the preprocessing result;

[0012] Based on the preprocessing results, a Cartesian grid is used to divide the mesh, resulting in a three-dimensional geometric model after mesh division.

[0013] Based on the three-dimensional geometric model after meshing, and combined with the governing equations, turbulence model, VOF-to-DPM atomization model, and dynamic mesh model, a three-dimensional numerical simulation model is established and numerical simulation is performed on the three-dimensional numerical simulation model. The dynamic mesh model represents the coupling effect between the atomization flow field and the moving parts of the needle nozzle. After numerical simulation, the dynamic characteristics of the needle nozzle are obtained, including flow response characteristics and dynamic spray flow field characteristics.

[0014] In one possible implementation, the design variables include the geometric variables and adjustment condition variables of the needle-plug injector; the flow response characteristics of the needle-plug injector include at least one of adjustment accuracy, response time, rise time, and overshoot; and the atomization characteristics of the needle-plug injector include at least one of Solta average particle size and spray cone angle.

[0015] In one possible implementation, the step of training the response surface model using the simulation dataset to obtain an intelligent agent model for the needle-plug injector includes:

[0016] A response surface model was selected and established by comparing the Kriging model, the radial basis function model, and the neural network. The response surface model takes the geometric variables and the regulation condition variables as inputs, and the output of the response surface model includes key parameters of the flow response characteristic index and the atomization characteristic index.

[0017] The response surface model is trained using the simulation dataset so that it can learn and capture nonlinear relationships, resulting in an intelligent proxy model for the needle-plug injector.

[0018] In one possible implementation, the intelligent agent model of the needle-plug injector is optimized based on the simulation dataset by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector, including:

[0019] Based on the simulation dataset, the intelligent agent model of the needle-plug injector is optimized by combining the intelligent agent model of the needle-plug injector with a deep reinforcement learning optimization algorithm. The deep neural network corresponding to the deep reinforcement learning optimization algorithm and the intelligent agent model of the needle-plug injector are used as the environment to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector. The deep reinforcement learning optimization algorithm includes the proximal policy optimization (PPO) algorithm.

[0020] In one possible implementation, obtaining the design structural parameters for the flow-adjustable type applied to a variable-thrust liquid rocket engine includes:

[0021] Acquire flow regulation condition data, oxygen-fuel ratio parameters, and combustion chamber pressure parameters. Perform thermodynamic calculations based on the flow regulation condition data and rocket engine principles to obtain thermodynamic calculation results. Determine flow data including characteristic velocity, ground theoretical specific impulse, combustion chamber diameter, total thrust chamber mass flow rate corresponding to each regulation condition point, and mass flow rates of oxygen and combustion paths based on the oxygen-fuel ratio parameters, combustion chamber pressure parameters, and thermodynamic calculation results.

[0022] The injection ring area of ​​the needle-plug injector and the specified flow coefficients of the oxygen and combustion paths are obtained. The pressure drop of the oxygen and combustion paths is calculated based on the specified flow coefficients, the liquid density flowing into the needle-plug injector and the injection ring area of ​​the needle-plug injector, so that the operating conditions and pressure drop form an upward convex curve relationship.

[0023] The length-to-diameter ratio of the needle length and diameter of the needle injector, and the diameter ratio of the combustion chamber diameter and the needle diameter of the needle injector are obtained. Based on the length-to-diameter ratio and the diameter ratio, the structural parameters of the needle injector, including the needle diameter, needle length and skip distance, are determined. Based on the structural parameters, the structural design data of the needle injector is obtained.

[0024] Based on the simulation results of the flow characteristic curve corresponding to the flow data, the specified flow coefficient and the injection annular gap area are corrected to obtain the correction results. Based on the correction results and the structural design data, the design structural parameters of the flow-adjustable needle-plug injector applied to the variable thrust liquid rocket engine are obtained.

[0025] In one possible implementation, calculating the pressure drop in the oxygen and combustion circuits based on the specified flow coefficient, the liquid density flowing into the needle injector, and the injection annular area of ​​the needle injector includes:

[0026] Based on the specified flow coefficient, the liquid density flowing into the needle injector, and the injection annular area of ​​the needle injector, the pressure drop in the oxygen and combustion circuits is calculated using the following formula:

[0027] ;

[0028] in, For the pressure drop in the oxygen and combustion circuits, A specified flow coefficient between 0 and 1. The density of the liquid flowing into the needle syringe. A The area of ​​the injection circumferential joint is given. For mass flow rate.

[0029] Secondly, this application provides an optimized device for a needle-plug injector of a variable-thrust liquid rocket engine, comprising:

[0030] The acquisition module is used to acquire various design structural parameters of the flow-adjustable needle-plug injector applied to variable-thrust liquid rocket engines;

[0031] A module is established to build a three-dimensional transient numerical simulation model based on the design structural parameters; wherein, the three-dimensional transient numerical simulation model is used to characterize the dynamic characteristics of the needle injector during the needle movement process, and the dynamic characteristics include flow response characteristics and dynamic spray flow field characteristics;

[0032] The training module is used to perform simulation calculations of the continuous adjustment dynamic process of the needle plunger based on the three-dimensional transient numerical simulation model, obtain a simulation dataset, and use the simulation dataset to train the response surface model, resulting in an intelligent proxy model for the needle plunger injector. The response surface model is used to map the design variables, which contain the design structural parameters and the operating condition data of the needle plunger injector, to the performance parameters of the needle plunger injector, thereby completing the order reduction process from high-dimensional input to low-dimensional output.

[0033] The optimization module is used to optimize the intelligent agent model of the needle-plug injector based on the simulation dataset by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm, so as to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector.

[0034] Thirdly, this application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method described in the first aspect above.

[0035] Fourthly, this application also provides a computer-readable storage medium storing computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method described in the first aspect above.

[0036] This application brings the following beneficial effects:

[0037] This application provides a method and apparatus for optimizing a needle-plug injector for a variable-thrust liquid rocket engine. The method acquires various design structural parameters of a flow-adjustable needle-plug injector applied to a variable-thrust liquid rocket engine. A three-dimensional transient numerical simulation model is established based on these parameters. This model characterizes the dynamic characteristics of the needle-plug injector during needle-plug movement, including flow response characteristics and dynamic spray flow field characteristics. Simulation calculations of the continuous adjustment dynamic process of the needle-plug are performed based on the three-dimensional transient numerical simulation model to obtain a simulation dataset. The simulation dataset is then used to train a response surface model, resulting in an intelligent surrogate model for the needle-plug injector. The response surface model maps design variables containing the design structural parameters and operating condition data to the performance parameters of the needle-plug injector, thus achieving a reduction in order from high-dimensional input to low-dimensional output. The intelligent surrogate model is optimized using a combination of the intelligent surrogate model and a specified reinforcement learning optimization algorithm to obtain the optimized design variables of the intelligent surrogate model. In this scheme, the establishment of a needle-plug injector structural design method and a three-dimensional transient numerical simulation model prepares a dataset for the subsequent establishment of a surrogate model. Based on the established three-dimensional transient numerical simulation model, simulation calculations of the continuous adjustment dynamic process of the needle plug are carried out, a dataset is established, and an intelligent surrogate model is built. Then, based on the reinforcement learning optimization design method, the needle-plug injector surrogate model is optimized, and the optimization results of the design variables of the intelligent surrogate model are obtained. This realizes the needle-plug injector optimization design method based on the intelligent surrogate model and optimization algorithm, improves the design efficiency of high-performance needle-plug injectors, shortens the development cycle of needle-plug engines, and solves the technical problem of low design efficiency of high-performance needle-plug injectors.

[0038] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of a circumferential slit-type needle-bolt injector.

[0041] Figure 2 A flowchart illustrating the optimization method for the needle-plug injector of a variable thrust liquid rocket engine provided in this application embodiment;

[0042] Figure 3 Another flowchart illustrating the method for optimizing the needle-plug injector of a variable-thrust liquid rocket engine provided in this application embodiment;

[0043] Figure 4 The needle-plug injector structural design flowchart is provided in the variable thrust liquid rocket engine needle-plug injector optimization method provided in the embodiments of this application.

[0044] Figure 5 The numerical simulation calculation flowchart is provided in the variable thrust liquid rocket engine needle-plug injector optimization method provided in the embodiments of this application.

[0045] Figure 6 The optimization design algorithm schematic diagram is shown in the variable thrust liquid rocket engine needle-plug injector optimization method provided in the embodiments of this application.

[0046] Figure 7 A schematic diagram of a variable thrust liquid rocket engine needle-plug injector optimization device provided in this application embodiment;

[0047] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] The terms "comprising" and "having," and any variations thereof, used in the embodiments of this application, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0050] Currently, wide-range thrust adjustment technology for variable-thrust liquid rocket engines is a key technology in the aerospace field. The needle-bolt injector is a crucial actuator for thrust adjustment. Inside the needle-bolt injector is a movable needle bolt connected to a movable sleeve. By changing the position of the sleeve, the injection area is altered, thereby achieving thrust adjustment. Its operating characteristics directly determine the precision of the adjustment. A schematic diagram of the annular-slit needle-bolt injector structure is shown below. Figure 1 As shown, numerical simulation is the primary method for evaluating the dynamic characteristics of flow atomization in needle-plug injectors. For needle-plug injectors in liquid rocket engines, current numerical simulation models are mainly one-dimensional and three-dimensional, using ordinary variable cross-section throttling orifices and models designed for fixed cross-section needle-plug injectors. These models significantly simplify the dynamic adjustment process, reducing it to a steady-state or quasi-steady-state injection atomization process, resulting in insufficient model accuracy. Furthermore, in existing technologies, the design methods for variable-thrust liquid rocket engine needle-plug injectors are mainly based on traditional trial-and-error methods or experience-based optimization methods. This approach requires designers to optimize the injection atomization performance and flow regulation performance of the needle-plug injector through multiple experiments and adjustments. The design process relies on empirical formulas and repeated experiments, limiting the iteration speed of needle-plug injector development and leading to high trial-and-error costs.

[0051] The dynamic adjustment process of needle-plug injectors is quite complex, involving changes in needle position and spray area. One-dimensional simulation models neglect the complex dynamic response characteristics during the dynamic adjustment process, resulting in insufficient accuracy in evaluating their dynamic characteristics and inadequate ability to characterize nonlinear phenomena. Even with a significantly simplified three-dimensional numerical simulation model, which reduces the dynamic continuous adjustment process of the needle-plug injector to a steady-state process with a fixed cross-section, problems remain, including insufficient computational accuracy, unclear coupling between needle movement and atomized flow field changes, and unclear pressure-flow coupling.

[0052] Therefore, traditional design methods for high-performance needle-operated injectors are typically inefficient and struggle to fully account for the complex nonlinear flow atomization characteristics within the injector. Furthermore, design methods relying on empirical formulas may lack universality when facing different needle-operated injection conditions.

[0053] Based on this, the present application provides a method and apparatus for optimizing the needle-plug injector of a variable thrust liquid rocket engine. This method can solve the technical problem of low design efficiency of high-performance needle-plug injectors.

[0054] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0055] Figure 2 This is a flowchart illustrating an optimization method for a needle-plug injector in a variable-thrust liquid rocket engine, provided as an embodiment of this application. Figure 2 As shown, the method includes:

[0056] Step S110: Obtain the design structural parameters of the flow-adjustable needle-type injector applied to a variable-thrust liquid rocket engine.

[0057] like Figure 3 As shown. First, the design data for the needle-plug injector structure, that is, the optimized design of the needle-plug injector, requires designing a flow-adjustable needle-plug injector structure suitable for variable thrust liquid rocket engines.

[0058] In an optional implementation, step S110 may specifically include the following steps:

[0059] Acquire flow regulation condition data, oxygen-fuel ratio parameters, and combustion chamber pressure parameters. Perform thermodynamic calculations based on the flow regulation condition data and rocket engine principles to obtain thermodynamic calculation results. Determine flow data including characteristic velocity, theoretical ground specific impulse, combustion chamber diameter, total thrust chamber mass flow rate corresponding to each regulation condition point, and mass flow rates of oxygen and combustion paths based on the oxygen-fuel ratio parameters, combustion chamber pressure parameters, and thermodynamic calculation results.

[0060] Obtain the injection ring area of ​​the needle-plug injector, as well as the specified flow coefficients of the oxygen and combustion paths. Calculate the pressure drop of the oxygen and combustion paths based on the specified flow coefficients, the liquid density flowing into the needle-plug injector, and the injection ring area of ​​the needle-plug injector, so that the operating conditions and pressure drop form an upward convex curve relationship.

[0061] Obtain the length-to-diameter ratio of the needle plug length and diameter, and the diameter ratio of the combustion chamber diameter to the needle plug diameter of the needle plug injector. Based on the length-to-diameter ratio and the diameter ratio, determine the structural parameters of the needle plug injector, including the needle plug diameter, needle plug length, and skip distance. Based on the structural parameters, obtain the structural design data of the needle plug injector.

[0062] Based on the simulation results of the flow characteristic curve corresponding to the flow data, the specified flow coefficient and the area of ​​the injection annular gap are corrected to obtain the correction results. Based on the correction results and structural design data, the design structural parameters of the flow-adjustable needle-plug injector applied to the variable thrust liquid rocket engine are obtained.

[0063] Furthermore, the calculation of the pressure drop in the oxygen and combustion circuits based on the specified flow coefficient, the liquid density flowing into the needle injector, and the injection annular gap area of ​​the needle injector can specifically include the following steps: Calculate the pressure drop in the oxygen and combustion circuits using the following formula, based on the specified flow coefficient, the liquid density flowing into the needle injector, and the injection annular gap area of ​​the needle injector: ;in, For the pressure drop in the oxygen and combustion circuits, A flow coefficient between 0 and 1 The density of the liquid flowing into the needle syringe. A For the area of ​​the sprayed circumferential joint, For mass flow rate.

[0064] For example, for the structural design of a circumferential-type needle-plug injector, a needle-plug injector structural design method based on the flow-circumferential-pressure drop relationship can be adopted. The specific process is as follows: Figure 4 As shown, firstly, the flow rate regulation conditions are determined (e.g., 100%~50%, divided into 6 operating points). Then, thermodynamic calculations are performed based on rocket engine principles. According to parameters such as oxygen-fuel ratio and combustion chamber pressure, the total thrust chamber mass flow rate, as well as the mass flow rates of the oxygen and combustion paths, are determined for each regulation operating point, considering the characteristic velocity, theoretical ground specific impulse, and combustion chamber diameter. It is assumed that the operating condition and flow rate have a linear relationship. Then, flow coefficients for the oxygen and combustion paths are assumed based on experience. The coefficient can be a constant value or a concave curve relationship between the operating condition and the flow rate coefficient. The flow rate coefficient will be subsequently corrected based on simulation results. Next, the injection annular gap area and gap width will be initially designed, and the above formula will be used... The pressure drop in the oxygen and combustion circuits was calculated. This process establishes a convex-shaped curve relationship between the operating condition and pressure drop. Then, based on empirical values ​​such as the diameter ratio (the ratio of the combustion chamber diameter to the needle plug diameter) and the length-to-diameter ratio (the ratio of the needle plug length to the diameter), key structural parameters such as the needle plug diameter, needle plug length, and skip distance are determined, completing the structural design. Finally, based on the simulation results of the flow characteristic curve, the flow coefficient and the injection annular gap area are corrected.

[0065] In this embodiment, a needle-plug injector structural design method based on the flow-annular gap-pressure drop relationship is proposed, taking into account the moving parts and adjustable flow rate of the needle-plug injector, making the needle-plug injector structural design more accurate and comprehensive.

[0066] Step S120: Establish a three-dimensional transient numerical simulation model based on the design structural parameters.

[0067] The three-dimensional transient numerical simulation model is used to characterize the dynamic characteristics of the needle-plug injector during needle-plug movement, including flow response characteristics and dynamic spray flow field characteristics. For establishing a high-precision, degradable numerical simulation model, for example, a high-precision three-dimensional numerical simulation model is built by combining governing equations, turbulence models, VOF-to-DPM atomization models, and dynamic mesh models. This model aims to comprehensively consider the dynamic characteristics of the needle-plug injector during needle-plug movement. Figure 3 As shown, the structural design method of the needle-plug injector and the establishment of the three-dimensional transient numerical simulation model are intended to prepare a dataset for the subsequent establishment of a surrogate model.

[0068] In an optional implementation, step S120 may specifically include the following steps:

[0069] A three-dimensional geometric model is established based on the structural parameters of each design, and fluid domain preprocessing is performed based on the three-dimensional geometric model to obtain the preprocessing result; based on the preprocessing result, a Cartesian mesh is used to generate a mesh, and a meshed three-dimensional geometric model is obtained.

[0070] Based on the meshed 3D geometric model, a 3D numerical simulation model is established by combining the governing equations, turbulence model, VOF-to-DPM atomization model, and dynamic mesh model. The 3D numerical simulation model is then used for numerical simulation. The dynamic mesh model represents the coupling effect between the atomization flow field and the moving parts of the needle nozzle. After solving the numerical simulation, the dynamic characteristics of the needle nozzle are obtained, including the flow response characteristics and the dynamic spray flow field characteristics.

[0071] For example, to establish a high-precision, degradable 3D numerical simulation model for a needle-plug injector, an advanced dynamic mesh model and a VOF-to-DPM atomization model can be used, combined with a turbulence model, to build a 3D transient numerical simulation model. This 3D transient numerical simulation model can not only accurately simulate the actual working dynamic process of the needle-plug injector, but also consider the coupling effect between the flow atomization and the moving parts. By reasonably parameterizing the design parameters of the needle-plug injector, the model is ensured to have degradability, providing a feasible foundation for subsequent intelligent optimization design. The numerical simulation calculation process is as follows: Figure 5 As shown.

[0072] In practical applications, standards The model is only suitable for simulating fully turbulent flow processes, while RNG... The model takes into account low Reynolds number flows, improving its accuracy. Therefore, this application embodiment selects an RNG-based model. Model.

[0073] Since the needle-type injector adjusts thrust through the movement of the needle, which involves moving parts, the atomized flow field region will change during the movement. This embodiment of the application uses a dynamic mesh model to consider the coupling effect between the atomized flow field and the moving parts, and uses a spring smoothing method to update the fluid domain mesh that deforms due to the movement.

[0074] It should be noted that the mesh generation is based on an adaptive Cartesian mesh, which can flexibly handle arbitrary three-dimensional geometries and precisely capture the flow field through adaptive mesh refinement technology. Mesh elements affected by the needle-plug injector's movement are also automatically generated based on a Cartesian mesh. Furthermore, in this embodiment, the motion mode of the needle-plug injector's dynamic adjustment process is explicitly defined, also known as active motion, meaning the equation for the motion velocity is predetermined by external conditions.

[0075] The aforementioned VOF-to-DPM atomization model combines the advantages of both VOF and DPM models. In densely sprayed regions, it can better predict the primary and secondary breakup processes of the liquid film. In sparsely sprayed regions, it can track droplet trajectories and predict particle size distribution. For example, in this embodiment, the particle size range in the conversion criterion of the VOF-to-DPM model is set to 0~0.35mm, the conversion frequency is 5, and the sphericity coefficient is set to a default value of 0.5 based on the calculation of the normal standard deviation and the orthogonality of the radius surface.

[0076] By establishing a three-dimensional transient numerical simulation model that couples fluid flow and component motion, and considering the micron-level annular gap flow structure, this model can more comprehensively and accurately characterize the dynamic characteristics of the flow response, internal flow field features, and spray distribution features of the needle-plug injector compared to a one-dimensional numerical simulation model.

[0077] Step S130: Simulation calculation of the dynamic process of continuous adjustment of the needle plug is performed based on the three-dimensional transient numerical simulation model to obtain the simulation dataset. The response surface model is then trained using the simulation dataset to obtain the intelligent agent model of the needle plug injector.

[0078] The response surface model is used to map design variables, which contain various design structural parameters and needle injector operating condition data, to the performance parameters of the needle injector, so as to complete the order reduction process from high-dimensional input to low-dimensional output.

[0079] In establishing the intelligent agent model for the aforementioned needle-operated injector, firstly, based on the established three-dimensional transient numerical simulation model, simulation calculations of the continuous adjustment dynamic process of the needle-operated injector were carried out, and a dataset was established, such as... Figure 3As shown, an intelligent proxy model for the needle-plug injector is constructed for optimized design. This involves introducing an intelligent proxy model based on a high-precision numerical simulation model to achieve more efficient optimization. The core of this step is to establish a response surface model, which maps the design variables of the needle-plug injector structure to its performance parameters, realizing a reduction in order from high-dimensional input to low-dimensional output.

[0080] In one possible implementation, the design variables include the geometric variables of the needle-plug injector and the regulation condition variables; the flow response characteristic indicators corresponding to the needle-plug injector include at least one of regulation accuracy, response time, rise time, and overshoot; the atomization characteristic indicators corresponding to the needle-plug injector include at least one of Solta average particle size and spray cone angle. For example, before establishing the above-mentioned surrogate model, parametric modeling of the needle-plug injector is performed, and calculation examples are planned to ensure comprehensive and effective coverage of the design space. This step is the foundation of the entire optimization design process. Through reasonable parametric modeling and range planning, sufficient information can be ensured for the training and inference of the intelligent surrogate model. In this embodiment, the geometric dimensions of the needle-plug injector (such as circumferential gap width, dimensionless skip distance, dimensionless circumferential gap width) and regulation conditions (such as needle displacement and needle movement speed) are selected as optimization variables. The flow response characteristic indicators of the needle-plug injector, such as regulation accuracy, response time, rise time, and overshoot, and the atomization characteristic indicators, such as Solta average particle size and spray cone angle, are used as performance indicators of the needle-plug injector. For experimental design, different experimental design methods (such as central composite design, Box-Behnken design, etc.) can be selected and compared, and the optimal experimental design method can be selected based on the final goodness-of-fit evaluation.

[0081] The specific process of simulating the dynamic process of continuous needle thimble adjustment based on a three-dimensional transient numerical simulation model in step S130 above, and obtaining the simulation dataset, exemplarily involves establishing a large-scale numerical simulation model and dataset. Specifically, to train the surrogate model, large-scale numerical simulations are relied upon, utilizing a high-performance computer cluster to solve numerous examples within the design parameter range, ensuring the accuracy and comprehensiveness of the numerical simulation model. Post-processing the solved model establishes a large and diverse dataset, providing ample training samples for the surrogate model.

[0082] As an optional implementation, the step S130 above, which involves training the response surface model using a simulation dataset to obtain an intelligent proxy model for the needle-operated injector, may specifically include the following steps:

[0083] A response surface model was selected and established by comparing the Kriging model, radial basis function model, and neural network. The response surface model takes geometric variables and operating condition variables as inputs, and its output includes key parameters of flow response characteristics and atomization characteristics. The response surface model was trained using a simulation dataset to learn and capture nonlinear relationships, resulting in an intelligent proxy model for the needle-plug injector.

[0084] For example, in establishing an intelligent surrogate model for optimizing the design of a needle-operated injector, a response surface model is first built. This step allows for the selection and comparison of various response surface models (such as the Kriging model, radial basis function model, neural network, etc.). This model takes the geometry and operating conditions of the needle-operated injector as input and outputs key parameters related to flow response and atomization characteristics. Through training on a large dataset, the model learns and captures complex nonlinear relationships, thereby establishing a more accurate intelligent surrogate model.

[0085] Step S140: Based on the simulation dataset, the intelligent agent model of the needle-plug injector is optimized by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector.

[0086] In this step, such as Figure 3 As shown, based on the intelligent agent model of the needle-plug injector trained in step S130 above, the agent model of the needle-plug injector is optimized using a reinforcement learning optimization design method. As an example, step S140 may specifically include the following steps:

[0087] Based on a simulation dataset, this study combines an intelligent agent model for needle-plug injectors with a deep reinforcement learning optimization algorithm. Using the deep neural network corresponding to the deep reinforcement learning optimization algorithm and the intelligent agent model for needle-plug injectors as the environment, the intelligent agent model for needle-plug injectors is optimized to obtain the design variable optimization results of the intelligent agent model for needle-plug injectors. Among them, the deep reinforcement learning optimization algorithm includes the proximal policy optimization (PPO) algorithm.

[0088] For example, in the structural optimization design of a needle-plug injector based on an intelligent agent model, the structural optimization design of the needle-plug injector is carried out by combining the established intelligent agent model with an appropriate optimization algorithm. This process mainly focuses on improving the atomization performance, efficiency, and stability of the needle-plug injector. Through iterative optimization, an optimized design scheme that meets the design requirements and performance expectations is obtained. The structural optimization design of the needle-plug injector is carried out by drawing on deep reinforcement learning methods (e.g., the PPO algorithm), using the established intelligent agent model as the environment and a deep neural network as the environment. The optimization design is completed through the interaction between the two. The principle of the optimization design algorithm is as follows: Figure 6 As shown.

[0089] It should be noted that the PPO algorithm is an improvement on the Trust Region Policy Optimization (TRPO) algorithm, using a simpler and more efficient method to enforce the policy. and Similar. The PPO algorithm is as follows: .

[0090] By organically combining the above series of steps, the needle-plug injection device optimization design method proposed in this application embodiment has significant advantages in terms of improving design efficiency, accuracy, and versatility.

[0091] The method provided in this application can serve as an optimization design method for a slit-type needle-plug injector in a variable-thrust liquid rocket engine. By establishing a needle-plug injector structural design method and a three-dimensional transient numerical simulation model, it can prepare a dataset for the subsequent establishment of a surrogate model. Based on the established three-dimensional transient numerical simulation model, simulation calculations of the continuous adjustment dynamic process of the needle plug are carried out, a dataset is established, and an intelligent surrogate model is built. Then, based on the reinforcement learning optimization design method, the needle-plug injector surrogate model is optimized, and the optimization results of the design variables of the intelligent surrogate model are obtained. This realizes an optimization design method for needle-plug injectors based on an intelligent surrogate model and optimization algorithm, improves the design efficiency of high-performance needle-plug injectors, shortens the development cycle of needle-plug engines, and solves the technical problem of low design efficiency of high-performance needle-plug injectors.

[0092] Figure 7 A schematic diagram of an optimized device for a needle-plug injector in a variable-thrust liquid rocket engine is provided. Figure 7 As shown, the variable thrust liquid rocket engine needle plug injector optimization device 700 includes:

[0093] The acquisition module 701 is used to acquire various design structural parameters of the flow-adjustable needle-plug injector applied to a variable-thrust liquid rocket engine;

[0094] Module 702 is used to establish a three-dimensional transient numerical simulation model based on the design structural parameters; wherein, the three-dimensional transient numerical simulation model is used to characterize the dynamic characteristics of the needle injector during the needle movement process, and the dynamic characteristics include flow response characteristics and dynamic spray flow field characteristics;

[0095] Training module 703 is used to perform simulation calculations of the continuous adjustment dynamic process of the needle plunger based on the three-dimensional transient numerical simulation model, obtain a simulation dataset, and use the simulation dataset to train the response surface model, resulting in an intelligent proxy model for the needle plunger injector; wherein, the response surface model is used to map the design variables containing the design structural parameters and the operating condition data of the needle plunger injector to the performance parameters of the needle plunger injector, so as to complete the order reduction process from high-dimensional input to low-dimensional output;

[0096] The optimization module 704 is used to optimize the intelligent agent model of the needle-plug injector based on the simulation dataset by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm, so as to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector.

[0097] The variable thrust liquid rocket engine needle-plug injector optimization device provided in this application embodiment has the same technical features as the variable thrust liquid rocket engine needle-plug injector optimization method provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0098] An electronic device provided in this application embodiment, such as Figure 8 As shown, the electronic device 800 includes a processor 802 and a memory 801. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the method provided in the above embodiments.

[0099] See Figure 8 The electronic device also includes a bus 803 and a communication interface 804. The processor 802, the communication interface 804 and the memory 801 are connected through the bus 803. The processor 802 is used to execute executable modules, such as computer programs, stored in the memory 801.

[0100] The memory 801 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 804 (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network.

[0101] Bus 803 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 8 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0102] The memory 801 is used to store programs. After receiving an execution instruction, the processor 802 executes the program. The method executed by the apparatus defined by the process disclosed in any of the preceding embodiments of this application can be applied to the processor 802 or implemented by the processor 802.

[0103] The processor 802 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 802 or by instructions in software form. The processor 802 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 801, and processor 802 reads the information from memory 801 and, in conjunction with its hardware, completes the steps of the above method.

[0104] Corresponding to the above-described method for optimizing the needle-plug injector of a variable-thrust liquid rocket engine, this application also provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are invoked and executed by a processor, they cause the processor to perform the steps of the above-described method for optimizing the needle-plug injector of a variable-thrust liquid rocket engine.

[0105] The variable thrust liquid rocket engine needle-plug injector optimization device provided in this application embodiment can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in this application embodiment are the same as those in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the above method embodiments, and will not be repeated here.

[0106] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0107] For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0108] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0109] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0110] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the variable thrust liquid rocket engine needle-plug injector optimization method described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0111] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0112] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. An optimization method for a needle-plug injector in a variable-thrust liquid rocket engine, characterized in that, The method includes: Obtain the design structural parameters of the flow-adjustable needle-plug injector for use in variable-thrust liquid rocket engines; A three-dimensional transient numerical simulation model is established based on the aforementioned design structural parameters; wherein, the three-dimensional transient numerical simulation model is used to characterize the dynamic characteristics of the needle-plug injector during the needle-plug movement process, the dynamic characteristics including flow response characteristics and dynamic spray flow field characteristics; The simulation calculation of the continuous adjustment dynamic process of the needle plunger is performed based on the three-dimensional transient numerical simulation model to obtain a simulation dataset. The response surface model is then trained using the simulation dataset to obtain an intelligent agent model for the needle plunger injector. The response surface model is used to map the design variables, which contain the design structural parameters and the operating condition data of the needle plunger injector, to the performance parameters of the needle plunger injector, so as to complete the order reduction process from high-dimensional input to low-dimensional output. Based on the simulation dataset, the intelligent agent model of the needle-plug injector is optimized by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm, so as to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector.

2. The method according to claim 1, characterized in that, The establishment of a three-dimensional transient numerical simulation model based on the aforementioned design structural parameters includes: A three-dimensional geometric model is established based on the design structural parameters, and fluid domain preprocessing is performed based on the three-dimensional geometric model to obtain the preprocessing result; Based on the preprocessing results, a Cartesian grid is used to divide the mesh, resulting in a three-dimensional geometric model after mesh division. Based on the three-dimensional geometric model after meshing, and combined with the governing equations, turbulence model, VOF-to-DPM atomization model, and dynamic mesh model, a three-dimensional numerical simulation model is established and numerical simulation is performed on the three-dimensional numerical simulation model. The dynamic mesh model represents the coupling effect between the atomization flow field and the moving parts of the needle nozzle. After numerical simulation, the dynamic characteristics of the needle nozzle are obtained, including flow response characteristics and dynamic spray flow field characteristics.

3. The method according to claim 1, characterized in that, The design variables include the geometric variables and adjustment condition variables of the needle-plug injector; the flow response characteristic index of the needle-plug injector includes at least one of adjustment accuracy, response time, rise time, and overshoot; and the atomization characteristic index of the needle-plug injector includes at least one of Solta average particle size and spray cone angle.

4. The method according to claim 3, characterized in that, The process of training the response surface model using the simulation dataset to obtain an intelligent proxy model for the needle-operated injector includes: A response surface model was selected and established by comparing the Kriging model, the radial basis function model, and the neural network. The response surface model takes the geometric variables and the regulation condition variables as inputs, and the output of the response surface model includes key parameters of the flow response characteristic index and the atomization characteristic index. The response surface model is trained using the simulation dataset so that it can learn and capture nonlinear relationships, resulting in an intelligent proxy model for the needle-plug injector.

5. The method according to claim 1, characterized in that, Based on the simulation dataset, the intelligent agent model of the needle-plug injector is optimized by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm, resulting in the optimization results of the design variables of the intelligent agent model of the needle-plug injector, including: Based on the simulation dataset, the intelligent agent model of the needle-plug injector is optimized by combining the intelligent agent model of the needle-plug injector with a deep reinforcement learning optimization algorithm. The deep neural network corresponding to the deep reinforcement learning optimization algorithm and the intelligent agent model of the needle-plug injector are used as the environment to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector. The deep reinforcement learning optimization algorithm includes the proximal policy optimization (PPO) algorithm.

6. The method according to claim 1, characterized in that, The acquisition of various design structural parameters for a flow-adjustable needle-plug injector applied to a variable-thrust liquid rocket engine includes: Acquire flow regulation condition data, oxygen-fuel ratio parameters, and combustion chamber pressure parameters. Perform thermodynamic calculations based on the flow regulation condition data and rocket engine principles to obtain thermodynamic calculation results. Determine flow data including characteristic velocity, ground theoretical specific impulse, combustion chamber diameter, total thrust chamber mass flow rate corresponding to each regulation condition point, and mass flow rates of oxygen and combustion paths based on the oxygen-fuel ratio parameters, combustion chamber pressure parameters, and thermodynamic calculation results. The injection ring area of ​​the needle-plug injector and the specified flow coefficients of the oxygen and combustion paths are obtained. The pressure drop of the oxygen and combustion paths is calculated based on the specified flow coefficients, the liquid density flowing into the needle-plug injector and the injection ring area of ​​the needle-plug injector, so that the operating conditions and pressure drop form an upward convex curve relationship. The length-to-diameter ratio of the needle length and diameter of the needle injector, and the diameter ratio of the combustion chamber diameter and the needle diameter of the needle injector are obtained. Based on the length-to-diameter ratio and the diameter ratio, the structural parameters of the needle injector, including the needle diameter, needle length and skip distance, are determined. Based on the structural parameters, the structural design data of the needle injector is obtained. Based on the simulation results of the flow characteristic curve corresponding to the flow data, the specified flow coefficient and the injection annular gap area are corrected to obtain the correction results. Based on the correction results and the structural design data, the design structural parameters of the flow-adjustable needle-plug injector applied to the variable thrust liquid rocket engine are obtained.

7. The method according to claim 6, characterized in that, The calculation of the pressure drop in the oxygen and combustion circuits based on the specified flow coefficient, the liquid density flowing into the needle injector, and the injection annular area of ​​the needle injector includes: Based on the specified flow coefficient, the liquid density flowing into the needle injector, and the injection annular area of ​​the needle injector, the pressure drop in the oxygen and combustion circuits is calculated using the following formula: ； in, For the pressure drop in the oxygen and combustion circuits, A flow coefficient between 0 and 1 The density of the liquid flowing into the needle syringe. A The area of ​​the injection circumferential joint is given. For mass flow rate.

8. An optimization device for a needle-plug injector of a variable-thrust liquid rocket engine, characterized in that, include: The acquisition module is used to acquire various design structural parameters of the flow-adjustable needle-plug injector applied to variable-thrust liquid rocket engines; A module is established to build a three-dimensional transient numerical simulation model based on the design structural parameters; wherein, the three-dimensional transient numerical simulation model is used to characterize the dynamic characteristics of the needle injector during the needle movement process, and the dynamic characteristics include flow response characteristics and dynamic spray flow field characteristics; The training module is used to perform simulation calculations of the continuous adjustment dynamic process of the needle plunger based on the three-dimensional transient numerical simulation model, obtain a simulation dataset, and use the simulation dataset to train the response surface model, resulting in an intelligent proxy model for the needle plunger injector. The response surface model is used to map the design variables, which contain the design structural parameters and the operating condition data of the needle plunger injector, to the performance parameters of the needle plunger injector, thereby completing the order reduction process from high-dimensional input to low-dimensional output. The optimization module is used to optimize the intelligent agent model of the needle-plug injector based on the simulation dataset by combining the intelligent agent model of the needle-plug injector with a specified reinforcement learning optimization algorithm, so as to obtain the optimization results of the design variables of the intelligent agent model of the needle-plug injector.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

Images