为了建立基于实际机组运行数据的再热汽温模型，采用基于M次不相关激励和汇总智能优化（MUEAIO）的多变量系统辨识方法。该方法的优点在于引入汇总优化的方式，有效地降低了模型误差，提高了辨识精度。以安徽田集电厂660 MW锅炉的再热器为例，分别选取一段或三段现场运行历史数据作为模型辨识依据，并用另一段运行数据进行模型验证，实验结果表明:基于MUEAIO方法辨识得到的再热汽温模型能够正确表征系统的实际动态特性，且精确度较高，可以为再热汽温系统控制方案的设计和控制器参数的优化奠定基础。

In order to establish a reheated steam temperature model based on actual unit operation data,a multivariable system identification method based on Mtimes Uncorrelated Excitation and Assembly Intelligent Optimization (MUEAIO) is proposed.The advantage of this method is that the introduction of assembly optimization can effectively reduce the model error and improve the identification accuracy.With the reheater of 660 MW boiler in Tianji Power Plant of Anhui Province as an example,the historical data of one or three section of field operation are selected as the identification basis of the model,and the model is validated by another section of operation data.The experimental results show that the reheated steam temperature model identified by MUEAIO method can correctly represent the actual dynamic characteristics of the system,and has high accuracy.It can lay a foundation for the design of control scheme and the optimization of controller parameters for reheat steam temperature system.

TP273

上海市科技创新行动计划"高新技术领域项目（17511109400）;上海市科学技术委员会工程技术研究中心项目（14DZ2251100）