风能发电系统关键技术研究

发布时间：2018-05-18 09:58

本文选题：风能发电 + 风速预测　；参考：《合肥工业大学》2014年博士论文

【摘要】：近年来,随着经济的高速发展,能源问题变得越来越严重,寻找新的替代能源已成为当务之急,风能发电以其资源丰富和技术上的优势,成为当前国内外研究和发展的重点。中国已成为世界上的风能大国,但是对风能发电的研究起步较晚,特别是控制系统方面,与此相关的理论研究和应用技术仍然很薄弱。因此,研究风能发电技术的主要关键技术和相关理论方法问题具有重要的意义。本文以风能发电系统为研究对象,对风能发电系统的一些关键问题,特别是在控制系统相关的问题方面进行了深入的研究。全文的主要研究内容如下： 1.研究了风速的预测问题。通过分析风速序列高度的非线性、大间歇非平稳性和混沌特性,构建了基于相空间重构小波神经网络的风速预测模型。 (1)分析了风速序列的非线性,采用自相关函数检验法判定了风速序列的非平稳性,采用功率谱法判定了风速序列的混沌特性,构建了一种风速预测模型。 (2)针对风速序列的混沌特性,采用G-P (Grassberger-Procaccia)法计算相关维数,由Takens嵌入定理求得嵌入维数,将相空间重构嵌入维数作为模型时序数据输入维数,避免输入节点数选取的任意性使得到的预测结果不一定是其解空间的表达问题。 (3)利用小波变换对非平稳性的分解,结合神经网络良好的非线性映射能力,有效地解决风速序列的非线性和非平稳性问题,给出了模型结构和学习算法,并进行了仿真实验验证。 (4)采用我国浙江杭州湾入海口(北纬90°,东经120°)多组半小时实测风速数据,对本文构建的风速预测模型进行工程验证,结果表明本文所建模型的精度明显优于传统方法。 2.研究了变速恒频双馈风电机组的功率控制问题。基于双馈风力发电系统的特性,提出了一种基于指数趋近率的积分变结构控制器,实现了双馈风电机组有功功率和无功功率的解耦控制。 (1)基于最大功率反馈控制策略以及定子电压定向的矢量控制策略,实现了双馈风力发电系统的最大功率跟踪和双馈电机有功功率和无功功率的解耦控制。 (2)针对传统的双馈电机矢量控制策略结构复杂,鲁棒性不强等问题,在双馈电机定子电压定向矢量控制的基础上,提出了一种变结构控制策略。通过直接在转子静止abc坐标系实现转子侧电流的无静差控制,避免了对转子电流实施坐标变换,简化了控制结构。且滑模变结构控制对于电机参数变化及外部扰动具有很好的鲁棒性,可以有效地抑制电机参数变化及外部扰动对双馈风能发电系统的影响。 (3)在变结构控制的滑动面引入积分环节,可消除电机参数变化及外部扰动造成的稳态误差,又可通过调节积分环节的参数,改善滑动运动的动态品质。通过采用指数趋近律可保证滑动模态到达过程的动态品质,又可消弱变结构控制本质上的开关不连续性造成的系统抖振。 (4)通过理论推导和仿真实验验证了本文控制策略的可行性和有效性,并进行了工程应用验证。 3.研究了风电机组的偏航控制问题。基于偏航控制系统的工作特性,采用V-HC (Vane-Hill Climbing)算法实现了偏航最大功率跟踪；提出了一种滑模速度控制器,实现了偏航感应电机转子磁链定向矢量控制。 (1)针对基于风向传感器的风向微变化时风向偏航控制角度难以精确化问题,在风向角小范围变化时采用基于功率检测的爬山算法进行最大发电功率跟踪,以解决小范围内风向传感器难以确定风向角的问题,提高了偏航控制的精度。 (2)针对风速检测和偏航控制具有滞后性,采用基于相空间重构的小波神经网络预测模型进行风速短时预测,根据预测风速对风电机组进行超前偏航控制,即可实现风电机组的最佳发电,又能减小对风力发电机组的风力冲击,保护和延长风力发电机的寿命。 (3)针对偏航控制系统抗较大外部扰动的鲁棒性问题,在偏航感应电机转子磁链定向矢量控制的基础上,将滑模变结构控制和矢量控制相结合,提出了一种滑模速度控制器用于偏航控制系统的速度控制,给出了滑模速度控制器的设计方法,分析了滑模速度控制器的有效性。滑模变结构控制对于电机参数变化及外部扰动具有很好的鲁棒性,可以有效地提高偏航系统的稳定性和抗扰动性。 (4)通过对比实验和工程验证,表明本文的控制策略是可行有效的。 4.设计了一套中小型风能发电机组控制装置,实际运行结果表明,本文的控制方法和策略是可行有效的。 (1)实现了中小型风能发电系统的信号采集和处理、偏航控制、自动解缆、并网发电控制、停机制动控制、安全保护、本地监控、远程监控。 (2)给出了偏航控制和功率控制主要功能的软硬件实现方法。 (3)对设计应用的系统进行实验和运行测试,通过实测数据分析,验证了控制装置和相关算法的有效性。
[Abstract]:In recent years, with the rapid development of economy, the problem of energy has become more and more serious. Looking for new alternative energy has become a top priority. Wind power generation has become the focus of research and development at home and abroad with its rich resources and technical advantages. China has become a big power of wind power in the world, but the research on wind power generation is late, In particular, the theoretical research and application technology related to the control system are still very weak. Therefore, it is of great significance to study the key technologies and related theoretical methods of wind power generation technology.
This paper takes the wind power generation system as the research object, and studies some key problems of the wind power generation system, especially the related problems in the control system. The main contents of this paper are as follows:
1. the prediction of wind speed is studied. By analyzing the nonlinearity of the height of the wind speed sequence, the large intermittent nonstationary and chaotic characteristics, a wind speed prediction model based on the phase space reconstruction wavelet neural network is constructed.
(1) the nonlinearity of the wind speed sequence is analyzed. The nonstationary of the wind speed sequence is determined by the autocorrelation function test method. The chaotic characteristics of the wind speed series are determined by the power spectrum method, and a wind speed prediction model is constructed.
(2) in view of the chaotic characteristics of the wind speed sequence, the correlation dimension is calculated by G-P (Grassberger-Procaccia) method. The embedding dimension is obtained by the Takens embedding theorem. The embedding dimension of the phase space is used as the input dimension of the model time series data to avoid the arbitrariness of the selection of the number of input nodes. The result of the prediction is not necessarily the expression of the solution space. Question.
(3) using the wavelet transform to decompose the nonstationary and combining the good nonlinear mapping ability of the neural network, the nonlinear and non-stationary problems of the wind speed sequence are solved effectively. The model structure and learning algorithm are given, and the simulation experiment is carried out.
(4) to verify the wind speed prediction model constructed in this paper, the results of the wind speed prediction model constructed in this paper are verified by using the data of the measured wind speed in the mouth of the Hangzhou Bay of Zhejiang (90 degrees north latitude and 120 degrees east) in this paper. The results show that the precision of the model is obviously superior to the traditional method.
2. the power control problem of the variable speed constant frequency doubly fed wind turbine is studied. Based on the characteristics of the doubly fed wind power generation system, an integral variable structure controller based on the exponential convergence rate is proposed, which realizes the decoupling control of the active and reactive power of the doubly fed wind turbine.
(1) based on the maximum power feedback control strategy and the stator voltage oriented vector control strategy, the decoupling control of the maximum power tracking of the doubly fed wind power generation system and the active power and reactive power of the doubly fed machine is realized.
(2) a variable structure control strategy is proposed on the basis of the stator voltage directional vector control of the doubly fed motor based on the complex vector control strategy of the traditional doubly fed motor vector control strategy. The static difference control of the rotor side current is realized directly in the rotor static ABC coordinate system, which avoids the coordinate change of the rotor current. In addition, the control structure is simplified, and the sliding mode variable structure control has good robustness for the change of motor parameters and external disturbances. It can effectively restrain the change of motor parameters and the influence of external disturbance on the doubly fed wind power generation system.
(3) the integral link is introduced into the sliding surface of the variable structure control, which can eliminate the steady-state error caused by the change of motor parameters and the external disturbance, and can also improve the dynamic quality of the sliding motion by adjusting the parameters of the integral link. By adopting the exponential approach law, the dynamic quality of the sliding mode can be guaranteed and the essence of the variable structure control can be weakened. A system buffeting caused by discontinuity on the switch.
(4) the feasibility and effectiveness of the proposed control strategy are verified by theoretical deduction and simulation experiments, and verified by engineering application.
3. the yaw control problem of the wind turbine is studied. Based on the working characteristics of the yaw control system, the V-HC (Vane-Hill Climbing) algorithm is used to achieve the maximum power tracking of the yaw. A sliding mode speed controller is proposed to realize the rotor flux orientation vector control of the yaw induction motor.
(1) in order to solve the problem that the wind direction yaw control angle is difficult to be precise when the wind direction sensor based on the wind direction change, the mountain climbing algorithm based on power detection is used to track the maximum power generation power when the wind direction angle is small, so as to solve the problem that the wind direction sensor is difficult to determine the wind direction in the small range and improve the precision of the yaw control.
(2) in view of the hysteresis of wind speed detection and yaw control, the wavelet neural network prediction model based on phase space reconstruction is used to predict wind speed in short time. The optimal power generation of wind turbine can be realized, and the wind power generation can be reduced, the protection and extension can be reduced. The life of the wind turbine.
(3) in view of the robustness of the yaw control system against large external disturbances, on the basis of the directional vector control of the rotor flux linkage of the yaw induction motor, a sliding mode speed controller is proposed for the speed control of the yaw control system, and the design of the sliding mode speed controller is given. The validity of the sliding mode speed controller is analyzed. The sliding mode variable structure control has good robustness for the change of the motor parameters and the external disturbance, and it can effectively improve the stability and disturbance resistance of the yaw system.
(4) through comparison experiment and engineering verification, it shows that the control strategy is feasible and effective.
4. design a control device for small and medium sized wind turbines. The actual operation results show that the control method and strategy are feasible and effective.
(1) realizing the signal acquisition and processing of small and medium wind power generation system, yaw control, automatic cable release, grid connected power generation control, stop brake control, safety protection, local monitoring and remote monitoring.
(2) the hardware and software implementation of the main functions of yaw control and power control are given.
(3) experiment and operation test of the designed application system. The validity of the control device and related algorithm is verified through the analysis of measured data.
【学位授予单位】：合肥工业大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM614

【参考文献】