阅读说明：本技术 一种风力发电t-s模糊鲁棒调度容错控制方法 (Wind power generation T-S fuzzy robust scheduling fault-tolerant control method ) 是由 游国栋 徐涛 苏虹霖 王军 沈延新 于 2019-08-08 设计创作，主要内容包括：本发明涉及一种风力发电T-S模糊鲁棒调度容错控制方法,其技术特点是：建立执行器故障的参数不确定非线性系统的T-S模糊模型；基于非线性T-S模糊模型采用模糊专用观测器对系统状态进行有效估计,针对系统的不确定性、不可测量状态和执行器故障构建模糊比例积分观测器,用于故障信号的精确重构,然后,利用平行分布补偿方法构建基于观测器故障重构的模糊调度容错控制器；通过基于观测器故障重构的模糊调度容错控制器进行风力发电T-S模糊鲁棒调度容错控制。本发明设计合理,其通过建立执行器故障影响下的非线性T-S模糊鲁棒调度容错控制方法的数学模型,实现了风力发电T-S模糊鲁棒调度容错控制功能,具有良好稳态和动态性能。(the invention relates to a wind power generation T-S fuzzy robust scheduling fault-tolerant control method, which is technically characterized by comprising the following steps: establishing a T-S fuzzy model of the actuator fault parameter uncertain nonlinear system; based on a nonlinear T-S fuzzy model, a special fuzzy observer is adopted to effectively estimate the state of the system, a fuzzy proportional integral observer is constructed aiming at the uncertainty, the unmeasured state and the actuator fault of the system and is used for accurately reconstructing fault signals, and then a fuzzy scheduling fault-tolerant controller based on observer fault reconstruction is constructed by utilizing a parallel distribution compensation method; and carrying out T-S fuzzy robust scheduling fault-tolerant control on wind power generation through a fuzzy scheduling fault-tolerant controller based on observer fault reconstruction. The method is reasonable in design, achieves the T-S fuzzy robust scheduling fault-tolerant control function of the wind power generation by establishing a mathematical model of the nonlinear T-S fuzzy robust scheduling fault-tolerant control method under the influence of the actuator fault, and has good steady-state and dynamic performances.)

1. A wind power generation T-S fuzzy robust scheduling fault-tolerant control method is characterized by comprising the following steps:

Step 1, establishing a T-S fuzzy model of an uncertain parameter nonlinear system of an actuator fault;

step 2, effectively estimating the state of the system by adopting a special fuzzy observer based on a nonlinear T-S fuzzy model, constructing a fuzzy proportional integral observer aiming at the uncertainty, the unmeasured state and the actuator fault of the system for accurately reconstructing fault signals, and then constructing a fuzzy scheduling fault-tolerant controller based on observer fault reconstruction by utilizing a parallel distribution compensation method;

And 3, carrying out T-S fuzzy robust scheduling fault-tolerant control on the wind power generation through a fuzzy scheduling fault-tolerant controller based on observer fault reconstruction.

2. the wind power generation T-S fuzzy robust scheduling fault-tolerant control method according to claim 1, characterized in that: the specific implementation method of the step 1 comprises the following steps: aiming at a nonlinear system with uncertainty, establishing a series of fuzzy rules, wherein each rule represents one subsystem, and considering the fault of a system actuator, a T-S fuzzy model of the whole parameter uncertainty nonlinear system is as follows:

wherein R is_iIs the ith fuzzy inferenceRule, z (t) ═ z₁(t)z₂(t)...z_k(t)]^TA pre-condition variable is represented by a pre-condition variable,Is a fuzzy set; i 1,2.. R is the number of the fuzzy inference rules of the system, j 1,2.. k, x (t) is belonged to RⁿIs the system state, u (t) is e to R^mrepresenting a control input, y (t) e R^prepresents the system output, A_i∈R^n×n，B_i∈R^n×mAnd C_i∈R^p×nis a parameter matrix of the system, Δ A_iAnd Δ B_iis a matrix of values of the real numbers that are uncertain,Representing a known actuator fault matrix, d (t) e R^q×1Is a time-varying signal of actuator failure (q < n), a precondition variable z₁(t)z₂(t)...z_k(t) is measurable and fault-independent.

3. The wind power generation T-S fuzzy robust scheduling fault-tolerant control method according to claim 1, characterized in that: the fuzzy proportional integral observer is as follows:

Wherein the content of the first and second substances,representing the estimated state of the unknown fuzzy observer, K_iIs an observer error matrix, G_iFor product to be designedfractional gain, y (t) represents the output vector,The final output of the FPIO is represented,representing the output estimation error; the estimated value of the actuator fault time-varying signal d (t) can be expressed as:

The fuzzy special observer is as follows:

Wherein the content of the first and second substances,Is the estimated state of the fuzzy special purpose observer,Is the estimated output of a fuzzy special observer, N_i∈K^n×1is an observer gain matrix;

The fuzzy scheduling fault-tolerant controller comprises:

wherein g (t) ═ g₁(t),g₂(t),...,g_k(t)]As a precondition variable, L_j∈R^m×nFor the feedback gain matrix of rule j, r (t) e k^m×1Is a reference input.

Technical Field

The invention belongs to the technical field of wind power generation, and particularly relates to a T-S fuzzy robust scheduling fault-tolerant control method for wind power generation.

background

With the development of scientific technology, many high-level comprehensive large-scale complex nonlinear systems are applied in a large scale, so that certain fault tolerance is provided for faults occurring in a control system, and the guarantee of the safety and the reliability of the control system becomes more and more important. Wind energy is an important green energy source, and the development and utilization of the wind energy become the focus of attention in the global energy field. Wind Energy Conversion Systems (WECS) exhibit random, switching behavior under random, intermittent wind forces, and are typically large complex nonlinear systems.

Currently, in the fault-tolerant control of the wind energy conversion system, scholars such as Stanton and the like research an H infinite fault-tolerant control method of the wind energy conversion system based on a random PWA model, and solve the problems of modeling and fault-tolerant control of the wind energy conversion system under the action of random wind load. The students such as Wuzhongqiang and Yang research a state feedback fault-tolerant controller designed based on an adaptive fault observer, and the good performance of normal operation can be still maintained when the system fails. Foreign scholars Bakri A E and the like perform T-S modeling on the doubly-fed wind power system, and fault-tolerant controllers are established respectively based on a fuzzy synovial observer and a fuzzy observer aiming at faults of an actuator and a sensor of the system. However, in the above control methods, the uncertainty generated by the system is not considered, so that the robust stability of the system is not sufficiently guaranteed.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides a wind power generation T-S fuzzy robust scheduling fault-tolerant control method which is reasonable in design and has good steady-state and dynamic performances.

The technical problem to be solved by the invention is realized by adopting the following technical scheme:

A wind power generation T-S fuzzy robust scheduling fault-tolerant control method comprises the following steps:

Step 1, establishing a T-S fuzzy model of an uncertain parameter nonlinear system of an actuator fault;

Step 2, effectively estimating the state of the system by adopting a special fuzzy observer based on a nonlinear T-S fuzzy model, constructing a fuzzy proportional integral observer aiming at the uncertainty, the unmeasured state and the actuator fault of the system for accurately reconstructing fault signals, and then constructing a fuzzy scheduling fault-tolerant controller based on observer fault reconstruction by utilizing a parallel distribution compensation method;

And 3, carrying out T-S fuzzy robust scheduling fault-tolerant control on the wind power generation through a fuzzy scheduling fault-tolerant controller based on observer fault reconstruction.

the specific implementation method of the step 1 comprises the following steps: aiming at a nonlinear system with uncertainty, establishing a series of fuzzy rules, wherein each rule represents one subsystem, and considering the fault of a system actuator, a T-S fuzzy model of the whole parameter uncertainty nonlinear system is as follows:

Wherein R is_iIs the ith fuzzy inference rule, z (t) ═ z₁(t)z₂(t)...z_k(t)]^Ta pre-condition variable is represented by a pre-condition variable,Is a fuzzy set; i 1,2.. R is the number of the fuzzy inference rules of the system, j 1,2.. k, x (t) is belonged to RⁿIs the system state, u (t) is e to R^mRepresenting a control input, y (t) e R^prepresents the system output, A_i∈R^n×n，B_i∈R^n×mAnd C_i∈R^p×nis a parameter matrix of the system, Δ A_iAnd Δ B_iIs a matrix of values of the real numbers that are uncertain,representing a known actuator fault matrix, d (t) e R^q×1Is a time-varying signal of actuator failure (q < n), a precondition variable z₁(t)z₂(t)...z_k(t) is measurable and fault-independent.

The fuzzy proportional integral observer is as follows:

Wherein the content of the first and second substances,representing the estimated state of the unknown fuzzy observer, K_iis an observer error matrix, G_iFor the integral gain to be designed, y (t) represents the output vector,The final output of the FPIO is represented,Representing the output estimation error; the estimated value of the actuator fault time-varying signal d (t) can be expressed as:

The fuzzy special observer is as follows:

Wherein the content of the first and second substances,is the estimated state of the fuzzy special purpose observer,Is the estimated output of a fuzzy special observer, N_i∈K^n×1Is an observer gain matrix;

the fuzzy scheduling fault-tolerant controller comprises:

Wherein g (t) ═ g₁(t),g₂(t),...,g_k(t)]As a precondition variable, L_j∈R^m×nFor the feedback gain matrix of rule j, r (t) e k^m×1Is a reference input. The invention has the advantages and positive effects that:

The invention has reasonable design, provides a robust fault-tolerant control idea based on system state estimation and fault reconstruction aiming at a parameter uncertain nonlinear system with an actuator fault, describes the nonlinear system by adopting a T-S fuzzy model through the topology of fault-tolerant control of a wind energy conversion system, simultaneously considers the uncertainty and the unmeasured state variable of the system, establishes a mathematical model of a nonlinear T-S fuzzy robust scheduling fault-tolerant control method under the influence of the actuator fault, realizes the T-S fuzzy robust scheduling fault-tolerant control function of wind power generation, and has good steady-state and dynamic performances.

Drawings

FIG. 1 is a block diagram of the overall structure of a Wind Energy Conversion System (WECS) with a doubly-fed induction generator (DFIG);

FIG. 2 is a membership function z₁(t) schematic representation.

Detailed Description

The embodiments of the present invention will be described in detail with reference to the accompanying drawings.

the design idea of the invention is as follows: the invention designs a fuzzy robust scheduling fault-tolerant control strategy (FRSFTC) based on a T-S fuzzy model, which takes the unmeasured state variable and uncertainty in a system into consideration, effectively estimates the system state through a fuzzy observer based on the T-S fuzzy model of the system, designs a fuzzy PI observer aiming at the actuator fault of the system, realizes the accurate reconstruction of a fault signal, adopts compensation control according to the estimated fault information, and utilizes a Parallel Distributed Compensation (PDC) method to construct a fuzzy scheduling fault-tolerant controller, thereby realizing the purposes of actively fault-tolerant the actuator fault and ensuring the robust stability of the system.

a wind power generation T-S fuzzy robust scheduling fault-tolerant control method comprises the following steps:

step 1, establishing a T-S fuzzy model of the actuator fault parameter uncertain nonlinear system:

in this step, a series of fuzzy rules of the non-linear system with uncertainty is established for the non-linear system with uncertainty. Each rule represents a subsystem, and thus, the T-S model structure of the uncertain parameter fuzzy system equation (1) is described as follows:

Wherein R isⁱRepresenting the ith fuzzy inference rule, z (t) ═ z₁(t)z₂(t)...z_k(t)]^Ta pre-condition variable is represented by a pre-condition variable,is a fuzzy set, i is 1,2, R represents the number of fuzzy inference rules of the system, j is 1,2ⁿis the state of the system, u (t) is e.g. R^mRepresenting a control input, y (t) e R^pis the system output, A_i∈R^n×n、B_i∈R^n×mAnd C_i∈R^p×nis a parameter matrix of the system, Δ A_iAnd Δ B_iIs an indeterminate real-valued matrix.

Assuming that the uncertainty norm of the system is bounded, after defuzzification, the entire state equation for the entire fuzzy T-S system equation can be obtained:

Wherein Denotes z_j(t) in fuzzy setsMembership function of, h_i(z (t)) is the weight of rule i, h_i(z (t)) and u_i(z (t)) satisfies the following condition:

In view of the failure of the system actuator, the system model (2) is rewritten as follows:

WhereinRepresenting a known actuator fault matrix, d (t) e R^q×1Is a time-varying signal of actuator failure (q < n), a precondition variable z₁(t)z₂(t)...z_k(t) is measurable and fault-independent.

the invention makes

Furthermore, the system model (3) can be expressed as

The system uncertainty fuzzy rule is described as follows:

The output of the ambiguity uncertainty can be expressed as:

Wherein

s＝2^crepresenting the number of fuzzy rules; scalar c represents the number of uncertain elements in deltaa and deltab,AndAnd is given by:

defining fuzzy weights:

For convenience of simplification, we respectively turn h_l(ΔA,ΔB) And u_i(z (t)) is written as h_land u_ifrom (4), (5) and (6), the system model (3) becomes:

and 2, effectively estimating the state of the system by adopting a fuzzy special observer (FDO) based on a nonlinear T-S fuzzy model, constructing a Fuzzy Proportional Integral Observer (FPIO) aiming at the uncertainty, the unmeasured state and the actuator fault of the system, realizing the accurate reconstruction of a fault signal, and then constructing a fuzzy scheduling fault-tolerant controller based on observer fault reconstruction.

for T-S fuzzy system actuator faults, a Fuzzy Proportional Integral Observer (FPIO) is designed based on a T-S fuzzy model. Wherein the ith fuzzy rule is:

the estimated value of the actuator fault time-varying signal d (t) can be expressed as:

Whereinrepresenting the estimated state of FPIO, K_iRepresenting observer error matrix, G_idenotes the integral gain to be designed, y (t) denotes the output vector,The final output of the FPIO is represented,Indicating the output estimation error.

After deblurring, the final output of the FPIO (12) is described as follows:

Since the system state variables cannot be directly measured, it is necessary to design a fuzzy observer to reconstruct the system state. Assuming that the state of the system model (1) is observable, the following fuzzy special observer (FDOS) is designed based on the T-S fuzzy model:

WhereinIs the estimated state of a fuzzy special purpose observer (FDOS),Is the estimated output of FDOS, N_i∈K^n×1Is the observer gain matrix. The fuzzy observer available after defuzzification is described as:

a local state feedback controller based on a T-S fuzzy model (1) is designed for each subsystem based on a Parallel Distributed Compensation (PDC) principle on the assumption that the state of the fuzzy system (1) is locally controllable. The jth rule of the controller input is:

Rj:Ifg₁(t)isM_1jand g₂(t)isM_2j...and g_k(t)is M_kj,then

Wherein g (t) ═ g₁(t),g₂(t),...,g_k(t)]as a precondition variable, L_j∈R^m×nFor the feedback gain matrix of rule j, r (t) e k^m×1for reference input, after defuzzification, the overall Fuzzy Robust Scheduling Fault Tolerant Controller (FRSFTC) can be expressed as:

the preconditions of the designed fuzzy robust scheduling fault-tolerant controller are the same as the preconditions of the uncertainty of the system model. The rule is expressed as:

u_j(g (t)) is abbreviated to u_jthe control output of the designed FRSFTC is described as:

and 3, realizing the T-S fuzzy robust scheduling fault-tolerant control function of the wind power generation through the fuzzy robust scheduling fault-tolerant controller, and achieving the purposes of actively fault-tolerant of the actuator faults and ensuring the robust stability of the system.

The gains of the controller and the observer are obtained by solving an LMI (linear matrix inequality) method, and the stability of the wind energy conversion system is proved.

Closed loop equation (20) and estimation error equation (19) defining the system:

Substituting equation (18) into equation (20), the closed loop equation of system equation (20) in the event of actuator failure can be expressed as:

Order to

According to the formulae (19) and (22), it is readily obtained:

the system state error is estimated as:

Order to

Then:

Assuming d (t) as a time-varying fault signal, we have:

combining equations (23), (24), (26) and (27) results in a new augmented blur system (28):

WhereinS_i＝[B_i 0 0 0]，Y＝[0 0 0 I]，And

the essential condition for ensuring the closed loop stability of the system by the effectiveness of the method is proved by the Taylor series and Lyapunov (Lyapunov) function stability theory.

Theorem 1. for uncertain parameters, actuator fault fuzzy control system (28) if inequality μ [ t H_ijlΤ^-1]≤-||ΤΔH_ijlΤ^-1If | max- τ is true, the system (28) is globally asymptotically stable, where τ is designed to be a positive value, a suitable dimensional transformation symmetric matrix.

And (3) proving that: for the system (28), it can be obtained according to Taylor's formula

Wherein λ_max(.) represents the largest eigenvalue and denotes the conjugate transpose. It is assumed that the fault is bounded,0≤d_amaxless than + ∞. From now onwe can therefore get:

||φ(t)||≤d_amax,0≤d_amax＜+∞ (31)

If the following inequality

μ[ΤH_ijlΤ^-1]≤-||ΤΔH_ijlΤ^-1||max-τ (32)

τ is a non-zero positive constant. According to equations (29) and (31), it is possible to obtain:

t₀< t is an arbitrary initial time. If (31) is satisfied, the system (28) becomes stable in the global asymptote when t → ∞. Assuming that r (t) is 0, Φ (t) is 0 and r (t) ≠ 0, Φ (t) ≠ 0, from (31) and (32), it can be easily seen that:

Whereinwhen (35) is bounded, we can get that the system is also bounded when r (t) and φ (t) are bounded, so the system is stable.

Theorem 1 for the fuzzy control system shown in System (28), assume that there is a matrix X_i、M_a11、W_jand O_iAnd the controller and observer gains of the fuzzy system are set to L_j＝W_jlM_a11 ^-1、Andand satisfyThe following inequality:

the closed loop system (28) then settles asymptotically globally.

and (3) proving that: fault tolerant control aims at finding the gain L of the controller and observer_j、N_i、K_iAnd G_iTo asymptotically converge towards zero, if r (t) ≠ 0, φ (t) ≠ 0 and to ensure a bounded state based on (31), if r (t) ≠ 0, φ (t) ≠ 0, this problem is translated into finding the matrix P to verify V (t) < 0.

The Lyapunov function is defined as follows:

V(t)＝X(t)^TPX(t) (37)

From the system (28), matrix H_ijl、S_i、p and Y can be represented as:

Thus, the matrix (37) can be re-expressed as:

Equations (39) and (40) are a set of non-linear matrix inequalities, which are not a linear matrix inequality, assuming P₁＝diag(P_a11,P_a22) Application variable W_j＝M_a11L_jvariable O_i＝P_a22N_iand variablesOn the left, multiplied by(39) and right multiplication byan LMIS (36) in the subscription is available.

We will now specifically describe the analysis of Wind Energy Conversion System (WECS) actuator faults as an example.

first, studying a dynamic mathematical model of a Wind Energy Conversion System (WECS) according to Betz theory, the mechanical power captured by a wind turbine from wind energy is:

P_wt＝0.5ρR²V³C_p(λ,β) (41)

Where ρ represents air density, R represents wind turbine rotor radius, V is wind speed, β is pitch angle, λ is Tip Speed Ratio (TSR), Ω_ris the turbine speed of the low-speed shaft, C_pthe converted power coefficient wind energy is converted into mechanical energy. TSR λ is the ratio of the linear blade tip speed to the wind turbine wind speed, which is defined as λ Ω_rAnd R/V. Output torque T of wind turbine_wtCan be expressed as:

T_wt＝P_wt/Ω_r＝0.5ρR²V³C_p(λ,β)/Ω_r (42)

when the wind speed is constant, the mechanical power output of the wind turbine depends only on the power coefficient C_p. If the pitch angle beta remains constant, the power coefficient C_pDetermined only by TSR λ. For a particular wind turbine, there is a single optimum TSR λ_opt. At this time, C_pmaxIs defined as the maximum wind energy capture coefficient. The maximum capture of wind energy can be achieved by adjusting the electromagnetic torque of the generator to follow the change in wind speed to a maximum value. With fixed pitch control β -0 at rated wind speed, i.e. TSRC_p(λ,β)＝C_p(λ)、C_pmaxAnd 0.48, the optimum TSR.

a Wind Energy Conversion System (WECS) mainly comprises a wind driven generator, a transmission system, a generator and a power grid. Wind turbines capture wind energy and convert it into mechanical energy through the rotation of the wind turbine, which drives the rotor of a doubly-fed induction motor through a drive train to generate electricity through a converter to a grid.

By using the kinematic equation of the transmission system, a dynamic model of the wind power generation system can be obtained:

WhereinD_rand D_grepresenting the damping constants, τ, of the rotor and generator, respectively_gis the time constant, K_lsRepresenting the equivalent torsional stiffness of the low speed shaft, D_lsis the damping constant, T, of the equivalent low speed shaft_hIs high-speed shaft torque, J_rIs the moment of inertia of the rotor, J_gIs the generator moment of inertia, T_gIs the generator torque, T_g,refIs the torque required by the generator, omega_gis the mechanical generator speed, n_bis the transmission ratio.

from the kinematic model (43), the standard form of the WECS equation of state can be expressed as:

Wherein x (t) ═ x₁ x₂ x₃ x₄]^T＝[Ω_r Ω_g T_h T_g]^T，u(t)＝T_g,ref。

the following description of the application of T-S obfuscation to WECS follows. By looking at the function A (x) of the system matrix, z is defined₁(t)＝Ω_rAnd z₂(t)＝Ω_gAre the precondition variables. Then select z₁(t) and z₂(t) membership functionAnd

For simplicity, the membership functions of the two fuzzy subsets can be represented by the formula (45):

wherein the variable z_jtFrom its upper limit value z_jminAnd a lower limit value z_jmaxand (4) limiting. Member function z₁(t) As shown in FIG. 2, each member function also represents the model uncertainty for each subsystem, and the member function z₂(t) is achieved in the same manner.

The T-S fuzzy model of the parameter uncertainty and actuator failure of the wind energy conversion system (44) can be represented by the following 4 rules:

Rule 1:Ifz₁(t)isF₁andz₂(t)isF₂，then

Rule 2:Ifz₁(t)isF₁andz₂(t)isthen

Rule 3:Ifz₁(t)isandz₂(t)isF₂，then

Rule 4:Ifz₁(t)isandz₂(t)isthen

Thus, the global fuzzy model of the wind energy conversion system may be expressed as:

ΔA_iand Δ B_iand Δ B_i0 represents a bounded uncertainty for the system parameter. Taking into account the parameter uncertainty at the nominal value Δ J_gIn the range of 30% to 50%. From formula (6), we can get c-6 and s-64. Thus, the ambiguity uncertainty is given byIt is given. Then, equation (28) of the WECS fuzzy model can be expressed as:

As can be seen from equation (18), the system control output is:

In conclusion, a wind power generation T-S fuzzy robust scheduling fault-tolerant control system can be obtained, as shown in FIG. 1.

The invention can realize the T-S fuzzy robust scheduling wind power generation system fault-tolerant control technology by considering the interference of various factors such as external interference, uncertainty of system parameters and the like on the wind power generation system. The method takes the unmeasured state variable and uncertainty in the system into consideration, effectively estimates the system state through a fuzzy observer based on a system T-S fuzzy model, designs a fuzzy PI observer aiming at the actuator fault of the system, realizes the accurate reconstruction of a fault signal, adopts compensation control according to the estimated fault information, designs a robust scheduling fault-tolerant controller by using a Parallel Distributed Compensation (PDC) method, and realizes the purposes of actively fault-tolerant the actuator fault and ensuring the robust stability of the system.

It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.