Fuzzy finite time optimal synchronous control method of fractional order permanent magnet synchronous generator
阅读说明：本技术 分数阶永磁同步发电机的模糊有限时间最优同步控制方法 (Fuzzy finite time optimal synchronous control method of fractional order permanent magnet synchronous generator ) 是由 罗绍华 李俊阳 李�昊 李少波 于 2021-10-26 设计创作，主要内容包括：本发明涉及一种分数阶永磁同步发电机的模糊有限时间最优同步控制方法,属于发电机技术领域。建立了具有电容电阻耦合的分数阶主动、从动永磁同步发电机之间的同步模型。通过动力学分析充分揭示了系统存在包括混沌振荡在内的丰富动力学行为,并通过设计数值方法给出了稳定性和非稳定性边界。然后,在分数阶反步控制理论框架下,提出了一种融合分层二型模糊神经网络、有限时间命令滤波器、有限时间预设性能函数的模糊有限时间最优同步控制方案。稳定性分析证明了闭环系统的所有信号在成本函数最小的情况下是有界的。最后,数值模拟结果验证了本发明的可行性和优越性。(The invention relates to a fuzzy finite time optimal synchronization control method for a fractional order permanent magnet synchronous generator, and belongs to the technical field of generators. A synchronous model between the fractional order driving permanent magnet synchronous generator and the fractional order driven permanent magnet synchronous generator with capacitance-resistance coupling is established. The dynamic analysis fully reveals that the system has rich dynamic behaviors including chaotic oscillation, and gives a stability and non-stability boundary by designing a numerical method. Then, under the framework of a fractional order backstepping control theory, a fuzzy finite time optimal synchronous control scheme which integrates a hierarchical two-type fuzzy neural network, a finite time command filter and a finite time preset performance function is provided. The stability analysis demonstrates that all signals of the closed loop system are bounded with a minimum cost function. Finally, the feasibility and the superiority of the invention are verified by the numerical simulation result.)
1. The fuzzy finite time optimal synchronous control method of the fractional order permanent magnet synchronous generator is characterized by comprising the following steps of: the method comprises the following steps:
s1: modeling a system;
s2: designing a numerical method and applying the numerical method to solving a nonlinear fractional order system;
s3: and establishing a hierarchical two-type fuzzy neural network and designing a controller.
2. The fuzzy finite time optimal synchronization control method of fractional order permanent magnet synchronous generator according to claim 1, characterized by: the S1 specifically includes: the wind energy conversion system consists of a wind turbine, a permanent magnet synchronous generator and three converters; the system comprises a diode bridge rectifier, a dc/dc boost converter and an inverter in sequence; the electric energy generated by the permanent magnet synchronous generator is transmitted to a power grid through a converter;
in combination with local aerodynamic properties, the power produced by a wind turbine is expressed as:
in the formula, rho, R, omegar，β，ωrAnd vwExpressed as air density, turbine radius, speed, blade angle, speed, wind speed, Cp(ωrR/vwBeta) is expressed as a turbine power coefficient;
according to the rotation law, a mechanical fractional order model of the permanent magnet synchronous generator is given:
in the formula, alpha, J, Tt，Tg，And b represents fractional order coefficient, system inertia, turbine torque, generator torque, time, and viscous friction coefficient, respectively;
defining the torque of the electromagnetic generator:
in the formula LdAnd LqRepresenting d-and q-axis inductances, idAnd iqRepresenting d-axis and q-axis stator currents, and p and phi representing the number of pole pairs and magnetic flux of permanent magnets in the three-phase permanent magnet synchronous generator;
the fractional order model of the permanent magnet synchronous generator in the synchronous rotating d-q reference system is expressed as:
in the formula Rs，VdAnd VqRepresenting stator resistance, d-axis and q-axis stator voltages;
in the Laplace domain, the fractional order integrator in the form of a linear approximation with zero pole pairs is represented by a transfer function with a slope of-20 m dB/decade in a Bode plot:
in the formulaPf，ωmaxAnd dfRespectively representing angular frequency, bandwidth and difference between the actual line and the approximate line, QiAnd PiRepresenting the zero and the pole of the singular function;
the stator winding is symmetrical to obtain L ═ Ld＝Lq(ii) a The fractional order model of the permanent magnet synchronous generator is defined by the formula (2) and the formula (4) as follows:
in the formula(ii) a Caputo fractional derivative representing when α >0 and the origin is at the origin;
by introducing a new variable x1＝Lωr/Rs，x2＝pLφiq/bRs，x3＝pLφid/bRsThe normalized fractional order model of the permanent magnet synchronous main motor is written as:
in the formulaμ＝-pφ2/bRs，σ＝3Lb/2JRs，ρ＝bL/JRs，x1，x2，x3，t，uq，udAnd TLRespectively representing normalized angular velocity, q-axis current, d-axis current, time, q-axis voltage, d-axis voltage and load torque, wherein sigma, rho and mu represent system parameters;
using the Heaviside function H (T-T)g) Establishing a fractional order model of the driven permanent magnet synchronous generator:
in the formula tg，κ1And kappa2Respectively representing the initial synchronization time, the capacitive coupling and the resistive coupling, u2And u3Represents a control input;
defining the synchronization error as e1＝y1-x1，e2＝y2-x2And e3＝y3-x3(ii) a Subtracting the equation (7) from the equation (8) to obtain:
definition 1: for a sufficiently differentiable function F (t), the Caputo fractional derivative is written as:
in the formulaRepresenting Euler gamma function, n-1 < alpha < n,
the laplace transform is performed on (10) to obtain:
for any continuous function, when 0 < alpha < 1 and F1(t) and F2(t) in the interval [0, t ]V]When inside, exist
When F is present1(t)＝F2(t) deriving the following inequality:
introduction 1: for anyIf constant cs＞0，ds＞0，Is an arbitrary real function, then the non-column equation holds
Definition 2: the minimum performance cost function is as follows:
in the formulaU，Respectively representing a penalty function, an optimal control input and an N-order matrix,
3. the fuzzy finite time optimal synchronization control method of the fractional order permanent magnet synchronous generator according to claim 2, characterized in that: the S2 specifically includes:
for a fractional order differential equation with a given function g (·,)yfThe fractional derivative quotient of (t) is defined as an infinite series, i.e.:
wherein h >0 andrepresents a step size and a coefficient satisfying a condition based on a Euler-Gamma function of
In the interval [0T]Middle definition equidistant grid tnNh, N ═ 0,1, …, N, where N ═ T/h; then obtaining the value at t ═ tnIntegral expression of time:
let constantIn each subinterval [ t ]j,tj+1]Vector field of up-approximation(18) The numerical values of formula are rewritten as:
the system parameters of the master/slave permanent magnet synchronous generator fractional order model are set to be three working conditions:
working condition 1: α is 0.99, σ is 3, ρ is 4, μ is 25, TLInitial condition x is 01(0)＝0.1，x2(0)＝0.9，x3(0)＝20；
Working condition 2: α is 0.99, σ is 17, ρ is 16, μ is 25, TLInitial condition x is 01(0)＝1.5，x2(0)＝0.5，x3(0)＝20；
Working condition 3: α is 0.99, σ is 5.5, ρ is 5.5, μ is 20, TLInitial condition x is 01(0)＝0.1，x2(0)＝0.1，x3(0)＝3；
By introducing the formula (5), approximate transfer functions of different fractional orders are given
The error values in the formula are respectively less than or equal to 0.1dB, less than or equal to 0.2dB and less than or equal to 0.4 dB.
4. The fuzzy finite time optimal synchronization control method of the fractional order permanent magnet synchronous generator according to claim 3, characterized in that: the S3 specifically includes:
the hierarchical two-type fuzzy neural network consists of five layers, namely an input layer, a membership layer, a rule layer, a hierarchical layer and an output layer;
at the second level, the upper and lower membership grades are written as:
in the formulaRespectively representing the center, upper width and lower width of the jth membership grade of the ith input;
in the rule layer, the up/down trigger rule is calculated as:
in the fourth layer, YRAnd YLExpressed as a set reduced order center:
in the formulaAndreferred to as the weight of the image to be displayed,T jandthe upper and lower triggering degrees of the jth rule are shown, M shows the number of fuzzy rules,
in the output layer, the fuzzy output in vector form is derived:
wherein w ≡ wR wL]，
ξ≡[ξR ξL]
By calling (26), the hierarchical two-type fuzzy neural network realizes high-precision approximation to any unknown but bounded function on a compact set, then
In the formulaN denotes the input number, e (X) >0 denotes the approximation error, ΩwAnd DXA tight set of appropriate bounds for w and X, respectively; introducing an optimal parameter w*The parameter satisfies the condition: represents an approximation of h; definition ofWherein w*Representing the amount of labor for analysis;
in order to avoid exponential increase of the rule number, two criteria of tracking error reduction rate and completeness of a fuzzy rule epsilon are adopted to execute structural adjustment of the hierarchical two-type fuzzy neural network; the tracking error reduction rate corresponds to the derivative of the square of the tracking error between the drive system output and the response system; the completeness of the fuzzy rule is defined as that 'at least one fuzzy rule ensures that the trigger strength in the operation range is not less than epsilon'; if the membership grade is equal to or greater than 0.5 in the hierarchy, saving them; otherwise, deleting the unimportant rules in the self-structure algorithm;
in order to improve the solving speed and simplify the structure, the hierarchical two-type fuzzy neural network is transformed as follows:
wTξ(X)≤ζξT(X)ξ(X)/2b2+b2/2 (28)
in the formula, zeta | | w | | non-woven phosphor2B is greater than 0; exist ofWhereinAn estimated value of ζ is represented,
to avoid system performance degradation and suppress the convergence behavior of error variables, a positive and strictly monotonically decreasing finite time preset performance function:
in the formula aiWhere i is 0, …,3 denotes a design parameter, T0Andrepresenting a convergence time and a convergence boundary; the performance function satisfies the following constraints:
in the formula beta0Representing an initial value of a finite time preset performance function;
when the constraint signal is matched with the convergence rate of the preset performance function, limiting the larger overshoot of the control output by the limiting condition in the initial stage; for a predetermined parameter, β0，T0Andby properly selecting four design parameters, a satisfactory finite time preset performance function can be easily obtained;
to achieve a faster response, a fractional order time-limited command filter based on a first order Levant differentiator is presented:
in the formulaAndwhich is indicative of the state of the command filter,input signal representing a command filter, ci,1And ci,2Denotes a positive design constant, a positive constant Gi,1And Gi,2Satisfy | Zi,1-αr|≤Gi,1Andconditions;
obviously, by appropriate selection of ci,1，ci,2When the input noise is completely suppressed during a transient of limited time, there is Zi,1＝αrAnd
introducing error variables
In the formulaAnd Zi,1I is 2,3 same and represents the output of the fractional order command filter;
the compensated tracking error is expressed as:
vi＝zi-θi,i＝1,2,3 (33)
in the formula [ theta ]iRepresenting a compensation signal between the virtual control and filtered signals;
the error variables have the following inequalities:
wherein 0 <ρ,
Defining a smooth and derivable function S1(z1) The above inequality can be rewritten as an unconstrained form:
e1＝β(t)S1(z1) (35)
it is to be noted that it is preferable that,belonging to one of the above smooth and derivable functions;
the inverse of (35) is then written as:
in the formula1(t)＝e1(t)/β(t)；
(36) The fractional derivative of equation is derived as:
in the formula
By using the formulas (9) and (37), there are:
due to disturbances from wind speed, generator temperature, stator resistance, friction coefficient and work load, etc., where f1＝-ρe1，f2＝-e2-y1y3+x1x3-(κ1e1+κ2e2)H(t-tg)+μe1，f3＝-e3+y1y2-x1x2Are all considered unknown non-linear functions;
based on a fractional order backstepping control principle, the controller is designed to consist of three steps;
the first step is as follows: considering f1Is estimated on a compact set, i.e. using the above hierarchical type 2 fuzzy neural network, for the convenience of controller design
Wherein (·) represents (x)1,x2,x3) Abbreviations of (a);
the first Lyapunov candidate function is chosen as:
get V1The fractional derivative of (a) yields:
in the formulaAndrepresenting virtual control and σ Upper bound, ζ1＝||w1||2And b1＞0；
By definition 2, the following cost function is designed to achieve its minimum;
in the formula uoi，And kappaiRespectively representing optimal control input, normal number and design parameters; to compensate for the estimation error of the hierarchical two-type fuzzy neural network, the following inequality is used
The optimal control input is designed asWhere P isiBelonging to algebraic Riccati equationkiSolutions of > 0;
deriving optimal control inputs:
selecting virtual control and its adaptive law and compensation signal
In the formula k1＞0，l1＞0，γ1＞0，s1＞0，0＜γ＜1；
By substituting (14) for (44) - (46) to (41) to obtain
In the formulaAnd
the second step is that: to solve the aforementioned unknown non-linear function f2It is approximated with high precision using a hierarchical two-type fuzzy neural network in the form:
selecting a second Lyapunov candidate function:
V2the fractional order derivative of (a) is written as:
zeta in the formula2＝||w2||2And b2＞0；
The q-axis control input and the adaptive law and compensation signals are designed as follows:
in the formula k2＞0，l2＞0，γ2>0 and s2＞0；
Utilization sum-, further simplified to:
in the formulaAnd
the third step: for processing unknown non-linear functions f3By usingA hierarchical two-type fuzzy neural network with very high accuracy and repeatability to estimate:
due to the adoption of the directional control of the magnetic field,is equal to zero; defining the last Lyapunov candidate function:
calculating V3Has a fractional order derivative of
Zeta in the formula3＝||w3||2And b3＞0；
The d-axis control input, adaptive law and compensation signals are designed as follows:
in the formula k3＞0，l3＞0，γ3>0 and s3＞0；
Obtaining:
in the formulaAnd
Technical Field
The invention belongs to the technical field of generators and relates to a fuzzy finite time optimal synchronous control method of a fractional order permanent magnet synchronous generator.
Background
The permanent magnet synchronous generator is one of the most important electric power devices, and can convert wind energy into mechanical work, and the magnetic rotor is driven to rotate by the mechanical work to output alternating current. At present, it is widely used for high-speed power generation, renewable energy power generation, wind power generation, and the like. Permanent magnet synchronous generators of wind power plants are typically operated in harsh environments, and fluctuations from local wind speeds, generator temperatures, magnetic fields, friction and workload are unavoidable. The fluctuations may cause various nonlinear dynamics phenomena such as chaotic oscillation, multi-stability, bifurcation and the like of the permanent magnet synchronous generator. These fluctuations can lead to a degradation of the performance of the permanent magnet synchronous generator and even to a damage of the generator without taking effective measures. In addition, with the improvement of living standard, people have higher and higher requirements on the reliability and safety of the permanent magnet synchronous generator system. Therefore, how to accurately model, reveal the stable and unstable boundaries of the system motion, and provide a fuzzy finite time optimal synchronous control scheme between the fractional order driving and driven permanent magnet synchronous generators in a preset performance range has important significance and challenge.
Compared with the integral order calculus, the fractional order calculus can establish a more accurate model, more truly describe the dynamic characteristics of the actual engineering system, and simultaneously, the optimal dynamic response can be obtained by adjusting the fractional order value of the system. Westerlunand Ekstam experimentally demonstrated fractional order values for various capacitor media and found a close fractional order relationship between current and voltage. Luo et al experimentally verified the control performance of a fractional order system using a fractional order capacitive membrane model. For modeling and dynamic analysis of fractional order generators, there have been sporadic literature reports over the past few decades. Xu et al established a fractional order model of the hydro-generator set system and performed nonlinear dynamical analysis using six typical fractional order values. Dynamic analysis is carried out on the fractional order permanent magnet synchronous generator by Borah and Roy, and a single-state prediction control method is provided. Ardjal et al established a fractional order model of a wind energy conversion system and designed a nonlinear cooperative controller of a generator and a grid converter. These important findings have a certain heuristic effect on the current research. However, these models are limited to analyzing a single generator, do not consider the coupling strength between generators, and the corresponding kinetic analysis does not give stable and unstable boundaries and regions, and does not consider a system control problem such as finite time, uncertainty, performance constraints and optimization.
The problem of synchronization of intercoupled systems is of great concern in nonlinear science, for example in the field of power generation where synchronous control allows multiple generators to achieve their precise frequency coincidence in parallel mode. To better address this problem, a great deal of valuable work such as robust synchronization, adaptive synchronization, and optimal synchronization control continues to emerge. However, these works are limited to integer order systems that do not take into account the strength of the bi-directional and unidirectional coupling, and these systems do not produce complex non-linear dynamics when conditions fluctuate, deviating from the subject of the synchronization control in question. Sadeghi et al solved the problem of smooth synchronization of brushless doubly fed induction generators using machine models. Zhu et al solve the synchronization problem of the chaotic oscillation permanent magnet synchronous generator network by using a self-adaptive pulse control method. The above document does not disclose chaos and period intervals in an integer order generator. At the same time, due to perturbation and uncertainty of system modeling parameters, failure of these schemes without pre-configured performance is inevitable.
As an effective control method, the backstepping control is widely applied to integer-order uncertain nonlinear systems by fusing a fuzzy logic system or a neural network. Some researchers have extended this to the area of fractional order. In order to solve the inherent 'complexity explosion' problem of the backstepping control, whether the backstepping control belongs to an integer order or a fractional order, a first-order low-pass filter, a tracking differentiator and an observer are generally adopted to solve the problem. However, the coupled permanent magnet synchronous generator is completely different from a nonlinear mathematical model, the mathematical model cannot generate undesirable dynamic behaviors such as chaotic oscillation and multi-stability, and meanwhile, the fuzzy finite time optimal synchronous control problem is not mentioned in the work. Optimal control is an important research topic because it consumes less resources. Some scholars introduce the method into the backstepping control of the nonlinear system, and fully exert the advantages of the method. The finite time control of the nonlinear system has the advantages of high response speed, high convergence accuracy and the like, and is a research subject of interest. In a wind farm, a single-direction coupled fractional order permanent magnet synchronous generator system is a highly complex system with surrounding generator couplings. How to fuse effective methods such as optimal control, backstepping control and finite time control and the like, and provide a fuzzy finite time optimal synchronous control scheme of a unidirectional coupling permanent magnet synchronous generator system to achieve specified performance is still a prominent problem in the field of fractional order control.
Disclosure of Invention
In view of the above, the present invention provides a fuzzy finite time optimal synchronization control method for a fractional order permanent magnet synchronous generator.
In order to achieve the purpose, the invention provides the following technical scheme:
the fuzzy finite time optimal synchronous control method of the fractional order permanent magnet synchronous generator comprises the following steps:
s1: modeling a system;
s2: designing a numerical method and applying the numerical method to solving a nonlinear fractional order system;
s3: and establishing a hierarchical two-type fuzzy neural network and designing a controller.
Optionally, the S1 specifically includes: the wind energy conversion system consists of a wind turbine, a permanent magnet synchronous generator and three converters; the system comprises a diode bridge rectifier, a dc/dc boost converter and an inverter in sequence; the electric energy generated by the permanent magnet synchronous generator is transmitted to a power grid through a converter;
in combination with local aerodynamic properties, the power produced by a wind turbine is expressed as:
in the formula, rho, R, omegar，β，ωrAnd vwExpressed as air density, turbine radius, speed, blade angle, speed, wind speed, Cp(ωrR/vwBeta) is expressed as a turbine power coefficient;
according to the rotation law, a mechanical fractional order model of the permanent magnet synchronous generator is given:
in the formula, alpha, J, Tt，Tg，And b represents fractional order coefficient, system inertia, turbine torque, generator torque, time, and viscous friction coefficient, respectively;
defining the torque of the electromagnetic generator:
in the formula LdAnd LqRepresenting d-and q-axis inductances, idAnd iqRepresenting d-axis and q-axis stator currents, and p and phi representing the number of pole pairs and magnetic flux of permanent magnets in the three-phase permanent magnet synchronous generator;
the fractional order model of the permanent magnet synchronous generator in the synchronous rotating d-q reference system is expressed as:
in the formula Rs，VdAnd VqRepresenting stator resistance, d-axis and q-axis stator voltages;
in the Laplace domain, the fractional order integrator in the form of a linear approximation with zero pole pairs is represented by a transfer function with a slope of-20 m dB/decade in a Bode plot:
in the formulaPf，ωmaxAnd dfRespectively representing angular frequency, bandwidth and difference between the actual line and the approximate line, QiAnd PiRepresenting the zero and the pole of the singular function;
the stator winding is symmetrical to obtain L ═ Ld＝Lq(ii) a The fractional order model of the permanent magnet synchronous generator is defined by the formula (2) and the formula (4) as follows:
in the formula(ii) a Caputo fractional derivative representing when α >0 and the origin is at the origin;
by introducing a new variable x1＝Lωr/Rs，x2＝pLφiq/bRs，x3＝pLφid/bRsThe normalized fractional order model of the permanent magnet synchronous main motor is written as:
in the formulaμ＝-pφ2/bRs，σ＝3Lb/2JRs，ρ＝bL/JRs，x1，x2，x3，t，uq，udAnd TLRespectively represent normalizationChanging angular velocity, q-axis current, d-axis current, time, q-axis voltage, d-axis voltage and load torque, wherein sigma, rho and mu represent system parameters;
using the Heaviside function H (T-T)g) Establishing a fractional order model of the driven permanent magnet synchronous generator:
in the formula tg，κ1And kappa2Respectively representing the initial synchronization time, the capacitive coupling and the resistive coupling, u2And u3Represents a control input;
defining the synchronization error as e1＝y1-x1，e2＝y2-x2And e3＝y3-x3(ii) a Subtracting the equation (7) from the equation (8) to obtain:
definition 1: for a sufficiently differentiable function F (t), the Caputo fractional derivative is written as:
in the formulaRepresenting Euler gamma function, n-1 < alpha < n,
the laplace transform is performed on (10) to obtain:
for an arbitrary continuous function, whenAlpha is more than 0 and less than 1 and F1(t) and F2(t) in the interval [0, t ]V]When inside, exist
When F is present1(t)＝F2(t) deriving the following inequality:
introduction 1: for any xs，If constant cs＞0，ds＞0，Is an arbitrary real function, then the non-column equation holds
Definition 2: the minimum performance cost function is as follows:
in the formulaU，Respectively representing a penalty function, an optimal control input and an N-order matrix,
optionally, the S2 specifically includes:
for a fractional order differential equation with a given function g (·,)yfThe fractional derivative quotient of (t) is defined as an infinite series, i.e.:
wherein h >0 andrepresents a step size and a coefficient satisfying a condition based on a Euler-Gamma function of
In the interval [0T]Middle definition equidistant grid tnNh, N ═ 0,1, …, N, where N ═ T/h; then obtaining the value at t ═ tnIntegral expression of time:
let constantIn each subinterval [ t ]j,tj+1]Up-approximated vector field g (τ, y)f(τ)); (18) the numerical values of formula are rewritten as:
the system parameters of the master/slave permanent magnet synchronous generator fractional order model are set to be three working conditions:
working condition 1: α is 0.99, σ is 3, ρ is 4, μ is 25, TLInitial condition x is 01(0)＝0.1，x2(0)＝0.9，x3(0)＝20；
Working condition 2: α is 0.99, σ is 17, ρ is 16, μ is 25, TLInitial condition x is 01(0)＝1.5，x2(0)＝0.5，x3(0)＝20；
Working condition 3: α is 0.99, σ is 5.5, ρ is 5.5, μ is 20, TLInitial condition x is 01(0)＝0.1，x2(0)＝0.1，x3(0)＝3；
By introducing the formula (5), approximate transfer functions of different fractional orders are given
The error values in the formula are respectively less than or equal to 0.1dB, less than or equal to 0.2dB and less than or equal to 0.4 dB.
Optionally, the S3 specifically includes:
the hierarchical two-type fuzzy neural network consists of five layers, namely an input layer, a membership layer, a rule layer, a hierarchical layer and an output layer;
at the second level, the upper and lower membership grades are written as:
in the formulaRespectively representing the center, upper width and lower width of the jth membership grade of the ith input;
in the rule layer, the up/down trigger rule is calculated as:
in the fourth layer, YRAnd YLExpressed as a set reduced order center:
in the formulaAndreferred to as the weight of the image to be displayed, jTandthe upper and lower triggering degrees of the jth rule are shown, M shows the number of fuzzy rules,
in the output layer, the fuzzy output in vector form is derived:
wherein w ≡ wR wL]，
ξ≡[xR xL].
By calling (26), the hierarchical two-type fuzzy neural network realizes high-precision approximation to any unknown but bounded function on a compact set, then
In the formulaN denotes the input number, e (X) >0 denotes the approximation error, ΩwAnd DXA tight set of appropriate bounds for w and X, respectively; introducing an optimal parameter w*The parameter satisfies the condition: represents an approximation of h; definition ofWherein w*Representing the amount of labor for analysis;
in order to avoid exponential increase of the rule number, two criteria of tracking error reduction rate and completeness of the fuzzy rule e are adopted to execute structural adjustment of the hierarchical two-type fuzzy neural network. The tracking error reduction rate corresponds to the derivative of the square of the tracking error between the drive system output and the response system. The completeness of a fuzzy rule is defined as "at least one fuzzy rule guarantees that the trigger strength in the operating range is not less than e". If the membership level is equal to or greater than 0.5 in the hierarchy, they are saved. Otherwise, these unimportant rules in the self-structuring algorithm are deleted.
In order to improve the solving speed and simplify the structure, the hierarchical two-type fuzzy neural network is transformed as follows:
wTξ(X)≤ζξT(X)ξ(X)/2b2+b2/2 (28)
in the formula, zeta | | w | | non-woven phosphor2B is greater than 0; exist ofWhereinAn estimated value of ζ is represented,
to avoid system performance degradation and suppress the convergence behavior of error variables, a positive and strictly monotonically decreasing finite time preset performance function:
in the formula aiWhere i is 0, …,3 denotes a design parameter, T0Andrepresenting a convergence time and a convergence boundary; the performance function satisfies the following constraints:
in the formula beta0Representing an initial value of a finite time preset performance function;
when the constraint signal is matched with the convergence rate of the preset performance function, limiting the larger overshoot of the control output by the limiting condition in the initial stage; for a predetermined parameter, β0，T0Andby properly selecting four design parameters, a satisfactory finite time preset performance function can be easily obtained;
to achieve a faster response, a fractional order time-limited command filter based on a first order Levant differentiator is presented:
in the formulaAndwhich is indicative of the state of the command filter,input signal representing a command filter, ci,1And ci,2Representing a positive design constant, normal constant Γi,1And gammai,2Satisfy | Zi,1-αr|≤Γi,1Andconditions;
obviously, by appropriate selection of ci,1，ci,2When the input noise is completely suppressed during a transient of limited time, there is Zi,1＝αrAnd
introducing error variables
In the formulai-2, 3 and Zi,1I is 2,3 same and represents the output of the fractional order command filter;
the compensated tracking error is expressed as:
vi＝zi-θi,i＝1,2,3 (33)
in the formula [ theta ]iRepresenting compensation signals between virtual control and filtered signalsNumber;
the error variables have the following inequalities:
in the formula
Defining a smooth and derivable function S1(z1) The above inequality can be rewritten as an unconstrained form:
e1＝β(t)S1(z1) (35)
it is to be noted that it is preferable that,belonging to one of the above smooth and derivable functions;
the inverse of (35) is then written as:
in the formula1(t)＝e1(t)/β(t)；
(36) The fractional derivative of equation is derived as:
in the formula
By using the formulas (9) and (37), there are:
due to the wind speed, the generator temperature and the statorDisturbance of the sub-resistance, coefficient of friction and working load, etc., where f1＝-ρe1，f2＝-e2-y1y3+x1x3-(κ1e1+κ2e2)H(t-tg)+μe1，f3＝-e3+y1y2-x1x2Are all considered unknown non-linear functions;
based on a fractional order backstepping control principle, the controller is designed to consist of three steps;
the first step is as follows: considering f1Is estimated on a compact set, i.e. using the above hierarchical type 2 fuzzy neural network, for the convenience of controller design
Wherein (·) represents (x)1,x2,x3) Abbreviations of (a);
the first Lyapunov candidate function is chosen as:
get V1The fractional derivative of (a) yields:
in the formulaAndrepresenting virtual control and σ Upper bound, ζ1＝||w1||2And b1＞0；
By definition 2, the following cost function is designed to achieve its minimum;
in the formula uoi，And kappaiRespectively representing optimal control input, normal number and design parameters; to compensate for the estimation error of the hierarchical two-type fuzzy neural network, the following inequality is used
The second step is that: to solve the aforementioned unknown non-linear function f2It is approximated with high precision using a hierarchical two-type fuzzy neural network in the form:
in the formulaAnd
the third step: to process an unknown non-linear function f3The estimation is carried out by utilizing a hierarchical two-type fuzzy neural network with high precision and repeatability:
obtaining:
in the formulaAnd
the invention has the beneficial effects that:
firstly, different from an integer order model of a single isolated permanent magnet synchronous generator, the invention establishes a one-way coupling fractional order permanent magnet synchronous generator synchronous model capable of realizing ordered and coordinated motion. The dynamic characteristics of the system can be accurately described, and the freedom degree of design is increased. Meanwhile, under a designed numerical method, the dynamic analysis reveals the stable and unstable boundaries of the system along with the change of time, and explains the chaos and regular motion trend of the system along with the change of parameters.
Secondly, the invention solves the problem of 'complexity explosion' of the traditional backstepping control of the nonlinear system (even in a high-order system), processes the problem of infinite time convergence in an exponential form and the problem of non-adjustable convergence speed between an initial stage and a steady-state stage, and simultaneously provides greater flexibility and better functional characteristics compared with a common fuzzy logic system and a neural network.
Thirdly, under a fractional order backstepping framework, fuzzy finite time optimal synchronous control of the unidirectional coupling fractional order permanent magnet synchronous generator system is achieved, and the problems of finite time convergence, unknown system nonlinear function, preset constraint conditions, minimum cost function and the like are solved.
Fourth, additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a schematic view of a wind energy conversion system;
FIG. 2 is a diagram of the relationship between Lyapunov exponent and system parameters (σ, ρ), fractional order and time under working condition 1;
FIG. 3 is a diagram of the relationship between Lyapunov exponent and system parameters (σ, ρ), fractional order and time under condition 2;
FIG. 4 is a contour map of the maximum Lyapunov exponent of chaotic oscillation on the parameter plane of alpha-rho, alpha-sigma and sigma-rho under the working condition 1;
FIG. 5 is a contour map of the maximum Lyapunov exponent of chaotic oscillation on the parameter plane of alpha-rho, alpha-sigma and sigma-rho under the working condition 2;
FIG. 6 is a graph of synchronization error with/without a finite time preset performance function under the same conditions;
FIG. 7 shows synchronization errors at different fractional orders;
FIG. 8 is a graph of compensated tracking error for different coupling parameters;
FIG. 9 is a graph of transition weights for different system parameters;
FIG. 10 is a graph of transition weights at different fractional orders;
FIG. 11 is a graph of the performance of a fractional order finite time command filter under different operating conditions;
FIG. 12 is a graph of q and d axis control inputs and optimal control inputs with no generator load at start up;
fig. 13 shows the chaos suppression capability.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
1. System modeling
As shown in fig. 1, a wind energy conversion system is typically made up of a wind turbine, a permanent magnet synchronous generator, and three converters (in turn, a diode bridge rectifier, a dc/dc boost converter, and an inverter). The electric energy generated by the permanent magnet synchronous generator is transmitted to the power grid through the converter.
In combination with local aerodynamic properties, the power produced by a wind turbine is expressed as:
in the formula, rho, R, omegar，β，ωrAnd vwExpressed as air density, turbine radius, speed, blade angle, speed, wind speed, Cp(ωrR/vwAnd β) is expressed as a turbine power coefficient.
According to the rotation law, a mechanical fractional order model of the permanent magnet synchronous generator is given:
in the formula, alpha, J, Tt，Tg，And b represents fractional order coefficient, system inertia, turbine torque, generator torque, time, and viscous friction coefficient, respectively.
Defining the torque of the electromagnetic generator:
in the formula LdAnd LqRepresenting d-and q-axis inductances, idAnd iqThe d-axis and q-axis stator currents are shown, and p and phi show the number of pole pairs and the magnetic flux of permanent magnets in the three-phase permanent magnet synchronous generator.
The fractional order model of the permanent magnet synchronous generator in the synchronous rotating d-q reference system is expressed as:
in the formula Rs，VdAnd VqRepresenting the stator resistance, d-axis and q-axis stator voltages.
In the Laplace domain, the fractional order integrator in the form of a linear approximation with zero pole pairs is represented by a transfer function with a slope of-20 m dB/decade in a Bode plot:
in the formulaPf，ωmaxAnd dfRespectively representing angular frequency, bandwidth and difference between the actual line and the approximate line, QiAnd PiRepresenting the zeros and poles of the singular function.
The stator winding is symmetrical to obtain L ═ Ld＝Lq. The fractional order model of the permanent magnet synchronous generator is defined by the formula (2) and the formula (4) as follows:
in the formulaRepresenting the Caputo fractional derivative when α >0 and the origin is at the origin.
By introducing a new variable x1＝Lωr/Rs，x2＝pLφiq/bRs，x3＝pLφid/bRsThe normalized fractional order model of the permanent magnet synchronous main motor can be written as:
in the formulaμ＝-pφ2/bRs，σ＝3Lb/2JRs，ρ＝bL/JRs，x1，x2，x3，t，uq，udAnd TLThe normalized angular velocity is represented, q-axis current, d-axis current, time, q-axis voltage, d-axis voltage, and load torque are represented, respectively, and σ, ρ, and μ represent system parameters.
Using the Heaviside function H (T-T)g) And establishing a fractional order model of the driven permanent magnet synchronous generator:
in the formula tg，κ1And kappa2Respectively representing the initial synchronization time, the capacitive coupling and the resistive coupling, u2And u3Representing a control input.
Defining the synchronization error as e1＝y1-x1，e2＝y2-x2And e3＝y3-x3. Subtracting the equation (7) from the equation (8) to obtain:
definition 1: for a sufficiently differentiable function F (t), the Caputo fractional derivative is written as:
in the formulaRepresenting Euler gamma function, n-1 < alpha < n,
the laplace transform is performed on (10) to obtain:
for any continuous function, when 0 < alpha < 1 and F1(t) and F2(t) in the interval [0, t ]V]When inside, exist
When F is present1(t)＝F2(t) deriving the following inequality:
introduction 1: for any xs,If constant cs＞0，ds＞0，θ(xs,ys) Is an arbitrary real function, then the non-column equation holds
Definition 2: the minimum performance cost function is as follows:
in the formulaU，Respectively representing a penalty function, an optimal control input and an N-order matrix,
2. kinetic analysis and problem formulation
In order to reveal the dynamic characteristics of the fractional order master/slave permanent magnet synchronous generator, this section performs a dynamic analysis on it. At present, it is difficult to obtain an explicit solution to the nonlinear fractional order differential equation. By utilizing the integration rule, the invention designs a numerical method and applies the numerical method to solve the nonlinear fractional order system.
For a general fractional order differential equation with a given function g (·,)t≥0，yfThe fractional derivative quotient of (t) is defined as an infinite series, i.e.:
wherein h >0 andrepresents the step size and the coefficient while satisfying the condition based on the Euler-Gamma function
In the interval [0T]In (1) define an equidistant grid tnN-nh, N-0, 1, …, N, wherein N-T/h. Then obtaining the value at t ═ tnIntegral expression of time
Let constantIn each subinterval [ t ]j,tj+1]Up-approximated vector field g (τ, y)f(τ)). (18) The numerical values of formula are approximately rewritten as:
the system parameters of the master/slave permanent magnet synchronous generator fractional order model are set to be three working conditions:
working condition 1: α is 0.99, σ is 3, ρ is 4, μ is 25, TLInitial condition x is 01(0)＝0.1，x2(0)＝0.9，x3(0)＝20；
Working condition 2: α is 0.99, σ is 17, ρ is 16, μ is 25, TLInitial condition x is 01(0)＝1.5，x2(0)＝0.5，x3(0)＝20；
Working condition 3: α is 0.99, σ is 5.5, ρ is 5.5, μ is 20, TLInitial condition x is 01(0)＝0.1，x2(0)＝0.1，x3(0)＝3。
By introducing the formula (5), approximate transfer functions of different fractional orders are given
The error values in the formula are respectively less than or equal to 0.1dB, less than or equal to 0.2dB and less than or equal to 0.4 dB.
The permanent magnet synchronous generator produces chaotic oscillations with two attractors. In order to show the advantages of the fractional order model, the dynamic characteristics of the permanent magnet synchronous generator related to the fractional order value are disclosed at the same time. The step value is set to 0.96, and the periodic motion of the permanent magnet synchronous generator is immediately switched to a chaotic orbit with an attractor. In the working condition 1, when alpha is 0.97, the system has two motion states of T-time periodic behavior and chaotic attractor. In the operating condition 2, when α is 0.98, the motion behavior of the permanent magnet synchronous generator becomes chaotic oscillation with two attractors. Finally, regardless of condition 1 or condition 2, once α increases to 0.992 or 0.998, the system motion is maintained in a chaotic oscillating state with two attractors.
FIGS. 2-3 illustrate the relationship between Lyapunov exponent and system parameters (σ, ρ), fractional order and time under conditions 1-2. In the first sub-diagram of fig. 2, when ρ ∈ (26.5), the permanent magnet synchronous generator undergoes chaotic oscillation. In its second sub-diagram, the permanent magnet synchronous generator falls into a chaotic state when σ is set to the interval [2.16 ]. The third graph of fig. 2 reveals that chaotic oscillation can be triggered when the fractional order is in the interval (0.9121). As can be seen from the last subgraphs of FIGS. 2-3, the permanent magnet synchronous generator is in a chaotic state all the time in the whole process. In the first sub-diagram of fig. 3, the permanent magnet synchronous generator is in a chaotic state when ρ ∈ (216). It is noted that there is a switching point between the periodic state and chaotic oscillation. In the second diagram of fig. 3, when σ is in the interval [12.420], chaotic oscillation occurs in the permanent magnet synchronous generator. As seen from the third graph of fig. 3, the fractional order value when α >0.928 causes chaotic oscillation of the permanent magnet synchronous generator.
FIGS. 4-5 show maximum Lyapunov exponent contour plots in the α - ρ, α - σ, and σ - ρ parameter planes for system stability and instability boundaries under conditions 1-2. The color bars of each subgraph represent the intervals between which chaos occurs between these two parameters, e.g., α - ρ, α - σ, and σ - ρ. It is clear that the lyapunov index is maximal in the lemon yellow region. That is, when the combination of the three parameters falls into the lemon yellow area, the permanent magnet synchronous generator generates chaotic oscillation. Chaotic oscillation has many application scenarios. However, in the case of a permanent magnet synchronous generator, such oscillations can lead to a reduction in the performance of the operating system and even to the burning of the motor and its surrounding components. Therefore, it is necessary to provide an effective method for suppressing the chaotic oscillation. Meanwhile, when the high-dimensional hyper-chaotic system is faced, the fuzzy neural network has the defect of poor approximation precision. When minimizing the cost function, how to further improve the stability time and dynamic performance of the unidirectional coupling fractional order permanent magnet synchronous generator is still a difficult problem.
3. Fuzzy finite time optimal controller design
3.1 hierarchical two-type fuzzy neural network
The hierarchical two-type fuzzy neural network is a derivative of the one-type fuzzy neural network and has stronger learning capability, function approximation capability and fault-tolerant capability. The system consists of five layers, namely an input layer, a membership layer, a rule layer, a descending layer and an output layer. At the second level, the upper and lower membership grades are written as:
in the formulaRespectively representing the center, top width and bottom width of the jth membership level of the ith input.
In the rule layer, the up/down trigger rule is calculated as:
in the fourth layer, YRAnd YLExpressed as a set reduced order center:
in the formulaAndreferred to as the weight of the image to be displayed,T jandthe upper and lower triggering degrees of the jth rule are shown, M shows the number of fuzzy rules,
in the output layer, the fuzzy output in vector form is derived:
wherein w ≡ wR wL]，
ξ≡[xR xL].
By calling (26), the hierarchical two-type fuzzy neural network realizes high-precision approximation to any unknown but bounded function on a compact set, then
In the formulaN denotes the input number, e (X) >0 denotes the approximation error, ΩwAnd DXAre a tight set of appropriate bounds for w and X, respectively. Introducing an optimal parameter w*The parameter satisfies the condition: an approximation of h is shown. Definition ofWherein w*Representing the amount of labor for analysis.
In order to avoid exponential increase of the rule number, two criteria of tracking error reduction rate and completeness of the fuzzy rule e are adopted to perform structural adjustment of the hierarchical two-type fuzzy neural network. The tracking error reduction rate corresponds to the derivative of the square of the tracking error between the drive system output and the response system. The completeness of a fuzzy rule is defined as "at least one fuzzy rule guarantees that the trigger strength in the operating range is not less than e". If the membership level is equal to or greater than 0.5 in the hierarchy, they are saved. Otherwise, these unimportant rules in the self-structuring algorithm are deleted.
In order to improve the solving speed and simplify the structure, the hierarchical two-type fuzzy neural network is transformed as follows:
wTξ(X)≤ζξT(X)ξ(X)/2b2+b2/2 (28)
in the formula, zeta | | w | | non-woven phosphor2And b is greater than 0. Exist ofWhereinRepresenting an estimate of ζ.
Remarks 1: compared to Grunwald-Letnikov and Riemann-Liouville derivatives, the initial conditions of the fractional order differential equation Caputo derivatives have the same form as the initial conditions of the integer order differential equation. When the Caputo derivative is applied to a constant, its value is zero. Thus, directly obtain
3.2 controller design
In order to avoid the performance degradation of the system and inhibit the convergence characteristic of the error variable, a positive and strictly monotonically decreasing finite-time preset performance function is designed:
in the formula aiWhere i is 0, …,3 denotes a design parameter, T0Andthe convergence time and the convergence boundary are indicated. Meanwhile, the performance function satisfies the following constraint conditions:
in the formula beta0Representing an initial value of a finite time preset performance function.
Remarks 2: when the constraint signal matches the convergence rate of the preset performance function, the limiting condition suppresses a large overshoot of the control output at the initial stage. For a predetermined parameter, e.g. beta0，T0Andby properly selecting the four design parameters, it is easy toA satisfactory finite time preset performance function is obtained.
To achieve a faster response, a fractional order time-limited command filter based on a first order Levant differentiator is presented:
in the formulaAndwhich is indicative of the state of the command filter,input signal representing a command filter, ci,1And ci,2Representing a positive design constant, normal constant Γi,1And gammai,2Satisfy | Zi,1-αr|≤Γi,1Andand (4) conditions.
Obviously, by appropriate selection of ci,1，ci,2When the input noise is completely suppressed during a transient of limited time, there is Zi,1＝αrAnd
introducing error variables
In the formula of alphai cI is 2,3 and Zi,1And i is equal to 2 and 3, and represents the output of the fractional order command filter.
The compensated tracking error is expressed as:
ui＝zi-θi,i＝1,2,3 (33)
in the formula [ theta ]iRepresenting a compensation signal between the virtual control and the filtered signal.
The error variables have the following inequalities:
wherein 0 is more than rho,
defining a smooth and derivable function S1(z1) The above inequality can be rewritten as an unconstrained form:
e1＝β(t)S1(z1) (35)
it is to be noted that it is preferable that,belonging to one of the smooth and derivable functions mentioned above.
The inverse of (35) is then written as:
in the formula1(t)＝e1(t)/β(t)。
(36) The fractional derivative of equation is derived as:
in the formula
By using the formulas (9) and (37), there are:
due to disturbances from wind speed, generator temperature, stator resistance, friction coefficient and work load, etc., where f1＝-ρe1，f2＝-e2-y1y3+x1x3-(κ1e1+κ2e2)H(t-tg)+μe1，f3＝-e3+y1y2-x1x2Are considered unknown non-linear functions.
Based on the fractional order backstepping control principle, the controller is designed to consist of three steps.
The first step is as follows: considering f1Is estimated on a compact set, i.e. using the above hierarchical type 2 fuzzy neural network, for the convenience of controller design
Wherein (·) represents (x)1,x2,x3) Abbreviations of (a).
The first Lyapunov candidate function is chosen as:
get V1The fractional derivative of (a) yields:
in the formulaAndrepresenting virtual control and sigma upBoundary, ζ1＝||w1||2And b1＞0。
By definition 2, the following cost function is designed to achieve its minimum.
In the formula uoi，And kappaiRespectively representing the optimal control input, normal and design parameters. To compensate for the estimation error of the hierarchical two-type fuzzy neural network, the following inequality is used
The optimal control input is designed asWhere P isiBelonging to algebraic Riccati equationkiSolutions of >0.
The optimal control inputs are then further derived:
the virtual control and the self-adapting law and compensation signal thereof are selected
In the formula k1＞0，l1＞0，γ1＞0，s1＞0，0＜γ＜1。
By substituting (14) for (44) - (46) to (41) to obtain
In the formulaAnd
the second step is that: to solve the aforementioned unknown non-linear function f2It is approximated with high precision using a hierarchical two-type fuzzy neural network in the form:
selecting a second Lyapunov candidate function:
V2the fractional order derivative of (a) is written as:
zeta in the formula2＝||w2||2And b2＞0。
The q-axis control input and the adaptive law and compensation signals are designed as follows:
in the formula k2＞0，l2＞0，γ2>0 and s2＞0。
Utilization sum-, further simplified to:
in the formulaAnd
the third step: also, to process the unknown non-linear function f3The estimation is carried out by utilizing a hierarchical two-type fuzzy neural network with high precision and repeatability:
due to the adoption of magnetic field orientation control, thereforeEqual to zero. Defining the last Lyapunov candidate function:
calculating V3Has a fractional order derivative of
Zeta in the formula3＝||w3||2And b3＞0.
The d-axis control input, adaptive law and compensation signals are designed as follows:
in the formula k3＞0，l3＞0，γ3>0 and s3＞0。
The following can be obtained:
in the formulaAnd
remarks 3: as the order of the complex nonlinear system increases, the filtering error of the fractional order finite time command filter becomes difficult to control. In this case, it is difficult to obtain a minimum tracking error. Therefore, it is necessary and meaningful to solve this problem and achieve a limited time convergence by constructing the compensation signal.
4. Stability analysis
Theorem 1: for the problem of fuzzy finite time optimal synchronization control of a unidirectional coupled fractional order permanent magnet synchronous generator with chaotic oscillation, if q-axis and d-axis control inputs (51) and (58) with adaptive laws (52), (59) and compensation signals (53), (60) are designed, all signals of a closed-loop system are bounded, and synchronization errors meet preset performance requirements. Meanwhile, the purpose of fuzzy finite time optimal synchronous control is achieved, and the cost function is minimized.
And (3) proving that: defining entire Lyapunov function candidates
Calculating (62) a fractional derivative
In the formulabe＝2min{d1,b1,d2,b2,d3,b3}，
ui，And thetaiIn tight set omegaeInternal and finite time T0Stability is achieved by
In the formula De＝min{ce/(1-λe)ae,(ce/(1-λe)be)2/(γ+1)And 0 < lambdae＜1。
Finite time T0Comprises the following steps:
T0≤max{t1,t2} (65)
in the formula
After the syndrome is confirmed.
5. Results and analysis of the experiments
And constructing a numerical method for solving the nonlinear fractional order system based on an integral rule, and realizing the numerical method by programming. The coupling parameter of the driving permanent magnet synchronous generator and the driven permanent magnet synchronous generator is kappa1＝0.1，κ2-0.1. The parameters of the finite time preset performance function are selected as follows: a is0＝0.4，a1＝0，a2＝-4.9，a3＝6.3，βT00.02. The limited time is set to T0Fractional order finite time command filter parameter set to c 0.41,1＝6，c1,25. Parameter setting of smooth invertible functionρ＝0.1，Controller parameter set to k1＝k2＝k3＝10，γ1＝γ2＝2，γ3＝8，l1＝l2＝l3＝10，s1＝s2＝s3＝0.5，Y is 0.4 and k1＝k2＝k31. Slavery to be associated with hierarchical two-type fuzzy neural networksThe attribute grade centers are uniformly distributed in the interval [ -11 ]]And selecting the upper and lower widths of the membership grade asAnd
fig. 6 shows the synchronization error of the driving and driven unidirectional coupling fractional order permanent magnet synchronous generator with and without a finite time preset performance function. It is clear that conventional solutions without a finite time preset performance function have a large overshoot and a long oscillation period during the whole operation. The designed scheme realizes the rapid synchronization of the driving permanent magnet synchronous generator and the driven permanent magnet synchronous generator, and the error is very small in limited time. Therefore, under the same conditions, the designed scheme is obviously superior to the traditional scheme without the limited time preset performance function.
Real-time changes in the grid caused by the surrounding environment through the transmission buses and feeders of the wind farm can have an inevitable effect on the operating state of each generator, which can lead to inaccurate modeling and system parameter perturbations. As can be seen from fig. 7, the design scheme overcomes these adverse effects well, and allows the ac master and slave permanent magnet synchronous generators to achieve precise frequency coincidence under normal load. The motion state of the driven permanent magnet synchronous generator is absorbed into the attractor of the driving permanent magnet synchronous generator in the shortest time possible. At the same time, the synchronization error is controlled within a pre-configured performance range.
The size of the coupling coefficient directly determines synchronous/asynchronous motion and harmonic oscillation/periodic motion when the permanent magnet synchronous generator runs in parallel. Three combinations are given, such as I: kappa1＝0.1，κ2＝-0.1；II：κ1＝0.2，κ2＝-0.15；III：κ1＝0.15，κ2-0.2. Under all three conditions, the permanent magnet synchronous generator generates oscillation behavior before the method is applied. Fig. 8 illustrates the compensated tracking error after the designed scheme is adopted, and it is obvious that several curves are overlapped all the time. Description of the inventionThe designed method has stronger coupling coefficient disturbance resistance.
The hierarchical two-type fuzzy neural network with conversion compensates and approximates functions for the flutter problem caused by wind speed, generator temperature, stator resistance, friction coefficient, work load and the like. Three sets of system disturbance parameters are defined: a1: ρ is 5, μ is 19; a2: ρ is 5.5, μ is 20; a3: ρ is 6 and μ is 21. From fig. 9-10, it can be seen that the conversion weights of the hierarchical two-type fuzzy neural network can be converged into a small neighborhood of the origin rapidly in the self-structure algorithm. At the same time, the three curves always overlap, whether in different system parameters or in different fractional orders.
The fractional order finite time instruction filter matched with the compensation signal achieves the purposes of high response speed, high estimation precision and finite time convergence. Fig. 11 fully demonstrates that the fractional order finite time instruction filter maintains its high performance regardless of the presence of variable coupling coefficients, system parameter perturbations or fractional order variations, etc.
FIG. 12 is a control input and an optimal control input at start-up with and without an external load. It can be seen from the figure that the unidirectional coupling fractional order permanent magnet synchronous generator system rapidly realizes stable operation after transient fluctuation when a load exists. Therefore, the designed controller has good anti-interference capability.
As can be seen from fig. 13, when the proposed control method is not yet introduced, the permanent magnet synchronous generator may generate a large amount of dynamic behavior including chaotic oscillation. Once the proposed fuzzy finite time optimal control scheme is used, each permanent magnet synchronous generator switches to periodic motion immediately within 0.4 seconds, where the corresponding chaotic oscillations are completely suppressed.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.