浮子式波浪发电系统的模型预测控制

发布时间：2018-05-26 20:56

本文选题：波浪发电 + 模型预测控制　；参考：《天津大学》2016年硕士论文

【摘要】：波浪能可以利用波能转换装置(Wave Energy Converter,WEC)将其转换成可利用的电能。为了有效、充分的利用波浪能,就必须采取有效的控制策略来提高波能转换器的转换效率,实现最大波能捕获。由于波浪的随机性和系统模型的非线性等原因,使传统的控制方法在波浪发电系统控制当中有很大的局限性。鉴于此种情况,本文在控制方法中进行了创新与改进,提出了将智能控制方法中的模型预测控制方法(Model Predictive Control,MPC)应用到浮子式波浪发电系统当中,并且针对现有的文献中仅仅对浮子运动部分来分析,获得最大波能捕获的局限性,文中将浮子位置、浮子速度和电机电流共同作为预测控制的状态变量。由于模型预测控制方法具有对模型精度要求不高,包含对外部输入的预测,并且能够将系统的限制条件考虑进去等优点,使其在波浪发电系统中具有很大的优势,因此模型预测控制方法的研究在波浪发电系统中具有广阔的前景和重要的现实意义。本文针对点吸式波浪发电系统的模型预测控制方法进行了深入研究,主要从以下几方面展开工作:(1)分析了波浪发电技术的研究现状和研究成果,总结了浮子式波浪发电系统在整个海洋波浪能源利用领域中的优势和研究意义,分析了现有典型波能发电装置各自的优缺点以及工作原理,并对浮子式波浪发电系统的控制策略进行了系统的研究和详细的总结。(2)研究浮子式波浪发电系统的工作运行状态,以及其中每个部分的具体运动情况,对浮子式波浪发电系统的入射波进行了模拟。以单浮子的波浪发电系统为研究对象,对浮子的受力情况进行了具体的分析与建模,在系统中选择了直线电机作为功率输出装置,并对其进行了数学模型的推导,进而得到单浮子式波浪发电系统波浪输出部分的数学模型。(3)分别以浮子速度、直线电机的交轴电流作为最优控制信号,将电机电流、浮子速度和浮子位置作为状态变量,入射波所产生的波浪力作为输入变量,电机力作为控制变量进行建模,获得了模型预测控制所需的状态空间模型。将其进一步离散化,利用预测公式和目标函数进行滚动优化和反馈校正。基于MATLAB对两种不同目标函数下的控制方法进行了仿真。对比了以两种不同信号作为参考时的控制效果,以及MPC和PI控制器的系统控制效果,仿真结果证明了MPC应用在点吸式波浪发电系统的可行性和优越性。(4)对双浮子波浪发电系统的模型预测控制算法进行了研究,建立了预测控制的模型,有别于单浮子系统的是没有参考轨迹,目标函数的建立时采用了直接功率最大化的方法,最后以规则波作为输入进行了仿真分析,证明了模型预测控制方法在双浮子的波浪发电系统中也有良好的应用效果。
[Abstract]:Wave energy can be converted into usable electric energy by using wave Energy converter. In order to make full use of wave energy, it is necessary to adopt effective control strategy to improve the conversion efficiency of wave energy converter and achieve maximum wave energy capture. Because of the randomness of wave and the nonlinearity of system model, the traditional control method has great limitation in the control of wave power generation system. In view of this situation, this paper has carried on the innovation and the improvement in the control method, proposed the model predictive control method in the intelligent control method to apply to the float type wave power generation system. Aiming at the limitation of obtaining the maximum wave energy, the position of the float, the speed of the float and the current of the motor are taken as the state variables of predictive control. Because the model predictive control method has the advantages of not demanding the precision of the model, including the prediction of the external input, and taking into account the limited conditions of the system, it has great advantages in the wave power generation system. Therefore, the study of model predictive control has broad prospects and important practical significance in wave power generation systems. In this paper, the model predictive control method of point-absorbing wave power generation system is deeply studied, and the research status and results of wave power generation technology are analyzed from the following aspects: 1. The advantages and research significance of floating wave power generation system in the field of ocean wave energy utilization are summarized. The advantages and disadvantages of existing typical wave power generation devices and their working principles are analyzed. The control strategy of the floating wave power generation system is systematically studied and the detailed summary is given. The operating state of the floating wave power generation system and the specific motion of each part of the system are studied. The incident wave of floating wave power generation system is simulated. Taking the wave power generation system of the single float as the research object, the concrete analysis and modeling of the force condition of the float are carried out. In the system, the linear motor is selected as the power output device, and the mathematical model is derived. Then the mathematical model of wave output part of single float wave power generation system is obtained. The speed of float and the cross axis current of linear motor are taken as the optimal control signals respectively. The current of motor, the velocity of float and the position of float are taken as state variables, respectively. The wave force generated by the incident wave is used as input variable and the motor force as control variable. The state space model for predictive control is obtained. It is further discretized for rolling optimization and feedback correction using prediction formulas and objective functions. Two control methods under different objective functions are simulated based on MATLAB. The control effect when two different signals are used as reference, and the control effect of MPC and Pi controller are compared. The simulation results show that the feasibility and superiority of MPC applied in the point suction wave power generation system is proved. The model predictive control algorithm of the double float wave power generation system is studied, and the predictive control model is established. Different from the single float system, there is no reference track, and the method of direct power maximization is used in the establishment of the objective function. Finally, the regular wave is used as the input for the simulation analysis. It is proved that the model predictive control method also has good application effect in the wave power generation system with double floats.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：P743.2;P742

【相似文献】