On-board online performance analysis and planning method for turbine engine transient state

Abstract

The present invention discloses an onboard online performance analysis and planning method for a turbine engine transition state, and relates to the technical field of aircraft engines. The method comprises: establishing an onboard engine system, which is capable of converting acceleration control instructions into control instructions for fuel quantity and tail nozzle area. For the engine system in the transition state, the state equation is discretized. The present application uses acceleration as an optimization variable to design a transition state control plan. The horizontal coordinate of the control plan is time, and the vertical coordinates are the accelerations of the high and low pressure rotors, respectively. The present method has the following advantages: achieving direct decoupling of the acceleration of the dual rotors; and being able to control the consistency of the dual rotor acceleration process by controlling the integral area.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft engines, and specifically to an on-board online performance analysis and planning method for the transition state of a turbine engine. Background Art

[0002] During the working states of a fighter taking off, landing, accelerating or decelerating in the air, the engine is in a dynamic process of state switching. The state parameters are likely to exceed the safety limits, reducing the service life of the engine and affecting the working safety. Therefore, it is necessary to reasonably design a transition state control plan to reduce the transition state operation time under the constraints of aerodynamic load, thermal load and mechanical load. After the control plan curve is reduced in dependence on operating conditions through similarity conversion, the formed control law can be applied to the transition state control of the entire envelope. Therefore, how to design the transition state control plan is the focus of research.

[0003] A large amount of research has been done on the method for establishing the engine transition state control plan at home and abroad. In 1958, Gerus first established a control law curve of acceleration with respect to rotor speed. In 1975, Merrill added a fuel compensation loop on the basis of the steady-state fuel control plan and set a fuel quantity limiter to ensure that the turbine temperature and the stability constraints of the compressor would not be exceeded. In 1982, Howlett developed an adaptive acceleration plan that automatically compensates for the loss of surge margin and can adjust the pre-programmed acceleration online to avoid surge.

[0004] The current methods for establishing the transition state control plan are still imperfect and have poor real-time performance, making it difficult to apply them to the online generation of on-board control plans. The power extraction method can design the control plan for a single-shaft engine, but it cannot handle the coupling effect of the rotors of a mixed-exhaust turbofan engine and coordinate the remaining power of a two-shaft engine, thus making it difficult to design the control plan forward. The control plan design method based on the dynamic stability method also faces the problem of the acceleration coupling of the two rotors. It is necessary to continuously establish the dynamic coupling relationship of acceleration during the design process to carry out the forward design of the control plan, which is very difficult. The control plan design method based on the dynamic model to determine the state requires pre-establishing a nozzle area adjustment plan, sacrificing the advantage of multi-variable coordinated adjustment. Moreover, the adjustment plan of a single fuel quantity cannot coordinate the acceleration process of the two rotors, which extends the acceleration time to a certain extent. The fuel quantity and tail nozzle control plan designed using the optimization algorithm, with the maximization of the remaining power of a single shaft as the optimization goal, also cannot plan the acceleration process of the two shafts, optimize the transition state time as a whole, and needs to solve the multi-variable optimization problem with boundary constraints step by step based on the dynamic model. The solution process is relatively complex, restricting its engineering application.

[0005] The coupling relationship of the dual-rotor acceleration is the difficulty restricting the forward design of the transient control plan for the mixed-flow turbofan engine. The current design method aims to maximize the single-shaft surplus power, which can only ensure that the single shaft reaches the target speed fastest, while the other shaft is in an uncontrolled state and takes longer to stabilize to the target speed at the end of the acceleration process. Therefore, the optimization goal of maximizing the single-shaft surplus power cannot guarantee the optimal acceleration time. A more reasonable optimization goal is to reduce the total operating time in the transient state on the premise of ensuring that the dual rotors reach the target stable speed simultaneously. However, the current design method of the control plan with the control quantity as the optimization variable is difficult to decouple the relationship between the control quantity and the dual-rotor acceleration, so it is impossible to coordinate the acceleration process of the dual rotors, and there is room for further optimization of the acceleration time.

[0006] Based on this, a method for on-board online performance analysis and planning of the transient state of a turbine engine is now provided, which can eliminate the drawbacks of the existing technology. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for on-board online performance analysis and planning of the transient state of a turbine engine to solve the problems in the background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A method for on-board online performance analysis and planning of the transient state of a turbine engine, including: establishing an online control system that can convert the acceleration control instruction into control instructions for the fuel quantity and the nozzle area for use by the actual engine. For the engine system in the transient state, discretize its state equation as shown in Equation (1), where x = [N L , N H , y = [Tt 3 , Tt 4 , Pt 3 , Pt 4 , SM Fan , SM HPC , u = [W f , A 8 , and k represents the kth discrete time:

[0010]

[0011] Assume that the acceleration instruction at time k is It is necessary to solve u(k) and x(k) so that Since x(k) and u(k) are nonlinearly constrained by the common working equation, it is difficult to directly determine u(k). Therefore, the time interval can be reduced, and x(k - 1) can be used to replace x(k). x(k - 1) is determined by the engine monitoring values. The system state equation is transformed into Equation (2). Drawing on the idea of the dynamic stability method, only the dual-rotor acceleration residual quantity needs to be added to the common equation, and the fuel quantity and nozzle area are iterated to turn the dynamic state into a static state and solve for u(k).

[0012]

[0013] After the pilot pushes the throttle lever to demand thrust, the system searches for a steady-state operating point that meets the thrust demand and has the optimal comprehensive performance. Based on the nonlinear adaptive engine model, a dual-rotor acceleration control plan is generated. The command is updated at a certain period during the transient state. Based on the dynamic stability method, the control quantity is calculated from the acceleration command value and the rotational speed monitoring value. Considering the deviation between the model and the actual situation, an additional PI control loop is added to adjust the fuel quantity to ensure that the high-pressure rotor acceleration follows the control command.

[0014] Based on the above technical solutions, the present invention also provides the following alternative technical solutions:

[0015] In the alternative solution: The solution model of the dynamic stability method is used not only to solve the control quantity in the online control system but also to design the acceleration control plan. At the transient state k, the dual-rotor acceleration command is The integral area of the control plan can determine the state quantity [N L , N H , the iterative variables of the common working equation are [W, α, β Fan , β HPC , π HPT , π LPT , A 8 , W f , the residual quantity is as shown in Equation (3), and the Newton-Simpson algorithm can be used to solve the common working equation and determine the system state;

[0016]

[0017] In the alternative solution: The method for generating a control plan with equal rotational speed increments:

[0018] The rotational speed increase amount during the acceleration process is evenly divided into m segments. Within each segment, the rotational speeds of the high-pressure and low-pressure rotors increase by Corresponding to the acceleration control plan, each acceleration plan and the abscissa form a trapezoid, and its integral area is the rotational speed increment of each segment. The abscissa span is the duration of each segment, as Figure 2As shown in the figure; the piece-by-piece design of the control plan can be carried out. By adjusting the final acceleration value of each section, the transient performance can be adjusted, and the abscissa span of the double rotors can be controlled to be the same to keep the acceleration time of the double rotors consistent;

[0019] To illustrate that the design variable of each section is the final acceleration of the high-pressure rotor in this design section, the accelerations of the high- and low-pressure rotors at the end of the kth design section are denoted as It can be determined by as shown in Equations (4) and (5).

[0020]

[0021]

[0022] At the end of the kth design section, the rotational speeds of the high- and low-pressure rotors are determined by Equations (6) and (7).

[0023]

[0024]

[0025] After the rotational speed and acceleration in the kth section are determined, the state of the system at time k can be determined by the dynamic stability method. Among them, N L，k , N H，k are both determined values, It is determined by . Therefore, the final state of the kth design section is determined by as shown in Equation (8).

[0026]

[0027] When the rotational speed increment of the rotor is small enough, y k varies monotonically in each design section. Therefore, as long as y k-1 and y k meet the boundary constraints, it can be ensured that the design section meets the boundary constraints. Designing piece by piece from the starting acceleration state, a control plan that meets the boundary constraints and has the optimal acceleration time can be generated. The boundary constraints are shown in Equation (9).

[0028]

[0029] In the alternative solutions: to correct the part of the acceleration plan that exceeds the boundary constraint limit, assume that state 1 reaches a certain constraint limit and state 2 exceeds the boundary constraint limit. The rotational speed increment of the double rotors from state 1 to state 2 is a fixed value, and the state equation can be listed as shown in Equation (10):

[0030]

[0031] It can be seen from Equation (10) that after the rotational speed increment of the double rotors is determined, y2 Determined by u 2 Under the premise of ensuring equal duration of the dual-rotor within the process, adjust the acceleration value of the medium and high-pressure rotor, then y 2 can meet the boundary constraints. According to the principle of residual power, the change rate of the monitored quantity y has a positive relationship with the change rate of the dual-rotor acceleration. After correcting the curve trend of the dual-rotor acceleration control plan, it basically ensures that y meets the boundary constraints.

[0032] In the alternative solution: The design process of the phased control plan is as follows:

[0033] Step 1: To keep A 8 unchanged, correct the solution model based on the dynamic stability method, reduce the acceleration residual of the low-pressure rotor and the A 8 iteration variable. First, adjust the acceleration slope of the high-pressure rotor to meet Then adjust the end state of the first stage to meet SM HPC = SM HPC，max ;

[0034] Step 2: Keep A 8 and unchanged and extend for a certain time. The fuel quantity will continue to rise, and at the same time, the acceleration of the low-pressure rotor will rise and remain until it meets

[0035] Step 3: Make the initial acceleration of the dual-rotor in stage 5 equal to the end-state acceleration of stage 2. Based on the solution model of the dynamic stability method, under the condition of meeting , reduce the common duration of the dual-rotor;

[0036] Step 4: Calculate the duration of the yellow dashed line state based on the rotor speed difference between the initial state of stage 5 and the end state of stage 2. The previous control plan design steps can make the duration of the dual-rotor in the yellow dashed line state the same;

[0037] Step 5: Based on the solution model of the dynamic stability method, simulate stage K and stage 5. During this period, Tt 4 , SM HPC will have overshoot and fall back. Locate the critical moment when Tt 4 = Tt 4，max , SM HPC = SM HPC，max , and calculate the dual-rotor speed increment in the overshoot stage of the first overshoot state based on the integral area;

[0038] Step 6: Modify the acceleration plan of State 1 within the overshoot range, calculate the Δt of Stage K after modification, and the modification method and control plan design method can satisfy the equal duration of the dual rotors within Stage K. Based on the dynamic stability method, a solution model is obtained, and the critical point where State 2 reaches the boundary constraint within Stage K is recalculated.

[0039] Step 7: Adopt the same control plan modification method as in Step 6, calculate Δt from the rotational speed increment of State 2 within the overshoot critical point, and finally generate a control plan that satisfies the boundary constraint.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] The present invention takes the dual-rotor acceleration as the optimization variable and designs a transient control plan. The abscissa of the control plan is time, and the ordinates are the high- and low-pressure rotor accelerations respectively. This method has the following advantages: realizing the direct decoupling of the dual-rotor accelerations; being able to control the consistency of the dual-rotor acceleration process by controlling the integral area; due to the direct correlation between acceleration and remaining power, being able to coordinately adjust the dual-rotor acceleration plan to regulate the transient performance; the system state at a certain moment in the transient state can be directly determined by the ordinate and the integral area, avoiding calling the dynamic model for time-step integration one by one, reducing the number of non-linear model solutions, and improving the real-time performance of generating the control plan.

[0042] The control plan establishment method proposed by the present invention aims to be applied to on-board online establishment of the transient control plan. Therefore, it needs to meet engineering practicability and calculation real-time performance. The core difficulty lies in the handling of boundary constraints and the coordinated planning of the dual-rotor acceleration process. Brief Description of the Drawings

[0043] Figure 1 It is the principle block diagram of the engine system in the transient state of the present invention.

[0044] Figure 2 It is the control chart of the rotational speed increment of the present invention.

[0045] Figure 3a It is the schematic diagram of the acceleration parameter structure of the present invention.

[0046] Figure 3b It is the schematic diagram of the thermal parameters of the present invention.

[0047] Figure 3c It is the schematic diagram of the surge margin of the present invention.

[0048] Figure 3d It is the schematic diagram of the control variables of the present invention.

[0049] Figure 4 It is the schematic diagram of the constraint-based control progress division of the present invention.

[0050] Figure 5 Schematic diagram of progress modification for constraint-based acceleration control according to the present invention.

[0051] Figure 6 Process of stage control progress design according to the present invention.

[0052] Figure 7a Schematic diagram of acceleration control plan for Example 1 according to the present invention.

[0053] Figure 7b Schematic diagram of thermal parameters for Example 1 according to the present invention.

[0054] Figure 7c Schematic diagram of surge margin for Example 1 according to the present invention.

[0055] Figure 7d Schematic diagram of control variables for Example 1 according to the present invention.

[0056] Figure 8a Control variable w for Example 2 according to the present invention f Schematic diagram.

[0057] Figure 8b Control variable A for Example 2 according to the present invention 8 Schematic diagram.

[0058] Figure 8c Schematic diagram of HPC surge margin for Example 2 according to the present invention.

[0059] Figure 8d Schematic diagram of thermal parameters for Example 2 according to the present invention.

[0060] Figure 8e Schematic diagram of shaft speed for Example 2 according to the present invention Detailed implementation manners

[0061] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0062] In one embodiment, as Figure 1 - shown in Figure 3, for the method of on-board online performance analysis and planning of a turbofan engine during the transient state, an online control system needs to be established, which can convert the acceleration control command into control commands for fuel quantity and nozzle area for use in a real engine. For the engine system in the transient state, its state equation is discretized as shown in Equation (1). Where x = [N L , N H , y = [Tt 3 , Tt 4 , Pt 3 , Pt 4 , SM Fan , SM HPC, u = [W f , A 8 , where k represents the k-th discrete time instant.

[0063]

[0064] Assume that the acceleration command at time k is It is necessary to solve for u(k) and x(k) such that Since x(k) and u(k) are nonlinearly constrained by the common working equations, it is difficult to directly determine u(k). Therefore, the time interval can be reduced, and x(k - 1) is used to replace x(k), where x(k - 1) is determined by the engine monitoring values. The system state equation is transformed into Equation (2). Drawing on the idea of the dynamic stability method, it is only necessary to incorporate the dual-rotor acceleration residual in the common equation, and iterate the fuel quantity and nozzle area to turn the dynamic problem into a static one and solve for u(k).

[0065]

[0066] The control system architecture is shown in Figure (1). After the pilot pushes the throttle lever to demand thrust, the system searches for a steady-state operating point that meets the thrust demand and has the optimal comprehensive performance. Based on the nonlinear adaptive engine model, a dual-rotor acceleration control plan is generated and the command is updated at a certain period during the transient state. Based on the dynamic stability method, the control quantity is calculated from the acceleration command value and the rotational speed monitoring value. Considering the deviation between the model and the actual situation, an additional PI control loop is added to adjust the fuel quantity to ensure that the high-pressure rotor acceleration follows the control command. The engine dynamic simulation model is used to replace the real engine to verify the effectiveness of the control plan;

[0067] The solution model of the dynamic stability method is used both in the online control system to solve for the control quantity and in the design of the acceleration control plan. At the transient time instant k, the dual-rotor acceleration command is The integrated area of the control plan can determine the state quantity [N L , N H , the iterative variables of the common working equation are [W, α, β Fan , β HPC , π HPT , π LPT , A 8 , W f , the residual is shown in Equation (3), and the Newton-Simpson algorithm can be used to solve the common working equation to determine the system state.

[0068]

[0069] Dual-rotor acceleration plan generation method: It is necessary to study the generation method of the transition-state dual-rotor acceleration control plan to reduce the acceleration time while satisfying the maximum boundary constraints. Therefore, it is necessary to plan the rotational speed increments of the dual rotors throughout the process to ensure that the dual rotors reach the target rotational speed stably at the end of the acceleration. Two control plan design methods are proposed in this paper, and the one with better real-time performance is selected for the online generation of the airborne control plan;

[0070] Control plan generation method with equal rotational speed increments: The starting rotational speed of the transition-state rotor [N L，initial , N H，initial , and the target rotational speed [N L，final , N H，final . For the control plan generation method with equal rotational speed increments, the rotational speed increase during the acceleration process is evenly divided into m segments. The rotational speeds of the high- and low-pressure rotors increase by respectively in each segment. Corresponding to the acceleration control plan, each acceleration plan and the abscissa form a trapezoid, and the integral area is the rotational speed increment of each segment. The abscissa span is the duration of each segment, as shown in Figure 2 . The control plan can be designed segment by segment. By adjusting the final-state acceleration value of each segment, the transition-state performance can be adjusted, and the abscissa spans of the dual rotors are controlled to be the same to keep the acceleration times of the dual rotors consistent;

[0071] To illustrate that the design variable of each segment is the acceleration of the high-pressure rotor at the end of the design segment, let the accelerations of the high- and low-pressure rotors at the end of the kth design segment be respectively, which can be determined by , as shown in Eqs. (4) and (5).

[0072]

[0073]

[0074] At the end of the kth design segment, the rotational speeds of the high- and low-pressure rotors are determined by Eqs. (6) and (7).

[0075]

[0076]

[0077] After the rotational speed and acceleration in the kth segment are determined, the state of the system at time k can be determined by the dynamic stability method. Among them, N L，k N H，k are all determined values, which is determined by . Therefore, the final-state state of the kth design segment is determined by , as shown in Eq. (8).

[0078]

[0079] When the rotational speed increment of the rotor is small enough, y k varies monotonically in each design segment. Therefore, it is only necessary to ensure that y k-1 and y k satisfy the boundary constraints, then it can be ensured that the design segment meets the boundary constraints. Designing segment by segment from the acceleration starting state can generate a control plan that meets the boundary constraints and has the optimal acceleration time. The boundary constraints are shown in Equation (9);

[0080]

[0081] This method has two advantages: First, co-design the rotational speed increments of the dual rotors to ensure that the dual rotors reach the target rotational speed stably at the same time; Second, the division of intervals only needs to ensure the monotonicity of y within the interval. According to the principle of residual power, when the rotational speed is constant, there is a positive relationship between y and the acceleration of the dual rotors. Under the requirement of coordinated rotational speed increments, the acceleration of the dual rotors is affected by control. Therefore, the change trend of y in each design segment has a positive relationship with the linear change trend. Therefore, only by ensuring the monotonicity of y within the design interval, the rotational speed increment within each design interval can be increased, and the number of calls to the nonlinear model can be reduced. In contrast, for the control plan design method based on the open-loop control quantity, the value of y depends on the step-by-step integration of the dynamic model in time. To ensure the integration accuracy, the upper limit of the time interval is low, and it is impossible to further reduce the number of calls to the nonlinear model and improve the real-time performance.

[0082] This method has strong versatility and reasonable acceleration process coordination, but still faces problems such as acceleration interval division, boundary constraint switching, multiple calls to the nonlinear model, and constrained optimization, which cause certain difficulties in developing an automatic control plan generation program. Based on this method, this paper proposes a second method that better meets the engineering practicality;

[0083] Control plan generation method with staged optimization: To reduce the number of calls to the nonlinear model and the difficulty of compiling the automatic design software, it is necessary to further improve the control plan design method. A better approach is to first establish an ideal initial control plan, simulate the transient process based on the fixed state method, divide the control plan according to the types of constraint boundaries, and correct the dual rotor control plan for the part where the parameters exceed the limit. It is found in actual simulation that there is a significant connection between the curve trend of the acceleration control plan and the trend of the monitored parameters. Taking the dynamic simulation under a certain initial plan as an example, Table 1 lists the key parameters of the acceleration process;

[0084] Table 1Main parameters of acceleration

[0085]

[0086] The initial control plan does not consider boundary constraints. It is trapezoidal in shape. The acceleration rise and fall times of both rotors are 0.5 s, and the total duration is the same. The maximum acceleration value is calculated from the rotational speed difference between the initial and final states of the dual rotors and the total duration. Figure 3 shows the initial acceleration plan and the variations of thermodynamic and aerodynamic parameters;

[0087] As can be seen from Figure 3, there is a significant connection between the parameter variations and the trend of the acceleration plan graph. In the initial stage, the acceleration of the dual rotors rises sharply, and the variation rates of W f and A 8 are large. At the same time, to meet the acceleration requirement of the low-pressure rotor, A 8 increases to increase the remaining power of the low-pressure rotor. In the constant acceleration stage of the dual rotors, since the fuel quantity at the initial and final moments deviates the most from the steady-state fuel quantity corresponding to this state, the surge margin of the high-pressure compressor at the initial and final moments reaches an extreme value. Analyzing the variations of the thermodynamic parameters, it can be seen that except for Tt 4 , the overshoots of other thermodynamic parameters are relatively small because Tt 4 is directly affected by the overshoot of the fuel quantity. At the end state moment of the constant acceleration stage of the dual rotors, the calculated value of the fuel quantity reaches a peak and far exceeds the steady-state fuel quantity corresponding to this state, resulting in a large overshoot of Tt 4 . At the end of the acceleration process, the acceleration of the dual rotors returns to zero. The variation rates of the fuel quantity and the nozzle area are higher than those in the constant acceleration process of the dual rotors, and all thermodynamic parameters return to the target steady state;

[0088] Therefore, the boundary constraints can be determined according to the parameter variation characteristics of the acceleration process, and the acceleration plan can be divided according to the boundary constraints, as Figure 4 shown. In stage 1, it is necessary to keep the nozzle area unchanged, control the fuel variation rate to meet the boundary constraints, and the surge margin of the high-pressure compressor at the end state meets the boundary constraints. In stage 2, continue to keep the nozzle area unchanged, maintain the acceleration of the high-pressure rotor unchanged, and the acceleration of the low-pressure rotor continues to rise until the remaining rotational speed increments of the dual rotors are coordinated. In stage 3, keep the accelerations of the high- and low-pressure rotors unchanged, then the surge margin of the high-pressure compressor remains basically unchanged. In stage 4, adjust the acceleration plan curve of the dual rotors so that the thermodynamic parameters meet the boundary limits. In stage 5, control the variation rates of the fuel quantity and the nozzle area to meet the boundary constraints;

[0089] To correct the part of the acceleration plan that exceeds the boundary constraint limits, assume that state 1 reaches a certain constraint limit and state 2 exceeds the boundary constraint limits. The rotational speed increment of the dual rotors from state 1 to state 2 is a fixed value, and the state equation can be listed as shown in Equation (10).

[0090]

[0091] As can be seen from Equation (10), after the rotational speed increment of the dual rotors is determined, y 2 is determined by u2 It is decided that, on the premise of ensuring that the duration of the double rotor in the process is equal, adjust the acceleration value of the medium-high pressure rotor, so that y 2 meets the boundary constraints, such as Figure 5 shown. According to the principle of residual power, there is a positive relationship between the change rate of the monitored quantity y and the change rate of the double rotor acceleration. After correcting the trend of the double rotor acceleration control plan graph, it is basically ensured that y meets the boundary constraints.

[0092] The design process of the phased control plan is as Figure 6 shown:

[0093] Step 1: To keep A 8 unchanged, correct the solution model based on the dynamic stability method to reduce the acceleration residual of the low-pressure rotor and the A 8 iteration variable. First, adjust the acceleration slope of the high-pressure rotor to meet Then adjust the end state of the first stage to meet SM HPC = SM HPC，max .

[0094] Step 2: Keep A 8 and unchanged and extend for a certain time. The fuel quantity will continue to rise, and at the same time, the acceleration of the low-pressure rotor will rise and remain until it meets

[0095] Step 3: Make the initial acceleration of the double rotor in stage 5 equal to the end-state acceleration of stage 2. Based on the solution model of the dynamic stability method, under the condition of meeting reduce the common duration of the double rotor.

[0096] Step 4: Calculate the duration of the yellow dotted line state based on the difference in rotor speeds between the initial state of stage 5 and the end state of stage 2. The previous control plan design steps can make the duration of the double rotor in the yellow dotted line state the same.

[0097] Step 5: Based on the solution model of the dynamic stability method, simulate stage K and stage 5. During this period, Tt 4 , SM HPC will have overshoot and fall back. Locate Tt 4 = Tt 4，max , SM HPC = SM HPC，max of the critical moment, and calculate the double rotor speed increment in the overshoot stage of the first overshoot state based on the integral area.

[0098] Step 6: Correct the acceleration plan of state 1 in the overshoot interval. The correction method is as Figure 5As shown, calculate the Δt of the corrected Stage K. The correction method and the control plan design method can ensure that the durations of the two rotors in Stage K are equal. Based on the dynamic stability method, a solution model is obtained, and the critical point where the state 2 reaches the boundary constraint within Stage K is recalculated.

[0099] Step 7: Use the same control plan correction method as in Step 6 to calculate Δt from the rotational speed increment of state 2 within the overshoot critical point, and finally generate a control plan that meets the boundary constraints.

[0100] Conclusion:

[0101] Example 1

[0102] After specifying the overshoot limit boundary of the thermal parameters and the surge margin, based on the phased acceleration control plan design method, the generated acceleration control plan and the parameter changes during the transient process are shown in Figure 7:

[0103] As can be seen from Figure 7, the acceleration control plan design method proposed in this paper can keep the engine system parameters within the boundary constraints, and both the fuel quantity and the nozzle area have a smooth transition. Among them, the acceleration of the two rotors in Stage 3 of the control plan remains unchanged, and the surge margin is slightly higher than the boundary constraint, which can be used for reference. Figure 5 The proposed control plan correction method can be used to coordinately correct the acceleration of the two rotors corresponding to the maximum value of the surge margin during this process to further reduce the acceleration time. However, considering the real-time requirement for generating the on-board control plan, the control plan for this process is not further optimized in this paper.

[0104] As shown in Figures 8(a) and 8(b), the nozzle area remains unchanged in the OA section under both acceleration plans. Limited by the maximum fuel change rate, the AB section is restricted by the maximum surge margin boundary constraint, and the BC section is restricted by the combustion chamber outlet temperature. The difference lies in the CD section. For the control plan based on open-loop fuel, the high-pressure rotor has reached the target speed at point C, so the fuel quantity needs to be adjusted to keep the high-pressure rotor speed constant, and the process constraint is the high-pressure rotor speed. For the phased acceleration control plan, neither rotor has reached the target speed in the CD section, and the process constraint is the change rate of W f ,A 8 ;

[0105] Table 2 Comparison of two acceleration control parameters

[0106]

[0107] The comparison results of the acceleration process are sorted into Table 2. Method 1 is the control plan design method with a fixed state, and Method 2 is the phased control plan design method proposed in this paper. In terms of acceleration time, for the control plan based on open-loop fuel, the duration of the OC section is 5 s, and the duration of the CD section is 1.05 s. For the acceleration control plan, the duration of the OC section is 5.2 s, and the duration of the CD section is 0.3 s. Therefore, the total time of the transient process based on the acceleration plan can be saved by 0.55 s, accounting for 10% of the total time. In terms of overshoot, this method reduces the overshoot of the acceleration process due to the planned acceleration process.

[0108] The reason for the reduction of the acceleration time by the method in this paper is that for the control plan based on the open-loop fuel quantity, at the CD section, the nozzle area reaches the target area, but the fuel quantity exceeds the target steady-state fuel quantity, and the fuel quantity needs to be adjusted to control the overshoot of the high-pressure rotor speed. Limited by the change rate of the fuel quantity, the high-pressure rotor speed cannot be fixed during this process, and oscillation will occur. Therefore, the fuel quantity will also overshoot and callback, prolonging the acceleration time. At the same time, since the open-loop fuel control plan cannot control the speeds of the two rotors simultaneously, even when the fuel quantity at D reaches the target steady state, the deviation of the low-pressure rotor speed accounts for 2% of the total speed increment, and it is necessary to continue to extend the time by 1 s to maintain an accuracy of 1%. Therefore, for the control law design method based on the open-loop fuel quantity, due to the inability to coordinate the speed increments of the two rotors and avoid the problem of overshoot of the control variables, its acceleration time is not the optimal solution. For the control plan based on acceleration, when approaching the target state, it can adjust the two control variables in advance, control the speeds of the two rotors to be stable simultaneously, and further optimize the acceleration time.

[0109] In summary, this paper studies the method for establishing the engine transient control plan, proposes a method for establishing a phased acceleration control plan, and the following conclusions can be obtained through simulation research:

[0110] (1) The control plan proposed in this paper coordinates the acceleration process of the dual rotors of the mixed-exhaust turbofan engine and further optimizes the acceleration time.

[0111] (2) The control plan proposed in this paper, by finding the overshoot critical point and correcting the control plan within the collaborative correction critical point, not only satisfies the boundary constraints but also avoids solving the constrained optimization problem, meeting the requirements of engineering practicability.

[0112] (3) Based on the solution model of the dynamic stability method, instead of using the dynamic simulation model for control plan design, the number of calls to the nonlinear model is reduced, improving the calculation real-time performance.

[0113] Based on the method proposed in this paper, it is expected to establish an automatic control plan calculation system for online calculation of the airborne transient control plan.

[0114] As described above, it is only the specific implementation manner of the present disclosure. However, the protection scope of the present disclosure is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present disclosure can easily think of changes or substitutions, which should all be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.

Claims

1. A method for on-board online performance analysis and planning of a turbine engine in transition state, characterized in that: include: An online control system is established to convert the acceleration control command into the control command of the fuel quantity and the tail nozzle area. For the engine system in the transition state, its state equation is discretized as shown in formula (1), where x = [N L ,N H ],y=[Tt3,Tt4,Pt3,Pt4,SM Fan ,SM HPC ],u=[W f , A8], k represents the kth discrete moment: Assume that the acceleration command at time k is We need to solve u(k) and x(k) so that Since x(k) and u(k) are subject to nonlinear constraints of the common working equation, it is difficult to directly determine u(k). Therefore, the time interval can be reduced and x(k) can be replaced by x(k-1). x(k-1) is determined by the engine monitoring value. The system state equation is transformed into equation (2). Drawing on the idea of ​​the dynamic stability method, it is only necessary to include the rotor acceleration residual in the common equation, iterate the fuel amount and the nozzle area, and then the dynamic can be turned into static to solve u(k). After the pilot pushes the throttle lever to request thrust, the system searches for a steady-state operating point that meets the thrust requirement and has the best overall performance. Based on the nonlinear adaptive overall performance model, a control plan is generated. In the transition state, the command is updated at a certain period. Based on the dynamic stability method, the control quantity is calculated from the acceleration command value and the speed monitoring value. Considering the deviation between the model and the actual situation, a PI control loop is added to adjust the fuel quantity to ensure that the high-pressure rotor acceleration follows the control command. The solution model of the dynamic stability method is used to solve the control quantity in the online control system; at the transition state k moment, the acceleration command is The control plan integral area can determine the state quantity [N L , N H ], the iterative variables of the common working equation are [W, α, β Fan , β HPC , π HPT , π LPT , A8, W f ], the residual is shown in formula (3), and the Newton-Simpson algorithm can be used to solve the common working equation and determine the system state; 2. The method for on-board online performance analysis and planning of a turbine engine in transient state according to claim 1, characterized in that: Method for generating a control plan with constant speed increments: The speed increase during the acceleration process is divided into m sections, and the speed of the high and low pressure rotors in each section increases respectively. Corresponding to the acceleration control plan, each acceleration plan and the horizontal axis form a trapezoid, the integral area is the speed increment of each section, and the horizontal axis span is the duration of each section. The control plan can be designed section by section, and the transition state performance can be adjusted by adjusting the final state acceleration value of each section. The horizontal axis span of the dual rotors is controlled to be the same to keep the acceleration time of the dual rotors consistent; To illustrate that the design variables of each section are the high-pressure rotor acceleration at the end of the design section, the high-pressure and low-pressure rotor accelerations at the end of the kth design section are can be Determine, as shown in formula (4) and (5); At the end of the kth design stage, the high and low pressure rotor speeds are determined by equations (6) and (7); After the speed and acceleration of the kth segment are determined, the state of the system at time k can be determined by the dynamic stability method, where N L,k , N H,k are all definite values, Depend on Therefore, the final state of the kth design segment is determined by Determine, as shown in formula (8); When the rotor speed increment is small enough, y k In each design segment, it changes monotonically, so we only need to ensure that y k-1 With y k Satisfying the boundary constraints can ensure that the design segment satisfies the boundary constraints; designing segment by segment from the acceleration start state can generate a control plan that satisfies the boundary constraints and has the optimal acceleration time; the boundary constraints are shown in formula (9); 3. The method for on-board online performance analysis and planning of a turbine engine in transient state according to claim 1, characterized in that ； In order to correct the part of the acceleration plan that exceeds the boundary constraint limit, it is assumed that state 1 reaches a certain constraint limit, state 2 exceeds the boundary constraint limit, and the double rotor speed increment from state 1 to state 2 is a fixed value. The state equation can be listed as shown in formula (10): It can be seen from formula (10) that after the dual rotor speed increment is determined, y2 is determined by u2. Under the premise of ensuring that the dual rotor duration is equal in the process, adjust The acceleration value of the medium and high voltage rotors can make y2 meet the boundary constraints. According to the residual power principle, the change rate of the monitored quantity y is positively correlated with the change rate of the dual rotor acceleration. After correcting the graph trend of the dual rotor acceleration control plan, it is basically guaranteed that y meets the boundary constraints.

4. The method for on-board online performance analysis and planning of a turbine engine in transient state according to claim 3, characterized in that: The phased control plan design process is as follows: Step 1: To keep A8 unchanged, modify the solution model based on the dynamic stability method, reduce the low-pressure rotor acceleration residual and A8 iteration variables, and first adjust the high-pressure rotor acceleration slope to meet Then adjust the final state of stage one Meet SM HPC =SM HPC,max ; Step 2: Keep A8 and If the fuel level remains unchanged for a certain period of time, the fuel volume will continue to increase, and the low-pressure rotor acceleration will increase, and this will be maintained until the Step 3: Let the initial acceleration of the dual rotors in stage 5 be equal to the final acceleration in stage 2, and solve the model based on the dynamic stability method. Under the condition of , reduce the duration of the dual rotors together; Step 4: Based on the difference between the rotor speed at the initial state of stage 5 and the rotor speed at the final state of stage 2, calculate the duration of the yellow dashed line state. The previous control plan design steps can make the duration of the two rotors in the yellow dashed line state the same; Step 5: Solving the model based on the dynamic stability method, simulating phases K and 5, during Tt4, SM HPC There will be overshoot and fallback, positioning Tt4 = Tt 4，max , S MHPC =S MHPC，max The critical moment of the overshoot state is calculated based on the integral area, and the double rotor speed increment in the overshoot stage when the overshoot state first occurs is calculated; Step 6: Correct the acceleration plan of state 1 in the overshoot interval, calculate the corrected Δt of stage K, and the correction method and control plan design method can satisfy the equal duration of the dual rotors in stage K. The solution model is obtained based on the dynamic stability method, and the critical point of state 2 reaching the boundary constraint in stage K is recalculated; Step 7: Using the same control plan correction method as step 6, calculate Δt from the speed increment of state 2 within the overshoot critical point, and finally generate a control plan that meets the boundary constraints.

Images