Nonlinear control method and system suitable for airborne voltage type three-phase rectifier

Abstract

The invention provides a nonlinear control method and system suitable for an airborne voltage type three-phase rectifier. The method and system mainly improve the dynamic adjustment performance of a backstepping control algorithm. The method comprises the following steps: performing abc-dq conversion according to an alternating-current side phase voltage and an alternating-current side line current to obtain a current direct-axis component, a current quadrature-axis component, a voltage direct-axis component and a voltage quadrature-axis component, and constructing an average state space model; converting the average state space model into a differential equation, calculating a state variable of total energy stored in the system and a state variable of input power of the system so as to construct a disturbance observer and a backstepping controller, and finally obtaining a duty ratio direct-axis component and a duty ratio quadrature-axis component; after the disturbance observation technology is adopted, when interference is added, the convergence speed of the output voltage is higher, overshoot is smaller, the transient response is higher, and the dynamic regulation performance is higher.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of power electronic switching power supply control, and mainly relates to a nonlinear control method and system suitable for an airborne voltage-type three-phase rectifier. BACKGROUND

[0002] With the rapid development of electric propulsion aircraft (EPA), airborne electronic devices gradually replace the traditional aircraft mechanical, aerodynamic, hydraulic and other facilities, and become the main electrical equipment in the airborne power system (APS). The main advantages of the APS in the EPA include low fuel consumption, light weight, low operating cost and less air pollution. The aircraft generator and auxiliary power unit (APU) have alternating current output voltage characteristics, while the aviation lighting and heater are typical direct current loads. Therefore, in order to ensure the energy conversion efficiency, the EPA adopts a hybrid AC / DC APS containing at least an alternating current voltage bus and a direct current voltage bus as the main power transmission network, and adopts a three-phase AC / DC rectifier to connect between the alternating current bus and the direct current bus. In the hybrid APS of the EPA, the motor driver, electric actuator and air conditioning system and other strictly regulated loads are characterized as constant power loads (CPLs). These converter-based loads seriously threaten the stability of the APS due to their negative incremental impedance and nonlinear characteristics, further reducing the safety of flight. Therefore, it is necessary to explore the control strategy of the AC / DC rectifier to eliminate the influence of the CPLs on the system stability.

[0003] Currently, linear and nonlinear methods are usually used to deal with this kind of problem. The linear method based on proportional-integral (PI) is commonly used in the double-loop structure control of three-phase AC-DC rectifiers. However, the design of the PI controller can only guarantee the stability in a small region near the given operating point. Larger disturbances, such as larger CPL changes, larger DC bus voltage changes, larger AC frequency / voltage steps, etc., will make the actual operating point deviate from the stable region, resulting in the failure of the PI controller. In order to overcome the shortcomings of the linear method, several nonlinear control methods based on large signals are proposed to stabilize the DC bus voltage, such as model predictive control (MPC), sliding mode control (SMC), fuzzy logic control (FLC) and backstepping control (BSC). However, the model predictive control method has the defect that the online calculation burden increases exponentially with the increase of the system size; the sliding mode control method requires an additional high-cost output current sensor, and the DC bus voltage chattering caused by the sliding mode control is inevitable; the control performance of the fuzzy logic control method is highly dependent on the design and selection of the experience rules, and has the shortcomings of poor interoperability and complex calculation; compared with the above control methods, the backstepping control method has stronger robustness and stability, but the dynamic adjustment performance of the backstepping control method is slightly poor. SUMMARY

[0004] In order to overcome the shortcomings of the prior art and improve the dynamic adjustment performance of the backstepping control algorithm, the present application provides a nonlinear control method suitable for an airborne voltage-type three-phase rectifier, which further compensates the dynamic performance of the backstepping control algorithm by using disturbance observation technology, and ensures the fast dynamic adjustment performance of the nonlinear control algorithm.

[0005] A nonlinear control method suitable for an airborne voltage-type three-phase rectifier, comprising the following steps: Step 1: obtaining the AC side phase voltage and the AC side line current , performing abc-dq transformation on the AC side phase voltage and the AC side line current to obtain the current direct-axis component, the current quadrature-axis component, the voltage direct-axis component and the voltage quadrature-axis component, and constructing an average state space model; Step 2: converting the average state space model into a differential equation, calculating the state variable of the total energy stored in the system and the state variable of the input power of the system; Step 3, constructing a disturbance observer and a backstepping controller according to the state variable of the total energy stored in the system and the state variable of the input power of the system, obtaining an intermediate control law and a virtual control law; Step 4: based on the virtual control law k and the average state space model, obtaining the duty cycle direct axis component, the duty cycle quadrature axis component and the auxiliary control law through the control law transformation module; Step 5: performing dq-abc transformation on the duty cycle direct axis component and the duty cycle quadrature axis component, inputting the transformed results into a PWM generator to obtain a pulse width modulation signal, taking the pulse width modulation signal as a driving signal of the rectifier, and adjusting the output voltage of the rectifier through real-time control of the on-off of the switching tube.

[0006] Further, the step of constructing the average state space model is: The average state space model of the rectifier circuit described in the d-q axis is: (1) wherein, is the alternating current side line current the current direct axis component obtained after coordinate transformation; is the alternating current side line current the current quadrature axis component obtained after coordinate transformation; is the alternating current side phase voltage the voltage direct axis component obtained after coordinate transformation; is the alternating current side phase voltage the voltage quadrature axis component obtained after coordinate transformation; is the duty cycle the duty cycle direct axis component obtained after coordinate transformation; duty cycle the duty cycle quadrature axis component obtained after coordinate transformation; is time; is the alternating current side phase voltage angular frequency, ; is the direct current side output voltage; is the direct current side load power; is the alternating current side filter inductance; is the direct current side filter capacitance; is the alternating current side line resistance; .

[0007] Further, the step of converting the average state space model into a differential equation to calculate the state variable of the total energy stored in the system and the state variable of the input power of the system is: To obtain a general nonlinear controller, the average state space model under d-q axis is converted into differential equations suitable for extended disturbance observer and backstepping controller: (2) wherein, represents a state variable of total energy stored in the system, represents a state variable of input power of the system; is a derivative of , is a derivative of ; , ; is an unmeasurable disturbance quantity, ; is an unknown disturbance quantity; is a virtual control law; the virtual control law is used for the design of an auxiliary control strategy to achieve the actual control objective.

[0008] Further, a disturbance observer and a backstepping controller are constructed according to the state variable of total energy stored in the system and the state variable of input power of the system to obtain the steps of intermediate control law and virtual control law k . The disturbance observer is: (3) (4) wherein, is an observation value of disturbance quantity ; is an observation value of disturbance quantity ; is an observation value of first derivative of disturbance quantity ; is an observation value of first derivative of disturbance quantity ; , , and are intermediate observation quantities in the observer; , , and are positive adjustable observer gains, and the convergence time of the observer is adjusted by changing the gain parameters; According to the nonlinear backstepping control theory, a backstepping controller based on large signal stability is constructed; the backstepping controller is based on state variable error function and disturbance observer error function to construct Lyapunov error function equation, thereby deducing intermediate control law and virtual control law k； Firstly, the state variable error function and the disturbance observer error function are constructed: (5) wherein, is the energy state variable error; is the power state variable error; is the non-memorized disturbance error; is the unknown disturbance error; is the state variable reference value of the DC bus voltage; is the state variable reference value of the DC bus current; wherein, the reference value of the DC bus voltage is: (6) wherein is the DC reference voltage; The Lyapunov error function equation is constructed based on the state variable error function and the disturbance observer error function V is: (7) The derivative of the Lyapunov error function equation is obtained: (8) According to the nonlinear backstepping control theory, in order to make V converge to zero, the intermediate control law and the virtual control law k are respectively: (9) wherein, and are positive adjustable gains of the backstepping controller.

[0009] Further, based on the virtual control law k and the average state space model, the steps of obtaining the duty cycle direct axis component, the duty cycle quadrature axis component and the auxiliary control law through the control law transformation module are: Based on the virtual control law k and the average state space model, the actual control law, i.e. the duty cycle direct axis component and the duty cycle quadrature axis component is obtained through the control law transformation module: (10) wherein, is the auxiliary control law, ; and and Custom positive adjustable gain for power factor controller Reference value for line current quadrature axis component Reference value for line current quadrature axis component Set to zero to ensure unity power factor Adjust with changes in until .

[0010] The system applies a nonlinear control method for an airborne voltage-type three-phase rectifier, including a current transformation module, a voltage transformation module, a state variable calculation module, an extended disturbance observer module, a reference value calculation module, a backstepping controller module, a control law transformation module, a power factor control module, and a PWM generator module. The current transformation module is used to obtain and after abc-dq transformation of the AC side line current; the voltage transformation module is used to obtain and after abc-dq transformation of the AC side phase voltage; the state variable calculation module converts the average state space model into a differential equation to calculate the state variable of the total energy stored in the system and the state variable of the input power of the system; the reference value calculation module calculates the reference value of the state variable of the total energy stored in the system; the extended disturbance observer module and the backstepping controller module construct a disturbance observer and a backstepping controller, respectively, based on the state variable of the total energy stored in the system and the state variable of the input power of the system to obtain an intermediate control law and a virtual control law; the power factor control module adjusts the output auxiliary control law based on real-time adjustment of the line current quadrature axis component to control ; the control law transformation module obtains the duty cycle direct axis component and the duty cycle quadrature axis component based on the virtual control law k and the average state space model through the control law transformation module; the duty cycle direct axis component and the duty cycle quadrature axis component are dq-abc transformed, and the transformed results are input into the PWM generator module to obtain a pulse width modulation signal; the pulse width modulation signal is used as the driving signal of the rectifier to adjust the output voltage of the rectifier through real-time control of the on-off of the switching tube.

[0011] The beneficial effects of this invention are: by employing a backstepping control method based on an extended disturbance observer, the anti-interference capability of the rectifier can be significantly improved, and the adverse effects of large-signal disturbances such as CPL changes, large DC bus voltage changes, and large AC frequency / voltage steps on the system can be effectively suppressed. A simulation model built in Matlab / Simulink compares the output voltage control effects before and after using the disturbance observation technique. With the disturbance observation technique, the output voltage converges faster, has less overshoot, stronger transient response, and better dynamic adjustment performance when disturbances are introduced. Attached Figure Description

[0012] Figure 1 This is a structural diagram of the control scheme for APS-fed constant power load in the EPA of this invention embodiment; Figure 2 This is a control principle diagram of the backstepping control method based on an extended disturbance observer according to an embodiment of the present invention; Figure 3 The diagram shows the output voltage control effect of the control method of the present invention when a resistive load undergoes a sudden change; (a) is a diagram of the phase voltage change on the AC side; (b) is a diagram of the voltage control effect of the control method of the present invention when a resistive load undergoes a sudden change; (c) is a diagram of the load power estimate and the actual load power change obtained from ENDO. Figure 4 The diagram shows the output voltage control effect of the control method of the present invention when the constant power load changes; (a) is a diagram of the AC side phase voltage change; (b) is a diagram of the voltage control effect of the control method of the present invention when the constant power load changes; (c) is a diagram of the load power estimate and the actual load power change obtained by ENDO. Figure 5 The diagram shows the output voltage control effect of the control method of the present invention when the DC reference voltage changes abruptly; (a) is a diagram of the AC side phase voltage change; (b) is a diagram of the voltage control effect of the control method of the present invention when the DC reference voltage changes abruptly; (c) is a diagram of the load power estimate and actual load power change obtained by ENDO. Figure 6 These are diagrams showing the output voltage control effect of the control method of the present invention when the input AC side phase voltage changes; (a) is a diagram showing the AC side phase voltage change; (b) is a diagram showing the voltage control effect of the control method of the present invention when the input AC side phase voltage changes; (c) is a diagram showing the load power estimate and actual load power change obtained from ENDO. Figure 7is a comparison chart of output voltage control effect of the control method of the present application and the traditional PI control method under constant power load change; (a) is a chart of load power estimation value and actual load power obtained by ENDO; (b) is a chart of voltage control effect of the method proposed in the present application; (c) is a chart of voltage control effect under the traditional PI control method. DETAILED DESCRIPTION

[0013] The solution adopted by the present application comprises the following steps: Step 1: obtaining the alternating side phase voltage and the alternating side line current , performing abc-dq transformation on the alternating side phase voltage and the alternating side line current to obtain the current direct-axis component, the current quadrature-axis component, the voltage direct-axis component and the voltage quadrature-axis component; describing the rectifier circuit as an average state space model under d-q axis is: (1) wherein, is the current direct-axis component obtained after coordinate transformation of the alternating side line current ; is the current quadrature-axis component obtained after coordinate transformation of the alternating side line current ; is the voltage direct-axis component obtained after coordinate transformation of the alternating side phase voltage ; is the voltage quadrature-axis component obtained after coordinate transformation of the alternating side phase voltage ; is the duty ratio obtained after coordinate transformation of the duty ratio; is the duty ratio obtained after coordinate transformation of the duty ratio; is the alternating side phase voltage angular frequency, ; is the direct current side output voltage; is the direct current side load power; is the alternating side filter inductance; is the direct current side filter capacitance; is the alternating side line resistance; Step 2: converting the average state space model into differential equations, calculating the state variable of total energy stored in the system and the state variable of input power of the system; in order to obtain a general type of nonlinear controller, the average state space model under d-q axis is converted into differential equations suitable for extended disturbance observer and backstepping controller: (2) wherein, is a state variable representing the total energy stored in the system, is a state variable representing the input power of the system; and are the derivatives of and respectively; , ; is an unmeasurable disturbance quantity, ; is an unknown disturbance quantity; is a virtual control law; the virtual control law is used to assist in the design of the control strategy to achieve the actual control objective; Step 3, constructing a disturbance observer and a backstepping controller according to the state variable of the total energy stored in the system and the state variable of the input power of the system, obtaining an intermediate control law and a virtual control law k ; In order to improve the performance of the nonlinear controller, the disturbance observer is used to eliminate the external additional disturbance and the internal uncertainty disturbance of the system, and to help provide a fast dynamic response to the internal and external disturbances, and to reduce the number of sensors; The disturbance observer is: (3) (4) wherein, is the observation value of the disturbance quantity ; is the observation value of the disturbance quantity ; is the observation value of the first derivative of the disturbance quantity ; is the observation value of the first derivative of the disturbance quantity ; , , , are all intermediate observation quantities in the observer; , , , are all positive adjustable observer gains, and the convergence time of the observer is adjusted by changing the gain parameters; According to the nonlinear backstepping control theory, a backstepping controller based on large signal stability is constructed; the backstepping controller is based on the state variable error function and the disturbance observer error function, and a Lyapunov error function equation is constructed, so as to derive an intermediate control law and a virtual control lawk； Firstly, the state variable error function and the disturbance observer error function are constructed: (5) wherein, is the energy state variable error; is the power state variable error; is the non-memorized disturbance error; is the unknown disturbance error; is the state variable reference value of ; is the state variable reference value of ; wherein, the reference value of is: (6) wherein is the DC reference voltage; The Lyapunov error function equation is constructed based on the state variable error function and the disturbance observer error function V is: (7) The derivative of the Lyapunov error function equation is obtained: (8) According to the nonlinear backstepping control theory, in order to make V converge to zero, the intermediate control law and the virtual control law k are respectively: (9) wherein, and are positive adjustable gains of the backstepping controller; Step 4: Based on the virtual control law k and the average state space model, the actual control law, i.e. the duty ratio and is obtained through the control law transformation module: Based on the virtual control law k and the average state space model, the actual control law, i.e. the duty ratio direct-axis component and the duty ratio quadrature-axis component is obtained through the control law transformation module: (10) wherein, is the auxiliary control law, ; , ​Customizing positive adjustable gain for power factor controller; The line current quadrature axis component reference value is set to zero to ensure unit power factor, With adjustment until ; Step 5: Perform dq-abc transformation on the duty ratio direct axis component and the duty ratio quadrature axis component , input the transformed results into a PWM generator to obtain a pulse width modulation signal, use the pulse width modulation signal as a driving signal of the rectifier, and adjust the output voltage of the rectifier through real-time control of the on-off of the switching tube; The system applying the nonlinear control method for the airborne voltage type three-phase rectifier includes a current transformation module, a voltage transformation module, a state variable calculation module, an extended disturbance observer module, a reference value calculation module, a backstepping controller module, a control law transformation module, a power factor control module, and a PWM generator module. The current transformation module is used to obtain and after abc-dq transformation of the AC side line current; the voltage transformation module is used to obtain and after abc-dq transformation of the AC side phase voltage; the state variable calculation module converts the average state space model into a differential equation, calculates the state variable of the total energy stored in the system and the state variable of the input power of the system; the reference value calculation module calculates the reference value of the state variable of the total energy stored in the system; the extended disturbance observer module and the backstepping controller module construct a disturbance observer and a backstepping controller, respectively, according to the state variable of the total energy stored in the system and the state variable of the input power of the system, to obtain an intermediate control law and a virtual control law; the control law transformation module obtains the duty ratio direct axis component k and the duty ratio quadrature axis component by the control law transformation module according to the virtual control law and the average state space model; the power factor control module controls by real-time adjustment of the output of the line current quadrature axis component, so as to realize control of the power factor; the duty ratio direct axis component and the duty ratio quadrature axis component are subjected to dq-abc transformation, the transformed results are input into the PWM generator module to obtain a pulse width modulation signal, the pulse width modulation signal is used as a driving signal of the rectifier, and the output voltage of the rectifier is adjusted through real-time control of the on-off of the switching tube.

[0014] The present application will be further described in conjunction with the embodiments and drawings. The effects of the present application can be illustrated by the following simulation experiments.

[0015] Simulation conditions: The application is simulated on an Intel(R) Core(TM) Ultra 7 155H 1.40GHz central processing unit, 32G memory, WINDOWS 10 operating system, using MATLAB / Simulink software.

[0016] Example 1 The APS feed constant power load based on the extended disturbance observer backstepping control method used in the EPA has the following steps: S01, obtaining the d-q components of the quantities obtained by abc-dq transformation of the alternating side phase voltage.

[0017] S02, using the sampled , the filter inductance and its equivalent series resistance, filter capacitance are known to establish the rectifier circuit state space equation required to meet the design of the extended disturbance observer and the design of the backstepping controller.

[0018] S03, for the system state space model, design the extended disturbance observer to estimate the disturbance, design the intermediate control law and the virtual control law, combine the extended disturbance observer with the backstepping controller to optimize the effect of control.

[0019] S04, according to the line current quadrature axis component, design the power factor controller to obtain the auxiliary control law. Transform the state equation to obtain the actual control law, that is, the duty cycle of the converter in the synchronous rotating coordinate system. Dq-abc transform the control law as input, and then input it into the PWM generator to obtain the pulse width modulation signal as the driving signal of the converter. The output voltage of the converter is adjusted by real-time control of the switch tube on-off.

[0020] In specific embodiments, the average state space model under d-q axis is used in the method, and its expression is as follows: (1) In formula (1): is the alternating side line current the current direct axis component and the current quadrature axis component obtained after coordinate transformation; is the alternating side phase voltage the voltage direct axis component and the voltage quadrature axis component obtained after coordinate transformation; is the duty cycle the direct axis component and the quadrature axis component obtained after coordinate transformation; VAC for the AC side phase voltage angle frequency; VDC for the DC side output voltage; PDC for the DC side load power; L for the AC side filter inductance; C for the DC side filter capacitance; R for the AC side line resistance; To obtain a general nonlinear controller, the original average state space model in d-q axis is transformed into a differential equation suitable for the design of extended disturbance observer and backstepping controller: (2) In equation (2): is a newly defined state variable; represents the total energy stored in the system, represents the input power of the system; and are the derivatives of and respectively; is an unmeasurable disturbance quantity; is an unknown disturbance quantity; is a virtual control law used to assist in the design of the control strategy to achieve control of the target; Further, to improve the performance of the nonlinear controller, a disturbance observer is used to eliminate external additional disturbances and internal uncertain disturbances of the system, and to provide a fast dynamic response to internal and external disturbances, and to reduce the number of sensors. Therefore, the design of the disturbance observer is as follows: (3) (4) In equation (3) and equation (4): is the observation value of the disturbance quantity ; is the observation value of the first derivative of the disturbance quantity ; is the intermediate observation quantity in the observer; is a positive adjustable observer gain, which can adjust the convergence time of the observer by changing the gain parameter; Further, according to the nonlinear backstepping control theory, a backstepping controller based on large signal stability is designed. First, define the state variable error function and the disturbance observer error function: (5) In equation (5): and respectively, are energy state variable error and power state variable error; and respectively, are unknown disturbance error and unmeasurable disturbance error; and is the state variable and is the reference value.

[0021] wherein the state variable reference value can be calculated by disturbance quantity observation value and sampling value: (6) wherein is the direct current reference voltage; the Lyapunov error function equation is constructed V as follows: (7) derivation of equation (7) is as follows: (8) According to the nonlinear backstepping control theory, in order to make V converge to zero, the designed intermediate control law and the virtual control law k are respectively as follows: (9) in equation (9): and are positive adjustable gains of the backstepping controller; the actual control law, i.e. the duty cycle k and can be obtained by using the virtual control law and combining the circuit state space equation (1) of the three-phase voltage type rectifier: (10) wherein:

[0022] the reference value of the line current quadrature axis component is set to zero to ensure the unity power factor, which is adjusted with the change of until ; , is the self-defined positive adjustable gain of the power factor controller.

[0023] the obtained and in the above equation are the switching period duty cycles of the converter in the synchronous rotating coordinate system, and the switching functions of the abc three-phase can be obtained by inverse coordinate transformation, and The input of the PWM generator can obtain a pulse width modulation signal as a driving signal of the converter.

[0024] Embodiment 2 Corresponding to the above-mentioned embodiment, the present application also provides an extended disturbance observer-based backstepping control system for an APS-fed constant power load in an EPA. The control system comprises a current transformation module, a voltage transformation module, a state variable calculation module, an extended disturbance observer module, a reference value calculation module, a backstepping controller module, a power factor control module, a control law transformation module and a PWM generation module. The details will be described in combination with Figure 1 Figure 1 The differential equation in the above-mentioned embodiment corresponds to the differential equation (2) suitable for designing the extended disturbance observer and the backstepping controller in the present application; the extended disturbance observer module corresponds to the formula (3) and the formula (4); the state estimation value corresponds to the formula (6); the backstepping controller corresponds to the formula (7), the formula (8) and the formula (9); the actual control input corresponds to the formula (10); and the power factor controller corresponds to the auxiliary control law The control method of the system comprises the following steps: Step 1, using the sampled and the known circuit element parameters to design the required rectifier circuit state space equation.

[0025] Step 2, for the established voltage-type rectifier model, the original state space equation is changed into a state space form suitable for designing the extended disturbance observer and the backstepping controller while considering the circuit parameter perturbation. Step 3, for the state space equation obtained in step 2, the extended disturbance observer is designed; the reference value of the state variable is calculated based on the estimation of the observer, which is used for backstepping control; and the virtual control law and the auxiliary control law are designed according to the backstepping control theory.

[0026] Step 4, according to the line current cross-axis component, the power factor controller is designed to obtain the auxiliary control law; the virtual control law and the auxiliary control law obtained in step 3 are transformed to obtain the actual control law of the voltage-type rectifier, which is input into the PWM generator to obtain a pulse width modulation signal as a driving signal of the converter, and the output voltage of the converter is adjusted through real-time control of the on-off of the switching tube.

[0027] Step 1: In the method, the average state space model of the rectifier circuit is usually used in the d-q axis, and the expression thereof is obtained through the current transformation module and the voltage transformation module as follows: ​​ (1) In formula (1): is the line current of the AC side is the current direct-axis component and the current quadrature-axis component obtained after coordinate transformation; is the phase voltage of the AC side is the voltage direct-axis component and the voltage quadrature-axis component obtained after coordinate transformation; is the duty ratio is the direct-axis component and the quadrature-axis component obtained after coordinate transformation; is the angular frequency of the phase voltage of the AC side; is the output voltage of the DC side; is the load power of the DC side; is the filter inductance of the AC side; is the filter capacitance of the DC side; is the line resistance of the AC side; Step 2: The disturbance observer and the backstepping controller are designed based on the following form: (2) In formula (2): is a newly defined state variable; represents the total energy stored in the system, represents the input power of the system; and are the derivatives of and respectively; is calculated by the state variable calculation module ; is an unmeasurable disturbance quantity; is an unknown disturbance quantity; is a virtual control law, used for the design of the auxiliary control strategy to achieve control of the target; Step 3: In order to improve the performance of the nonlinear controller, the disturbance observer module is used to eliminate external additional disturbances and internal uncertain disturbances of the system, and to provide a fast dynamic response to internal and external disturbances, and to reduce the number of sensors. Therefore, the disturbance observer is designed as follows: (3) (4) In formulas (3) and (4): the observation of the disturbance quantity; the observation of the first derivative of the disturbance quantity the observation of the first derivative of the disturbance quantity is an intermediate observation in the observer; is a positive adjustable observer gain, and the convergence time of the observer can be adjusted by changing the gain parameter; Further, according to the nonlinear backstepping control theory, a backstepping controller based on large signal stability is designed. First, the state variable error function and the disturbance observer error function are defined: (5) In formula (5): and are the energy state variable error and the power state variable error, respectively; and are the non-memorized disturbance error and the unknown disturbance error, respectively; and are the reference values of the state variables and .

[0028] wherein the state variable reference value can be calculated by the disturbance quantity observation and the sampling value through the reference value calculation module: (6) wherein is the direct current reference voltage; The Lyapunov error function equation is constructed as: V (7) Derivation of (7) is obtained: (8) According to the nonlinear backstepping control theory, in order to make V converge to zero, the designed intermediate control law is calculated through the backstepping controller module k and the virtual control law are respectively: (9) In formula (9): and are positive adjustable gains of the backstepping controller; Step 4: The actual control law, i.e. the duty cycle k can be obtained through the control law transformation module by combining the virtual control law ​and : (10) wherein:

[0029] The line current quadrature axis component reference value is set to zero to ensure unity power factor, adjustment is made with the change of ; ; , is a user-defined positive adjustable gain of the power factor controller.

[0030] The and obtained in the last step are the switching period duty ratios of the converter in the synchronous rotating coordinate system, and the switching functions of the three-phase abc can be obtained by inverse coordinate transformation , and , which are input into the PWM generation module to obtain the pulse width modulation signal as the driving signal of the converter.

[0031] In Matlab / Simulink, the APS feed constant power load based on extended disturbance observer backstepping control method proposed in the application for use in EPA is simulated to verify its effectiveness. All simulations are carried out under the condition of voltage type rectifier feeding constant power load.

[0032] Firstly, the case of resistance load mutation is verified. As shown in Figure 3 , before t=2.0 s, the original resistance load is 20 Ω. Then, at t=2.0 s, the resistance load is reduced to 10 Ω, and at t=3.0 s, the resistance load is increased to 20 Ω again. From these simulation results, it can be easily seen from Figure 3 that in the presence of resistance load step, ENDO can accurately estimate the load power. At the same time, the DC voltage can quickly recover to the reference value, and the steady-state error is zero.

[0033] Then, the case of constant power load mutation is verified. Figure 4 The dynamic response when CPL changes is shown. Specifically, at t=2.0 s, it jumps from 12.0 kW to 18.0 kW, and then at t=3.0 s, it is further increased to 24.0 kW. In the case of large CPL change, after a short transient period, the DC bus voltage is adjusted to the reference voltage. It can be noted that when the APS is adjusted by the composite nonlinear controller, the adverse effects of large CPL change on the DC bus voltage are eliminated within 50.0 ms.

[0034] Figure 5 The dynamic response of the proposed controller to the change of DC link reference voltage is shown. Specifically, the DC bus reference voltage is decreased from 540 V to 500 V at t = 2.0 s, and then increased to 570 V at t = 3.0 s. From these results, it can be seen that when the voltage reference value steps, the actual DC voltage will change slightly, but after a short transient time, the actual DC voltage will eventually be adjusted to the required voltage by the proposed controller. In addition, the actual load power will also change accordingly due to the change of DC link reference voltage. From the third subgraph, ENDO can quickly and accurately estimate the load power.

[0035] Figure 6 The dynamic response of the proposed controller to the change of DC link reference voltage is shown. Specifically, the DC bus reference voltage is decreased from 540 V to 500 V at t = 2.0 s, and then increased to 570 V at t = 3.0 s. From these results, it can be seen that when the voltage reference value steps, the actual DC voltage will change slightly, but after a short transient time, the actual DC voltage will eventually be adjusted to the required voltage by the proposed controller. In addition, the actual load power will also change accordingly due to the change of DC link reference voltage. From the third subgraph, ENDO can quickly and accurately estimate the load power.

[0036] The proposed control method is then compared with the traditional PI controller. Figure 7 The output response of the voltage source rectifier fed constant power load using the proposed method and the traditional PI control is shown under the same constant power load power change. It can be seen that the constant power load is first increased from 12 kW to 18 kW, and then to 24 kW at t = 2.0 s and t = 3.0 s, respectively. The voltage fluctuation of the voltage source rectifier controlled by the traditional PI controller becomes larger at t = 2 s, and loses output stability at t = 3 s, while the controller designed by the proposed method maintains the output stability of the voltage source rectifier. Compared with PI control, the system output using the proposed controller has a faster response speed.

Claims

1. A nonlinear control method applicable to airborne voltage-type three-phase rectifiers, characterized in that, Includes the following steps: Step 1: Obtain the AC phase voltage and AC side current For AC side phase voltage and AC side current The abc-dq transformation is performed to obtain the direct-axis component of the current, the quadrature-axis component of the current, the direct-axis component of the voltage, and the quadrature-axis component of the voltage, and an average state-space model is constructed. The rectifier circuit can be described as an average state-space model under the dq axis as follows: （1） in, AC side current The direct-axis component of the current obtained after coordinate transformation; AC side current The current cross-axis component obtained after coordinate transformation; AC side phase voltage The voltage direct-axis component obtained after coordinate transformation; AC side phase voltage The cross-axis component of voltage obtained after coordinate transformation; Duty cycle The duty cycle direct axis component obtained after coordinate transformation; Duty cycle The duty cycle cross-axis component obtained after coordinate transformation; For time; The phase voltage angular frequency on the AC side. ; This is the DC-side output voltage; This refers to the DC-side load power. For AC side filter inductance; For DC-side filter capacitors; For AC side line resistance; . Step 2: Transform the average state-space model into differential equations, and calculate the state variables of the total energy stored in the system and the state variables of the system's input power; Step 3: Construct a disturbance observer and a backstepping controller based on the state variables of the total energy stored in the system and the state variables of the system's input power to obtain the intermediate control law and the virtual control law; Step 4: Based on virtual control law k The duty cycle direct-axis component, duty cycle quadrature-axis component, and auxiliary control law are obtained from the average state-space model through the control law transformation module. Step 5: Perform dq-abc transformation on the direct-axis component and the quadrature-axis component of the duty cycle, input the transformed result into the PWM generator to obtain the pulse width modulation signal, use the pulse width modulation signal as the drive signal of the rectifier, and adjust the output voltage of the rectifier by real-time control of the switching transistor.

2. The nonlinear control method for an airborne voltage-type three-phase rectifier according to claim 1, characterized in that: The steps to transform the average state-space model into differential equations and calculate the state variables of the total energy stored in the system and the input power of the system are as follows: To obtain a general-purpose nonlinear controller, the average state-space model under the dq axis is transformed into differential equations suitable for extended disturbance observers and backstepping controllers: （2） in, State variables representing the total energy stored in the system. State variables representing the system's input power; for The derivative of for The derivative; , ; For unmeasurable disturbances, ; The disturbance quantity is unknown. This is a virtual control law; virtual control law Used to assist in the design of control strategies to achieve actual control objectives.

3. The nonlinear control method for an airborne voltage-type three-phase rectifier according to claim 1, characterized in that: Based on the state variables of the total energy stored in the system and the input power of the system, a disturbance observer and a backstepping controller are constructed to obtain the intermediate control law. and virtual control law k The steps are as follows: The disturbance observer is: （3） （4） in, For disturbance quantity Observed values; For disturbance quantity Observed values; The first derivative of the disturbance Observed values; The first derivative of the disturbance Observed values; , , and All are intermediate observations of the observer; , , and All are positively adjustable observer gains, and the convergence time of the observer is adjusted by changing the gain parameter; Based on nonlinear backstepping control theory, a backstepping controller based on large-signal stability is constructed. The backstepping controller uses the state variable error function and the disturbance observer error function as its foundation to construct the Lyapunov error function equation, thereby deriving the intermediate control law. and virtual control law k； First, construct the state variable error function and the disturbance observer error function: （5） In the formula, Error for energy state variables; For power state variable error; This is an unmemorable perturbation error; This is an unknown disturbance error; for State variable reference value; for Reference values ​​for state variables; in, Reference value for: （6） in This is the DC reference voltage; Lyapunov error function equations are constructed based on the state variable error function and the disturbance observer error function. V for: （7） Differentiating the Lyapunov error function equation yields: （8） According to nonlinear backstepping control theory, in order to make V If it converges to zero, then the intermediate control law... and virtual control law k They are respectively: （9） in, and All are positive adjustable gains of the backstepping controller.

4. The nonlinear control method for an airborne voltage-type three-phase rectifier according to claim 1, characterized in that: Based on virtual control law k The steps for obtaining the duty cycle direct-axis component, duty cycle quadrature-axis component, and auxiliary control law through the control law transformation module, based on the average state-space model, are as follows: Based on virtual control law k The duty cycle direct-axis component is obtained from the average state-space model through the control law transformation module. and duty cycle cross axis component : (10) in, To assist the control law, ; and All are custom positive adjustable gains of the power factor controller; Reference value of quadrature axis component of line current Set to zero to ensure a power factor of one unit. Follow Adjustments were made to the changes until... .

5. A system applying a nonlinear control method suitable for airborne voltage-type three-phase rectifiers, characterized in that: It includes a current conversion module, a voltage conversion module, a state variable calculation module, an extended disturbance observer module, a reference value calculation module, a backstepping controller module, a control law conversion module, a power factor control module, and a PWM generator module; The current transformation module is used to obtain the AC side current after undergoing the abc-dq transformation. and The voltage conversion module is used to obtain the AC phase voltage after the abc-dq conversion. and ; The state variable calculation module transforms the average state-space model into differential equations and calculates the state variables of the total energy stored in the system and the state variables of the system's input power. The reference value calculation module calculates the reference value of the state variable of the total energy stored in the system; The extended disturbance observer module and the parametric backstepping controller module construct the disturbance observer and backstepping controller based on the state variables of the total energy and the input power stored in the system, respectively, to obtain the intermediate control law and the virtual control law; the power factor control module adjusts the output auxiliary control law in real time based on the quadrature-axis component of the line current. The control law transformation module then uses the virtual control law. k The duty cycle direct-axis component is obtained from the average state-space model through the control law transformation module. and duty cycle cross axis component ; The direct axis component of duty cycle and duty cycle cross axis component A dq-abc conversion is performed, and the result is input into the PWM generator module to obtain a pulse width modulation signal. The pulse width modulation signal is used as the drive signal for the rectifier, and the output voltage of the rectifier is adjusted by real-time control of the switching transistors.

Images