【摘要】 参数辨识是系统辨识的重要内容,其方法的有效性决定了建模的最终精度。水轮机调节系统是一类复杂非线性系统,其参数会随运行工况的变化而改变。同时,实际测量信号受环境的影响而夹杂着噪声等无关信息,这将增加参数辨识的难度。因此,对于不同工况和测试条件,参数辨识方法的适用性和精度,以及参数的可辨识性值得进一步研究。以国网水口电站机组为研究对象,基于多种智能优化算法并考虑多种试验工况,对比研究了在不同辨识场景下各辨识算法的有效性和精度,以及参数的可辨识性。结果表明,生物地理学优化算法在各种辨识条件下均具有较好且稳定的性能。更多还原

【Abstract】 Parameter estimation is an important part of system identification, and the final accuracy of modeling is determined by the effectiveness of its method. Hydraulic turbine regulating system is a complex non-linear system, and its parameters will change with the operating conditions. At the same time, the actual measurement signal is affected by the environment and is mixed with irrelevant information such as noise, which will increase the difficulty of parameter identification. Therefore, the applicability and accuracy of the parameter identification method for different operating and testing conditions, and the identifiability of the parameters are worthy of a further study. Based on a variety of intelligent optimization algorithms and two experimental conditions, the effectiveness and accuracy of each identification algorithm under various identification scenarios as well as the identifiability of parameters are compared and studied in this paper. The results show that the biogeography-based optimization algorithm has high and stable performance under various identification conditions.更多还原