Active-disturbance-rejection control system of variable-cycle engine

Abstract

The invention belongs to the field of aero-engine control, and particularly relates to a variable cycle engine active disturbance rejection control system which comprises a tracking differentiator, an expansion state observer, an error nonlinear feedback unit and a disturbance compensation unit. The tracking differentiator can obtain the target rotating speed or pressure ratio of the engine and generate the tracking trajectory of the target. The extended state observer can collect the state quantity of the engine and the control quantity of the disturbance compensation unit to generate disturbance estimation and state estimation. The method does not depend on an object accurate model, and only needs to estimate the system dynamics online through an extended state observer. Compared with a traditional PID which needs to repeatedly set parameters (such as a proportionality coefficient and integral time) for different working conditions, control parameters (such as the tracking speed of TD, the bandwidth of an extended state observer and the gain of nonlinear feedback) of the algorithm only need to be calibrated based on the typical working conditions of the engine, and the design complexity and the maintenance cost of the controller are greatly reduced.

Description

Technical Field

[0001] This application belongs to the field of aero-engine control, and specifically relates to a variable cycle engine active disturbance rejection control system. Background Technology

[0002] Variable cycle aero-engines are a type of complex multivariable nonlinear system with strong coupling relationships between parameters. Traditional turbofan engines often employ single-variable PID control algorithms. However, due to the changing dominant characteristics of variable cycle engines, a single PID control algorithm is insufficient to meet the engine control performance requirements. This is because the control process of a variable cycle engine needs to consider multiple steady-state control performances, transient performance during multi-mode switching, and smooth transitions between turbojet and turbofan modes.

[0003] The single-variable PID control algorithm is commonly used in aero-engines. The rationale behind classic PID lies in designing a feedback law based on the combined behavior of the past (I), present (P), and future (D) errors, with its control mechanism completely independent of the mathematical model of the object. This is the fundamental reason for the widespread application of PID in process control. The drawback of PID is its simple use of a linear weighted sum of the proportional, derivative, and integral components of the error. This linear configuration makes it difficult to resolve the contradiction between speed and overshoot, leading to difficulties in effectively controlling quality.

[0004] Therefore, ensuring control quality during the transition state and mode transition of a variable cycle engine is a problem that needs to be solved. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a variable cycle engine active disturbance rejection control system to solve the problem of difficulty in controlling quality during the transition state and mode transition processes of existing variable cycle engines.

[0006] The technical solution of this application is: a variable cycle engine active disturbance rejection control system, including a tracking differentiator, an extended state observer, an error nonlinear feedback unit, and a disturbance compensation unit;

[0007] The tracking differentiator can acquire the target speed or pressure ratio of the engine and generate the target tracking trajectory;

[0008] The extended state observer can collect the state variables of the engine and the control variables of the disturbance compensation unit to generate disturbance estimates and state estimates.

[0009] The tracking trajectory and state estimation are calculated to obtain the tracking error, and the error nonlinear feedback unit can generate an error control quantity based on the tracking error.

[0010] The disturbance control quantity is obtained after the error control quantity and disturbance estimation are calculated. The disturbance compensation unit can process the disturbance control quantity, generate a control quantity to control the engine, and send the control quantity to the extended state observer.

[0011] Preferably, the formula for calculating the target tracking trajectory generated by the tracking differentiator is:

[0012] ;

[0013] In the formula, , For process variables, For the input limit value, To track the coefficients of the differentiator, State 1 of the discrete system State 2 of the discrete system For output, , For process variables, To track the differentiator, It is a symbolic function.

[0014] Preferably, the controller of the extended state observer is:

[0015] ;

[0016] The equation for the extended state observer is:

[0017] ;

[0018] In the formula, The controller represents the extended state observer. As the initial value, For proportional parameters, Observer reference state, Differential parameters, , , For the observer's state variables, The observer's perturbation, This is an estimated constant value for the system within its operating range. It is the deviation of the extended state observer from the state and the disturbance;

[0019] Perturbation and state estimates are generated based on the extended state observer equation.

[0020] Preferably, the control law of the error nonlinear feedback unit adopts a nonlinear feedback structure of PID information of the error.

[0021] Preferably, the control rate of the error nonlinear feedback unit is:

[0022] ;

[0023] In the formula, K I For the integral coefficient, K P For proportionality coefficient, K D For the differential coefficients, α I α P α D These represent the degrees of error in the integral, proportional, and differential terms, respectively.

[0024] Preferably, the control quantities include fuel flow rate and nozzle area. The expanded state observer collects measurable state variables, generates state estimates, and combines them with the given speed collected by the tracking differentiator to generate nonlinear feedback units for speed deviation and pressure ratio deviation errors. The nonlinear feedback units estimate the speed deviation and pressure ratio deviation respectively to obtain pressure ratio control quantities and speed control quantities. Simultaneously, the expanded state observer generates speed control disturbances and pressure ratio control disturbances based on the measurable state variables. The speed control disturbances are combined with the speed control quantities to generate a new fuel flow rate, and the pressure ratio control disturbances are combined with the pressure ratio control quantities to generate a new nozzle area. The engine is then controlled for the next cycle using the new fuel flow rate and nozzle area.

[0025] The variable cycle engine active disturbance rejection control system of this application has the following advantages:

[0026] Without relying on an exact model of the object, the system dynamics can be estimated online only through an extended state observer. Compared to traditional PID controllers, which require repeated tuning of parameters (such as proportional coefficient and integral time) for different operating conditions, the control parameters of this algorithm (such as the tracking speed of the TD, the bandwidth of the extended state observer, and the gain of the nonlinear feedback) only need to be calibrated based on typical engine operating conditions. This significantly reduces the complexity of controller design and maintenance costs, making it more suitable for complex systems such as variable cycle engines that dynamically change with flight conditions.

[0027] Through the innovative design of the active disturbance rejection control framework, the problems of multivariable coupling, non-smooth transition state, insufficient steady-state accuracy and weak disturbance rejection capability of variable cycle engines are systematically solved. While improving the control quality, the engineering practicality is enhanced, and key technical support is provided for the high-performance control of variable cycle engines.

[0028] In existing aero-engine loop PID controllers, the linear combination of errors creates a trade-off between overshoot and speed. The ADRC (Advanced Dynamic Controller) employs nonlinear error feedback, using nonlinear combinations of errors to achieve a "large error, small gain; small error, large gain" effect, thus improving response speed and reducing overshoot. To estimate external and internal disturbances, the ADRC also uses an extended state observer to monitor system disturbances. Attached Figure Description

[0029] Figure 1 This is a block diagram of the active disturbance rejection control structure for the variable cycle engine in this application;

[0030] Figure 2 This is a schematic diagram of the logic structure of the active disturbance rejection controller in this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are only some, not all, of the embodiments of this application. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0032] The first aspect of this application provides an active disturbance rejection control system for a variable cycle engine. For a series integral type controlled object, providing the system with a maximum acceleration input does not necessarily result in the fastest, overshoot-free attainment of the desired value. Therefore, traditional PID control algorithms suffer from a trade-off between overshoot and speed. To achieve the goal of fastest, overshoot-free input tracking, the original input needs to be processed, i.e., the feedback error is gradually changed by arranging a transient process. This is the active disturbance rejection control algorithm proposed in this application, thereby resolving the trade-off between speed and overshoot.

[0033] like Figures 1-2 It includes a tracking differentiator, an extended state observer, an error nonlinear feedback unit, and a disturbance compensation unit.

[0034] The tracking differentiator can obtain the target speed or pressure ratio of the engine and generate the target tracking trajectory.

[0035] The extended state observer can collect the state variables of the engine and the control variables of the disturbance compensation unit to generate disturbance estimates and state estimates.

[0036] The tracking error is obtained after the tracking trajectory and state estimation are calculated. The error nonlinear feedback unit can generate error control quantity based on the tracking error.

[0037] The disturbance control quantity is obtained after the error control quantity and disturbance estimation are calculated. The disturbance compensation unit can process the disturbance control quantity, generate a control quantity to control the engine, and send the control quantity to the extended state observer.

[0038] Without relying on an accurate object model, the system dynamics can be estimated online using an extended state observer. When the engine speed and pressure ratio deviate, the extended state observer can quickly identify and correct the speed and pressure ratio, thereby achieving self-feedback regulation and self-disturbance rejection control, and improving control quality.

[0039] Preferably, the specific design of the tracking trajectory generated by the tracking differentiator is as follows:

[0040] For discrete systems

[0041]

[0042] The fastest control synthesis function is derived as follows:

[0043]

[0044] The algorithm formula is as follows:

[0045]

[0046] In the formula, , For process variables, For the input limit value, To track the coefficients of the differentiator, State 1 of the discrete system State 2 of the discrete system For output, , For process variables, To track the differentiator, It is a symbolic function.

[0047] Preferably, the extended state observer is specifically designed as follows:

[0048] The control quantity of the variable cycle engine active disturbance rejection controller is the main fuel flow rate. and tail nozzle area Its output is the high-voltage rotor speed. and turbine pressure ratio Therefore, the dynamics of the engine system can be expressed as:

[0049]

[0050] In the formula, and These are the internal disturbances during core machine operation. and These are external disturbances to the two control loops, respectively. and These represent the mutual influence and coupling effects between the two loops. Therefore, the parameters can be estimated. and Then, the dynamic system above can be formally expressed as follows:

[0051]

[0052] here and This is an estimated constant value for the system within its operating range, while in reality, a control system is a complex nonlinear dynamic equation. and It will not be a fixed constant value; the two cannot be completely similar throughout the entire process. Here, we only need to ensure that the two values ​​are relatively similar during system operation.

[0053] Let the total disturbances in the two loops be respectively and The original system can then be simplified to the following form:

[0054]

[0055] The above state equations are then extended using state observer design, with the total perturbation treated as a new one-dimensional state for state extension, to obtain the following: Taking the state equation as an example:

[0056]

[0057] In the formula, Then an extended state observer can be established to estimate the state and total disturbance.

[0058]

[0059] in

[0060]

[0061] The mean square error e of the fal function and The exponential term, This represents the boundary point with respect to the piecewise error function. By designing appropriate extended stater parameters... , Make , And by performing real-time compensation for the total disturbance, the compensated system can obtain

[0062]

[0063] Similarly, we can apply this to π. TDynamically design an extended state observer to estimate the state and total disturbance:

[0064]

[0065] The parameter definitions are similar to those above, and the observer state... , The compensated system ,in , Since both are control quantities calculated by the error feedback controller, decoupling can be achieved using the estimated state and extended state. Furthermore, to simplify the design, we can take... , The simplified ESO is obtained as follows:

[0066] , .

[0067] It can be seen that, with appropriate compensation, a large number of internal uncertainties and disturbances caused by external changes can be resolved, greatly simplifying the system model. At the same time, by appropriately adjusting the parameters of each set of extended state observers, the measured data can be filtered to ensure that the control effect is to a certain extent unaffected by external noise.

[0068] For the above extended state observer:

[0069]

[0070] Right now:

[0071]

[0072] The characteristic equation of the observer system is obtained by solving the equation. It can be configured near the approximate observer bandwidth. Then its configuration parameters can be obtained, and the configurations of the two parameters can be unified, i.e. , .

[0073] The first dimension is the state estimate, and the second dimension, the output, is the estimate of the total disturbance. The concept of total disturbance originated from active disturbance rejection control (ADRC) technology and is an important tool for real-time disturbance compensation in ADRC systems. In the construction of the extended state observer described above, all nonlinear, uncertain, and disturbance components are factored into the total disturbance.

[0074] The observer-based controller is designed as follows:

[0075] ;

[0076] The state observer equation is:

[0077] ;

[0078] In the formula, The controller represents the extended state observer. As the initial value, For proportional parameters, Observer reference state, Differential parameters, , , For the observer's state variables, The observer's perturbation, This is an estimated constant value for the system within its operating range. It is the deviation of the extended state observer from the state and the disturbance.

[0079] Preferably, the control law of the error nonlinear feedback unit adopts a nonlinear feedback structure based on the PID information of the error. The control law of the error nonlinear feedback unit is:

[0080] ;

[0081] In the formula, K I For the integral coefficient, K P For proportionality coefficient, K D For the differential coefficients, α I α P α D These represent the degrees of error in the integral, proportional, and differential terms, respectively.

[0082] This replaces the classic linear weighted sum form. Here, α describes the degree of nonlinearity and is an adjustable parameter. Using this nonlinear combination instead of the classic linear weighted sum form can greatly increase the design freedom and achieve better control. For example, by adjusting the adjustable parameter K... I K P K D α I α P α D This allows the nonlinear PID controller to achieve a significantly faster step response than the traditional fixed-gain PID controller, with a reduced transition time and no overshoot. This superior dynamic and static control performance is attributed to the nonlinear gain. Furthermore, although the nonlinear combination of feedback introduces computational complexity, high-performance hardware computing units can effectively address this issue.

[0083] Preferably, the control variables include fuel flow rate and nozzle area. An extended state observer collects measurable state variables and generates state estimates. Combined with a given engine speed collected by a tracking differentiator, a nonlinear feedback unit for speed deviation and pressure ratio deviation errors is generated. This nonlinear feedback unit estimates the speed deviation and pressure ratio deviation, respectively, to obtain the pressure ratio control variable and speed control variable. Simultaneously, the extended state observer generates speed control disturbances and pressure ratio control disturbances based on the measurable state variables. The speed control disturbance is combined with the speed control variable to generate a new fuel flow rate, and the pressure ratio control disturbance is combined with the pressure ratio control variable to generate a new nozzle area. The engine is then controlled for the next cycle using the new fuel flow rate and nozzle area. Through iterative control, continuous correction of the fuel flow rate and nozzle area is achieved, ensuring control quality.

[0084] In summary, this application has the following advantages:

[0085] Without relying on an exact model of the object, the system dynamics can be estimated online only through an extended state observer. Compared to traditional PID controllers, which require repeated tuning of parameters (such as proportional coefficient and integral time) for different operating conditions, the control parameters of this algorithm (such as the tracking speed of the TD, the bandwidth of the extended state observer, and the gain of the nonlinear feedback) only need to be calibrated based on typical engine operating conditions. This significantly reduces the complexity of controller design and maintenance costs, making it more suitable for complex systems such as variable cycle engines that dynamically change with flight conditions.

[0086] Through the innovative design of the active disturbance rejection control framework, the problems of multivariable coupling, non-smooth transition state, insufficient steady-state accuracy and weak disturbance rejection capability of variable cycle engines are systematically solved. While improving the control quality, the engineering practicality is enhanced, and key technical support is provided for the high-performance control of variable cycle engines.

[0087] In existing aero-engine loop PID controllers, the linear combination of errors creates a trade-off between overshoot and speed. The ADRC (Advanced Dynamic Controller) employs nonlinear error feedback, using nonlinear combinations of errors to achieve a "large error, small gain; small error, large gain" effect, thus improving response speed and reducing overshoot. To estimate external and internal disturbances, the ADRC also uses an extended state observer to monitor system disturbances.

[0088] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A variable cycle engine active disturbance rejection control system, characterized in that, It includes a tracking differentiator, an extended state observer, an error nonlinear feedback unit, and a disturbance compensation unit; The tracking differentiator can acquire the target speed or pressure ratio of the engine and generate the target tracking trajectory; The extended state observer can collect the state variables of the engine and the control variables of the disturbance compensation unit to generate disturbance estimates and state estimates. The tracking trajectory and state estimation are calculated to obtain the tracking error, and the error nonlinear feedback unit can generate an error control quantity based on the tracking error. The disturbance control quantity is obtained after the error control quantity and disturbance estimation are calculated. The disturbance compensation unit can process the disturbance control quantity, generate a control quantity to control the engine, and send the control quantity to the extended state observer.

2. The variable cycle engine active disturbance rejection control system as described in claim 1, characterized in that, The formula for calculating the target tracking trajectory generated by the tracking differentiator is as follows: ； In the formula, , For process variables, For the input limit value, To track the coefficients of the differentiator, State 1 of the discrete system State 2 of the discrete system For output, , For process variables, To track the differentiator, It is a symbolic function.

3. The variable cycle engine active disturbance rejection control system as described in claim 2, characterized in that, The controller of the extended state observer is: ； The equation for the extended state observer is: ； In the formula, The controller represents the extended state observer. As the initial value, For proportional parameters, Observer reference state, Differential parameters, , , For the observer's state variables, The observer's perturbation, This is an estimated constant value for the system within its operating range. It is the deviation of the extended state observer from the state and the disturbance; Perturbation and state estimates are generated based on the extended state observer equation.

4. The variable cycle engine active disturbance rejection control system as described in claim 3, characterized in that, The control law of the error nonlinear feedback unit adopts a nonlinear feedback structure based on the PID information of the error.

5. The variable cycle engine active disturbance rejection control system as described in claim 4, characterized in that, The control rate of the error nonlinear feedback unit is: ； In the formula, K I For the integral coefficient, K P For proportionality coefficient, K D For the differential coefficients, α I α P α D These represent the degrees of error in the integral, proportional, and differential terms, respectively.

6. The variable cycle engine active disturbance rejection control system as described in claim 5, characterized in that, The control quantities include fuel flow rate and nozzle area. The extended state observer collects measurable state variables and generates state estimates. Combined with the given speed collected by the tracking differentiator, it generates nonlinear feedback units for speed deviation and pressure ratio deviation errors. The nonlinear feedback units estimate the speed deviation and pressure ratio deviation respectively to obtain the pressure ratio control quantity and speed control quantity. At the same time, the extended state observer generates speed control disturbance and pressure ratio control disturbance based on the measurable state variables. The speed control disturbance and the speed control quantity are combined to generate a new fuel flow rate, and the pressure ratio control disturbance and the pressure ratio control quantity are combined to generate a new nozzle area. The engine is then controlled for the next cycle using the new fuel flow rate and nozzle area.

Images