阅读说明：本技术 一种基于改进自抗扰神经控制器的无刷双馈风力发电系统 (Brushless double-fed wind power generation system based on improved auto-disturbance rejection neural controller ) 是由 侯晓鑫 张志强 游国栋 于 2019-08-29 设计创作，主要内容包括：本发明属于无刷双馈风力发电技术领域,涉及一种基于改进自抗扰神经网络控制器的无刷双馈风力发电系统并网运行控制方案。针对无刷双馈风力发电系统的非线性时变特征,磁路饱和和温升引起电感和电阻的变化,给依赖于发电机参数的传统的无刷双馈发电机直接功率控制带来困难。控制方案采用功率外环和电流内环,外环通过自抗扰控制技术实现功率绕组网侧的有功和无功功率控制,并针对自抗扰控制结构整定参数多而复杂的问题,将鸽群优化算法你和演化博弈理论相结合。进一步针对改进的自抗扰控制器对精确数学模型的依赖,内环结合神经网络逆控制思想,构建神经网络控制器,通过神经网络实现控制绕组dq轴定子电流控制,达到对风力发电系统功率的最优控制。(The invention belongs to the technical field of brushless double-fed wind power generation, and relates to a brushless double-fed wind power generation system grid-connected operation control scheme based on an improved active disturbance rejection neural network controller. Aiming at the nonlinear time-varying characteristic of the brushless doubly-fed wind power generation system, the magnetic circuit saturation and the temperature rise cause the change of inductance and resistance, and the difficulty is brought to the direct power control of the traditional brushless doubly-fed generator depending on the generator parameters. The control scheme adopts a power outer ring and a current inner ring, the outer ring realizes active power and reactive power control on the power winding network side through an active disturbance rejection control technology, and a pigeon group optimization algorithm is combined with an evolutionary game theory aiming at the problem that an active disturbance rejection control structure has more and complicated setting parameters. Further aiming at the dependence of the improved active disturbance rejection controller on an accurate mathematical model, the inner ring is combined with the inverse control idea of the neural network to construct the neural network controller, and the control of the stator current of the dq axis of the control winding is realized through the neural network, so that the optimal control of the power of the wind power generation system is achieved.)

1. A brushless double-fed wind power generation system direct power control based on an improved auto-disturbance rejection neural controller is characterized by comprising the following steps:

(1) detecting the voltage and the current of a three-phase stator of a power winding through a voltage sensor and a current sensor;

(2) the three-phase voltage and current are subjected to Clark conversion and park conversion to obtain dq-axis voltageAnd current

(3) Calculating the active power P and the reactive power Q from the converted voltage and current, with a given value of P^*And Q^*Outputting active components and reactive components of stator current of control winding through pigeon group optimized evolution game principle active disturbance rejection controller as difference valuesCurrent of outputThen outputs voltage through a neural network controllerAnd obtaining the three-phase voltage of the control winding through inverse park transformation, and outputting a control signal to the inverter through SVPWM.

(4) The active power and the reactive power of the brushless doubly-fed generator are accurately controlled through an optimization algorithm, the robustness to the internal disturbance and the external disturbance of the system is strong, and the grid-connected system can stably operate under the condition that parameters of the brushless doubly-fed generator are not accurate.

Technical Field

The invention relates to a grid-connected control method of a brushless doubly-fed wind generator, in particular to a control method of direct power control of a brushless doubly-fed wind power generation system based on improved active disturbance rejection nerves.

Background

A power winding of a traditional brushless doubly-fed generator system is directly connected with a power grid, a control winding is connected with the power grid through an inverter, the traditional brushless doubly-fed generator control system adopts a double closed-loop structure, an inner loop adopts closed-loop control of a current PI control method, an outer loop adopts closed-loop control of an active power PI control method and a reactive power PI control method, but the PI control method is sensitive to interference, and the performance of the control system is greatly influenced by the change of parameters (inductance and resistance) of the brushless doubly-fed generator.

Disclosure of Invention

Aiming at the problem that a traditional brushless double-fed generator control system is sensitive to interference, an active-disturbance-rejection winding and a neural network are combined, a control scheme for grid-connected operation of a brushless double-fed wind power generation system is provided, aiming at the defect that parameter setting of an active-disturbance-rejection algorithm is complex and time-consuming, the control scheme is used for dynamically adjusting parameters of an active-disturbance-rejection controller, the control scheme is provided with an inner control ring and an outer control ring, the inner control ring realizes control of dq axis current of a control winding of a brushless double-fed wind power generator through the neural network, and the outer control ring realizes active power and reactive power control of the power side of the brushless double-fed wind power generator through an improved active-.

In order to achieve the purpose, the technical scheme adopted by the invention for solving the technical problem is as follows:

a brushless double-fed wind power generation system direct power control based on an improved auto-disturbance rejection neural controller is characterized by comprising the following steps:

(1) detecting the voltage and the current of a three-phase stator of a power winding through a voltage sensor and a current sensor;

(2) the three-phase voltage and current are subjected to Clark conversion and park conversion to obtain dq-axis voltageAnd current

(3) Calculating the active power P and the reactive power Q from the converted voltage and current, with a given value of P^*And Q^*Outputting control winding stator current through pigeon group optimized evolution game theory active disturbance rejection controller as difference valueActive and reactive componentsCurrent of outputThen outputs voltage through a neural network controllerAnd obtaining the three-phase voltage of the control winding through inverse park transformation, and outputting a control signal to the inverter through SVPWM.

(4) The active power and the reactive power of the brushless doubly-fed generator are accurately controlled through an optimization algorithm, the robustness to the internal disturbance and the external disturbance of the system is strong, and the grid-connected system can stably operate under the condition that parameters of the brushless doubly-fed generator are not accurate.

In the invention, the pigeon swarm optimization evolutionary game theory active disturbance rejection controller in the step (3) has the characteristics of independence on a controlled object model, high interference resistance, high precision and more accurate and efficient controller parameter setting, and further improves the rapidity and stability of the controller; the active disturbance rejection neural network control has higher tracking speed, and the operation performance of the brushless doubly-fed generator grid connection is improved.

The invention provides a brushless double-fed wind power generation control algorithm based on an improved active disturbance rejection neural controller, which consists of a controller, an inverter and a brushless double-fed generator, wherein the controller adopts an inner control ring and an outer control ring, the inner control ring realizes the dq axis current control of a control winding of the brushless double-fed wind power generator through a neural network, and the outer control ring realizes the active power and reactive power control of the power side of the brushless double-fed wind power generator through an improved active disturbance rejection technology, so that a brushless double-fed wind power generation system can run more stably and reliably.

Drawings

FIG. 1: control system schematic diagram of the invention

FIG. 2: active disturbance rejection controller structure diagram

Detailed Description

The invention relates to a brushless double-fed wind power generation system direct power control based on an improved active disturbance rejection neural controller, which is further described in detail in the following with reference to the attached drawings and embodiments.

FIG. 1 is a schematic diagram of a control system of the present invention, in which a controller sends a signal to a power device in a converter to realize direct power control of a brushless doubly-fed wind power generation system through the on and off of the power device. Fig. 2 is a structural diagram of an active disturbance rejection controller, which comprises a tracking differentiator, an extended state observer and a nonlinear combined control law. The neural network controller enables the objective functions of active power and reactive power to accurately track the given values by adjusting parameters on line.

The direct power control of the brushless doubly-fed wind power generation system based on the improved auto-disturbance rejection neural controller of the invention is described by combining with the figure 2, and comprises the following steps:

(1) detecting the voltage and the current of a three-phase stator of a power winding through a voltage sensor and a current sensor;

(2) the three-phase voltage and current are subjected to Clark conversion and park conversion to obtain dq-axis voltageAnd current

(3) Calculating the active power P and the reactive power Q from the converted voltage and current, with a given value of P^*And Q^*Outputting active components and reactive components of stator current of a control winding through a pigeon group optimized evolution game theory active disturbance rejection controller as difference valuesCurrent of outputThen outputs voltage through a neural network controllerThrough a reverse handkerchiefAnd the gram conversion is carried out to obtain the three-phase voltage of the control winding, and then the control signal is output to the inverter through SVPWM.

(4) The active power and the reactive power of the brushless doubly-fed generator are accurately controlled through an optimization algorithm, the robustness to the internal disturbance and the external disturbance of the system is strong, and the grid-connected system can stably operate under the condition that parameters of the brushless doubly-fed generator are not accurate.