论文部分内容阅读
单一仿真分析方法不能完整反映风力机各部分的耦合关系,影响结果准确性。针对这一问题,采用联合仿真技术,利用空气动力学理论计算气动力,运用多柔体动力学分析理论和软件实现整机的气弹相互耦合。该多柔体建模方法能较好地模拟风力机耦合振动特性,仿真结果更接近实际工作情况。由于多柔体模型考虑的自由度多,精度要求高,导致仿真时间增长,计算成本加大。因此引入人工神经网络方法对风电机组动力学性能进行分析预测。结果表明,采用联合仿真与神经网络相结合的方法,在保证预测精度的同时还能减少动力学仿真时间,弥补单独使用仿真分析方法的局限性。
A single simulation analysis method can not fully reflect the coupling relationship of various parts of the wind turbine, affecting the accuracy of the results. In response to this problem, the co-simulation technology, the use of aerodynamic theory to calculate aerodynamic force, the use of flexible multi-body dynamics analysis theory and software to achieve the whole air-to-air interaction. The multi-flexible modeling method can well simulate the wind turbine coupling vibration characteristics, the simulation results closer to the actual work. Due to the high degree of freedom considered by the multi-flexible body model, the precision is high, which leads to the increase of simulation time and the increase of calculation cost. Therefore, the artificial neural network is introduced to analyze and predict the dynamic performance of wind turbines. The results show that the combination of co-simulation and neural network can not only reduce the time of dynamic simulation but also make up for the limitations of the simulation analysis method.