论文部分内容阅读
在实际系统中,有许多系统是非线性系统,对非线性系统进行参数估计已势在必行。本文就是对此问题进行了偿试性研究,主要做了两项工作:(1)把非线性系统的参数辨识转换成线性差分模型的辨识,利用已成熟的线性系统方面的理论知识进行参数估计,然后再还原回非线性模型。(2)把跟踪引入最优控制中,本文虽也采用极大值原理,但针对离散和连续两种情况,根据建模的特殊性,提出新的性能指标,推导出适于该建模方法的最优控制。理论方法的是否可行主要靠实际来检验,本文以鞍钢第一初轧厂提供的计算机打印的二号炉第三号坑的数据验明上述思想。在此例中这种以各变量的实际值与趋势函数对应值之差为变量的建模方法,其精度比直接以各变量的实际值为变量的建模方法的精度有所改善。采用最优控制,可使炉温按理想的曲线进行烧钢,达到烧钢的标准化,以便保质保量地烧出钢。
In practical systems, many systems are nonlinear systems. It is imperative to estimate the parameters of nonlinear systems. This dissertation is a tentative research on this issue. The main work is done in two aspects: (1) The parameter identification of nonlinear system is transformed into the identification of linear difference model, and the parameter estimation is made using the theoretical knowledge of mature linear system , And then back to the nonlinear model. (2) The introduction of tracking into the optimal control, although this article also uses the maximum principle, but for both discrete and continuous cases, according to the particularity of the model, put forward new performance indicators, derive suitable for the modeling method The optimal control. The feasibility of the theoretical method depends mainly on the actual test. This paper verifies the above idea with the computer-printed No. 2 pit No. 2 pit provided by the first primary rolling mill in Ansteel. In this case, the accuracy of the modeling method that takes the difference between the actual value of each variable and the corresponding value of the trend function as a variable is improved over the accuracy of the modeling method that directly uses the actual value of each variable as a variable. The use of optimal control, the furnace temperature can be the ideal curve for burning steel, to achieve the standardization of burning steel, in order to burn the quality and quantity of steel.