利用最小二乘与最优控制的水轮机模型参数辨识

Identification of Hydraulic Turbine Model Parameters by Least Square Method and Optimal Control Method

为了改善一般常用水轮机模型基于实测数据参数进行辨识过程中底层编程繁琐的缺点,基于电力系统仿真中常用的理想水轮机数学模型,取134组模拟实测输入输出数据,用最小二乘与基于数值最优化技术的最优控制方法分别进行辨识从而获取参数,并利用两种方法对同一实测数据辨识结果进行仿真对比。算例仿真结果表明,在给定阶跃扰动下,利用最小二乘与最优控制方法进行参数辨识后的模型输出响应与实测响应结果(基于模拟样本数据)误差均低于0.04,且基于数值最优化技术的最优控制辨识效果与实测数据拟合程度更精确,验证了基于数值最优化技术的最优控制方法在线性水轮机模型参数辨识上应用的可行性,为水轮机模型参数辨识方法提供了一种可提升辨识效率的人机对话参考。

In order to improve the disadvantages of the low level programming in the process of parameter identification based on the measured data of common hydraulic turbine models,based on the ideal hydraulic turbine mathematical model commonly used in power system simulation,134 groups of input and output data of simulation measurement are used to identify the parameters with the least square method and the optimal control method based on numerical optimization technology,and the two methods are used to compare the simulation results of the same measured data after identification.The simulation results show that under the given step disturbance,the error between the model output response and the measured response(based on the simulated sample data)after the parameter identification of the least square and the optimal control method is less than 0.04,and the identification effect of the optimal control based on the numerical optimization technology and the measured data fit more accurately.The feasibility of the application of the optimal control method based on the numerical optimization technology in the parameter identification of the linear hydraulic turbine model is verified,which provides a man-machine conversation reference to improve the identification efficiency for the parameter identification method of the hydraulic turbine model.