CN109039180B - Fractional-order control method for grid-connected process of doubly-fed induction generator - Google Patents

Abstract

The invention discloses a fractional-order control method for a grid-connected process of a doubly-fed induction generator. Physical and chemical realization, that is, through a rational transfer function, it can effectively simulate fractional-order operators in amplitude-frequency characteristics and phase-frequency characteristics; establish a fractional-order PI ^λ controller: control the motor control model, control the doubly-fed induction power generation No-load grid connection process. The fractional-order PI ^λ controller can effectively reduce the active power and reactive power in the start-up stage, so that the system can enter the steady-state stage smoothly, and avoid the grid from being overly impacted; in the short-circuit fault state, the fractional-order PI ^λ control method can also be used to control the power grid. Reduce system vibration.

Description

Fractional-order control method for grid-connected process of doubly-fed induction generator

technical field

The invention relates to the technical field of control, in particular to a fractional-order control method for a grid-connected process of a doubly-fed induction generator.

Background technique

Doubly-fed induction generators (DFIG) are widely used in variable-speed constant-frequency wind power generation systems due to their flexible frequency conversion control, good regulation performance, good dynamic and steady-state characteristics, and decoupling control of active and reactive power. and ship electrical systems. The control of the doubly-fed induction generator is realized by the control of the rotor AC excitation converter. The stator and rotor system of the doubly-fed induction motor is a strong coupling system. In order to achieve decoupling control, the vector control method of stator flux linkage orientation is often used to decompose the alternating current into active power and reactive power, and the double closed-loop PI regulator is used to separate them. control it.

PI control is still the most commonly used control method in the doubly-fed power generation control system. It has the advantages of simple structure, convenient implementation and wide adaptability. However, this control method is a control method based on an accurate model, and the working conditions of the controlled object change. After that, its control performance will also decrease. Considering that it contains more complex dynamic parts, the DFIG system actually becomes a multi-variable nonlinear system. At this time, the traditional PI control system has the disadvantages of poor dynamic performance, large overshoot and weak anti-interference ability. protrude.

In view of the shortcomings of traditional PI control, the improvement schemes involved in domestic and foreign literatures mainly include:

1. Use the RBF neural network to adjust the PID controller parameters online to deal with the influence of system parameter uncertainty and external disturbance on the control performance, but it requires more historical data and the training process is long.

2. An integral sliding mode control strategy based on neural network is proposed. It is verified that the method has strong robustness to disturbances and good grid-connected performance. However, the chattering phenomenon of sliding mode control strategy is not can be solved.

3. A new excitation control strategy with variable parameters PI and neural network coordinated control is proposed. Its control effect does not depend on the parameters of the system, and it has good dynamic adjustment and online decoupling capabilities. Disadvantages such as lag.

4. The speed sensorless doubly-fed induction generator control technology based on model reference self-adaptation is studied, but this control idea lacks systematic design methods and the control precision is relatively low.

In the field of motors, some scholars have also applied fractional-order proportional-integral controller to the maximum power tracking in permanent magnet synchronous power generation system, and verified that the fractional-order PI controller has faster response speed and higher power output performance. Or design a fractional-order PID controller for the excitation system of permanent magnet synchronous generator, and use particle swarm algorithm to optimize the parameters. However, there are few reports on the improvement of PI control for doubly-fed induction generators.

SUMMARY OF THE INVENTION

In order to solve the deficiencies of the prior art, the present invention provides a fractional-order control method for the grid-connected process of a doubly-fed induction generator. The present invention introduces an adjustable integral order λ to increase its control dimension. It can effectively reduce the power oscillation of the motor under grid-connected transient process and fault disturbance, and has certain development potential and application value.

The fractional-order control method for the grid-connected process of the doubly-fed induction generator, including:

Constructing a motor control model, the model is established based on a synchronous rotating coordinate system and consists of a voltage equation, a flux linkage equation and an electromagnetic torque equation;

Constructing a fractional operator: Taking the Grunwald-Letnikov fractional calculus definition as the theoretical basis, determining the fractional differential calculation algorithm, obtaining the approximate value of the numerical differential of the function for the fractional differential calculation algorithm and proving its calculation accuracy, according to the Grunwald-Letnikov fraction The first-order calculus definition calculates its fractional derivative;

For the doubly-fed wind power generation system, the continuous transfer function method is used to rationally realize the fractional-order operator, that is, through a rational transfer function, it can effectively simulate the fractional-order operator in terms of amplitude-frequency characteristics and phase-frequency characteristics;

Establish a fractional-order PI ^λ controller: control the motor control model and control the no-load grid connection process of the doubly-fed induction generator.

In a further preferred technical solution, the fractional integral term I ^λ in the fractional PI ^λ controller is obtained by an improved Oustaloup filtering algorithm.

In a further preferred technical solution, an improved Oustaloup filtering algorithm is used to construct a fractional operator.

In a further preferred technical solution, the doubly-fed induction generator is based on the mathematical model of the synchronous rotating coordinate system:

Voltage equation:

Among them, ω _s is the slip angular velocity; ω _s =ω ₁ -ω _r , u _ds : d-axis component of stator voltage, u _qs : q-axis component of stator voltage, u _dr : d-axis component of rotor voltage, u _qr : rotor voltage q-axis component, i _ds : d-axis component of stator current, i _qs : q-axis component of stator current, i _dr : d-axis component of rotor current, i _qr : q-axis component of rotor current, R _s : stator-side resistance, R _r : Rotor side resistance, Ψ _qs : q-axis component of stator flux, Ψ _ds : d-axis component of stator flux, Ψ _qr : q-axis component of rotor flux, Ψ _dr : d-axis component of rotor flux, p: differential sign.

The flux linkage equation:

Electromagnetic torque equation:

T _e =n _p L _m ( _ids i _qr - i _qs i _dr ) (3).

In a further preferred technical solution, in the start-up stage of the doubly-fed wind turbine, no-load grid-connection control is required to make the motor smoothly cut into the grid, and the amplitude of each electrical component is required to be as small as possible. When the generator is no-load, the stator current components are all 0, that is, i _ds = i _qs = 0. The no-load mathematical model of the doubly-fed induction generator is as follows:

T _e = 0 (6)

In the formula, T ₀ —generator no-load torque.

In a further preferred technical solution, the simplified equation of the motor model ignoring the motor stator resistance is:

A further preferred technical solution, Grunwald-Letnikov fractional calculus definition:

In a further preferred technical solution, the fractional differential calculation algorithm is:

为函数的多项式系数，假设步长h足够小，根据(12)可直接求出函数数值微分的近似值，并可以证明其计算精度为o(h)，当f(t)的函数表达式确定时，可以直接由式(11)计算其出分数阶导数。In the formula,

is the polynomial coefficient of the function, assuming that the step size h is small enough, the approximate value of the numerical differentiation of the function can be directly obtained according to (12), and it can be proved that its calculation accuracy is o(h), when the function expression of f(t) is determined. , the fractional derivative can be calculated directly from equation (11).

In a further preferred technical solution, the PI ^λ controller function is:

where λ is the integration order.

In a further preferred technical solution, the control steps of the PI ^λ controller are: firstly detect the stator and rotor voltages and currents of the motor, and then carry out coordinate transformation to calculate the stator flux linkage Ψ ₁ , stator active power P and reactive power Q; It is actually necessary to set the stator active power command P* and the reactive power command Q*, and compare them, and the difference obtained is obtained through the fractional-order PI ^λ regulator to obtain the active component command i _qs * and the reactive component command i _ds * of the stator current, Then enter the current inner loop control.

Compared with the prior art, the beneficial effects of the present invention are:

(1) The fractional-order PI ^λ controller can make the control system of the doubly-fed wind power generation system have the flexibility to deal with nonlinear control objects, and can enhance the robustness in the dynamic process.

(2) The fractional-order PI ^λ controller can effectively reduce the active power and reactive power in the start-up stage, so that the system can enter the steady-state stage smoothly, and avoid the grid from being overly impacted; in the short-circuit fault state, the fractional-order PI ^λ control method is adopted. It can also reduce system vibration.

Description of drawings

The accompanying drawings that form a part of the present application are used to provide further understanding of the present application, and the schematic embodiments and descriptions of the present application are used to explain the present application and do not constitute improper limitations on the present application.

Figure 1 is a dual PWM converter;

Fig. 2 is the control block diagram of power outer loop PIλ;

Fig. 3 is the RSC control structure model of the doubly-fed induction motor;

Fig. 4 Active power in integer order and fractional order;

Fig. 5 Reactive power in integer order and fractional order;

Figure 6 shows the active power oscillation under the three-phase short-circuit fault state.

Detailed ways

It should be noted that the following detailed description is exemplary and intended to provide further explanation of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

It should be noted that the terminology used herein is for the purpose of describing specific embodiments only, and is not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural as well, furthermore, it is to be understood that when the terms "comprising" and/or "including" are used in this specification, it indicates that There are features, steps, operations, devices, components and/or combinations thereof.

In a typical embodiment of the present application, the steps of the present invention are:

(1) Construct the motor control model

The DFIG uses dual PWM converters for AC excitation, which are called grid-side PWM converters (GSC) and rotor-side PWM converters (RSC), respectively. As shown in Figure 1, from the circuit structure, GSC and RSC are decoupled and independent of each other through the DC bus, so the mathematical model and control strategy of the grid side and rotor side can be studied respectively.

GSC has the following functions: it can ensure that the waveform of the input current is close to sinusoidal, the use of GSC can ensure lower harmonic content and meet the required power factor, and provide a system power factor control method; control the input current to ensure the stability of the DC bus voltage.

The control goal of RSC is to provide excitation current for the motor rotor to adjust the reactive power output from the stator side, and to control the motor speed or the motor output active power by controlling the DFIG rotor current and torque current, so as to achieve the maximum wind energy tracking variable speed Constant frequency operation.

The fractional-order PI ^λ controller of this scheme is applied to the rotor-side power control outer loop in the doubly-fed asynchronous induction motor control strategy, so that the grid-connected dynamic process has better robustness. Therefore, this modeling focuses on RSC and no-load Grid-connected control.

Since the active power and reactive power output by the motor are closely related to the d and q-axis current components of the rotor, the d and q components of the rotor current are the main control targets.

The mathematical model of the doubly-fed induction generator based on the synchronous rotating coordinate system is as follows:

Voltage equation:

Among them, ω _s is the slip angular velocity; ω _s =ω ₁ -ω _r , u _ds : d-axis component of stator voltage, u _qs : q-axis component of stator voltage, u _dr : d-axis component of rotor voltage, u _qr : rotor voltage q-axis component, i _ds : d-axis component of stator current, i _qs : q-axis component of stator current, i _dr : d-axis component of rotor current, i _qr : q-axis component of rotor current, R _s : stator-side resistance, R _r : Rotor side resistance, Ψ _qs : q-axis component of stator flux, Ψ _ds : d-axis component of stator flux, Ψ _qr : q-axis component of rotor flux, Ψ _dr : d-axis component of rotor flux, p: differential sign.

The flux linkage equation:

L _s : stator self-inductance, L _r : rotor self-inductance, L _m : mutual inductance.

Electromagnetic torque equation:

T _e =n _p L _m ( _ids i _qr -i _qs i _dr ) (3)

T _e : electromagnetic torque, n _p : number of pole pairs.

In the start-up stage of the doubly-fed wind turbine, no-load grid-connection control is required to make the motor smoothly cut into the grid, and the amplitude of each electrical component is required to be as small as possible. When the generator is no-load, the stator current components are all 0, that is, i _ds = i _qs = 0, and the no-load mathematical model of DFIG is as follows:

T _e = 0 (6)

In the formula, T ₀ —generator no-load torque.

Since this vector control is the orientation of the stator flux linkage, and the voltage drop of the stator resistance at the power frequency is much lower than the voltage drop of the reactance and the back EMF of the motor, the motor stator resistance can be ignored in the calculation.

From this, the simplified equation can be obtained as follows:

From (8) to (10), the basic principle of DFIG no-load grid-connected control can be determined. After the doubly-fed generator is integrated into the power grid, the motor mainly adjusts the active and reactive power.

(2) Fractional operator and its rationalization

Precisely, fractional calculus should be called "non-integer order" integrals. The order of fractional calculus can be in the form of fractions, and theoretically it can be extended to complex and even irrational order. The application of fractional order is discussed.

This example uses the Grunwald-Letnikov fractional calculus definition as the theoretical basis:

Based on this definition, fractional order rationalization is carried out. According to formula (11), the calculation algorithm of fractional order differential is determined as:

为函数的多项式系数。假设步长h足够小，根据(12)可直接求出函数数值微分的近似值，并可以证明其计算精度为o(h)。当f(t)的函数表达式确定时，可以直接由式(11)计算其出分数阶导数。In the formula,

are the polynomial coefficients of the function. Assuming that the step size h is small enough, the approximate value of the numerical differentiation of the function can be directly obtained according to (12), and it can be proved that its calculation accuracy is o(h). When the functional expression of f(t) is determined, the fractional derivative can be calculated directly from equation (11).

Since the doubly-fed wind power generation system is a strongly coupled and nonlinear system, it is difficult to obtain its accurate functional expression. For the doubly-fed wind power generation system, the continuous transfer function method is used to rationally realize the fractional-order operator, that is, through a rational transfer function, it can effectively simulate the fractional-order operator in terms of amplitude-frequency characteristics and phase-frequency characteristics.

In this example, the improved Oustaloup filtering algorithm is used to construct a fractional operator:

The above fractional operator is the prior art, and the theoretical basis is taken from the series of papers by Oustaloup.

By convention, the weighting parameters take values b=10 and d=9. The zeros, poles and gains of the above fractional differential operator can be calculated as follows:

In order to ensure the stability of the algorithm, the cut-off frequency and parameter N are generally selected as [0.01, 100] and N=4, respectively.

(3) Fractional-order PI ^λ controller (use this fractional-order controller to control the grid-connected model of the fan)

The fractional-order PID controller can be written as PI ^λ D ^μ . After the differential and integral orders μ and λ are introduced into the fractional order controller, the adjustable parameters are expanded to make the setting range larger, and the controlled object can be controlled more flexibly. The transfer function of PI ^λ D ^μ is:

Because the differential term of PI ^λ D ^μ control is mainly for lead and lag correction, this correction is not needed in grid-connected control, so the PI ^λ controller is simplified and its control block diagram is shown in Figure 2. The fractional integral term I ^λ is implemented by the improved Oustaloup filtering algorithm in step (2).

The PI ^λ controller function is:

where λ is the integration order.

The whole system adopts double closed-loop control structure - power outer loop and current inner loop. First, the stator and rotor voltage and current of the motor are detected, and then the coordinates are transformed to calculate the stator flux linkage Ψ ₁ , stator active power P and reactive power Q; according to actual needs, set the stator active power command P* and reactive power The command Q* is compared, and the difference obtained is obtained through the fractional-order PI ^λ regulator to obtain the active component command i _qs * and the reactive component command i _ds * of the stator current, and then enter the current inner loop control.

A detailed RSC control model is constructed, and the Simulink block diagram is shown in Figure 3. The fractional-order control theory is applied to the no-load grid connection process of the doubly-fed induction generator to enhance its robustness.

In the PI controller of the rotor-side power control outer loop in the doubly-fed asynchronous induction motor control strategy, the adjustable integral term order λ is introduced to increase its control dimension, and it is designed as a fractional-order controller. The control system is made to deal with nonlinear control objects more flexibly, and the motor power oscillation under grid-connected transient process and fault disturbance can be effectively weakened.

Simulation

This application builds a grid-connected model of a doubly-fed induction motor under the MATLAB/simulink platform, and its RSC control structure model is shown in Figure 3. After the simulation experiment test, the results are shown in Table 1:

Table 1 Grid connection at different orders

After comprehensive comparison, it is finally determined that the fractional order control selection order is 0.8 order, and the grid connection process of the generator system is simulated. The active power and reactive power are shown in Figure 4 and Figure 5.

Figure 4 shows the active power graph in the process of grid connection. It can be seen from the figure that the fractional-order PI ^λ control is used. Although the time to fully enter the steady state is slightly higher than that of the integer-order PI ^λ control, the fractional-order control begins to trend at 0.17s. It is stable, and the integer-order PI control ends the large oscillation in 0.46s and begins to enter the steady state. The oscillation peak interval under the control of integer-order PI ^λ is [-4.47×106, 6.25×106], while the peak interval under the control of fractional-order PI ^λ is much lower than this, only [-3.11×106, 2.43×106]. Therefore, the system can enter the normal working state more smoothly, and reduce the impact on the power grid when it is connected to the grid.

Figure 5 is a graph of reactive power during grid connection. The state of the reactive power under the control of the integer-order and fractional-order reactive power PI ^λ is consistent with the active power. When the fractional-order PI ^λ control is used, the time to fully enter the steady state is slightly longer than that of the integer-order PI ^λ control, but the peak value of the oscillation decreases significantly, from [-4.77×106, 8.41×106] to [-1.018×104, 7.55 × 106].

Figure 6 shows the active power oscillation under the action of the two controllers when a three-phase short-circuit fault occurs in the steady-state operation of the system.

As shown in Figure 6, the active oscillation interval under the control of integer-order PI ^λ is [-1.349×106, 3.352×106], and under the control of fractional-order PI ^λ , the oscillation interval becomes [-1.349×106, 3.352×106] . This proves that the use of fractional order control can effectively suppress the oscillation.

The present application usually adopts double closed-loop PQ decoupling control for the rotor side of the traditional DFIG. The PI controller used in the power outer loop has shortcomings such as rough control and large overshoot, which affects the dynamic performance of the DFIG. In this paper, the fractional order theory is combined with the PI control technology, the adjustable order of the integral term is introduced, the control dimension is increased, the control performance of the doubly-fed induction motor is improved, and the grid-connected and short-circuit fault simulation research is carried out. The simulation results show that the fractional-order PI control method can effectively reduce the oscillation of active and reactive power in the startup phase, so that the system can enter the steady-state stage smoothly; in the short-circuit fault state, the fractional-order PI control method can also reduce the system oscillation.

The above descriptions are only preferred embodiments of the present application, and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims (8)

1. The fractional-order control method of the grid-connected process of the doubly-fed induction generator is characterized in that, comprising: Constructing a motor control model, the model is established based on a synchronous rotating coordinate system and consists of a voltage equation, a flux linkage equation and an electromagnetic torque equation; Constructing a fractional operator: Taking the Grunwald-Letnikov fractional calculus definition as the theoretical basis, determining the fractional differential calculation algorithm, obtaining the approximate value of the numerical differential of the function for the fractional differential calculation algorithm and proving its calculation accuracy, according to the Grunwald-Letnikov fraction The first-order calculus definition calculates its fractional derivative; For the doubly-fed wind power generation system, the continuous transfer function method is used to rationally realize the fractional-order operator, that is, through a rational transfer function, it can effectively simulate the fractional-order operator in terms of amplitude-frequency characteristics and phase-frequency characteristics; Establish a fractional-order PI ^λ controller: control the motor control model to control the no-load grid connection process of the doubly-fed induction generator; Among them, the doubly-fed induction generator is based on the mathematical model of the synchronous rotating coordinate system: Voltage equation:

Among them, ω _s is the slip angular velocity; ω _s =ω ₁ -ω _r , u _ds : d-axis component of stator voltage, u _qs : q-axis component of stator voltage, u _dr : d-axis component of rotor voltage, u _qr : rotor voltage q-axis component, i _ds : d-axis component of stator current, i _qs : q-axis component of stator current, i _dr : d-axis component of rotor current, i _qr : q-axis component of rotor current, R _s : stator-side resistance, R _r : Rotor side resistance, Ψ _qs : q-axis component of stator flux, Ψ _ds : d-axis component of stator flux, Ψ _qr : q-axis component of rotor flux, Ψ _dr : d-axis component of rotor flux, p: differential sign; In the start-up stage of the DFIG, it is necessary to carry out no-load grid-connection control to make the motor smoothly cut into the grid. The amplitude of each electrical component is required to be as small as possible, and the stator current component is 0 when the generator is no-load, that is, i _ds = i _qs = 0, the no-load mathematical model of the doubly-fed induction generator is as follows:

T _e = 0 (6)

In the formula, T ₀ —generator no-load torque, Te: electromagnetic torque, _{n p} : number of pole pairs, L _r : rotor self-inductance, L _m : mutual inductance. 2. The fractional-order control method for the grid-connected process of a doubly-fed induction generator as claimed in claim 1, wherein the fractional-order integral term I ^λ in the fractional-order PI ^λ controller is obtained by an improved Oustaloup filter algorithm. 3 . The fractional-order control method of the grid-connected process of the doubly-fed induction generator according to claim 1 , wherein the fractional-order operator is constructed by using an improved Oustaloup filtering algorithm. 4 . 4. the fractional order control method of the grid-connected process of the doubly-fed induction generator as claimed in claim 1, is characterized in that, the flux linkage equation:

In the formula, L _s : stator self-inductance. 5. the fractional order control method of the grid-connected process of the doubly-fed induction generator as claimed in claim 1, is characterized in that, the electromagnetic torque equation: T _e =n _p L _m ( _ids i _qr - i _qs i _dr ) (3). 6. the fractional order control method of the grid-connected process of the doubly-fed induction generator as claimed in claim 5, is characterized in that, the motor model simplified equation that ignores motor stator resistance is:

In the formula, Ψ ₁ is the stator flux linkage. 7. the fractional order control method of the grid-connected process of the doubly-fed induction generator as claimed in claim 1, is characterized in that, Grunwald-Letnikov fractional calculus definition:

The fractional differential calculation algorithm is:

In the formula, is the polynomial coefficient of the function, assuming that the step size h is small enough, the approximate value of the numerical differentiation of the function can be directly obtained according to (12), and it can be proved that its calculation accuracy is o(h), when the function expression of f(t) is determined. , the fractional derivative can be calculated directly from equation (11). 8. the fractional-order control method of the grid-connected process of the doubly-fed induction generator as claimed in claim 1, is characterized in that, the PI ^λ controller function is:

Among them, λ is the integration order; The control steps of the PI ^λ controller are: firstly detect the voltage and current of the stator and rotor of the motor, and then carry out coordinate transformation to calculate the stator flux linkage Ψ ₁ , stator active power P and reactive power Q; set the stator active power according to actual needs The power command P* and the reactive power command Q* are compared, and the difference obtained is obtained through the fractional-order PI ^λ regulator to obtain the active component command i _qs * and the reactive component command i _ds * of the stator current, and then enter the current inner loop control .

Images