自适应扰动估计的相关积分优化方法及应用
Correlation integral optimization method and application in adaptive disturbance estimation

针对传统相关积分优化方法, 当系统扰动与调优变量相关时, 在迭代优化的过程中, 目标函数难以收敛到 最优值的问题, 提出了一种改进的相关积分优化方法用于稳态操作调优. 基于数据驱动稳态模型, 构造了自适应扰 动估计器用来估计扰动均值, 对最小二乘方法计算得到的调优变量梯度均值进行补偿, 并修正调优变量, 确保目标 函数在调优的过程中收敛于最优值. 仿真对比及工业应用结果证实了所提方法的可行性和有效性.
In view of the traditional correlation integral algorithm, when the system disturbances are correlative with the decision variables, objective function can not converge to the optimal value in the process of iterative optimization. In this work, an improved method of correlation integral optimization is proposed. Based on the steady data driven model, an adaptive disturbance estimator is constructed to estimate the mean values of the disturbances and compensate the gradient values obtained by the traditional least square method. Based on the correlation integral optimization method, the modified optimization variables can be obtained to ensure the convergence of the optimization process.