双线性过程的鲁棒数据协调

Robust Data Reconciliation of Bilinear Process

针对基于鲁棒目标函数的双线性数据协调问题,提出了一种新的算法。首先利用两步法将双线性数据协调问题转化为两个线性数据协调问题,然后利用罚函数法把约束优化问题转化为无约束优化问题,最后给出了Huber函数的等价权,以及利用等价权法求解双线性鲁棒数据协调问题的表达式。在迭代求解过程中考虑了变量的上下限约束。仿真结果表明该方法能有效地求解双线性鲁棒数据协调问题,降低显著误差的影响。

A new algorithm is proposed to solve the bilinear data reconciliation problem based on the robust objective function.The data reconciliation problem of bilinear process is transformed into two linear data reconciliation problems with a two-step method.Penalty function is used to transform the constrained optimization problem into the unconstrained optimization problem.The equivalent weight of Huber function and the expression to solve the bilinear robust data reconciliation with the equivalent weight method are given.The upper and lower limits of variables are considered in iterative calculations.Simulation results verify that the proposed method can solve the bilinear robust data reconciliation problem efficiently and decrease the effect of gross errors.