Modal conversion process control plan design method considering air inlet / TBCC engine state matching

Abstract

The invention provides a modal conversion control plan design method considering air inlet / TBCC engine state matching, and the method comprises the steps: dividing a modal conversion control plan design flow into two steps: modal conversion starting point optimization and modal conversion process optimization. The optimal state of the mode conversion starting point is the initial starting point state of the mode conversion process. Compared with a control plan of a conventional engine level design, the method has the advantages that the temperature of an afterburner outlet can be reduced to 54K on the premise of keeping thrust not reduced, and the temperature boundary of limiting the mode conversion speed is effectively widened. The designed control plan can ensure that the temperature limiting boundary is not touched in the whole modal conversion process, modal conversion can be completed within 32 s, the time is shortened by 12 s compared with a conventional method, and meanwhile smooth transition of thrust and air flow can be ensured.

Description

Technical Field

[0001] The present invention relates to modeling and control of a combined power engine modal conversion process, and belongs to the field of aero-engine control, and in particular to a modal conversion process control plan design method considering inlet duct / TBCC engine state matching. Background Art

[0002] Turbine-based combined cycle (TBCC) engines, as a power source for hypersonic vehicles, have long been a hot topic of research in the field of combined propulsion. TBCC engines, structurally combining a turbine engine and a ramjet (or scramjet), offer flight Mach numbers ranging from subsonic and supersonic to hypersonic, and possess advantages such as horizontal takeoff and landing, wide range, and high economy. The modal transition process refers to the transition from a turbine engine (referred to as the turbine mode) to a ramjet engine (referred to as the ramjet mode). This ability to transition between these two modes enables TBCC engines to operate efficiently across a wide flight envelope. Modal transition control design involves designing the trajectory of the engine's adjustable mechanisms during this transition process to ensure a smooth and safe transition from turbofan to ramjet mode within parameter constraints. Since turbine engine components and ramjet engine components work together during the modal conversion process, the number of adjustable mechanisms is more than twice that of conventional engine adjustment mechanisms. The increase in the number of working parts and the increase in adjustable mechanisms greatly increases the difficulty of designing the process control plan.

[0003] Current TBCC engine mode switch control plan design methods typically focus on the engine itself, without much consideration of the matching adjustment between the inlet and the engine. Qiu Xiaojie of the Control Systems Research Institute of Aero-Engine Corporation of China used the SQP method to optimize the engine fuel flow, tail nozzle throat area, and various flow control valves during the mode switch process for a small parallel TBCC engine, resulting in a mode switch control plan [Qiu X., Su W., Tang Y. The mode switch control research of a small-type parallel TBCC engine based on the SQP method [C]. AIAA Modeling and Simulation Technologies Conference, AIAA 2015-2656, Dallas: 2015]. Zhang Mingyang from Northwestern Polytechnical University studied a small tandem TBCC engine as a research object to study the law of engine performance change. Based on the thrust and flow continuity criteria, he obtained the control plan of each engine adjustment parameter in the mode transition process [Zhang Mingyang, Wang Zhanxue, Zhang Xiaobo, et al. Simulation of windmill ramjet modal performance of tandem TBCC engine [J]. Journal of Aeronautics and Astronautics, 2018, 33(12): 2939-2949.]. Chen Min from Beijing University of Aeronautics and Astronautics studied the multi-objective control problem of tandem TBCC engine [Chen M., Tang H., Zhu Z. Goal Programming for Stable Mode Transition in TandemTurbo-ramjet Engines [J]. Chinese Journal of Aeronautics, 2009, 22(05): 486-492.] and used the Newton-Raphson method to study the mode transition control plan for the component-level model of the TBCC engine. The mode transition control plan obtained based on the above method can basically achieve a smooth transition of state parameters such as thrust during the mode transition process. However, as the modal conversion process is a dynamic process, we always hope that the engine can reach the target state and operate stably as quickly as possible. Therefore, under the premise of ensuring a smooth transition of the modal conversion, how to further reduce the time required for modal conversion is also one of the goals of the modal conversion process control plan design. Improving the modal conversion speed is actually an optimization problem that determines the shortest path that can achieve engine state transition under the premise of meeting the engine constraint parameter restrictions and achieving a smooth transition of the engine state. In the modal conversion control plan designed at the traditional engine level, the total temperature at the afterburner outlet is one of the important factors that restrict the acceleration of the modal conversion process. Due to the limitation of the total temperature at the afterburner outlet, the conversion speed that can be achieved by the traditional modal conversion control plan has basically reached its limit.

[0004] The modal conversion process occurs in the upper right area of ​​the turbofan engine operating envelope. At this time, the high-altitude and high Mach number flight conditions significantly enhance the coupling between the air inlet and the engine. Whether the air inlet and the engine can be well matched will greatly affect the conversion effect of the modal conversion process. Taking the matching characteristics of the air inlet and the engine into consideration when designing the modal conversion control plan will hopefully significantly improve the conversion speed of the modal conversion process. In view of the above analysis, the present invention proposes a modal conversion control plan design method that considers the matching of the air inlet and the TBCC engine state. The method includes two steps: modal conversion starting point optimization and modal conversion process optimization. The method obtains and maintains a good matching relationship between the air inlet and the engine during the modal conversion process, effectively widens the limit boundary of the total temperature at the afterburner outlet that restricts the speed of the modal conversion process, thereby significantly improving the conversion speed of the modal conversion process while ensuring a safe and smooth transition of the engine thrust. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the existing technology by designing a modal conversion control plan that can achieve better matching between the intake duct and the engine state, thereby effectively improving the speed of the TBCC engine modal conversion process while meeting the requirements of safety and stability.

[0006] Technical Solution: The present invention specifically provides a method for designing a modal conversion process control plan that considers intake duct / TBCC engine state matching. The method comprises the following steps:

[0007] Step 1: Based on the sequential quadratic programming (SQP) algorithm, steady-state optimization is performed. The supersonic inlet adjustable variables and the engine adjustable variables are selected as optimization variables. The modal conversion starting point is optimized to obtain the optimal state and proceed to the next step.

[0008] Step 2: Dynamic optimization is performed based on the particle swarm optimization (PSO) algorithm. The position of the inlet normal shock wave is constantly augmented into the optimization objective function to optimize the modal conversion process, thereby ensuring a good match between the inlet and the engine during the modal conversion process.

[0009] The optimal state of the modal conversion starting point obtained in step 1 is the starting state for initialization of the modal conversion process in step 2.

[0010] As an improvement, the mode conversion starting point optimization includes selecting the engine adjustment variable u engine and supersonic inlet control variable u inlet As the optimization variable, the engine minimum afterburner outlet total temperature T7 is taken as the optimization target to determine the engine state at the start of the mode conversion. The expression is as follows

[0011]

[0012] In formula 1, u is the optimization variable, u=[u inlet ,u engine ]=[β N2 ,L N2 ,A bl ,W fb ,W fa ,A8]W fb is the main fuel flow, W fa is the afterburner fuel flow, A8 is the tail nozzle throat area, β N2 The angle of the inclined plate can be adjusted in two levels, L N2 A is the length of the secondary adjustable inclined plate. bl is the deflation area of ​​the auxiliary deflation valve, F is the thrust, n L is the fan speed, n H is the compressor speed, T4 is the total temperature at the combustion chamber outlet, S mL is the fan surge margin, S mH is the compressor surge margin, RM is the return flow margin, x A is the position of the normal shock wave; the subscript “origin” represents the corresponding value of the parameter before optimization, and the subscripts “max” and “min” represent the maximum and minimum allowable values ​​of the corresponding parameters.

[0013] As an improvement, the present invention uses the sequential quadratic programming (SQP) optimization algorithm to optimize the performance of the modal conversion starting point. The SQP algorithm is an algorithm that converts complex nonlinear constrained optimization problems into relatively simple quadratic programming (QP) problems for solution. The Broyden-Fletcher-Goldforb-Shanno (BFGS) algorithm in the quasi-Newton method is used to optimize the B k To perform the calibration, the specific steps are as follows:

[0014] Step 1.1: Initialize parameters:

[0015] Initialize the objective function f(x), the constraint function g(x)≥0, h(x)=0; select the initial feasible point x0, satisfying g(x0≥0; initialize the Hessian inverse matrix B0=I, where I is the identity matrix; initialize the Lagrange multiplier λ0=0, μ0=0;

[0016] Step 1.2: Solve the QP subproblem (k-th search, k ≥ 0):

[0017] At the current point x k , solve the following quadratic programming problem:

[0018]

[0019] Where d is the search direction, B is the Hessian approximation matrix estimated by the BFGS quasi-Newton method, f(x) is the objective function, and g is i (x) is the inequality constraint, h j (x) is an equality constraint; Represents the gradient of the corresponding function; is the index set of inequality constraints, and ε is the index set of equality constraints; the Lagrange multiplier method is used to solve and output the current step search direction d k and the Lagrange multiplier vector λ of the inequality constraints k+1 and the Lagrange multiplier vector μ for the equality constraints k+1 ;

[0020] Step 1.3: Update parameters,

[0021] x k+1 =x k +α k d k Formula 3

[0022] where α k Can be determined by line search, the default is α k =1.

[0023] Step 1.4: Update the Hessian approximation matrix B using the BFGS method k :

[0024] Calculate parameter change s k and gradient change y k :

[0025] s k =x k+1 -x k ,y k =▽ x L(x k+1 ,λ k+1 ,μ k+1 )-▽ x L(x k ,λ k+1 ,μ k+1 ) Formula 4

[0026] The Lagrangian function is:

[0027]

[0028] Update B k :

[0029]

[0030] Step 1.5: Convergence determination:

[0031] If || d k ||<ε or the maximum number of iterations is reached, terminate and output x k ; Otherwise, let k = k + 1 and return to step 1.2 to continue iterating.

[0032] As an improvement, the objective function is optimized during the mode conversion process, including the fan speed n L Idle speed n Lidle , the specific expression is

[0033]

[0034] In formula 7, F before is the thrust obtained by optimizing the starting point of the modal conversion, x A,opt is the positive shock wave position of the inlet under mild supercritical state, u is the optimization variable, where u=[W fb ,W fa ,A8,A MSV ,A RVABI ], ω1, ω2, ω3 are the weight coefficients of the objective function.

[0035] As an improvement, the constraint boundary function set during the modal conversion process is formula 8:

[0036]

[0037] In formula 8, far4 is the fuel-air ratio of the main combustion chamber, n L is the fan speed, n Lidle is the slow speed, T7 is the total temperature at the ramjet combustion chamber outlet, S mL is the fan surge margin, S mH is the compressor surge margin, RM is the return flow margin, x A is the normal shock wave position, and |du| is the adjustment amount of the actuator within a single step.

[0038] As an improvement, the steps of the particle swarm algorithm are as follows: Assuming that the search space is D-dimensional, then the i-th particle X in the particle swarm i Expressed as,

[0039] X i =(x i,1 ,x i,2 ,L,x i,D ) T Formula 9

[0040] The particle group X is a population consisting of n particles.

[0041] X=(X1,X2,L,X n ) Formula 10

[0042] In the population, each particle is a possible solution to the problem, and X is calculated according to the objective function. i The fitness of the i-th particle, the update speed V i for,

[0043] V i =(v i,1 ,v i,2 ,L,v i,D ) T Formula 11

[0044] The optimal solution extreme values ​​of individuals and populations are,

[0045]

[0046] Where, P i is the individual's optimal solution extreme value, P g is the optimal solution extreme value of the population.

[0047] After the population is generated, particles update their speed and position according to individual extreme values ​​and group extreme values. In the kth iteration,

[0048]

[0049] In Formula 13, w is the inertia weight, d = 1, 2, ..., D, i = 1, 2, ..., n, c1 and c2 are acceleration factors, and r1 and r2 are random numbers.

[0050] Beneficial Effects: The present invention proposes a method for designing a modal transition control plan that takes into account the matching of the inlet duct / TBCC engine state. Compared with the control plan designed at the conventional engine level, this method can reduce the afterburner outlet temperature by up to 54K while maintaining the thrust, effectively widening the temperature boundary that limits the modal transition speed. The designed control plan can ensure that the temperature limit boundary is not touched during the entire modal transition process, and the modal transition can be completed within 32 seconds, which is 12 seconds less than the conventional method, while ensuring a smooth transition of thrust and air flow.

[0051] In addition, the present invention specifically adopts the following technical means to solve the technical problems:

[0052] 1. A method for optimizing the starting state of modal conversion based on sequential quadratic programming (SQP). engine and intake duct control variables u inlet They are used together as optimization variables to improve the common operating point position of the mode transition point inlet and the engine, and to broaden the afterburner outlet temperature boundary that limits the mode transition speed while ensuring that the thrust does not decrease.

[0053] 2. Optimize the modal conversion process using a particle swarm optimization (PSO) approach. The position of the inlet normal shock wave is augmented into the optimization objective function, ensuring that the matching operating point between the inlet and engine remains unchanged throughout the conversion process, maintaining a good match between the inlet and engine. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 Schematic diagram of the design steps of the mode conversion control plan based on SQP-PSO in the present invention.

[0055] Figure 2 This is the mode conversion control plan design process based on SQP-PSO of the present invention.

[0056] Figure 3 is the temperature limit of the mode conversion process of the present invention.

[0057] Figure 4 This is the comprehensive optimization principle for the present invention.

[0058] Figure 5 Optimizing the modal transition starting point performance of the TBCC propulsion system of the present invention: (a) thrust; (b) total temperature at the afterburner outlet; (c) positive shock wave position; and (d) total pressure recovery coefficient.

[0059] Figure 6 Optimizing the adjustable variable input and performance parameter changes for the modal conversion starting point of the present invention, (a) adjustable variables; (b) key performance parameters.

[0060] Figure 7 The control plan for the conversion process of the modal conversion point (Ma=2.8) of the present invention includes: (a) mode selection valve; (b) fuel flow in the main combustion chamber; (c) throat area of ​​the tail nozzle; (d) fuel flow in the afterburner; (e) front variable area ducted ejector; (f) rear variable area ducted ejector.

[0061] Figure 8 Figure 2 shows the changes in key state parameters during the conversion process of the modal conversion point (Ma=2.8) of the present invention, including: (a) thrust; (b) normal shock wave position; (c) rotational speed; (d) air flow rate; (e) total temperature at the main combustion chamber outlet; (f) total temperature at the afterburner combustion chamber outlet; (g) surge margin; and (h) backflow margin.

[0062] Figure 9 Comparison of the modal conversion control effects of the present invention, (a) thrust; (b) normal shock wave position; (c) backflow margin; (d) speed; (e) total temperature at the afterburner outlet; (f) air flow. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the present invention will be described clearly and completely below so that those skilled in the art can better understand the advantages and features of the present invention and thus more clearly define the scope of protection of the present invention. The embodiments described in the present invention are only some of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without making any creative work shall fall within the scope of protection of the present invention.

[0064] This paper proposes a modal transition control plan design method that considers the matching of the intake duct and the TBCC engine state. This method divides the modal transition control plan design process into two steps: modal transition starting point optimization and modal transition process optimization. Furthermore, for modal transition starting point optimization, a steady-state optimization based on a sequential quadratic programming (SQP) algorithm is performed to obtain the optimal intake duct / engine matching state at the modal transition starting state, effectively widening the temperature limit that restricts the modal transition speed.

[0065] For the optimization of the modal conversion process, dynamic optimization is performed based on the particle swarm optimization (PSO) algorithm to obtain the change trajectory of each control variable in the modal conversion process. By constantly augmenting the position of the inlet positive shock wave to the optimization objective function, the optimal matching state between the inlet and the engine is guaranteed to remain unchanged during the conversion process.

[0066] Compared to existing modal transition control plans, the control plan derived from the SQP-PSO method ensures a consistent and well-matched inlet and engine performance throughout the entire modal transition process. This significantly reduces the time required for the transition while ensuring a safe and smooth transition. Simulation results show that the afterburner outlet temperature does not reach the maximum temperature limit throughout the entire modal transition process, reducing the modal transition time by nearly 12 seconds and maintaining the maximum engine thrust fluctuation to less than 1%.

[0067] See Figure 1-2 , shown is the modal transition control plan design process based on SQP-PSO of the present invention, which specifically includes: first, starting point steady-state optimization is performed based on SQP to obtain the optimal air intake / engine matching state under modal transition flight conditions; on this basis, process optimization is performed based on PSO to maintain the optimal air intake / engine matching state during the modal transition process, and finally the control plan for each control variable of the engine during the modal transition process is found.

[0068] The two steps of modal conversion starting point optimization and modal conversion process optimization are described in detail below with reference to the accompanying drawings and specific implementations.

[0069] 1. Modal Conversion Starting Point Optimization Based on SQP

[0070] As a dynamic process, modal transitions are often accelerated to enable the propulsion system to reach its target state and stabilize operation as quickly as possible. However, to ensure a smooth and safe transition, the modal transition process is constrained by the engine's limiting parameters, preventing it from being infinitely accelerated. This section first investigates the optimization of the modal transition starting point, minimizing the engine parameters' proximity to the limiting boundary and providing sufficient margin for accelerating the modal transition process.

[0071] (1) Purpose and mathematical model of starting point optimization

[0072] At the beginning of the mode conversion process, the TBCC propulsion system works in the turbofan afterburner state. After the mode conversion begins, the fuel flow rate in the main combustion chamber W fb Gradually decrease, the fan, compressor and other rotor components gradually from the maximum state (n Lmax ,n Hmax ) decelerate to slow state (n Lidle ,n Hidle ); At the same time, the ramjet duct mode selection valve is opened, and the fuel flow rate of the afterburner is W fa Gradually increases to make up for the thrust loss caused by the decrease of main fuel. In this process, the air at the inlet of the air intake gradually transfers from the turbofan channel to the ramjet channel. If the speed of afterburner fuel increase is too fast, it will cause the total temperature T7 at the outlet of the afterburner to exceed the temperature. Figure 3 As shown in the figure, existing mode transition control plans typically use the turbofan's maximum or partial afterburner state as the transition starting point. The values ​​of the engine's adjustable variables are determined by the existing control plan for conventional turbofan engines (turbofan mode). Directly designing the mode transition control plan based on this starting point can slow the transition process due to the afterburner outlet total temperature limit. It can even cause the afterburner outlet total temperature to overheat during the transition process, posing serious safety risks.

[0073] Since the flight conditions at the mode transition point are close to the maximum operating Mach number of the turbofan mode, the total inlet temperature T2 increases significantly at this time, which reduces the engine rotor conversion speed. Under this condition, since the turbofan engine has deviated far from the optimal working state, the traditional performance optimization based solely on the engine adjustable variables cannot achieve satisfactory results. At this time, if the inlet adjustable variables are included in the optimization variables and the matching conditions between the inlet and the engine are improved, the total temperature at the afterburner outlet can be significantly reduced without reducing the thrust F, thereby widening the temperature limit boundary. Based on the above ideas, the starting state optimization of the mode transition process is carried out, and the engine adjustment variable u is selected. engine (Main fuel flow W fb , afterburner fuel flow W fa , tail nozzle throat area A8) and inlet duct adjustment variable u inlet (Secondary adjustable inclined plate angle βN2 , two-stage adjustable inclined plate length L N2 , auxiliary air release valve air release area A bl ) as the optimization variable, and the engine's minimum afterburner outlet total temperature T7 as the optimization target to determine the engine state at the starting point of the mode conversion.

[0074] In the present invention, Figure 4 It can be seen that the combined intake / engine optimization not only improves the total pressure recovery coefficient of the intake itself, but also obtains the optimal intake and engine matching operating point. The expression for establishing the starting point optimization is shown in Equation (1).

[0075]

[0076] Where u is the optimization variable, u=[u inlet ,u engine ]=[β N2 ,L N2 ,A bl ,W fb ,W fa ,A8], the subscript “origin” represents the corresponding value of the parameter before optimization.

[0077] As a specific embodiment of the present invention, the above optimization model parameters are selected as shown in Table 1 below:

[0078] Table 1 Example of optimized model parameters

[0079]

[0080] (2) SQP optimization algorithm

[0081] The present invention uses the sequential quadratic programming (SQP) optimization algorithm to optimize the performance of the modal conversion starting point. The SQP algorithm is an algorithm that converts complex nonlinear constraint optimization problems into relatively simple quadratic programming (QP) problems for solution. The Broyden-Fletcher-Goldforb-Shanno (BFGS) algorithm in the quasi-Newton method is used to optimize the B k The specific steps are described and introduced below in conjunction with Example 1.

[0082] Step 1.1: Initialize parameters:

[0083] Objective function The equality constraint h(x)=0 is FF origin =0; the inequality constraint function g(x)≥0 is:

[0084]

[0085] Initial feasible point x0=[βN2 ,L N2 ,A bl ,W fb ,W fa ,A8] = [19.749 0.218 0.323 0.120 1.6420.512]; the Hessian inverse matrix B0 = I. Initialize the Lagrange multiplier λ0 = 0.

[0086] Step 1.2: Use the Lagrange multiplier method to solve the QP subproblem (taking the first search as an example):

[0087] For the following quadratic programming problem:

[0088]

[0089] At the current point x0, the Lagrange multiplier method is used to solve. After one iteration, the output is d0 = [0.259 0.002 0.002 0.05 0.12 0.356]; λ1 = Δλ = [0.54 0.232 0.254 0.04 0.85 0.05 0.756 1.15 0.356]; μ1 = 1.25.

[0090] Step 1.3: Update parameters,

[0091] x1=x0+α0d0=[20.008 0.216 0.325 0.17 1.522 0.156]

[0092] Step 1.4: Update the Hessian approximation matrix B using the BFGS method k , calculate the parameter change s k and gradient change y k 、B k :

[0093] s1=x1-x0=[0.259 0.002 0.002 0.05 0.12 0.356]

[0094] y1=▽ x L(x1,λ1,μ1)-▽ x L(x0,λ1,μ1)=[0.855 0.01 1.002 0.85 1.14 0.356]

[0095]

[0096] Step 1.5: Convergence determination:

[0097] If || d k ||<∈ or the maximum number of iterations is reached, terminate and output x kOtherwise, let k = k + 1 and return to step 1.2 to continue the iteration. The calculation process summarized in the above steps is shown in Figure 2 In the blue box on the left.

[0098] (3) Analysis of the results of modal conversion starting point optimization data

[0099] The optimization result of the modal conversion starting point is as follows Figure 5 In order to reflect the advantages of performance optimization at the propulsion system level, the simulation process compared the optimization method of only considering the engine adjustable variables (the optimization amount is W fb ,A8,W fa ). The optimization command is entered at t=20s. From the optimization results, it can be seen that the engine thrust F remains basically unchanged before and after the optimization. Under the engine optimization scheme, the afterburner outlet temperature T7 can be reduced by about 18K, while under the forward comprehensive optimization scheme, the afterburner outlet temperature drops by 54K, which is significantly better than the engine optimization scheme. The original state is determined according to the original control plan of the turbofan engine. The total temperature at the afterburner outlet is high, and the positive shock wave position x A Deviation from the optimal working position x A,opt =1.1. In both optimization modes, the position of the inlet positive shock wave is towards x A,opt Closer, the work is closer to the critical state, making the total pressure recovery coefficient σ inlet However, the intake duct adjustment increases the value of the total pressure recovery coefficient in the critical state, improves the overall working efficiency, and thus obtains better optimization performance.

[0100] Figure 6 The changes of the propulsion system adjustable variables and other main limiting parameters and performance parameters after the starting point optimization are given. It can be seen that after the optimization instruction is cut in at 20s, all parameters change within a reasonable range. The fan speed n of the engine after optimization is L The temperature before the turbine T4 increased, but never exceeded the limit. in It is basically the same as before optimization, and in the comprehensive optimization mode, the installed thrust even increases slightly compared with before optimization, which is mainly due to the increase in the engine required air flow W a2 , resulting in the deflation flow rate W a,bl Descending, deflation resistance D bl Decrease.

[0101] 2. Modal Conversion Process Optimization Based on PSO

[0102] After determining the modal transition starting point, the next step was to optimize the modal transition process. During the modal transition, the rotor components of the TBCC propulsion system gradually decelerate, the ramjet duct gradually opens, and the inlet air flow gradually flows from the turbofan duct to the ramjet duct. Because this process involves many working components, the nonlinearity and number of adjustable variables are significantly increased compared to conventional turbofan and ramjet engines.

[0103] (1) Establishment of process optimization mathematical model

[0104] To maintain the inlet / exit matching relationship obtained by optimizing the starting point of the mode transition, it is necessary to continuously optimize the inlet and exhaust during the conversion process. The flight Mach number is approximately constant during the conversion, so no additional adjustment is required for the inlet, and its adjustable mechanism maintains the optimal position determined by the starting point optimization. Therefore, the adjustable variables of the TBCC propulsion system during the mode transition are engine-related parameters: the fuel flow rate W in the main combustion chamber fb , afterburner fuel flow W fa , tail nozzle throat area A8, mode selection valve opening area A MSV , rear adjustable duct area ejector outlet area A RVABI .

[0105] The primary goal of the modal conversion process optimization is to keep the total thrust of the propulsion system constant (F=Const). In addition, unlike the traditional method that only focuses on the constant engine thrust, in order to maintain the temperature margin obtained by the starting point optimization, it is necessary to ensure a good inlet / engine matching relationship during the conversion process. In the inlet / engine integrated propulsion system, whether the inlet and the engine are well matched is reflected in the position of the inlet positive shock wave. When the engine demand flow is equal to the inlet air supply (W a2 =W a1 =Const), the inlet works in a slightly supercritical state, and the positive shock wave is located just downstream of the throat (x A =x A,opt =1.1); when the demand increases (W a2 >W a1 ), the normal shock wave position moves downward (x A >x A,opt ), the intake duct works in a supercritical state; when the demand decreases (W a2 <W a1 ), the normal shock wave position moves forward (x A <x A,opt ), at this time, it is very easy for the air intake to enter the non-starting working state. Therefore, under the premise of not changing the intake volume, the normal shock wave position is always guaranteed to be at the normal shock wave position obtained by optimizing the starting point of the mode conversion during the mode conversion process (x A =x A,opt ) can ensure that the intake duct and the engine are always well matched during the conversion process.

[0106] In order to complete the mode conversion process as quickly as possible, the fan speed should be adjusted to the slow speed n as soon as possible. Lidle , the objective function of the optimization process should also include the fan speed n L According to the above analysis, the objective function expression of the TBCC modal conversion process is shown in formula (4),

[0107]

[0108] Where, F before is the thrust obtained by optimizing the starting point of the modal conversion, x A,opt is the positive shock wave position of the inlet under the mild supercritical state, and the optimization variable u=[W fb ,W fa ,A8,A MSV ,A RVABI ].

[0109] During the mode conversion process, the engine state constraint is set as the constraint boundary function. To prevent the afterburner outlet temperature from exceeding the material heat resistance limit and causing damage to the engine, the total temperature of the ramjet outlet needs to be limited (T7≤T 7max ); If the main fuel flow rate drops too quickly, causing the oil-air ratio to be less than a certain value, the main combustion chamber will be left with fuel and flameout, and the thrust will drop sharply. Therefore, it is necessary to constrain the oil-air ratio of the main combustion chamber (far4≥far 4min ); At the end of the mode conversion, the turbofan engine is in the slow state, and the fan speed is equal to the slow speed, so the minimum speed needs to be limited (n L ≥n Lmin ); At the same time, during the mode conversion process, it is also necessary to ensure that the engine does not gasp (S mL ≥S ML,min ,S mH ≥S mH,min ), the backflow margin is greater than 0 (RM>0), and the inlet positive shock wave position is located downstream of the throat (x A >1); the adjustment amount of the actuator within a single step is not greater than the maximum allowable adjustment amount (|du| <du max ). The constraint boundary function is set as shown in formula (13).

[0110]

[0111] At this point, a mathematical model for the engine mode transition control plan optimization problem has been established. This mathematical model is a multi-objective optimization problem with multiple constraints.

[0112] Example 2

[0113] The objective function is initialized as:

[0114]

[0115] ω1, ω2, and ω3 are the objective function weight coefficients. Preferably, as a specific embodiment of the present invention, after repeated trial and error, the objective function weight coefficients are selected as ω1 = 0.42, ω2 = 0.3, and ω3 = 0.28. Since the goal is to minimize the objective function, the objective function is directly used as the fitness function F(x) for evaluating each particle.

[0116] The inequality constraint g(x)≥0 is:

[0117]

[0118] Set the population size n = 200, the search space dimension D = 5, the inertia weight w = 0.8, the acceleration factors c1 = 0.2, c2 = 0.2, the maximum number of iterations M = 200, and the threshold limit ξ = 0.1.

[0119] During a single-step iteration, the optimal value of the fitness function decreases to [15.05 2.350.140.14 0.02]. When the optimal value of the fitness function is less than the given threshold ξ = 0.1, the optimization requirement is met, the search is stopped, and the optimization process for the next step is carried out.

[0120] The above optimization method is used to obtain the control plan of each adjustable variable of the engine during the mode conversion process as follows: Figure 7 As shown. The mode conversion starts at t=10s. In order to ensure sufficient backflow margin, the FVABI is closed in advance before the conversion begins. During the mode conversion, the MSV is gradually opened to open the ramjet channel. As the area of ​​the MSV increases, the fuel in the main combustion chamber decreases, causing the turbofan engine to deteriorate. The area of ​​the RVABI gradually increases to ensure that the outer flow can pass smoothly. During this period, the throat area of ​​the tail nozzle basically shows a trend of gradual increase. The afterburner fuel increases rapidly at first, and then shows a downward trend near the end of the mode conversion process. This is because the working points of the fan, compressor and other components deviate from the high-efficiency working area during the deceleration process. When the mode conversion reaches the end point (slow state), the working points of the fan and compressor components return to the higher efficiency area. At this time, in order to ensure constant thrust, the afterburner fuel decreases.

[0121] During the optimization process, the changes in the engine state parameters are as follows: Figure 8 As shown in the figure, it can be seen from the simulation results that:

[0122] 1) The optimized mode conversion control law can ensure constant thrust and smooth transition during the mode conversion. At the beginning of the conversion, the thrust fluctuates slightly. On the one hand, this is due to the interference of the MSV opening moment on the engine state. On the other hand, the thrust generated by the core engine decreases significantly due to the decrease in engine core speed. The afterburner fuel is unable to replenish the decreased thrust in time due to the adjustment step limit. Similar reasons cause the thrust to fluctuate slightly at the end of the conversion. However, the overall fluctuation amplitude of the thrust is less than 0.09% (ΔF<0.09%), meeting the requirement of basically constant thrust during the mode conversion process.

[0123] 2) The fluctuation amplitude of the normal shock wave position is less than 0.83% (Δx A <0.83%), and is always located at the optimal position x A,opt Therefore, the conversion process ensures the constancy of the engine demand flow and a good entry-exit matching relationship.

[0124] 3) Because the present invention maintains the good forward and backward matching relationship found at the start of the mode transition during the mode transition process, the total temperature T7 at the afterburner outlet remains within a reasonable range and does not reach the temperature limit. At its highest point, T7 still retains a temperature margin of nearly 10K.

[0125] 3. Comparison of modal conversion control effects

[0126] Next, the modal conversion control effects of the present invention's modal conversion control plan that considers the forward / backward state matching (referred to as the new control plan) and the conventional modal conversion control plan that does not consider the forward / backward state matching (referred to as the conventional control plan) are compared. The comparison results are shown in Figure 2. Figure 9 shown.

[0127] From the comparison results, it can be seen that under the two control plans, the thrust will not fluctuate significantly during the mode conversion process, and the engine state parameters, including speed, temperature, surge margin and backflow margin, all transition smoothly within a reasonable range, indicating that both methods can ensure the smoothness of the mode conversion process. Figure 9 (e) It can be seen that under the traditional control plan, the total temperature T7 at the afterburner outlet is high at the start of the mode transition, and the reserved temperature margin is small. Therefore, during the transition process, T7 touches the temperature limit boundary many times, which slows down the mode transition process. The transition time is Δt mt =44s. Under the new mode conversion control plan proposed in the present invention, T7 decreases significantly at the conversion starting point, and the position of the positive shock wave is always near the optimal working position during the conversion process, so that the afterburner outlet temperature margin is always large during the conversion process, and the afterburner outlet temperature T7 is always maintained at the limit temperature T during the entire process.7max After optimization, the conversion time is shortened to Δt mt =32s, it can be seen that the new mode conversion control plan proposed in the present invention can effectively improve the mode conversion speed under the premise of smooth mode conversion, achieving the expected effect.

[0128] The above-described embodiments merely illustrate several implementations of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A method for designing a mode conversion process control plan considering inlet port / TBCC engine state matching, characterized by: The steps of the method include Step 1: Based on the sequential quadratic programming (SQP) algorithm, steady-state optimization is performed. The supersonic inlet adjustable variables and the engine adjustable variables are selected as optimization variables. The modal conversion starting point is optimized to obtain the optimal state and proceed to the next step. Step 2: Based on the particle swarm optimization (PSO) algorithm, dynamic optimization is performed to constantly augment the position of the inlet normal shock wave into the optimization objective function, and the modal conversion process is optimized. After obtaining the optimal state, the modal conversion process is completed. The optimal state of the modal conversion starting point obtained in step 1 is the starting state for initialization of the modal conversion process in step 2.

2. The method for designing a mode conversion process control plan considering intake port / TBCC engine state matching according to claim 1, characterized in that: The mode conversion starting point optimization includes selecting the engine adjustment variable u engine and supersonic inlet control variable u inlet As the optimization variable, the engine minimum afterburner outlet total temperature T7 is taken as the optimization target to determine the engine state at the start of the mode transition. The expression is as follows: In formula 1, u is the optimization variable, u=[u inlet ,u engine ]=[β N2 ,L N2 ,A bl ,W fb ,W fa ,A8];W fb is the main fuel flow, W fa is the afterburner fuel flow, A8 is the tail nozzle throat area, β N2 The angle of the inclined plate can be adjusted in two levels, L N2 A is the length of the secondary adjustable inclined plate. bl is the deflation area of ​​the auxiliary deflation valve, F is the thrust, n L is the fan speed, n H is the compressor speed, T4 is the total temperature at the combustion chamber outlet, S mL is the fan surge margin, S mH is the compressor surge margin, RM is the return flow margin, x A is the position of the normal shock wave; the subscript origin represents the corresponding value of the parameter before optimization; the subscripts max and min represent the maximum and minimum allowable values ​​of the corresponding parameters.

3. The method for designing a mode conversion process control plan considering intake port / TBCC engine state matching according to claim 1, characterized in that: Modal conversion starting point optimization also includes steady-state optimization based on sequential quadratic programming SQP algorithm, and BFGS algorithm in quasi-Newton method for B k Perform correction; the specific steps are: Step 1.1: Initialize parameters: Initialize the objective function f(x), the constraint function g(x)≥0, h(x)=0; select the initial feasible point x0, satisfying g(x0)≥0, h(x0=0; initialize the Hessian inverse matrix B0=I, where I is the identity matrix, and initialize the Lagrange multiplier λ0=0, μ0=0; Step 1.2: Solve the QP subproblem, k-th search, k ≥ 0: At the current point x k , solve the following quadratic programming problem: Where d is the search direction, B is the Hessian approximation matrix estimated by the BFGS quasi-Newton method, f(x) is the objective function, and g is i (x) is the inequality constraint, h j (x) is an equality constraint; Represents the gradient of the corresponding function; is the index set of inequality constraints, and ε is the index set of equality constraints; the Lagrange multiplier method is used to solve and output the current step search direction d k and the Lagrange multiplier vector λ of the inequality constraints k+1 and the Lagrange multiplier vector μ for the equality constraints k+1 ; Step 1.3: Update parameters x k+1 =x k +α k d k Formula 3 where α k Determined by line search, default α k =1; Step 1.4: Update the Hessian approximation matrix B using the BFGS method k : Calculate parameter change s k and gradient change y k : The Lagrangian function is: Update B k : Step 1.5: Convergence determination: If || d k ||<∈ or the maximum number of iterations is reached, terminate and output x k ; Otherwise, let k = k + 1 and return to step 1.2 to continue iterating.

4. The method for designing a mode conversion process control plan considering intake port / TBCC engine state matching according to claim 2, characterized in that: In order to ensure that the inlet and engine states match during the mode conversion process, the optimization objective function includes the inlet positive shock wave position x A , the inlet normal shock wave position x under mild supercritical state A,opt To ensure the conversion speed, the optimization objective function also includes the fan speed n L , slow speed n Lidle , the specific expression is In formula 7, F before is the thrust obtained by optimizing the starting point of the modal conversion, u is the optimization variable, and ω1, ω2, and ω3 are the weight coefficients of the objective function.

5. The method for designing a mode conversion process control plan considering intake port / TBCC engine state matching according to claim 1 or 4, characterized in that: The constraint boundary function set during the modal conversion process is formula 8: In formula 8, far4 is the fuel-air ratio of the main combustion chamber, n L is the fan speed, n Lidle is the slow speed, T7 is the total temperature at the ramjet combustion chamber outlet, S mL is the fan surge margin, S mH is the compressor surge margin, RM is the return flow margin, x A is the normal shock wave position, and |du| is the adjustment amount of the actuator within a single step.

6. The method for designing a mode conversion process control plan considering intake port / TBCC engine state matching according to claim 1, characterized in that: The steps of the particle swarm algorithm are as follows: Assuming that the search space is D-dimensional, then the i-th particle X in the particle swarm is i Expressed as, X i =(x i,1 ,x i,2 ,L,x i,D ) T Formula 9 The particle group X is a population consisting of n particles. X=(X1,X2,L,X n ) Formula 10 In the population, each particle is a possible solution to the problem, and X is calculated according to the objective function. i The fitness of the i-th particle, the update speed V i for, V i =(v i,1 ,v i,2 ,L,v i,D ) T Formula 11 The optimal solution extreme values ​​of individuals and populations are, Where, P i is the individual's optimal solution extreme value, P g is the optimal solution extreme value of the population; After the population is generated, particles update their speed and position according to individual extreme values ​​and group extreme values. In the kth iteration, Where w is the inertia weight, d = 1, 2, ..., D, i = 1, 2, ..., n, c1, c2 are acceleration factors, and r1, r2 are random numbers.

Images